text stringlengths 454 608k | url stringlengths 17 896 | dump stringclasses 91
values | source stringclasses 1
value | word_count int64 101 114k | flesch_reading_ease float64 50 104 |
|---|---|---|---|---|---|
hard-coding?
All of the CS 111x courses provide a full introduction to programming sufficient to take additional CS courses. Students who took all versions perform comparably in CS 2110.
You might also consider taking CS 1511. CS 1511 is based off the successful CS Principles project and presents broader overview of CS topics. CS 111x will teach you programming skills and related topics; CS 1511 will teach you computational thinking and digital citizenship.
In order to come off the wait list, there has to be an open seat in BOTH the lecture and lab you are signed up for. If one or the other is not true, then SIS moves on to the next student who has the right combination.
Our main cap is the lab. The lab sessions CANNOT go over 46 due to fire code limits. A few seats in each lab are held back for a week or so to accommodate very special cases.
Some examples of special cases include:
If you feel you warrant special consideration and are currently on the waitlist, please fill out the form here:
All SEAS Students will be given a seat, once each, provided they fill out the above form. If you lose it by dropping and re-adding the course, we cannot re-accommodate you.
Consider the other CS1 courses:
We are offering three other CS1 courses this semester: CS 1112 with Jim Cohoon (for students with no programming experience) and CS 1113 with Qureshi Asma (for future engineers). These are both good options to consider and all count the same for prerequisites and major requirements! CS 1511 this semester is also an introduction to computer science, though with a broader scope (and thus less programming depth) than our other CS1 offerings.
We wish we could take everyone that wanted CS 1110 or CS 1111, but it’s simply not feasible with the room sizes and resources we have. Please do try again next semester if you can’t get in this time.
Let us know if you have any questions.
First, we have no control over this at all. It depends entirely on other students dropping the course.
Second, SIS will only tell you your lecture position, but your lab position is typically what matters to getting a seat. Meaning that the waiting list position number stated by SIS is close to meaningless.
In the past, we have had fairly steady drops from the first day of class up to the add deadline, generally with between 5 and 15% turnover. But the variance is high and we cannot predict which labs people will drop out of.
Yes, by dropping the class and re-adding the appropriate waiting list. Note that this will put you on the back of the waiting list. There is no way to change which lab you are waiting for without moving to the back of that lab’s list.
You are welcome to come to 1110 lectures in Wilson hall (the 10am and 2pm sections), but not to physically come to labs (the first week’s lab being an exception: you may come to that), though attempting the lab activities on your own is encouraged. 1111 lectures and the 1110 lecture in Rice hall typically fill the room, so you might be asked to leave the room to make space for those enrolled if you attend those lectures while on a waiting list.
We hope to also give you the ability to submit assignments while on the waiting list so that when/if you get off you are in no way behind on your work. No promises, though.
As this is a programming class, we do expect you to have access to a computer for the duration of the semester. If you are temporarily without a working machine, Python and PyCharm should be installed on publicly available machines in Alderman Library and other locations.
We expect you to make regular backups of your code so in the event of a failure you still have access to your assignments. We will not accept a computer failure as a reason to waive a late penalty for an assignment.
We highly suggest you look into using a cloud-based solution to make constant backups of particular directories on your computer, such as UVaBox found at or Dropbox found at. You can also do some basic assignments in an in-browser Python environment such as,,, etc.
If none of the above options work for you, the department has a small number of laptops it can loan out to students enrolled in CS classes. To gain access to one, talk to your professor (who has to make the request on your behalf).
Not easily… specifics follow based on your current status
Then can drop and re-add in SIS. But note that you have to drop first, so you can’t simply keep your space in a full lab section and swap lecture sections.
Don’t use SIS; this has to be handled by the registrar directly. Once the semester begins, email professor Tychonievich an email of the form
Please swap Jane Doe (mst3k) from CS 1110 lab section 103 (SIS id 16952) to lab section 105 (SIS id 17564)
Fill in the appropriate name, computing ID, lab/lecture sections and IDs. You can find SIS ids of courses on Lou’s List. We’ll then verify there is space and if so, forward the request on to the registrar.
These swaps cannot be made prior to the start of the semester. Sorry.
Have one (1) of the swapping parties email professor Tychonievich an email of the form
Please swap Jane Doe (mst3k) from CS 1110 lab section 103 (SIS id 16952) to lab section 105 (SIS id 17564)
Please swap John Doe (aa1a) from CS 1110 lab section 105 (SIS id 17564) to lab section 103 (SIS id 16952)
The email sender should CC the other parties in the swap.
Fill in the appropriate names, computing IDs, lab/lecture sections and IDs. You can find SIS ids of courses on Lou’s List. We’ll then verify that all students are enrolled in their from sections and that the swap does not change total enrollment size in any section and then forward the request on to the registrar.
See Can I switch which lab I’m waiting for above. The same logic applies to lectures.
You’ll have to wait for the restrictions to lift so that you could enroll before you can change section. See The course is listed as restricted above.
Fill out this form, which will log your computing ID (log in with your @virginia.edu google account). We only have a few reserved seats, so no guarantees, but we’ll try to accommodate those with genuine need.
Due to fire code limits, you cannot attend another lab session, even for just one week. We’re also doing group work, so you need to be there for your team.
If you need to do this because of conflict with another class or test, your other professor should provide you with an alternate time since you have a scheduled university class (this lab) at this time.
Missing one week in general will not affect your grade. Every student can miss one lab with no penalty (and you do not need to make up the work).
1111 students can attend an 1110 section on occasion if they like, though doing so will not excuse any missed attendance or participation activities. 1110 students should not attend 1111 lectures due to the size of the classroom. 1110 students can attend a different 1110 section on occasion as well, though if your instructor tracks attendance or participation those need to be done in your lecture, and if the Rice hall lecture fills up we’ll ask those not enrolled in that section to leave.
Absolutely not.
When we get closer to the end of the semester, we’ll have a form you’ll fill out to get a separate time. We will accommodate most all cases here with no issue. Please do not email us about this before the end of the semester as we will have no other information to tell you.
University policy does not provide any accommodations for travel. If you believe you are an exception, contact your dean; only deans may approve final exam rescheduling.
Know in advance that
I already paid for tickets is not a special case.
This could be a grading error, but could also be because you hard-coded those specific cases instead of solving the general problem (see the syllabus).
hard-coding?
Wikipedia defines it as
embedding […] an input or configuration data directly into the source code of a program. In this course, it most commonly appears when students solve the examples but not the general problem.
For example, suppose we ask for
a function called A correct solution would solve the general problem, like this:
sum that computes the sum of two numbers. For example,
sum(2, 3) should give
5 and
sum(-1.1, 1.0) should give
-0.1.
def sum(x, y): '''returns the mathematical sum of its arguments''' return x + y
Conversely, a solution with the example inputs hard-coded might look like this:
def sum(x, y): '''a hard-coded solution that returns the mathematical sum of (2, 3) and (-1.1, 1.0), but not most other values''' if x < 0: return -0.1 else: return 5
UVA makes three different wireless networks available to students. If one is down, try a different one
Cavalier – this network is encrypted (meaning it’s more secure) and tends to be best supported. It should be your default. See ITS’s guide for getting it set up.
Wahoo – this network is unencrypted and also usually unlisted. It also requires you register your device:
See ITS’s page on this network, but especially the device registration page.
You may have to manually enter the name
wahoo in your network manager for your computer to find this network.
Welcome – this network is unencrypted and intended for guest use, but you can register as your own guest:
Welcome_to_UVa_Wirelessnetwork
guestoption – you’re registering as your own guest.
Guest Wireless Passcode site.
Passcode:and click
It is uncommon for all three wireless networks to be down at the same time. In my experience,
Welcome_to_UVa_Wireless in particular is almost always up, in part because it is annoying to have to keep entering passcodes so few people use it.
We get so many of these requests that we cannot grant them all, and to avoid being unfair we generally do not grant any of them. If the announcement is purely academic in nature and there is a compelling reason why Introduction to Programming lecture is the right place to make it, email the professors; but we still make no guarantee we’ll accommodate you.
Raising your hand in class to make an announcement (rather than to ask or answer a question) is unprofessional behavior and will be treated as such. | http://cs1110.cs.virginia.edu/faq.html | CC-MAIN-2017-43 | refinedweb | 1,846 | 68.81 |
This is the documentation for older versions of Odoo (formerly OpenERP).
See the new Odoo user documentation.
See the new Odoo technical documentation.
Unit testing¶
Since version 4.2 of OpenERP, the XML api provides several features to test your modules. They allow you to
-
test the properties of your records, your class invariants etc.
-
test your methods
-
manipulate your objects to check your workflows and specific methods
This thus allows you to simulate user interaction and automatically test your modules.
Generalities¶
As you will see in the next pages, unit testing through OpenERP's XML can be done using three main tags: <assert>, <workflow> and <function>. All these tags share some common optional attributes:
These two attributes might be set on any of those tags (for <functions>, only the root <function> tag may accept it) or on the <data> tag itself. If you set a context attribute on both, they will be merged automatically.
Notice that Unit Testing tags will not be interpreted inside a <data> tag set in noupdate.
Using unit tests¶
You can declare unit tests in all your .XML files. We suggest you to name the files like this:
-
module_name_test.xml
If your tests are declared as demo data in the __openerp__.py, they will be checked at the installation of the system with demo data. Example of usage, testing the demo sale order produce a correct amount in the generated invoice.
If your tests are declared like init data, they will be checked at all installation of the software. Use it to test the consistency of the software after installation.
If your tests are declared in update sections, the tests are checked at the installation and also at all updates. Use it to tests consistencies, invariants of the module. Example: The sum of the credits must be equal to the sum of the debits for all non draft entries in the accounting module. Putting tests in update sections is very useful to check consistencies of migrations or new version upgrades.
Assert Tag¶
The assert tag allows you to define some assertions that have to be checked at boot time. Example :
<assert model="res.company" id="main_company" string="The main company name is Open sprl"> <test expr="name">Open sprl</test> </assert>
This assert will check that the company with id main_company has a name equal to "Open sprl". The expr field specifies a python expression to evaluate. The expression can access any field of the specified model and any python built-in function (such as sum, reduce etc.). The ref function, which gives the database id corresponding to a specified XML id, is also available (in the case that "ref" is also the name of an attribute of the specified model, you can use _ref instead). The resulting value is then compared with the text contained in the test tag. If the assertion fails, it is logged as a message containing the value of the string attribute and the test tag that failed.
For more complex tests it is not always sufficient to compare a result to a string. To do that you may instead omit the tag's content and just put an expression that must evaluate to True:
<assert model="res.company" id="main_company" string="The main company's currency is €" severity="warning"> <test expr="currency_id.code == 'eur'.upper()"/> </assert>
The severity attribute defines the level of the assertion: debug, info, warning, error or critical. The default is error. If an assertion of too high severity fails, an exception is thrown and the parsing stops. If that happens during server initialization, the server will stop. Else the exception will be transmitted to the client. The level at which a failure will throw an exception is by default at warning, but can be specified at server launch through the --assert-exit-level argument.
As sometimes you do not know the id when you're writing the test, you can use a search instead. So we can define another example, which will be always true:
<assert model="res.partner" search="[('name','=','Agrolait')]" string="The name of Agrolait is :Agrolait"> <test expr="name">Agrolait</test> </assert>
When you use the search, each resulting record is tested but the assertion is counted only once. Thus if an assertion fails, the remaining records won't be tested. In addition, if the search finds no record, nothing will be tested so the assertion will be considered successful. If you want to make sure that there are a certain number of results, you might use the count parameter:
<assert model="res.partner" search="[('name','=','Agrolait')]" string="The name of Agrolait is :Agrolait" count="1"> <test expr="name">Agrolait</test> </assert>
Require the version of a module.
<!-- modules requirement --> <assert model="ir.module.module" search="[('name','=','common')]" severity="critical" count="1"> <test expr="state == 'installed'" /> <!-- only check module version --> <test expr="'.'.join(installed_version.split('.')[3:]) >= '2.4'" /> </assert>
Workflow Tag¶
The workflow tag allows you to call for a transition in a workflow by sending a signal to it. It is generally used to simulate an interaction with a user (clicking on a button…) for test purposes:
<workflow model="sale.order" ref="test_order_1" action="order_confirm" />
This is the syntax to send the signal order_confirm to the sale order with id test_order_1.
Notice that workflow tags (as all other tags) are interpreted as root which might be a problem if the signals handling needs to use some particular property of the user (typically the user's company, while root does not belong to one). In that case you might specify a user to switch to before handling the signal, through the uid property:
<workflow model="sale.order" ref="test_order_1" action="manual_invoice" uid="base.user_admin" />
(here we had to specify the module base - from which user_admin comes - because this tag is supposed to be placed in an xml file of the sale module)
In some particular cases, when you write the test, you don't know the id of the object to manipulate through the workflow. It is thus allowed to replace the ref attribute with a value child tag:
<workflow model="account.invoice" action="invoice_open"> <value model="sale.order" eval="obj(ref('test_order_1')).invoice_ids[0].id" /> </workflow>
(notice that the eval part must evaluate to a valid database id)
Function Tag¶
The function tag allows to call some method of an object. The called method must have the following signature:
def mymethod(self, cr, uid [, …])
Where
-
cr is the database cursor
-
uid is the user id
Most of the methods defined in Tiny respect that signature as cr and uid are required for a lot of operations, including database access.
The function tag can then be used to call that method:
<function model="mypackage.myclass" name="mymethod" />
Most of the time you will want to call your method with additional arguments. Suppose the method has the following signature:
def mymethod(self, cr, uid, mynumber)
There are two ways to call that method:
-
either by using the eval attribute, which must be a python expression evaluating to the list of additional arguments:
<function model="mypackage.myclass" name="mymethod" eval="[42]" />
In that case you have access to all native python functions, to a function ref() that takes as its argument an XML id and returns the corresponding database id, and to a function obj() that takes a database id and returns an object with all fields loaded as well as related records.
-
or by putting a child node inside the function tag:
<function model="mypackage.myclass" name="mymethod"> <value eval="42" /> </function>
Only value and function tags have meaning as function child nodes (using other tags will give unspecified results). This means that you can use the returned result of a method call as an argument of another call. You can put as many child nodes as you want, each one being an argument of the method call (keeping them in order). You can also mix child nodes and the eval attribute. In that case the attribute will be evaluated first and child nodes will be appended to the resulting list.
Acceptance testing¶
This document describes all tests that are made each time someone install OpenERP on a computer. You can then assume that all these tests are valid as we must launch them before publishing a new module or a release of OpenERP.
Integrity tests on migrations¶
-
Sum credit = Sum debit
-
Balanced account chart
... Describe all integrity tests here | https://doc.odoo.com/6.0/ja/developer/5_20_unit_testing/ | CC-MAIN-2019-09 | refinedweb | 1,407 | 61.87 |
Manpage of GETRLIMIT
GETRLIMITSection: Linux Programmer's Manual (2)
Updated: 2016-10-08
Index
NAME)):
DESCRIPTIONThe getrlimit() and setrlimit() system calls get and set resource limits respectively. Each resource has an associated soft and hard limit, as defined by the rlimitstcapability) may make arbitrary changes to either limit value.
The value RLIM_INFINITYdenotes no limit on a resource (both in the structure returned by getrlimit() and in the structure passed to setrlimit()).
The resourceargument must be one of:
- RLIMIT_AS
- The maximum size of the process's virtual memory (address space) in bytes. This limit affects calls to brk(2), mmap(2), and mremap(2), which fail with the error ENOMEMupon exceeding this limit. Also automatic stack expansion will fail (and generate a SIGSEGVthat kills the process if no alternate stack has been made available via sigaltstack(2)). Since the value is a long, on machines with a 32-bit longeither this limit is at most 2 GiB, or this resource is unlimited.
- RLIMIT_CORE
- Maximum size of a corefile (see core(5)). When 0 no core dump files are created. When nonzero, larger dumps are truncated to this size.
- RLIMIT_CPU
- CPU time limit in seconds. When the process reaches the soft limit, it is sent a SIGXCPUsignal. The default action for this signal is to terminate the process. However, the signal can be caught, and the handler can return control to the main program. If the process continues to consume CPU time, it will be sent SIGXCPUonceupon encountering the soft limit of this resource.
- RLIMIT_FSIZE
- The maximum size of files that the process may create. Attempts to extend a file beyond this limit result in delivery of a SIGXFSZsignal.operation. Since Linux 2.6.9 it also affects the shmctl(2) SHM_LOCKoperation, where it sets a maximum on the total bytes in shared memory segments (see shmget(2)) that may be locked by the real user ID of the calling process. The shmctl(2) SHM_LOCKlocksris the mq_attrstructure specified as the fourth argument to mq_open(3), and the msg_msgand posix_msg_tree_nodestructurestypicallyon BSD.)
- RLIMIT_NPROC
- The maximum number of processes (or, more precisely on Linux, threads) that can be created for the real user ID of the calling process. Upon encountering this limit, fork(2) fails with the error EAGAIN. This limit is not enforced for processes that have either the CAP_SYS_ADMINor the CAP_SYS_RESOURCEcapability.
- RLIMIT_RSS
- Specifies the limit (in bytes)CPUsignal. If the process catches or ignores this signal and continues consuming CPU time, then SIGXCPUwill be generated once each second until the hard limit is reached, at which point the process is sent a SIGKILLsignal.signalargument has the same meaning as for setrlimit() and getrlimit().
If the new_limitargument is a not NULL, then the rlimitstructure to which it points is used to set new values for the soft and hard limits for resource. If the old_limitargument is a not NULL, then a successful call to prlimit() places thesoft and hard limits for resourcein the rlimitstructure pointed to by old_limit.
The pidargument specifies the ID of the process on which the call is to operate. If pidis 0, then the call applies to the calling process. To set or get the resources of a process other than itself, the caller must have the CAP_SYS_RESOURCEcapability in the user namespace of the process whose resource limits are being changed, or the real, effective, and saved set user IDs of the target process must match the real user ID of the caller andthe real, effective, and saved set group IDs of the target process must match the real group ID of the caller.
RETURN VALUEOn success, these system calls return 0. On error, -1 is returned, and errnois set appropriately.
ERRORS
- EFAULT
- A pointer argument points to a location outside the accessible address space.
- EINVAL
- The value specified in resourceis not valid; or, for setrlimit() or prlimit(): rlim->rlim_curwas greater than rlim->rlim_max.
- EPERM
- An unprivileged process tried to raise the hard limit; the CAP_SYS_RESOURCEcapability is required to do this.
- EPERM
- The caller tried to increase the hard RLIMIT_NOFILElimitand RLIMIT_NPROCderive from BSD and are not specified in POSIX.1; they are present on the BSDs and Linux, but on few other implementations. RLIMIT_RSSderives from BSD and is not specified in POSIX.1; it is nevertheless present on most implementations. RLIMIT_MSGQUEUE, RLIMIT_NICE, RLIMIT_RTPRIO, RLIMIT_RTTIME, and RLIMIT_SIGPENDINGare Linux-specific.
NOTESA child process created via fork(2) inherits its parent's resource limits. Resource limits are preserved across execve(2).
Lowering the soft limit for a resource below the process's current consumption of that resource will succeed (but will prevent the process from further increasing its consumption of the resource).
One can set the resource limits of the shell using the built-in ulimitcommand (limitand SIGKILLsignals delivered when a process encountered the soft and hard RLIMIT_CPUlimits were delivered one (CPU) second later than they should have been. This was fixed in kernel 2.6.8.
In 2.6.x kernels before 2.6.17, a RLIMIT_CPUlimit of 0 is wrongly treated as "no limit" (like RLIM_INFINITY). Since Linux 2.6.17, setting a limit of 0 does have an effect, but is actually treated as a limit of 1 second.
A kernel bug means that RLIMIT_RTPRIOdoes_CPUlimit_CPUsoft limit in this manner, and the Linux behavior is probably not standards conformant; portable applications should avoid relying on this Linux-specific behavior. The Linux-specific RLIMIT_RTTIMElimit exhibits the same behavior when the soft limit is encountered.
Kernels before 2.4.22 did not diagnose the error EINVALfor setrlimit() when rlim->rlim_curwas greater than rlim->rlim_max.
Representation of large resource limit values on 32-bit platformsThe glibc getrlimit() and setrlimit() wrapper functions use a 64-bit rlim_tdata type, even on 32-bit platforms. However, the rlim_tdata type used in the getrlimit() and setrlimit() system calls is a (32-bit) unsigned long. Furthermore, in Linux versions before 2.6.36,:
- *
-The limit */ if (prlimit(pid, RLIMIT_CPU, newp, &old) == -1) errExit("prlimit-1"); printf("); }
SEE ALSOprlimit(1), dup(2), fcntl(2), fork(2), getrusage(2), mlock(2), mmap(2), open(2), quotactl(2), sbrk(2), shmctl(2), malloc(3), sigqueue(3), ulimit(3), core(5), capabilities(7), cgroups(7), signal(7)
Index
- NAME
- SYNOPSIS
- DESCRIPTION
- RETURN VALUE
- ERRORS
- VERSIONS
- ATTRIBUTES
- CONFORMING TO
- NOTES
- BUGS
- EXAMPLE
- SEE ALSO
This document was created by man2html, using the manual pages.
Time: 16:30:14 GMT, October 09, 2016 Click Here! | https://www.linux.com/manpage/man3/vlimit.3.html | CC-MAIN-2016-44 | refinedweb | 1,060 | 54.52 |
Created on 2019-10-02 18:31 by Karl Kornel, last changed 2021-05-04 13:37 by Jelle Zijlstra. This issue is now closed.
Hello!
In, there is a note about the io and re classes not being included in typing.__all__. I am a relatively new user of typing, and I did `from typing import *` in my code. I ran the code through mypy first, which reported no problems, but then running Python 3.6 failed with a NameError (name 'IO' is not defined).
Reading through the typing source, it's clear that this was an intentional decision. So, instead of reporting a bug, I'd like to request a documentation enhancement!
The docs for typing make no mention of typing.io or typing.re. So, my request is: In the sections for the IO/TextIO/BinaryIO and Pattern/Match classes, include text warning the user that these types are not imported when you do `from typing import *`.
> So, my request is: In the sections for the IO/TextIO/BinaryIO and Pattern/Match classes, include text warning the user that these types are not imported when you do `from typing import *`.
I don't think this should really be a warning, probably just a note, but otherwise I totally agree. Would you like to make a PR?
Shadowing the real modules `re` and `io` by
from typing import *
would indeed be bad, but that argument IMHO doesn't hold for the types `IO`, `TextIO` and `BinaryIO`, yet they are not listed in `typing.__all__`. Is there a reason for that? And if not, could `IO`, `TextIO` and `BinaryIO` be added to `typing.__all__`?
Wait, is the OP maybe that there’s a difference between typeshed an the stdlib typing.py?
It turns out that IO, TextIO, BinaryIO, Match, and Pattern aren't in typing.__all__. As Walter points out above, there's no clear reason for this. I am submitting a PR to add them to __all__.
New changeset e1bcc88a502aa0239b6bcc4da3fe024307fd27f4 by Miss Islington (bot) in branch '3.10':
bpo-38352: Add to typing.__all__ (GH-25821) (#25884)
New changeset 00726e51ade10c7e3535811eb700418725244230 by Miss Islington (bot) in branch '3.9':
bpo-38352: Add to typing.__all__ (GH-25821) (#25885)
Fix merged to main (3.11), 3.10, and 3.9. Not applicable to older Pythons as they are security fixes only.
@Łukasz thanks for your merging spree!
I'm actually not sure this should go into 3.9. Seems potentially dangerous that people upgrading from an earlier 3.9 patch release will now see new names injected into their namespace if they do `from typing import *`. | https://bugs.python.org/issue38352 | CC-MAIN-2021-21 | refinedweb | 436 | 75.91 |
Masonite comes with bcrypt out of the box but leaves it up to the developer to actually encrypt things like passwords. You can opt to use any other hashing library but bcrypt is the standard of a lot of libraries and comes with some one way hashing algorithms with no known vulnerabilities. Many of hashing algorithms like SHA-1 and MD5 are not secure and you should not use them in your application.
You can read the bcrypt documentation here.
Also, we make sure that Javascript cannot read your cookies. It's important to know that although your website may be secure, you are susceptible to attacks if you import third party Javascript packages (since those libraries could be compromised) which can read all cookies on your website and send them to the hacker.
Other frameworks use cryptographic signing which attaches a special key to your cookies that prevents manipulation. This does't make sense as a major part of XSS protection is preventing third parties from reading cookies. It doesn't make sense to attach a digital signature to a plaintext cookie if you don't want third parties to see the cookie (such as a session id). Masonite takes this one step further and encrypts the entire string and can only be decrypted using your secret key (so make sure you keep it secret!).
In your
.env file, you will find a setting called
KEY=your-secret-key. This is the SALT that is used to encrypt and decrypt your cookies. It is important to change this key sometime before development. You can generate new secret keys by running:
$ craft key
This will generate a new key in your terminal which you can copy and paste into your
.env file. Your
config/application.py file uses this environment variable to set the
KEY configuration setting.
Additionally you can pass the
--store flag which will automatically set the
KEY= value in your
.env file for you:
$ craft key --store
Remember to not share this secret key as a loss of this key could lead to someone being able to decrypt any cookies set by your application. If you find that your secret key is compromised, just generate a new key.
You can use the same cryptographic signing that Masonite uses to encrypt cookies on any data you want. Just import the
masonite.sign.Sign class. A complete signing will look something like:
from masonite.auth import Signsign = Sign()signed = sign.sign('value') # PSJDUudbs87SB....sign.unsign(signed) # 'value'
By default,
Sign() uses the encryption key in your
config/application.py file but you could also pass in your own key.
from masonite.auth import Signencryption_key = b'SJS(839dhs...'sign = Sign(encryption_key)signed = sign.sign('value') # PSJDUudbs87SB....sign.unsign(signed) # 'value'
This feature uses pyca/cryptography for this kind of encryption. Because of this, we can generate keys using Fernet.
from masonite.auth import Signfrom cryptography.fernet import Fernetencryption_key = Fernet.generate_key()sign = Sign(encryption_key)signed = sign.sign('value') # PSJDUudbs87SB....sign.unsign(signed) # 'value'
Just remember to store the key you generated or you will not be able to decrypt any values that you encrypted in the first place.
Bcrypt is very easy to use and basically consists of a 1 way hash, and then a check to verify if that 1 way hash matches an input given to it.
It's important to note that any values passed to bcrypt need to be in bytes.
Again, all values passed into bcrypt need to be in bytes so we can pass in a password using the password helper function:
from masonite.helpers import passwordencrypted_password = password('secret')
Notice that the value passed in from the request was converted into bytes using the
bytes() Python function.
Once the password is hashed, we can just safely store it into our database
User.create(name=request.input('name'),password=password,email=request.input('email'),)
Do not store unhashed passwords in your database. Also, do not use unsafe encryption methods like MD5 or SHA-1.
In order to check if a password matches it's hashed form, such as trying to login a user, we can use the
bcrypt.checkpw() function:
bcrypt.checkpw(bytes('password', 'utf-8'), bytes(model.password, 'utf-8'))
This will return true if the string
'password' is equal to the models password.
More information on bcrypt can be found by reading it's documentation. | https://docs.masoniteproject.com/v/v2.2/security/encryption | CC-MAIN-2020-16 | refinedweb | 731 | 56.25 |
Last one for the day, then I go home.
You’ve read Test Methods are neither Methods nor Tests. You’ve dried off & are fully recovered.
This is the practical reason why NUnit should remove the ‘public’ requirements from test fixtures. I didn’t include it in the last post, because I didn’t want people to focus on the concrete impacts of my reasoning.
Today, NUnit sets you up to write like this:
class MyClass
{
//…
}
//————
// may be same file, different file in the same project, different netmodule, different assembly
[TestClass]
public class MyClassTests
{
[TestMethod]
public void Test1()
{
//…
}
}
I prefer to write like this:
class MyClass
{
//…
[TestClass]
/*public*/ class Tests
{
[TestMethod]
/*public*/ void Test1()
{
//…
}
}
}
It lets me normalize my naming. Tests for MyClass are called “MyClass.Tests”.
A Rename Refactoring of MyClass to YourClass, I won’t miss MyClassTests from the first example, because I’ve removed that duplication from my code.
BTW, I’m 0x1E today. Instead of taking the day off, I decided to spend the day doing only work I enjoy. Hence 8 full-featured blogs in one day.
Here here on normalized naming!
Here here to co-location of tests!
Happy birthday!
Jay – I agree..
I’ve always put test classes inside the parent class. It’s one of the ongoing fights with another architect here. I understand the argument that your normal assemblies don’t need to reference your test assemblies – but the advantages are so huge – otherwise you need an identical test hierarchy to your code hierarchy, you lose track of things etc etc etc
I always have the following:
> class blah
> {
> #region Tests
> [TestFixture]
> private class BlahTests
> {…}
> #endregion
> #region Support functions for others testing this
> public class Testing
> {
> public Blah CreateTestBlah(); {…}
> public Blah CreateTestBlahWithSomeCharacteristic(); {…}
> public void DeleteTestBlah( Blah testBlah);
> }
> #endregion
> … class body …
> }
This way, if someone is looking at the class, they always have the test class available. This is good because:
– if it’s maintained, the tests are visible and tend to get added to / updated
– the test hierarchy can build up with the class hierarchy
– the tests serve as examples of how to use the class
Also, the CreateTestBlah methods on Testing mean that people up the line don’t have to worry about setting up things. You want to test doing reimbursements against a vehicle package? You call VehiclePackage.Testing.CreateTestLivePackage – you then just issue a reimbursement – makes all the tests clean and simple.
To pull it off, I have my own TestEverything function which just gathers up all the testFixtures, test methods and runs them 🙂 | https://blogs.msdn.microsoft.com/jaybaz_ms/2004/07/01/if-test-fixtures-could-be-private/ | CC-MAIN-2016-40 | refinedweb | 426 | 61.06 |
many other, more sophisticated use cases where you can't use a list comprehension and a lambda function may be the shortest way to write something out.
Returning a function from another function
>>> def transform(n): ... return lambda x: x + n ... >>> f = transform(3) >>> f(4) 7
This is often used to create function wrappers, such as Python's decorators.
Combining elements of an iterable sequence with
reduce()
>>> reduce(lambda a, b: '{}, {}'.format(a, b), [1, 2, 3, 4, 5, 6, 7, 8, 9]) '1, 2, 3, 4, 5, 6, 7, 8, 9'
Sorting by an alternate key
>>> sorted([1, 2, 3, 4, 5, 6, 7, 8, 9], key=lambda x: abs(5-x)) [5, 4, 6, 3, 7, 2, 8, 1, 9]
I use lambda functions on a regular basis. It took me a while to get used to them, but eventually I came to understand that they're a very valuable part of the language.
lambda is just a fancy way of saying
function. Other than its name, there is nothing obscure, intimidating or cryptic about it. When you read the following line, replace
lambda by
function in your mind:
>>> f = lambda x: x + 1 >>> f(3) 4
It just defines a function of
x. Some other languages, like
R, say it explicitly:
> f = function(x) { x + 1 } > f(3) 4
You see? It's one of the most natural things to do in programming.
The two-line summary:
lambdakeyword: unnecessary, occasionally useful. If you find yourself doing anything remotely complex with it, put it away and define a real function..
I doubt lambda will go away. See Guido's post about finally giving up trying to remove it. Also see an outline of the conflict.
You might check out this post for more of a history about the deal behind Python's functional features:. :-)
My own two cents: Rarely is lambda worth it as far as clarity goes. Generally there is a more clear solution that doesn't include lambda.
lambdas are extremely useful in GUI programming. For example, lets say you're creating a group of buttons and you want to use a single paramaterized callback rather than a unique callback per button. Lambda lets you accomplish that with ease:
for value in ["one","two","three"]: b = tk.Button(label=value, command=lambda arg=value: my_callback(arg)) b.pack()
(Note: although this question is specifically asking about
lambda, you can also use functools.partial to get the same type of result)
The alternative is to create a separate callback for each button which can lead to duplicated code.
Pretty much anything you can do with
lambda you can do better with either named functions or list and generator expressions.
Consequently, for the most part you should just one of those in basically any situation (except maybe for scratch code written in the interactive interpreter).
I find lambda useful for a list of functions that do the same, but for different circumstances. Like the mozilla plural rules.
plural_rules = [ lambda n: 'all', lambda n: 'singular' if n == 1 else 'plural', lambda n: 'singular' if 0 <= n <= 1 else 'plural', ... ] # Call plural rule #1 with argument 4 to find out which sentence form to use. plural_rule[1](4) # returns 'plural'
If you'd have to define a function for all of those you'd go mad by the end of it. Also it wouldn't be nice with function names like plural_rule_1, plural_rule_2, etc. And you'd need to eval() it when you're depending on a variable function id.
In Python,
lambda is just a way of defining functions inline,
a = lambda x: x + 1 print a(1)
and..
def a(x): return x + 1 print a(1)
..are the exact same.
There is nothing you can do with lambda which you cannot do with a regular function - in Python functions are an object just like anything else, and lambdas simply define a function:
>>> a = lambda x: x + 1 >>> type(a) <type 'function'>
I honestly think the lambda keyword is redundant in Python - I have never had the need to use them (or seen one used where a regular function, a list-comprehension or one of the many builtin functions could have been better used instead)
For a completely random example, from the article "Python’s lambda is broken!":
To see how lambda is broken, try generating a list of functions
fs=[f0,...,f9]where
fi(n)=i+n. First attempt:
fs = [(lambda n: i + n) for i in range(10)] fs3 13
I would argue, even if that did work, it's horribly and "unpythonic", the same functionality could be written in countless other ways, for example:
>>> n = 4 >>> [i + n for i in range(10)] [4, 5, 6, 7, 8, 9, 10, 11, 12, 13]
Yes, it's not the same, but I have never seen a cause where generating a group of lambda functions in a list has been required.. It might make sense in other languages, but Python is not Haskell (or Lisp, or ...)
Please note that we can use lambda and still achieve the desired results in this way :
>>> fs = [(lambda n,i=i: i + n) for i in range(10)] >>> fs[3](4) 7
Edit:
There are a few cases where lambda is useful, for example it's often convenient when connecting up signals in PyQt applications, like this:
w = PyQt4.QtGui.QLineEdit() w.textChanged.connect(lambda event: dothing())
Just doing
w.textChanged.connect(dothing) would call the
dothing method with an extra
event argument and cause an error.. Using the lambda means we can tidily drop the argument without having to define a wrapping function
I've been using Python for a few years and I've never run in to a case where I've needed lambda. Really, as the tutorial states, it's just for syntactic sugar.
I can't speak to python's particular implementation of lambda, but in general lambda functions are really handy. They're a core technique (maybe even THE technique) of functional programming, and they're also very useuful in object-oriented programs. For certain types of problems, they're the best solution, so certainly shouldn't be forgotten!
I suggest you read up on closures and the map function (that links to python docs, but it exists in nearly every language that supports functional constructs) to see why it's useful.
Lambda function it's a non-bureaucratic way to create a function.
That's it. For example, let's supose you have your main function and need to square values. Let's see the traditional way and the lambda way to do this:
Traditional way:
def main(): ... ... y = square(some_number) ... return something def square(x): return x**2
The lambda way:
def main(): ... square = lambda x: x**2 y = square(some_number) return something
See the difference?
Lambda functions go very well with lists, like lists comprehensions or map. In fact, list comprehension it's a "pythonic" way to express yourself using lambda. Ex:
>>>a = [1,2,3,4] >>>[x**2 for x in a] [1,4,9,16]
Let's see what each elements of the syntax means:
[] : "Give me a list"
x**2 : "using this new-born function"
for x in a: "into each element in a"
That's convenient uh? Creating functions like this. Let's rewrite it using lambda:
>>> square = lambda x: x**2 >>> [square(s) for x in a] [1,4,9,16]
Now let's use map, which is the same thing, but more language-neutral. Maps takes 2 arguments:
(i) one function
(ii) an iterable
And gives you a list where each element it's the function applied to each element of the iterable.
So, using map we would have:
>>> a = [1,2,3,4] >>> squared_list = map(lambda x: x**2, a)
If you master lambdas and mapping, you will have a great power to manipulate data and in a concise way. Lambda functions are neither obscure nor take away code clarity. Don't confuse something hard with something new. Once you start using them, you will find it very clear.
One of the nice things about
lambda that's in my opinion understated is that it's way of deferring an evaluation for simple forms till the value is needed. Let me explain.
Many library routines are implemented so that they allow certain parameters to be callables (of whom lambda is one). The idea is that the actual value will be computed only at the time when it's going to be used (rather that when it's called). An (contrived) example might help to illustrate the point. Suppose you have a routine which which was going to do log a given timestamp. You want the routine to use the current time minus 30 minutes. You'd call it like so
log_timestamp(datetime.datetime.now() - datetime.timedelta(minutes = 30))
Now suppose the actual function is going to be called only when a certain event occurs and you want the timestamp to be computed only at that time. You can do this like so
log_timestamp(lambda : datetime.datetime.now() - datetime.timedelta(minutes = 30))
Assuming the
log_timestamp can handle callables like this, it will evaluate this when it needs it and you'll get the timestamp at that time.
There are of course alternate ways to do this (using the
operator module for example) but I hope I've conveyed the point.
Update: Here is a slightly more concrete real world example.
Update 2: I think this is an example of what is called a thunk.
As stated above, the lambda operator in Python defines an anonymous function, and in Python functions are closures. It is important not to confuse the concept of closures with the operator lambda, which is merely syntactic methadone for them.
When I started in Python a few years ago, I used lambdas a lot, thinking they were cool, along with list comprehensions. However, I wrote and have to maintain a big website written in Python, with on the order of several thousand function points. I've learnt from experience that lambdas might be OK to prototype things with, but offer nothing over inline functions (named closures) except for saving a few key-stokes, or sometimes not.
Basically this boils down to several points:
That's enough reason to round them up and convert them to named closures. However, I hold two other grudges against anonymous closures.
The first grudge is simply that they are just another unnecessary keyword cluttering up the language.
The second grudge is deeper and on the paradigm level, i.e. I do not like that they promote a functional-programming style, because that style is less flexible than the message passing, object oriented or procedural styles, because the lambda calculus is not Turing-complete (luckily in Python, we can still break out of that restriction even inside a lambda). The reasons I feel lambdas promote this style are:
There is an implicit return, i.e. they seem like they 'should' be functions.
They are an alternative state-hiding mechanism to another, more explicit, more readable, more reusable and more general mechanism: methods.
I try hard to write lambda-free Python, and remove lambdas on sight. I think Python would be a slightly better language without lambdas, but that's just my opinion.
Lambdas are actually very powerful constructs that stem from ideas in functional programming, and it is something that by no means will be easily revised, redefined or removed in the near future of Python. They help you write code that is more powerful as it allows you to pass functions as parameters, thus the idea of functions as first-class citizens.
Lambdas do tend to get confusing, but once a solid understanding is obtained, you can write clean elegant code like this:
squared = map(lambda x: x*x, [1, 2, 3, 4, 5])
The above line of code returns a list of the squares of the numbers in the list. Ofcourse, you could also do it like:
def square(x): return x*x squared = map(square, [1, 2, 3, 4, 5])
It is obvious the former code is shorter, and this is especially true if you intend to use the map function (or any similar function that takes a function as a parameter) in only one place. This also makes the code more intuitive and elegant.
Also, as @David Zaslavsky mentioned in his answer, list comprehensions are not always the way to go especially if your list has to get values from some obscure mathematical way.
From a more practical standpoint, one of the biggest advantages of lambdas for me recently has been in GUI and event-driven programming. If you take a look at callbacks in Tkinter, all they take as arguments are the event that triggered them. E.g.
def define_bindings(widget): widget.bind("<Button-1>", do-something-cool) def do-something-cool(event): #Your code to execute on the event trigger
Now what if you had some arguments to pass? Something as simple as passing 2 arguments to store the coordinates of a mouse-click. You can easily do it like this:
def main(): # define widgets and other imp stuff x, y = None, None widget.bind("<Button-1>", lambda event: do-something-cool(x, y)) def do-something-cool(event, x, y): x = event.x y = event.y #Do other cool stuff
Now you can argue that this can be done using global variables, but do you really want to bang your head worrying about memory management and leakage especially if the global variable will just be used in one particular place? That would be just poor programming style.
In short, lambdas are awesome and should never be underestimated. Python lambdas are not the same as LISP lambdas though (which are more powerful), but you can really do a lot of magical stuff with them.
Lambdas are deeply liked to functional programming style in general. The idea that you can solve problems by applying a function to a data, and merging the results, is what google uses to implement most of its algorithms. Programs written in functional rpogramming style, are easily parrallelized and hence are becoming more and more important with modern multiu core machiines. So in short, NO you should not forget them.
First congrats that managed to figure out lambda. In my opinion this is really powerful construct to act with. The trend these days towards functional programming languages is surely an indicator that it neither should be avoided nor it will be redefined in the near future.
You just have to think a little bit different. I'm sure soon you will love it. But be careful if you deal only with python. Because the lambda is not a real closure, it is "broken" somehow: pythons lambda is broken
I'm just beginning Python and ran head first into Lambda- which took me a while to figure out.
Note that this isn't a condemnation of anything. Everybody has a different set of things that don't come easily.
Is lambda one of those 'interesting' language items that in real life should be forgotten?
No.
I'm sure there are some edge cases where it might be needed, but given the obscurity of it,
It's not obscure. The past 2 teams I've worked on, everybody used this feature all the time.
the potential of it being redefined in future releases (my assumption based on the various definitions of it)
I've seen no serious proposals to redefine it in Python, beyond fixing the closure semantics a few years ago.
and the reduced coding clarity - should it be avoided?
It's not less clear, if you're using it right. On the contrary, having more language constructs available increases clarity.
This reminds me of overflowing (buffer overflow) of C types - pointing to the top variable and overloading to set the other field values...sort of a techie showmanship but maintenance coder nightmare..
Lambda is like buffer overflow? Wow. I can't imagine how you're using lambda if you think it's a "maintenance nightmare".
I started reading David Mertz's book today 'Text Processing in Python.' While he has a fairly terse description of Lambda's the examples in the first chapter combined with the explanation in Appendix A made them jump off the page for me (finally) and all of a sudden I understood their value. That is not to say his explanation will work for you and I am still at the discovery stage so I will not attempt to add to these responses other than the following: I am new to Python I am new to OOP Lambdas were a struggle for me Now that I read Mertz, I think I get them and I see them as very useful as I think they allow a cleaner approach to programming.
He reproduces the Zen of Python, one line of which is Simple is better than complex. As a non-OOP programmer reading code with lambdas (and until last week list comprehensions) I have thought-This is simple?. I finally realized today that actually these features make the code much more readable, and understandable than the alternative-which is invariably a loop of some sort. I also realized that like financial statements-Python was not designed for the novice user, rather it is designed for the user that wants to get educated. I can't believe how powerful this language is. When it dawned on me (finally) the purpose and value of lambdas I wanted to rip up about 30 programs and start over putting in lambdas where appropriate.
I can give you an example where I actually needed lambda serious. I'm making a graphical program, where the use right clicks on a file and assigns it one of three options. It turns out that in Tkinter (the GUI interfacing program I'm writing this in), when someone presses a button, it can't be assigned to a command that takes in arguments. So if I chose one of the options and wanted the result of my choice to be:
print 'hi there'
Then no big deal. But what if I need my choice to have a particular detail. For example, if I choose choice A, it calls a function that takes in some argument that is dependent on the choice A, B or C, TKinter could not support this. Lamda was the only option to get around this actually...
I'm a python beginner, so to getter a clear idea of lambda I compared it with a 'for' loop; in terms of efficiency. Here's the code (python 2.7) -
import time start = time.time() # Measure the time taken for execution def first(): squares = map(lambda x: x**2, range(10)) # ^ Lambda end = time.time() elapsed = end - start print elapsed + ' seconds' return elapsed # gives 0.0 seconds def second(): lst = [] for i in range(10): lst.append(i**2) # ^ a 'for' loop end = time.time() elapsed = end - start print elapsed + ' seconds' return elapsed # gives 0.0019998550415 seconds. print abs(second() - first()) # Gives 0.0019998550415 seconds!(duh)
I use it quite often, mainly as a null object or to partially bind parameters to a function.
Here are examples:
{ DATA_PACKET: self.handle_data_packets NET_PACKET: self.handle_hardware_packets }.get(packet_type, lambda x : None)(payload)
let say that I have the following API
def dump_hex(file, var) # some code pass class X(object): #... def packet_received(data): # some kind of preprocessing self.callback(data) #...
Then, when I wan't to quickly dump the recieved data to a file I do that:
dump_file = file('hex_dump.txt','w') X.callback = lambda (x): dump_hex(dump_file, x) ... dump_file.close()
A useful case for using lambdas is to improve the readability of long list comprehensions.
In this example
loop_dic is short for clarity but imagine
loop_dic being very long. If you would just use a plain value that includes
i instead of the lambda version of that value you would get a
NameError.
>>> lis = [{"name": "Peter"}, {"name": "Josef"}] >>> loop_dic = lambda i: {"name": i["name"] + " Wallace" } >>> new_lis = [loop_dic(i) for i in lis] >>> new_lis [{'name': 'Peter Wallace'}, {'name': 'Josef Wallace'}]
Instead of
>>> lis = [{"name": "Peter"}, {"name": "Josef"}] >>> new_lis = [{"name": i["name"] + " Wallace"} for i in lis] >>> new_lis [{'name': 'Peter Wallace'}, {'name': 'Josef Wallace'}]
I use lambdas to avoid code duplication. It would make the function easily comprehensible Eg:
def a_func() ... if some_conditon: ... call_some_big_func(arg1, arg2, arg3, arg4...) else ... call_some_big_func(arg1, arg2, arg3, arg4...)
I replace that with a temp lambda
def a_func() ... call_big_f = lambda args_that_change: call_some_big_func(arg1, arg2, arg3, args_that_change) if some_conditon: ... call_big_f(argX) else ... call_big_f(argY)
I use
lambda to create callbacks that include parameters. It's cleaner writing a lambda in one line than to write a method to perform the same functionality.
For example:
import imported.module def func(): return lambda: imported.module.method("foo", "bar")
as opposed to:
import imported.module def func(): def cb(): return imported.module.method("foo", "bar") return cb
Lambda is a procedure constructor. You can synthesize programs at run-time, although Python's lambda is not very powerful. Note that few people understand that kind of programming. | http://m.dlxedu.com/m/askdetail/3/7f14b0a5f09a8324e53c0eacb57cd6ed.html | CC-MAIN-2018-22 | refinedweb | 3,539 | 62.07 |
[SOLVED][Qt5.0.1 static] QtCreator doesn't see MySQL plugin
I build Qt with:
@
configure -static -platform win32-msvc2012 -qt-sql-mysql -plugin-sql-mysql -no-angle -no-icu -opengl desktop -nomake demos -nomake examples@
and now i'm trying to build the most simple program in Qt which will only print available sql plugins. I have no idea why, but this code:
@#include "mainwindow.h"
#include <QApplication>
#include <QtSql/QSqlDatabase>
#include <QMessageBox>
#include <QStringList>
#pragma comment(lib, "Qt5Sql.lib")
int main(int argc, char *argv[])
{
QApplication a(argc, argv);
QStringList lista = QSqlDatabase::drivers();
for(int i = 0; i < lista.length(); ++i)
QMessageBox::information(NULL, "asd", QApplication::tr("%1").arg(lista[i]));
MainWindow w;
w.show();
return a.exec();
}@
*.pro file:
greaterThan(QT_MAJOR_VERSION, 4): QT += widgets
TARGET = jakistam
TEMPLATE = app
SOURCES += main.cpp
mainwindow.cpp
HEADERS += mainwindow.h
FORMS += mainwindow.ui@
shows only qsqlite as available plugin. Of course in qtbase/plugins/sqldrivers i have got mysql libs. Any ideas what is going on?
- BelenMuñoz
I worked with ODBC and OCI and I had to add this "sql" to .pro file. Like this:
QT += core gui sql
Hope it helps!!
- SGaist Lifetime Qt Champion
Hi,
since you have build Qt static have a look at "the plugin doc": , it explains how to work with static plugins
Thank you SGaist. I have followed steps from this link and it turned out that i had to copy libmysql libs from MySQL folder to qtbase/lib. Now everything works perfectly fine. | https://forum.qt.io/topic/24740/solved-qt5-0-1-static-qtcreator-doesn-t-see-mysql-plugin | CC-MAIN-2018-13 | refinedweb | 249 | 68.06 |
Oleg Nesterov <oleg@tv-sign.ru> writes:> On 03/04, Roland McGrath wrote:>> Since it's after our own>> group_dead hit, the "ignored_task" check for our own group leader is>> redundant with that.>> Ah, good point. I didn't realize this when I was thinking about using> signal->live.>> So perhaps it's:>> >> do_each_pid_task(pgrp, PIDTYPE_PGID, p) {>> if (task_session(p->real_parent) == task_session(p) &&>> task_pgrp(p->real_parent) != pgrp &&>> atomic_read(&p->signal->live) > 0 &&>> task_tgid_nr_ns(p->real_parent, p->nsproxy->pid_ns) != 1)>> return 0;>> } while_each_pid_task(pgrp, PIDTYPE_PGID, p);>> I am hopeless, I can't understand orphaned pgrps.I will give it a quick try.When you login in text mode you get a fresh session (setsid).If you are using job control in your shell each job is assigneda separate process group (setpgrp).The shell and all process groups are in the same session.Intuitively a process group is considered orphaned when there isare no processes in the session that know about it and can wake itup. The goal is to prevent processes that will never wake up ifthey are stopped with ^Z.A process is defined as knowing about a process in a process groupwhen it is a parent of that process.The task_tgid_nr_ns(p->real_parent, p->nsproxy->pid_ns) == 1 check isthe proper check, as it handles namespaces properly. If we need toretain it.I don't believe we need to retain the check for init at all. sysvinitdoesn't use that feature and I would be surprised if any other initdid. Except for covering programming bugs which make testing harderI don't see how a version of init could care.Further as init=/bin/bash is a common idiom for getting into asimplified system and debugging it, there is a case for job controlto work properly for init. Unless I am misreading things the check for init prevent us from using job control from our first process.Which seems like it would make init=/bin/bash painful if job controlwas ever enabled.I believe that the only reason with a weird check for init like we areperforming that we are POSIX compliant is that our init process cancount as a special system process and can escape the definition.Therefore I think the code would be more maintainable, and the systemwould be less surprising, and more useful if we removed this specialcase processing of init altogether.I'm hoping that we can kill this check before pid namespaces arewidely deployed and a much larger number of programs can run as init.Eric | http://lkml.org/lkml/2008/3/5/532 | CC-MAIN-2017-34 | refinedweb | 420 | 64.91 |
This topic illustrates the use of Code Effects' data filtering capabilities using the LINQ to Object provider. The goal is to allow the end user to create a business rule that will be used to filter out unwanted items from an in-memory collection of source objects.
To begin, let's create three classes: one base and two descendants. We will use these classes as our source objects:
class ClassA
{
public int Id { get; set; }
public string FirstName { get; set; }
public string LastName { get; set; }
}
class ClassB : ClassA
{
public string Email { get; set; }
}
class ClassC : ClassA
{
public DateTime DOB { get; set; }
}
Then let's create a web page, add a Rule Editor, set its Mode property to Filter, and give it ClassA as a source object. If we were to build the project and deploy it to a server, the end user(s) could create and save a filter like this one (remember that filters are just evaluation type business rules):
Assuming that we saved this filter somewhere in a database, we can use it in any .NET code against any IEnumerable<> collection of ClassA instances to filter out those that don't satisfy the conditions in the filter:
using System;
using System.Linq;
using System.Collections.Generic;
using CodeEffects.Rule.Core;
class Test
{
public void Testing()
{
string rule = YouRuleStorage.GetRuleXml();
ClassA[] array = new[]
{
new ClassA { Id = 1, FirstName = "John", LastName = "Smith" },
new ClassB { Id = 2, FirstName = "Mike", LastName = "Doe" },
new ClassC { Id = 3, FirstName = "John", LastName = "Doe" },
new ClassB { Id = 6, Email = "test@test.test" },
new ClassC { Id = 8, DOB = DateTime.Now }
};
IEnumerable<ClassA> result = array.Filter(rule);
foreach(var item in result)
{
Console.WriteLine(item.Id);
}
// Produces:
// 3
// 6
// 8
}
}
As you can see, it takes very little effort to create very complex filters and use them against IEnumerable data of any size by employing the Filter extension method. LINQ To Object filtering in Code Effects is very efficient and extremely fast. | https://codeeffects.com/Doc/Business-Rule-Linq-To-Object-Support | CC-MAIN-2021-31 | refinedweb | 324 | 59.43 |
Introduction to Library Functions SHELL(3)
NAME
shell - a ksh library interface
SYNOPSIS
HEADERS/LIBRARIES
#include <shell.h>
libshell.a -lshell
DATA TYPES
Shell_t;
Shopt_t;
Shscope_t;
Shbltin_f;
Shinit_f;
Shwait_f;
FUNCTIONS
int sh_main(int argc, char *argv[], Sh_init fn);
Shell_t *sh_init(int argc, char *argv);
Shell_t *sh_getinterp(void);
Namval_t *sh_addbuiltin(const char *name,Sh_bltin_f fn,void *arg);
unsigned int sh_isoption(int option);
unsigned int sh_onoption(int option);
unsigned int sh_offoption(int option);
void *sh_parse(Shell_t *shp, Sfio_t *sp, int flags);
int sh_trap(const char *string, int mode);
int sh_eval(Sfio_t *sp,int mode);
int sh_fun(Namval_t *funnode, Namval_t *varnode, char *argv[]);
int sh_funscope(int argc,char *argv[],int(*fn)(void*),void *arg,int flags);
Shscope_t *sh_getscope(int index,int whence);
Shscope_t *sh_setscope(Shscope_t *scope);
int (*sh_fdnotify(int(*fn)(int,int)))(int,int);
char *sh_fmtq(const char *string);
void *sh_waitnotify(Shwait_f fn);
void sh_delay(double sec);
Sfio_t *sh_iogetiop(int fd, int mode);
int sh_sigcheck(void);
DESCRIPTION
The Shell library is a set of functions used for writing
extensions to ksh or writing commands that embed shell com-
mand processing. The include file <shell.h> contains the
type definitions, function prototypes and symbolic constants
defined by this interface. It also defines replacement
definitions for the standard library functions access(),
close(), dup(), exit(), fcntl(), lseek(), open(), pipe(),
read(), and write() that must be used with all code intended
to be compiled as built-in commands.
SunOS 5.10 Last change: 28 Feb 2003 1
Introduction to Library Functions SHELL(3)
The <shell.h> header includes <ast.h> which in turn includes
the standard include files, <stddef.h>, <stdlib.h>,
<stdarg.h>, <limits.h>, <stdio.h>, <string.h>, <unistd.h>,
<sys/types.h>, <fcntl.h>, and <locale.h>. The <shell.h>
header also includes the headers <cdt.h>, <cmd.h>, <sfio.h>,
<nval.h>, <stk.h>, and <error.h> so that in most cases, pro-
grams only require the single header <shell.h>.
Programs can use this library in one of the following ways:
1 To write builtin commands and/or other code that will
be loaded into the shell by loading dynamic libraries
at run time using the builtin(1) command. In this case
the shell will look for a function named lib_init in
your library and, if found, will execute this function
with argument 0 when the library is loaded. In addi-
tion, for each argument named on the builtin command
line, it will look for a function named b_name() in
your library and will name as a built-in.
2 To build separate a separate command that uses the
shell as a library at compile or run time. In this
case the sh_init() function must be called to initial-
ize this library before any other commands in this
library are invoked. The arguments argc and argv are
the number of arguments and the vector of arguments as
supplied by the shell. It returns a pointer the
Shell_t.
3 To build a new version of ksh with extended capabili-
ties, for example tksh(1). In this case, the user
writes a main() function that calls sh_main() with the
argc and argv arguments from main and pointer to func-
tion, fn as a third argument.. The function fn will be
invoked with argument 0 after ksh has done initializa-
tion, but before ksh has processed any start up files
or executed any commands. The function fn will be
invoked with an argument of 1 before ksh begins to exe-
cute a script that has been invoked by name since ksh
cleans up memory and long jumps back to the beginning
of the shell in this case. The function fn will be
called with argument -1 before the shell exits.
The Shell_t structure contains the following fields:
Shopt_t options; /* set -o options */
Dt_t *var_tree; /* shell variable dictionary */
Dt_t *fun_tree; /* shell function dictionary */
Dt_t *alias_tree; /* shell alias dictionary */
Dt_t *bltin_tree; /* shell built-in dictionary */
Shscope_t *topscope; /* pointer to top-level scope */
char *infile_name; /* path name of current input file*/
int inlineno; /* line number of current input file*/
int exitval; /* most recent exit value*/
This structure is returned by sh_init() but can also be
retrieved by a call to sh_getinterp().
SunOS 5.10 Last change: 28 Feb 2003 2
Introduction to Library Functions SHELL(3)
All built-in commands to the shell are invoked with three
arguments. The first two arguments give the number of argu-
ments and the argument list and uses the same conventions as
the main() function of a program. The third argument is a
pointer that can be associated with each built-in. The
sh_addbuiltin() function is used to add, replace or delete
built-in commands. It takes the name of the built-in, name,
a pointer to the function that implements the built-in, fn,
and a pointer that will be passed to the function when it is
invoked. If, fn is non-NULL the built-in command is added
or replaced. Otherwise, the given built-in command will be
deleted. The name argument can be in the format of a path-
name. It cannot be the name of any of the special built-in
commands. If name contains a /, the built-in is the
basename of the pathname and the built-in will only be exe-
cuted if the given pathname is encountered when performing a
path search. When adding or replacing a built-in,
sh_addbuiltin() function returns a pointer to the name-value
pair corresponding to the built-in on success and NULL if it
is unable to add or replace the built-in. When deleting a
built-in, NULL is returned on success or if not found, and
the name-value pair pointer is returned if the built-in can-
not be deleted.
The functions sh_onoption(), sh_offoption(), sh_isoption()
are used to set, unset, and test for shell options respec-
tively. The option argument can be any one of the follow-
ing:
SH_ALLEXPORT: The NV_EXPORT attribute is given to each
variable whose name is an identifier when a value is
assigned.
SH_BGNICE: Each background process is run at a lower
priority.
SH_ERREXIT: Causes a non-interactive shell to exit
when a command, other than a conditional command,
returns non-zero.
SH_EMACS: The emacs editing mode.
SH_GMACS: Same as the emacs editing mode except for
the behavior of CONTROL-T.
SH_HISTORY: Indicates that the history file has been
created and that commands can be logged.
SH_IGNOREEOF: Do not treat end-of-file as exit.
SH_INTERACTIVE:
SunOS 5.10 Last change: 28 Feb 2003 3
Introduction to Library Functions SHELL(3)
Set for interactive shells. Do not set or unset this
option. SH_MARKDIRS: A / is added to the end of each
directory generated by pathname expansion.
SH_MONITOR: Indicates that the monitor option is
enabled for job control.
SH_NOCLOBBER: The > redirection will fail if the file
exists. Each file created with > will have the O_EXCL
bit set as described in <fcntl.h>
SH_NOGLOB: Do not perform pathname expansion.
SH_NOLOG: Do not save function definitions in the his-
tory file.
SH_NOTIFY: Cause a message to be generated as soon as
each background job completes.
SH_NOUNSET: Cause the shell to fail with an error of
an unset variable is referenced.
SH_PRIVILEGED:
SH_VERBOSE: Cause each line to be echoed as it is read
by the parser.
SH_XTRACE: Cause each command to be displayed after
all expansions, but before execution.
SH_VI: The vi edit mode.
SH_VIRAW: Read character at a time rather than line at
a time when in vi edit mode.
The sh_trap() function can be used to compile and execute a
string or file. A value of 0 for mode indicates that name
refers to a string. A value of 1 for mode indicates that
name is an Sfio_t* to an open stream. A value of 2 for mode
indicates that name points to a parse tree that has been
returned by sh_parse(). The complete file associated with
the string or file is compiled and then executed so that
aliases defined within the string or file will not take
effect until the next command is executed.
The sh_eval() function executes a string or file stream sp.
If mode is non-zero and the history file has been created,
the stream defined by sp will be appended to the history
file as a command.
SunOS 5.10 Last change: 28 Feb 2003 4
Introduction to Library Functions SHELL(3)
The sh_parse() function takes a pointer to the shell inter-
preter shp, a pointer to a string or file stream sp, and
compilation flags, and returns a pointer to a parse tree of
the compiled stream. This pointer can be used in subsequent
calls to sh_trap(). The compilation flags can be zero or
more of the following:
SH_NL: Treat new-lines as ;.
SH_EOF: An end of file causes syntax error. By
default it will be treated as a new-line.
ksh executes each function defined with the function
reserved word in a separate scope. The Shscope_t type pro-
vides an interface to some of the information that is avail-
able on each scope. The structure contains the following
public members:
Sh_scope_t *par_scope;
int argc;
char **argv;
char *cmdname;
Dt_t *var_tree;
The sh_getscope() function can be used to the the scope
information associated with existing scope. Scopes are num-
bered from 0 for the global scope up to the current scope
level. The whence argument uses the symbolic constants
associated with lseek() to indicate whether the Iscope argu-
ment is absolute, relative to the current scope, or relative
to the topmost scope. Thesh_setscope() function can be used
to make a a known scope the current scope. It returns a
pointer to the old current scope.
The sh_funscope() function can be used to run a function in
a new scope. The arguments argc and argv are the number of
arguments and the list of arguments respectively. If fn is
non-NULL, then this function is invoked with argc, argv, and
arg as arguments.
The sh_fun() function can be called within a discipline
function or built-in extension to execute a discipline func-
tion script. The argument funnode is a pointer to the shell
function or built-in to execute. The argument varnode is a
pointer to the name value pair that has defined this discip-
line. The array argv is a NULL terminated list of arguments
that are passed to the function.
By default, ksh only records but does not act on signals
when running a built-in command. If a built-in takes a sub-
stantial amount of time to execute, then it should check for
interrupts periodically by calling sh_sigcheck(). If a sig-
nal is pending, sh_sigcheck() will exit the function you are
calling and return to the point where the most recent
SunOS 5.10 Last change: 28 Feb 2003 5
Introduction to Library Functions SHELL(3)
built-in was invoked, or where sh_eval() or sh_trap() was
called.
The sh_delay() function is used to cause the shell to sleep
for a period of time defined by sec.
The sh_fmtq() function can be used to convert a string into
a string that is quoted so that it can be reinput to the
shell. The quoted string returned by sh_fmtq may be returned
on the current stack, so that it must be saved or copied.
The sh_fdnotify() function causes the function fn to be
called whenever the shell duplicates or closes a file. It
is provided for extensions that need to keep track of file
descriptors that could be changed by shell commands. The
function fn is called with two arguments. The first argu-
ment is the original file descriptor. The second argument
is the new file descriptor for duplicating files, and
SH_FDCLOSE when a file has been closed. The previously
installed sh_fdnotify() function pointer is returned.
The sh_waitnotify() function causes the function fn to be
called whenever the shell is waiting for input from a slow
device or waiting for a process to complete. This function
can process events and run shell commands until there is
input, the timer is reached or a signal arises. It is
called with three arguments. The first is the file descrip-
tor from which the shell trying to read or -1 if the shell
is waiting for a process to complete. The second is a
timeout in milliseconds. A value of -1 for the timeout
means that no timeout should be set. The third argument is
0 for input file descriptors and 1 for output file descrip-
tor. The function needs to return a value >0 if there is
input on the file descriptor, and a value <0 if the timeout
is reached or a signal has occurred. A value of 0 indicates
that the function has returned without processing and that
the shell should wait for input or process completion. The
previous installed sh_waitnotify() function pointer is
returned.
The sh_iogetiop() function returns a pointer to the Sfio
stream corresponding to file descriptor number fd and the
given mode mode. The mode can be either SF_READ or
SF_WRITE. The fd argument can the number of an open file
descriptor or one of the following symbolic constants:
SH_IOCOPROCESS: The stream corresponding to the most
recent co-process.
SH_IOHISTFILE: The stream corresponding to the history
file. If no stream exists corresponding to fd or the
stream can not be accessed in the specified mode, NULL
SunOS 5.10 Last change: 28 Feb 2003 6
Introduction to Library Functions SHELL(3)
is returned.
SEE ALSO
builtin(1) cdt(3) error(3) nval(3) sfio(3) stk(3) tksh(1)
AUTHOR
David G. Korn (dgk at research dot att dot com).
SunOS 5.10 Last change: 28 Feb 2003 7
Page Last Modified: 01 May 2006
Privacy |
Trademarks |
Site Guidelines |
HelpYour use of this web site or any of its content or software indicates your agreement to be bound by these Terms of Use.Copyright © 1995-2008 Sun Microsystems, Inc. | http://www.opensolaris.org/os/project/ksh93-integration/docs/ksh93r/man/man3/shell/ | crawl-001 | refinedweb | 2,315 | 69.72 |
Red Hat Bugzilla – Full Text Bug Listing
Gabriel VLASIU reported [1] that yum-cron would install unsigned RPM packages that yum itself would refuse to install. The yum-cron code is based on that in yum-updatesd.py. This is due to the installUpdates() function (processPkgs() in yum-updatesd.py) failing to fully check the return code of the called sigCheckPkg() function. sigCheckPkg() is described thus:
def sigCheckPkg(self, po):
"""Verify the GPG signature of the given package object.
:param po: the package object to verify the signature of
:return: (result, error_string)
where result is::
0 = GPG signature verifies ok or verification is not required.
1 = GPG verification failed but installation of the right GPG key
might help.
2 = Fatal GPG verification error, give up.
"""
However, the processPkgs() and installUpdates() calling function do not account for return code 2:
def processPkgs(self, dlpkgs):
...
for po in dlpkgs:
result, err = self.updd.sigCheckPkg(po)
if result == 0:
continue
elif result == 1:
try:
self.updd.getKeyForPackage(po)
except yum.Errors.YumBaseError, errmsg:
self.failed([str(errmsg)])
and:
def installUpdates(self, emit):
...
for po in dlpkgs:
result, err = self.sigCheckPkg(po)
if result == 0:
continue
elif result == 1:
try:
self.getKeyForPackage(po)
except yum.Errors.YumBaseError, errmsg:
self.emitUpdateFailed(errmsg)
return False
yum-cron.py replaced yum-cron.sh in Fedora 19 (3.4.3-47); earlier versions of Fedora use yum-updatesd.
This has been corrected upstream [2] and in Fedora via yum-3.4.3-132.fc19 and yum-3.4.3-130.fc20.
This does not affect Red Hat Enterprise Linux 6 as it used neither yum-updatesd nor yum-cron; it used a shellscript that called yum itself to do updates.
[1]
[2];a=commitdiff;h=9df69e579496ccb6df5c3f5b5b7bab8d648b06b4
The comment 0 above explains that Red Hat Enterprise Linux 6 was not affected, as it did not include vulnerable version of yum-updatesd or yum-cron. This issue was resolved in yum-cron shipped as part of Red Hat Enterprise Linux 7 before its initial release.
Statement:
This issue did not affect the versions of yum as shipped with Red Hat Enterprise Linux 6 and 7.
It should also be noted that in their default configuration, yum-updatesd and yum-cron are not configured to automatically install available updates. They are configured to provide notification of updates availability. yum-cron is also configured to download updated packages, but not install them.
This issue has been addressed in following products:
Red Hat Enterprise Linux 5
Via RHSA-2014:1004
IssueDescription:. | https://bugzilla.redhat.com/show_bug.cgi?format=multiple&id=1057377 | CC-MAIN-2018-09 | refinedweb | 418 | 51.65 |
For any of you that are using DigitalCAT-formatted dictionaries (e.g. the one that comes with Stenomaster), I've fixed a major bug and I've also consolidated the files, so you don't need to use dccattest.py or dcatwinder.py anymore. Just download stenowinder.py and tktest.py at the GitHub, then open stenowinder.py in any text editor. The first few lines of the file look like this:
from ploverbd import exportDic
import unittest
import sys
# choose either 'dCAT' or 'Eclipse'
dictType = 'dCAT'
If it already says dCAT and you have a DigitalCAT-formatted dictionary, you're set. If it says Eclipse, change it to dCAT. Unless you have an Eclipse-formatted dictionary, in which case change it from dCAT to Eclipse. You get the idea. The default distributed dictionary, ploverbd.py, is in Eclipse format. Also, it's apparently grown too big to be directly accessed by Github's html parser and has to be downloaded by right click/save as (wait a while for it to muster its strength) or by getting the whole package by clicking the "download source" button.
New and exciting features, like functional file output and maybe even punctuation formatting, will with luck be added on Monday, (and sometime after that, I hope to add TX and Stentura protocol options into the program; I got the protocol specs today, but I can't make any sense of them, and they're second in priority to figuring out this file output thing.) If you guys find any more bugs, though, drop me a line and I'll try to patch them up as they come in.
3 comments:
Hi - I've tried this but it's not working and I can't use the dcattest file either. I think I need some more instruction....
Thanks Mirabai, I've got it working now. I think I missed a step before. | http://plover.stenoknight.com/2010/04/bug-fixes.html?showComment=1272733259146 | CC-MAIN-2021-10 | refinedweb | 318 | 73.47 |
I am working on an assignment where I have to create a program for users to input coordinates for a rectangle.
This program is intended to be a struct within a struct.
If invalid, I have to output an error message and let user try again, indefinitely, until the user gets it right.
The program should repeatedly ask users for coordinates of a point, and the program will quit when user enters 0 and 0 for x and y respectively.
The program has to say whether the point is inside of outside the rectangle. I also need to figure out where to put the main function, and what to put in it. Please let me know how exactly to complete this program, and I need to know ASAP. The program is due TONIGHT.
Here is my code:
Code:#define _CRT_SECURE_NO_DEPRECATE #include <stdio.h> typedef struct { int x; int y; } point_t; typedef struct { point_t upper_left; point_t lower_right; } rectangle_t; int is_inside (point_t* pPoint, rectangle_t* pRect) { return ((pPoint->x >= pRect->upper_left.x ) && (pPoint->x <= pRect->lower_right.x) && (pPoint->y >= pRect->upper_left.y ) && (pPoint->y <= pRect->lower_right.y)); } point_t get_point(char* prompt) { point_t pt; printf("Given a rectangle with a side parallel to the x axis and a series of points on the xy plane this program will say where each point lies in relation to the rectangle. It considers a point on the boundary of the rectangle to be inside the rectangle\n"); printf ("Enter coordinates for the upper left corner\n"); printf ("X: "); scanf ("%d", &pt.x); printf ("Y: "); scanf ("%d", &pt.y); return pt; } rectangle_t get_rect(char* prompt) { rectangle_t rect; printf (prompt); rect.upper_left = get_point("Upper left corner: \n"); rect.lower_right = get_point("Lower right corner: \n"); return rect; } | http://cboard.cprogramming.com/cplusplus-programming/114617-struct-program-find-point-rectangle.html | CC-MAIN-2016-40 | refinedweb | 287 | 65.93 |
19 November 2008 04:37 [Source: ICIS news]
SINGAPORE (ICIS news)--Shanghai Wujing Chemical Co raised its acetic acid operating rate marginally by nine percentage points to 54% of its combined 550,000 tonne/year capacity, following a stabilising of Chinese domestic acetic acid values, a company official said on Wednesday.
?xml:namespace>
The company resumed operations at its 300,000 tonne/year No 2 acetic acid line in Wujing, Shanghai, on 12 November following a two-week shutdown.
At the same time, it shut its 250,000 tonne/year No 1 line, bringing the operating rate to 54%, the company official said in Mandarin.
“We do not have any firm plans on the duration of the shutdown of the No 1 line,” he added.
Major acetic acid producers in Asia include Celanese, BP, CPDC, Jiangsu Sopo, Daicel Chemical Industries, Showa Denko and GNFC.
For more on acetic acid | http://www.icis.com/Articles/2008/11/19/9172683/chinas-shanghai-wujing-raises-acetic-acid-output.html | CC-MAIN-2014-35 | refinedweb | 149 | 56.39 |
So far, the appearance of the various elements has been fixed. But web pages should be able to override our style decisions and take on a unique character. This is done via CSS.
styleattribute
Different elements have different styles, like margins for paragraphs and borders for code blocks. Those styles are assigned by our browser, and it is good to have some defaults. But webpages should be able to override those choices.
The simplest mechanism for that is the
style attribute on elements. It looks like this:
<div style="margin-left:10px;margin-right;10px;"></div>
It’s a
<div> element with its
style attribute set. That attribute contains two key-value pairs, which set
margin-left and
margin-right to 10 pixels each.CSS allows spaces around the punctuation, but your attribtue parser may support it. We want store these pairs in a
style field on the
ElementNode so we could consult them during layout.
You should already have some attribute parsing code in your
ElementNode class to create the
attributes field. We just need to take out the
style attribute, split it on semicolons and then on the colon, and save the results:The
get method for dictionaries gets a value out of a dictionary, or uses a default value if it’s not present.
class ElementNode: def __init__(self, parent, tagname): # ... self.style = self.compute_style() def compute_style(self): style = {} style_value = self.attributes.get("style", "") for line in style_value.split(";"): prop, val = line.split(":") style[prop.lower().strip()] = val.strip() return style
To use this information, we’ll need to modify
BlockLayout:
def __init__(self, parent, node): # ... self.mt = px(self.style.get("margin-top", "0px")) # ... repeat for the right, bottom, and left edges # ... and for padding and border as well
where the
px function is this little helper:
Remember the write out the code to access the other 11 properties; the border one is called
border-top-width, not
border-top, but other than that, they’re very repetitive.
You’ll notice that I set the default for each of the property values to
0px. For now, let’s stick the per-element defaults at the top of
compute_style:
def compute_style(self): style = {} if self.tag == "p": style["margin-bottom"] = "16px" # ... other cases for ul, li, and pre # ...
Make sure the defaults come first in
compute_style so they can be overridden by values from the
style attribute.
The
style attribtue is set element by element. It’s good for one-off changes, but is a tedious way to change the style of, say, every paragraph on the page. Plus, if multiple web pages are supposed to share the same style, you’re liable to forget the
style attribute. In the early days of the web,I'm talking Netscape 3. The late 90s. this element-by-element approach was all there was.Though back then it wasn’t the
style attribute, it was a custom elements like
font and
center. CSS was invented to improve on this state of affairs:
To achieve that, extended the key-value
style attribute with two connected ideas: selectors and cascading. In CSS, you have blocks of key-value pairs, and those blocks apply to multiple elements, specified using a selector. Since that allows multiple key-values pairs to apply to one element, cascading resolves conflicts by using the most specific rule.
Those blocks look like this:
To support CSS in our browser, we’ll need to parse this kind of code. I'll use a traditional recursive-descent parser, which is is a bunch of parsing functions, each of which advances along the input and returns the parsed data as output.
Specifically, I’ll implement a
CSSParser class, which will store the input string. Each parsing function will take an index into the input and return a new index, plus it will return the parsed data.
Here’s the class:
So here, for example, is a parsing function for values:
Let’s pick this apart. First of all, it takes index
i pointing to the start of the value and returns index
j pointing to its end. It also returns the string between them, which in this case is the parsed data that we’re interested in.
The point of recursive-descent parsing is that it’s easy to build one parsing function by calling others. So here’s how to parse property-value pairs:
def pair(self, i): prop, i = self.value(i) _, i = self.whitespace(i) assert self.s[i] == ":" val, i = self.value(i + 1) return (prop, val), i
The
whitespace function increases
i until it sees a non-whitespace character (or the end of the document); you can write it yourself.
Note the
assert: that raises an error if you are trying to parse a pair but there isn’t one there. When we parse rule bodies, we can catch this error to skip property-value pairs that don’t parse:
def body(self, i): pairs = {} assert self.s[i] == "{" _, i = self.whitespace(i+1) while True: if self.s[i] == "}": break try: (prop, val), i = self.pair(i) pairs[prop] = val _, i = self.whitespace(i) assert self.s[i] == ";" _, i = self.whitespace(i+1) except AssertionError: while self.s[i] not in [";", "}"]: i += 1 if self.s[i] == ";": _, i = self.whitespace(i + 1) assert self.s[i] == "}" return pairs, i+1
I should stop and mention the importance of skipping code that causes parse errors. This is a double-edged sword. It hides error messages, making debugging CSS files more difficult, and also makes it harder to debug your parser.Try debugging without the
try block first. This makes “catch-all” error handling like this a code smell in most cases.
However, on the web there is an unusual benefit: it supports an ecosystem of multiple implementations. For example, different browsers may support different syntax for property values.Our browser does not support parentheses in property values, which are valid in real browsers, for example. Crashing on a parse error would mean web pages can’t use a feature until all browsers support it, while skipping parse errors means a feature is useful once a single browser supports it. This is variously called “Postel’s Law”,After a line in the specification of TCP, written by Jon Postel the “Digital Principle”,After a similar idea in circuit design. or the “Robustness Principle”: produce maximally supported output but accept unsupported input.
Finally, to parse a full CSS rule, we need to parse selectors. Selectors come in multiple types; for now, our browser will support three:
pselects all
<p>elements,
ulselects all
<ul>elements, and so on.
classattribute, which is a space-separated list of arbitrary names, so the
.fooselector selects the elements that have
fooin that list.
#mainselects the element with an
idvalue of
main.
We'll start by defining some data structures for selectors:I'm calling the
ClassSelector field
cls instead of
class because
class is a reserved word in Python.
class TagSelector: def __init__(self, tag): self.tag = tag class ClassSelector: def __init__(self, cls): self.cls = cls class IdSelector: def __init__(self, id): self.id = id
We now want parsing functions for each of these data structures. That’ll look like:
def selector(self, i): if self.s[i] == "#": name, i = self.value(i+1) return IdSelector(name), i elif self.s[i] == ".": name, i = self.value(i+1) return ClassSelector(name), i else: name, i = self.value(i) return TagSelector(name), i
Here I’m using
property for tag, class, and identifier names. This is a hack, since in fact tag names, classes, and identifiers have different allowed characters. Also tags are case-insensitive (as by the way are property names), while classes and identifiers are case-sensitive. I’m ignoring that but a real browser would not. Note the arithmetic with
i: we pass
i+1 to
value in the class and ID cases (to skip the hash or dot) but not in the tag case (since that first character is part of the tag).
I’ll leave it to you to finish up the parser, writing the
whitespace helper, the
rule function for parsing a selector followed by a body (making sure to skip rules with unknown selectors), and the
parse function, which unlike the others should not take an index input (it should start at 0) or produce an index output and should return a list of parsed selector/body pairs.
Now that we’ve parsed a CSS file, we need to apply it to the elements on the page.
Browsers get CSS code from two sources. First, each browser ships with a browser style sheet, which defines the default styles for all sorts of elements; second, browsers download CSS code from the web, as directed by web pages they browse to. Let’s start with the browser style sheet.
Our browser’s style sheet might look like this:
p { margin-bottom: 16px; } ul { margin-top: 16px; margin-bottom: 16px; padding-left: 20px; } li { margin-bottom: 8px; } pre { margin-top: 8px; margin-bottom: 8px; padding-top: 8px; padding-right: 8px; padding-bottom: 8px; padding-left: 8px; border-top-width: 1px; border-bottom-width: 1px; border-left-width: 1px; border-right-width: 1px; }
That moves code from
compute_style to a data file, let's call it
browser.css. Then we can run our CSS parser on it to extract the rules:
We now want to apply the rules to change the
style field of the
ElementNode objects on the page; let’s call the function that does that
style(tree, rules). Its logic is pretty simple:
ElementNodein the tree
Here's what the code would look like:
def style(node, rules): if not isinstance(node, ElementNode): return node.style = {} for selector, pairs in rules: if selector.matches(node): for property in pairs: node.style[property] = pairs[property] for child in node.children: style(child, rules)
Note that we're skipping
TextNode objects; that's because only elements can be selected in CSS.Well, there are also pseudo-elements, but we're not going to implement them… We’re also calling a currently-nonexistant
matches method on selectors. Here's how it looks like for
ClassSelector:
You can write
matches for
TagSelector and
IdSelector on your own. You can then call
style:
Once you’re done, you should be able to delete the default element styles from
compute_style and still see elements property styled.
This moves some code out of our browser into a plain CSS file, which is nice. But the goal is to go beyond just the browser styles; to do that, our browser needs to find website-specific CSS files, download them, and use them as well. Web pages call out their CSS files using the
link element, which looks like this:
<link rel="stylesheet" href="/main.css">
The
rel attribute here tells that browser that this is a link to a stylesheet; web pages can also link to a home-page, or a translation, or similar. Browsers mostly don't do anything with those other kinds of links, but search engines do, so
rel is widely used.
The
href attribute gives a location for the stylesheet in question. The browser is expected to make a GET request to that location, parse the stylesheet, and use it. Note that the location is not a full URL; it is something called a relative URL, which can come in three flavors:There are more flavors, including query-relative and scheme-relative URLs, which I’m skipping.
So to download CSS files, we're going to need to do three things:
<link>elements
For the first one, we'll need a recursive function that adds to a list:
def find_links(node, lst): if not isinstance(node, ElementNode): return if node.tag == "link" and \ node.attributes.get("rel", "") == "stylesheet" and \ "href" in node.attributes: lst.append(node.attributes["href"]) for child in node.children: find_links(child, lst) return lst
Then, to turn a relative URL into a full URL:
def relative_url(url, current): if url.startswith("http://"): return url elif url.startswith("/"): return "/".join(current.split("/")[:3]) + url else: return current.rsplit("/", 1)[0] + "/" + url
In the first case, the
[:3] and the
"/".join handle the two slashes that come after
http: in the URL, while in the last case, the logic ensures that a link to
foo.html on goes to, not.
We want to collect CSS rules from each of the linked files, and the browser style sheet, into one big list so we can apply each of them. So let’s add them onto the end of the
rules list:
for link in find_links(nodes, []): header, body = request(relative_url(link, url)) rules.extend(CSSParser(body).parse())
Put that block after you read the browser style, because user styles should to take priority over the browser style sheet,In reality this is handled by browser styles having a lower score than user styles in the cascade order, but our browser style sheet only has tag selectors in it, so every rule already has the lowest possible score. but before the call to
style, so you actually use the newly downloaded rules.
Our CSS engine should now change some margins and paddings as specified by the web page in question. For example, on this web page, you should see that the title of the page has moved down significantly.
So far,
style applies the rules in order, one after another. And furthermore, it overwrites existing styles, such as the styles that come from attribute values. We need to fix that.
In CSS, this is governed by the cascade order, which assigns a score to each selector; rules with higher-scoring selectors overwrite rules with lower-scoring selectors. Tag selectors get the lowest score; class selectors one higher; and id selectors higher still. The
style attribute has the highest-possible score, so it overwrites everything. So let's add the
score method to the selector classes that return this score. Maybe tag selectors have score 1, class selectors 16, id selectors 256.In this simplest implementation the exact numbers don't matter if they sort right, but choosing these numbers makes the exercises a little easier.
You can write the code for the other selector types.
We'll use the score to the rules before passing them to
style:
Here
x[0] refers to the selector half of a rule, and I'm calling the new
score method. In Python, the
sort function is stable, which means that things keep their relative order if possible. This means that in general, a later rule, from a later
<link> or just later in a file, will override an earlier one, which is what CSS does as well.
This works for rules with selectors. We also need inline styles to override linked stylesheets. We can just tack that on after the rules loop in
style:The
items() call is a Python way to get the key and value out of a dictionary as you iterate over it.
def style(node, rules): # for selector, pair in rules ... for property, value in node.compute_style().items(): node.style[property] = value # for child in node.children ...
Our CSS engine now correctly handles conflicts between different rules.
Right now, our CSS styles only affect the block layout mode. We'd like to extend CSS to affect inline layout mode as well, but there's a catch: inline layout is mostly concerned with text, but text nodes don't have any styles at all. How can that work?
The solution in CSS is inheritance. Inheritance means that if some node doesn't have a value for a certain property, it uses its parent's value instead. Some properties are inherited and some aren't; it depends on the property. Let's implement two inherited properties:
font-weight (which can be
normal or
bold) and
font-style (which can be
normal or
italicActually, it can also be
oblique. No one knows that that is, though some browsers will use that value to display pseudo-italics, that is, roman text that's been algorithmically slanted.). To inherit a property, we simple need to check, after all the rules and inline styles have been applied, whether the property is set and, if it isn't, to use the parent node's style:
INHERITED_PROPERTIES = [ "font-style", "font-weight" ] def style(node, rules): # handle inline styles for prop in INHERITED_PROPERTIES: if prop not in node.style and node.parent is None: node.style[prop] = "normal" # recurse into child nodes
This little loop has to come before the recursive calling of
style on the child nodes because getting the parent's value only makes sense if the parent has already inherited the correct property value.
On
TextNode objects we can do an even simpler trick, since it always inherits its styles from its parent:
Now that we have
font-weight and
font-style set on every node, we can use them in
InlineLayout to set the font:
self.bold = node.style.get("font-weight", "normal") == "bold" self.italic = node.style.get("font-style", "normal") == "italic"
Now that we have styles on both block and inline nodes, we can refactor
is_inline. Instead of testing directly for the
<b> and
<i>, we could test for a CSS property both share. The standard property is
display, which can be either
block or
inline; it basically tells you which of the two layout modes to use.Modern CSS adds some funny new values, like
run-in or
inline-block, and it has layout modes set by other properties, like
float and
position. Nothing gold can stay. So instead of…
… do …
With inheritance and the
display property, we can move some more code into the CSS file.
By the way—why move code to a data file? The advantage is that that the data file may be easier to write, especially independently of the rest of the code. So here, you could experiment with new browser default styles more quickly. But as a software design decision, it is not always a winner, since you have to maintain a new format (for the data file) and also code to parse the data and then to apply it as code. That’s true in general, but here in particular, we need a CSS parser and applier anyway, so the downsides do not apply, and the refactoring is very much worth it.
I think I’d like nodes to have a
font() method that returns the font to use.
This chapter was quite a lot of work! We implemented a rudimentary but complete layout engine, including a parser, selector matching, cascading, and even downloading and applying CSS files. Not only that, but the CSS engine should be relatively easy to extend, with new properties and selectors; our engine ignores selectors and properties it does not understand, so selectors and properties will immediately start working as they are implemented.
<title>element is hidden, as is everything inside the
<head>element (try adding a
<p>to the head in a real browser!). The way this works is that
displaycan have the value
none, in which case the element is not displayed and neither are any of its children. Implement
display: noneand use that in your browser style sheet to hide the
<head>element.
margin: 1pxto set all four margins to the same width; the same applies to
paddingand
border-width. You can also give multiple values to
margin, which distributes those values to the various sides: if there is one value, it is for all four sides; if there are two values, the first is for top and bottom and the second for left and right; if there are four, they are the top, right, bottom, and left values, in that unusual order; and finally if there are three values the middle one is both left and right. Implement shortcut properties. The best place to do this is in the parsing function
css_body, since that way it'll automatically happen whereever the rule is applied.
<b>and
<i>have
display: inline.)
span.announceselects elements that match both
spanand
.announce. Implement those, both in the parser and with a new
AndSelectorclass that combines multiple selectors into one. You're supposed to use lexicographic scoring for these
AndSelectorthings, but the easy thing to do is to sum the scores of the selectors being combined in
AndSelector.score. This will work fine as long as no strings more than 16 selectors together, if you used the scores suggested above.
ul strong, which selects all
<strong>elementsThis is basically a
<b>tag but with a hipper name. The idea was to drop all visual aspects from element names. with a
<ul>ancestor. Implement descendent selectors, both in parsing and with a new
DescendentSelectorclass. Scoring for descendent selectors works just like in
AndSelector. Make sure that something like
section .warningselects warnings inside sections, while
section.warningselects warnings that are sections. | https://browser.engineering/styles.html | CC-MAIN-2019-51 | refinedweb | 3,503 | 63.7 |
CCE 4 with Symbian SDKs
Why use GCCE 4
The SDKs for Symbian OS v.9.1+ development rely on the GCCE CSL Arm Toolchain to compile code for the real devices. However, the version of GCCE compiler distributed with the SDKs is quite old, 3.4.3. - originally published in November 2004.
Although GCC 3.4.3 is, in most cases, a good compiler, it nevertheless contains some problems. It is also quite slow. During the development of bigger applications, you may find out that a newer toolchain is preferable for the following reasons:
- GCC 4 compiler can be noticeably faster, especially in C code.
- GCC 4 seems to have fewer problems with optimisation.
- GCC 4 produces significantly smaller binaries (for example, in the testcase, the SIS with the full application shrank from 1.4 MB to 1 MB (Release)).
- GCC 4 seems to have fewer miscompilations.
- GCC 4 is stricter in producing errors and warnings, therefore resulting in better code quality.
Disclaimer
If you are planning to use GCCE 4 in your SDK, note that no functionality or correctness can be guaranteed. This is a "best-effort" HOWTO, provided with absolutely no warranty.
According to CodeSourcery:
"You are attempting to do something that we do not recommend you do with the zero-cost downloads. The Lite toolchains are provided at zero cost without support."
If you want to be safe, make a backup before you start.
Necessary files
First, a new version of the CSL Arm Toolchain must be downloaded from CodeSourcery. The path for downloading is here. If the direct link does not work, you can find the download from the 'Downloads' menu.
CodeSourcery produces many different packages. The package you need is "Sourcery G++ Lite Edition for ARM", with Symbian OS as the target OS. The package should be available free of charge.
There is also an archive of previous releases available. The releases are usually made twice a year. It is often recommended to choose an older release.
Versions
All packages have been tested with SDK for Series 60 3rd Edition, Maintenance Release (Symbian 9.1)
Replacement of the toolchain
After downloading the desired package version from CodeSourcery, install it. By default, it will install itself into the following directory:
c:\program files\CodeSourcery\Sourcery G++ Lite\
Backup the old toolchain
Before you start altering the toolchain, make a backup of the old version (with 3.4.3). In the case of a problem, you will not then need to reinstall the SDK.
The version supplied with the SDKs usually resides in the following directory:
C:\Program Files\CSL Arm Toolchain\
Create another directory, for example C:\Program Files\CSL-Backup\ and copy the whole content of the old version directory to the backup directory. If anything goes wrong, you will simply need to delete the altered contents of the C:\Program Files\CSL Arm Toolchain\ directory, and copy the backup back to its original location.
Copy the new toolchain
Remove the entire old content of the C:\Program Files\CSL Arm Toolchain\ directory and copy the content of the new toolchain directory from c:\program files\CodeSourcery\Sourcery G++ Lite\ into the old directory.
Having done this, you have the new toolchain in place. However, you still need to patch your SDK in order to work with the new toolchain.
Patching the SDK
Edit the compilation scripts
As far as known, all SDKs require this patch.
There are some files in the epoc32\tools subdirectory of the SDK which refer to the exact version of the toolchain, and therefore need patching. Namely:
cl_bpabi.pm cl_gcce.pm ide_cw.pm compilation_config/gcce.mk
Backup the above files into a different directory (foe example, "original_343"). Then edit all the files, and removing the"3.4.3" strings and substituting them with the GCC version you are using in the new toolchain (for example, "4.2.3").
Backup the files into a different directory - do not make backups in the same directory. The script which processes them in the IDE can mistake the backups for the real files. See the discussion here for more information.
Remove failing assertions
This patch concerns SDK for Series 60 3rd Edition, Maintenance Release. Your SDK may vary, and possibly may not need this patch at all.
Find the file d32locd.h in your SDK's include directory. Make a backup of this file. Then find and comment out the following lines:
__ASSERT_COMPILE(_FOFF(TLocalDriveCaps,iSize)%8 == 0); __ASSERT_COMPILE(_FOFF(TLocalDriveCapsV3,iFormatInfo.iCapacity) % 8 == 0);
These lines would trigger a compilation error with GCC 4 ("not a constant", etc.)
Correct va_lists
This patch concerns SDK for Series 60 3rd Edition, Maintenance Release. Your SDK may vary, and possibly may not need this patch at all.The type
va_listand related types are used in processing functions that have variable argument length (for example,
printffunction from the LIBC library). After the installation of the new toolchain, you will have multiple definitions of the necessary types, which will confuse the compiler into printing out errors each time you use
va_listand related types. To avoid this, find the file
gcce\gcce.hin your SDK's include directory. Make a backup of this file. In this file, find and comment out the lines:
typedef struct __va_list { void *__ap; } va_list; #define va_start(ap, parmN) __builtin_va_start(ap.__ap, parmN) #define va_arg(ap, type) __builtin_va_arg(ap.__ap, type) #define va_end(ap) __builtin_va_end(ap.__ap) typedef va_list __e32_va_list;
and add the following lines just before the old va_start definition:
typedef __builtin_va_list va_list; #define va_start(v,l) __builtin_va_start(v,l) #define va_arg(v,l) __builtin_va_arg(v,l) #define va_end(v) __builtin_va_end(v)
Alternative solution:
Comment out the above lines and add the following, just before the first commented-out line:
#include <libc/stdarg.h>
Note that because of this patch, each of your projects now de facto include the stdarg.h header from the standard C library. To contradict this, you must have the following line in the MMP file(s) of the project(s) you will be compiling with the new toolchain:
SYSTEMINCLUDE \epoc32\include\libcsince the
stdarg.hfile includes more files from that directory.
The first solution is preferred because the __builtin_va_xxx functions take arguments alignment into account. In particular, the libc/stdarg.h implementation will cause trouble if varargs contain intermixed 8-byte (like double) and 4-byte values. Also, the first solution does not require any modifications in the MMP file(s).
Remove extra qualifiers from the SDK's header files
This patch concerns SDK for Series 60 3rd Edition, Maintenance Release. Your SDK may vary, and possibly may not need this patch at all.
If you are using AVKON Query Dialog, find aknquerydialog.h in your SDK's include directory. Make a backup of this file. Then, find the following line:
CCoeControl* CAknQueryDialog::FindControlOnAnyPageWithControlType(TInt aControlType, TInt* aLineIndex=0, TInt* aPageIndex=0) const;
and remove the qualifier
CAknQueryDialog::
so that the line looks as follows:
CCoeControl* FindControlOnAnyPageWithControlType(TInt aControlType, TInt* aLineIndex=0, TInt* aPageIndex=0) const;
This prevents a compiler error from "extra qualification", which did not result in error in GCC 3.4.3, but is detected by GCC version 4.
If you are using ImageConversion.h, you will meet a similar problem in it. The offending line is
IMPORT_C static CIclRecognizerUtil* CIclRecognizerUtil::NewL();
Again, remove the extra qualification.
You may find more "extra qualification" errors in Symbian SDK include files. It seems to be a common problem. Next such error is in mmf\mmfcontrollerpluginresolver.h, then coecntrl.h, etc. They must all be patched. It should not be more than 4-5 errors.
The same applies to your own code as well - no extra qualifiers are allowed.
Solve linker errors for libsupc++. Detected with GCCE 4.3.2 only
During the linking phase, the GCCE 4.3.2 toolchain will complain about missing reference to <pr>_Unwind_GetTextRelBase</PR>.
This is caused by missing binary code in
CSL Arm Toolchain\lib\libsupc++.a
This file has about 7 kB in the 4.3.2 distribution.
There are two working solutions to this problem.
A simple solution is to substitute this file with the original file from some older toolchain (even 3.4.3). That file has about 15 kB. Just rewrite the newer file with the older file.
A more complicated solution is to write the missing functions yourself and add them to the source code of your project. The functions can be empty. This solution has nmot been tested here.
This problem was not detected with GCCE 4.2.0 or 4.2.3. So the solution might be to use them instead of 4.3.2.
Supply your own integer division routines
Some of the SDKs for Series 60 3rd Edition suffer from the following problem: the system libraries do not contain definitions of compiler helper functions __aeabi_uidiv and __aeabi_idiv. This was an omission on the part of Symbian, and it is not limited to the SDKs - the devices are affected as well.
If you do not intend to ever use integral division in your project, you may skip this section entirely.
To find out whether your SDK contains this error, enter the following code snippet somewhere in your code (not in an unreachable part), and try compiling the code with GCCE DEBUG:
#include <e32debug.h>
...
void tryDivision ()
{
TInt a, b, c;
a = 10;
b = 5;
c = a / b;
RDebug::Printf("Result of division is %d",c);
}
Next, call the tryDivision() function somewhere in your code, for example in ConstructL method of the AppUi class.
(The Printf line is necessary here because otherwise the compiler may throw out the useless division code from your binary, and you will be unable to detect possible errors.)
The code will always be compilable in WINSCW. However, in GCCE DEBUG, the build process may abort with the message "missing reference to __aeabi_idiv"
In this case, you have an incorrect SDK, and you will have to perform the following fix.
Create a new file in your project's src/ directory, named division.c. Don't use the extension .cpp because this will not work! The content of the file will be as follows:
// This code was suggested by Julian Brown from CodeSourcery. It is in public domain.
// Many thanks!
#if __GCCE__
#if __SERIES60_30__
extern unsigned int __aeabi_uidivmod(unsigned numerator, unsigned denominator);
int __aeabi_idiv(int numerator, int denominator)
{
int neg_result = (numerator ^ denominator) & 0x80000000;
int result = __aeabi_uidivmod ((numerator < 0) ? -numerator : numerator, (denominator < 0) ? -denominator : denominator);
return neg_result ? -result : result;
}
unsigned __aeabi_uidiv(unsigned numerator, unsigned denominator)
{
return __aeabi_uidivmod (numerator, denominator);
}
#endif // __SERIES60_30__
#endif // __GCCE__
The file will be automatically added to your MMP file. Adjust the preprocessor blocks (#if __SERIES60_30__) as needed.
Project cleaning
Run the following commands from the command in your project's group directory:
bldmake clean bldmake bldfiles abld build gcce udeb
This should clear and recreate your GCCE makefiles (which contain paths to the old toolchain).
Known migration issues
Warnings
GCCE 4 issues more warnings than GCCE 3 and as a result, you will find many warnings in previously warning-free code. This is, however, a feature rather than a bug as most of the warnings are useful.
Static initialization fiascos
GCCE 4 is more sensitive to "static initialization fiasco" than GCCE 3.
"Static initialization fiasco" means a situation where you initialize one static variable (or constant) using another static variable (or constant). Since the compiler or system does not guarantee the correct sequence of static initializations, you will encounter errors if you try to initialize something with a yet-uninitialized value.
To read more about the "static initialization fiasco", visit the C++ Faq Lite.
If you have "static initialization fiascos" in your Symbian code, chances are that GCCE 3.4.3 will not trigger any errors, while GCCE 4.x will. That is because the compilers produce different code, and the static initialization sequence will therefore differ.
This is an extremely frustrating situation, as it results in premature exits of your application. However, static initialization fiasco is a serious problem and which will not go away if you return to older GCCE. Chances are that once, in the future, if you need to compile your code with newer toolchain, you will be forced to do the repairs anyway.
A typical example is a static const descriptor initialized from another static const descriptor, or a static const struct, which uses static const descriptors, or a static const struct that uses resource ids. Solution: in this case, the offending structure is often used only locally, in a single class or a single function. It is therefore worth considering moving it there from a global header file.
The way how to detect "static initialization fiascos" is to run on/device debugging with a GCCE DEBUG build and watch for errors. The panic stack will navigate you to the precise structure which caused the problem (click on the "static initialization and destruction" line of the stack).
Potential portability errors found when trying to build WebKit (4.4-172)
While not being full bug reports, I include these two issues here for future reference, as they took me (mgroeber9110 ) quite a long time to figure out. Perhaps this rings a bell with someone running into a similar issue in the future.
The first set of bugs I encountered after switching to GCC 4.4 came from lines such as this:
TPtrC data = iVisualClipText ? iVisualClipText->Des() : KNullDesC();
which resulted in 'data' being wrongly initialized. Assigning iVisualClipText->Des() to a temporary variable fixed the issue. The same code worked fine with GCC 3.3.
I also had one bug that turned out to be caused by a missing "const" in front of the line
static int p05[3] = { 5, 25, 125 };
in DLL code - so it seems that initialized static data can still be somewhat problematic.
02 Oct
2009
A good article, which represents a short overview about GCCE(GNU Compiler Collection for Embedded) 4 for Symbian OS, the fundamental requirments about it's installation, the information about it's versions and much more; which can be essential to know before using it.
It provides useful material for beginners to study before getting started.
Seems dont work with S60v5 SDK. Application crashed at startup with Kern-exec. But works on S60v3 SDKs.
CodeSourcery has a new fully-validated toolchain on their site: Symbian ADT Sourcery G++ Lite 4.4-172. It worked for me perfectly with S60v5 SDK
11:23, 28 May 2010 (UTC)
It turns out that with division code it's better to use #ifdef __GCCE__ (tested with Nokia Qt SDK 1.1 and Nokia E72). With just the #if,#if pair the application wouldn't even load.
Jarek | http://developer.nokia.com/community/wiki/How_to_use_GCCE_4_with_Symbian_SDKs | CC-MAIN-2014-49 | refinedweb | 2,446 | 56.66 |
# Details
How often do you get to 404 pages? Usually, they are not styled and stay default. Recently I’ve found [test.do.am](http://test.do.am/) which interactive character attracts attention and livens up the error page.
Probably, there was just a cat picture, then they thought up eyes movement and developer implemented the idea.Now user visits the page and checks out the effect. It’s cool and pleasant small feature, it catches, then user discusses it with colleagues or friends and even repeats the feature. It could be this easy, if not:
1. Center point is not being renewed when user resizes the window. Open the browser window with small width viewport and resize to full screen, the cat looks not at the cursor.
2. Center point is placed on the left eye, not in binocular center of the circle.
3. When user hovers cursor between the eyes, apples of the eyes don’t get together and don’t focus. Eyes are looking to infinity, that’s why the cat looks not at user, it looks through him.
4. Eyes movements are immediate, they need to be smooth.
5. Apples' movements happen because of margin-left / margin-top changing. It’s incorrect, find explanation below.
6. Eyes don’t move if cursor is under footer.
**What I suggest**
For a start, let’s implement flawless eyes movement.
1. Prepare markup
```
```
2. Get links to eyes’ elements
```
const cat = document.querySelector('.cat');
const eyes = cat.querySelectorAll('.cat__eye');
const eye_left = eyes[0];
const eye_right = eyes[1];
```
3. Register mousemove event listener and get cursor coordinates:
```
let mouseX;
let mouseY;
window.addEventListener('mousemove', e => {
mouseX = e.clientX;
mouseY = e.clientY;
})
```
I add mousemove listener on window object, not document body, because I need to use all screen to get mouse coordinates.
4. Movement
Since I’m going to smoothen movements, I can’t manage them in mousemove handler.
Add update method that will be fetched by requestAnimationFrame which is synchronized with browser renewal. Usually renewals happen 60 times per second, therefore we see 60 pics per second every 16.6 ms.
If developer supposes user’s browser can’t support requestAnimationFrame, developer can use setTimeout fallback or ready-made [polyfill](https://gist.github.com/paulirish/1579671)
```
window.requestAnimationFrame = (function () {
return window.requestAnimationFrame ||
window.webkitRequestAnimationFrame ||
window.mozRequestAnimationFrame ||
function (callback) {
window.setTimeout(callback, 1000 / 60);
};
})();
```
In order to renew or stable fetching of update in time, I register started variable
```
let started = false;
let mouseX;
let mouseY;
window.addEventListener('mousemove', e => {
mouseX = e.clientX;
mouseY = e.clientY;
if(!started){
started = true;
update();
}
})
function update(){
// Here comes eyes movement magic
requestAnimationFrame(update);
}
```
This way I got constantly fetching update method and cursor coordinates. Then I need to get values of apples movements inside the eyes.
I try to move both eyes as single element
```
let dx = mouseX - eyesCenterX;
let dy = mouseY - eyesCenterY;
let angle = Math.atan2(dy, dx);
let distance = Math.sqrt(dx * dx + dy * dy);
distance = distance > EYES_RADIUS ? EYES_RADIUS : distance;
let x = Math.cos(angle) * distance;
let y = Math.sin(angle) * distance;
eye_left.style.transform = 'translate(' + x + 'px,' + y + 'px)';
eye_right.style.transform = 'translate(' + x + 'px,' + y + 'px)';
```
Pretty simple: find dx and dy, which are coordinate difference between eyes center and mouse, find angle from center to cursor, using Math.cos and Math.sin methods get movement value for horizontal and vertical. Use [ternary operator](https://en.wikipedia.org/wiki/%3F) and limit eyes movement area.
Y value is given first for Math.atan2 method, then x value. As a result user notices unnaturalness of eyes motions and no focusing.
Make each eyes move and watch without reference to each other.
```
// left eye
let left_dx = mouseX - eyesCenterX + 48;
let left_dy = mouseY - eyesCenterY;
let left_angle = Math.atan2(left_dy, left_dx);
let left_distance = Math.sqrt(left_dx * left_dx + left_dy * left_dy);
left_distance = left_distance > EYES_RADIUS ? EYES_RADIUS : left_distance;
let left_x = Math.cos(left_angle) * left_distance;
let left_y = Math.sin(left_angle) * left_distance;
eye_left.style.transform = 'translate(' + left_x + 'px,' + left_y + 'px)';
// right eye
let right_dx = mouseX - eyesCenterX - 48;
let right_dy = mouseY - eyesCenterY;
let right_angle = Math.atan2(right_dy, right_dx);
let right_distance = Math.sqrt(right_dx * right_dx + right_dy * right_dy);
right_distance = right_distance > EYES_RADIUS ? EYES_RADIUS : right_distance;
let right_x = Math.cos(right_angle) * right_distance;
let right_y = Math.sin(right_angle) * right_distance;
eye_right.style.transform = 'translate(' + right_x + 'px,' + right_y + 'px)';
```
Interesting but worse than previous result, eyes move up and down independently. So I used first demo as a movement mechanic basic and make apples of the eyes get together when cursor is about the center of character.
I’ll not describe entire code, please find hereby a result:
By trial and error I’ve matched needed parameters for eyes movement and focusing. So now I need smoothing.
**Smoothing**
Link [TweenMax library](https://greensock.com) and code something like this?
```
TweenMax.to( eye, 0.15, {x: x, y: y});
```
Linking entire lib for simple task does not make sense, therefore, I make smoothing from scratch.
Put the case that there is only one eye element on the page and its displacement area is not limited at all. To smoothen mouse coordinates values, I use this mechanics:
```
const SMOOTHING = 10;
x += (needX - x) / SMOOTHING;
y += (needY - y) / SMOOTHING;
eye.style.transform = 'translate3d(' + x + 'px,' + y + 'px,0)';
```
I use translate3d to separate eyes to another rendering stream and speed them up.
The trick is that every 16.6ms (60 pics per second) variable x and y tend to needed values. Each renew closes value to its needed one for 1/10 of difference.
```
let x = 0;
let needX = 100;
let SMOOTHING = 2;
function update(){
x += (needX - x) / SMOOTHING;
console.log(x);
}
```
Then every 16.6 ms renew we get simple smoothing and next x values (approx):
```
50
75
87.5
93.75
96.875
98.4375
99.21875
99.609375
100
```
A couple more unobvious tricks:
— Start this examination to optimize workload
```
if(x != needX || y != needY){
eye.style.transform = 'translate3d(' + x + 'px,' + y + 'px,0)';
}
```
But you have to equate x to needX when they get as close as eyes positions are almost the same
```
if(Math.abs(x - needX) < 0.25){
x = needX;
}
if(Math.abs(y - needY) < 0.25){
y = needY;
}
```
Otherwise x and y values will be reaching needX and needY too long; there will be no visual differences, but every screen change will affect eyes styles. Btw you can fiddle around with it yourself.
```
let x = 0;
let needX = 100;
let smoothing = 2;
function update(){
x += (needX - x) / smoothing;
if( Math.abs(x - needX) > 0.25 ){ // replace 0.25 with anything else and check number of x renewals.
window.requestAnimationFrame(update);
} else {
x = needX;
}
console.log( x.toString(10) );
}
update();
```
— If mechanics above is clear, you can create more complex effects, e.g. spring. The simplest smoothing and to cursor approximation looks like this:
```
x += (mouseX - x) / smoothing;
y += (mouseY - y) / smoothing;
```
Add smoothing a difference between needed and current coordinates values.
Sometimes approximation limitation makes sense. There is example above where value changes from 0 to 100, so in the 1st iteration value reaches “50”, it is pretty huge figure for 1 step. This mechanics kinda remind [paradox of Achilles and the tortoise](https://en.wikipedia.org/wiki/Zeno%27s_paradoxes#Achilles_and_the_tortoise)
**Winking**
Hide and show apples of eyes every 2-3 seconds. The most trivial method is «display: none;», «transform: scaleY(N)» with dynamic value of y-scale is a bit more complex.
Create 2 consts
const BLINK\_COUNTER\_LIMIT = 180; — number of renewals before start of blinking,
const BLINKED\_COUNTER\_LIMIT = 6; — number of renewals during one wink.
And 2 variables, which values will change every renewal.
```
let blinkCounter = 0;
let blinkedCounter = 0;
```
Code of winking
```
let blinkTransform = '';
blinkCounter++;
if(blinkCounter > BLINK_COUNTER_LIMIT){
blinkedCounter++
if(blinkedCounter > BLINKED_COUNTER_LIMIT){
blinkCounter = 0;
} else {
blinkTransform = ' scaleY(' + (blinkedCounter / BLINKED_COUNTER_LIMIT) + ')';
}
} else {
blinkedCounter = 0;
}
```
BlinkTransform is stroke variable that has empty value between winking and following ones during winking
```
' scaleY(0.17)'
' scaleY(0.33)'
' scaleY(0.50)'
' scaleY(0.67)'
' scaleY(0.83)'
' scaleY(1.00)'
```
All calculations give variable blinkTransform, value of which should be added to css code of eyes position transform. Thus empty string gets added in case of 3s down time and it doesn’t effect on eyes scale, css value gets added during blinking.
```
eye_left.style.transform = 'translate(' + xLeft + 'px,' + y + 'px)' + blinkTransform;
eye_right.style.transform = 'translate(' + xRight + 'px,' + y + 'px)' + blinkTransform;
```
**Lesson of the story**
Every day we meet things that seem simple and obvious and we even don't understand that this external simplicity hides a colossal amount of questions and improvements. In my opinion devil is in the details that form entire final result. Muhammad Ali the best boxer of 20th century raised heel of rear foot in the moment of straight punch. This manoeuvre increased effective distance of blow and gave him more chances to win. It always worked.
*P.S. I have no bearing on the website and hope its owners would not take offence at my comments. For convenience I named apple of the eye = eye in code.* | https://habr.com/ru/post/443720/ | null | null | 1,511 | 59.7 |
MAC_APPLICATION_MENU Services Hide %1 Hide Others Show All Preferences... Quit %1 About %1 AudioOutput <html>The audio playback device <b>%1</b> does not work.<br/>Falling back to <b>%2</b>.</html> <html>Switching to the audio playback device <b>%1</b><br/>which just became available and has higher preference.</html> Revert back to device '%1' Phonon:: Notifications Music Video Communication Games Accessibility Phonon::Gstreamer::Backend Warning: You do not seem to have the package gstreamer0.10-plugins-good installed. Some video features have been disabled. Warning: You do not seem to have the base GStreamer plugins installed. All audio and video support has been disabled Phonon::Gstreamer::MediaObject Cannot start playback. Check your Gstreamer installation and make sure you have libgstreamer-plugins-base installed. A required codec is missing. You need to install the following codec(s) to play this content: %0 Could not open media source. Invalid source type. Could not locate media source. Could not open audio device. The device is already in use. Could not decode media source. Phonon::VolumeSlider Volume: %1% Use this slider to adjust the volume. The leftmost position is 0%, the rightmost is %1% Q3Accel %1, %2 not defined Ambiguous %1 not handled Q3DataTable True False Insert Update Delete Q3FileDialog Copy or Move a File Read: %1 Write: %1 Cancel All Files (*) Name Nume Size Type Date Attributes &OK Look &in: File &name: File &type: Back One directory up Create New Folder List View Detail View Preview File Info Preview File Contents Read-write Read-only Write-only Inaccessible Symlink to File Symlink to Directory Symlink to Special File Dir Special Open Save As &Open &Save &Rename &Delete R&eload Sort by &Name Sort by &Size Sort by &Date &Unsorted Sort Show &hidden files the file the directory the symlink Delete %1 <qt>Are you sure you wish to delete %1 "%2"?</qt> &Yes &No New Folder 1 New Folder New Folder %1 Find Directory Directories Directory: Error %1 File not found. Check path and filename. Q3LocalFs Could not read directory %1 Could not create directory %1 Could not remove file or directory %1 Could not rename %1 to %2 Could not open %1 Could not write %1 Q3MainWindow Line up Customize... Q3NetworkProtocol Operation stopped by the user Q3ProgressDialog Cancel Q3TabDialog OK Apply Help Defaults Cancel Q3TextEdit &Undo &Redo Cu&t &Copy &Paste Clear Select All Q3TitleBar System Restore up Minimize Restore down Maximize Close Contains commands to manipulate the window Puts a minimized back to normal Moves the window out of the way Puts a maximized window back to normal Makes the window full screen Closes the window Displays the name of the window and contains controls to manipulate it Q3ToolBar More... Q3UrlOperator The protocol `%1' is not supported The protocol `%1' does not support listing directories The protocol `%1' does not support creating new directories The protocol `%1' does not support removing files or directories The protocol `%1' does not support renaming files or directories The protocol `%1' does not support getting files The protocol `%1' does not support putting files The protocol `%1' does not support copying or moving files or directories (unknown) Q3Wizard &Cancel < &Back &Next > &Finish &Help QAbstractSocket Host not found Connection refused Socket operation timed out Socket is not connected QAbstractSpinBox &Step up Step &down &Select All QApplication Activate Executable '%1' requires Qt %2, found Qt %3. Incompatible Qt Library Error QT_LAYOUT_DIRECTION Translate this string to the string 'LTR' in left-to-right languages or to 'RTL' in right-to-left languages (such as Hebrew and Arabic) to get proper widget layout. Activates the program's main window QCheckBox Uncheck Check Toggle QColorDialog Hu&e: &Sat: &Val: &Red: &Green: Bl&ue: A&lpha channel: &Basic colors &Custom colors &Add to Custom Colors Select color QComboBox Open False True Close QCoreApplication %1: permission denied QSystemSemaphore %1: already exists QSystemSemaphore %1: doesn't exists QSystemSemaphore %1: out of resources QSystemSemaphore %1: unknown error %2 QSystemSemaphore %1: key is empty QSystemSemaphore %1: unable to make key QSystemSemaphore %1: ftok failed QSystemSemaphore QDB2Driver Unable to connect Unable to commit transaction Unable to rollback transaction Unable to set autocommit QDB2Result Unable to execute statement Unable to prepare statement Unable to bind variable Unable to fetch record %1 Unable to fetch next Unable to fetch first QDateTimeEdit AM am PM pm QDial QDial SpeedoMeter SliderHandle QDialog What's This? Done QDialogButtonBox OK Save Open Cancel Close Apply Reset Help Don't Save Discard &Yes Yes to &All &No N&o to All Save All Abort Retry Ignore Restore Defaults Close without Saving &OK QDirModel Name Nume Size Kind Match OS X Finder Type All other platforms Date Modified QDockWidget Close Dock Float QDoubleSpinBox More Less QErrorMessage Debug Message: Warning: Fatal Error: &Show this message again &OK QFileDialog All Files (*) Directories &Open &Save Open %1 already exists. Do you want to replace it? %1 File not found. Please verify the correct file name was given. My Computer &Rename &Delete Show &hidden files Back Parent Directory List View Detail View Files of type: Directory: %1 Directory not found. Please verify the correct directory name was given. '%1' is write protected. Do you want to delete it anyway? Are sure you want to delete '%1'? Could not delete directory. Save As Drive File Unknown Find Directory Show Forward New Folder &New Folder &Choose Remove File &name: Look in: Create New Folder QFileSystemModel Invalid filename <b>The name "%1" can not be used.</b><p>Try using another name, with fewer characters or no punctuations marks. Name Nume Size Kind Match OS X Finder Type All other platforms Date Modified My Computer Computer %1 TB %1 GB %1 MB %1 KB %1 bytes QFontDatabase Normal Bold Demi Bold Black Demi Light Italic Oblique Any Latin Greek Cyrillic Armenian Georgian Khmer Simplified Chinese Traditional Chinese Japanese Korean Vietnamese Symbol Ogham Runic QFontDialog &Font Font st&yle &Size Effects Stri&keout &Underline Sample Wr&iting System Select Font QFtp Not connected Host %1 not found Connection refused to host %1 Connected to host %1 Connection refused for data connection Unknown error Connecting to host failed: %1 Login failed: %1 Listing directory failed: %1 Changing directory failed: %1 Downloading file failed: %1 Uploading file failed: %1 Removing file failed: %1 Creating directory failed: %1 Removing directory failed: %1 Connection closed Host %1 found Connection to %1 closed Host found Connected to host QHostInfo Unknown error QHostInfoAgent Host not found Unknown address type Unknown error QHttp Unknown error Request aborted No server set to connect to Wrong content length Server closed connection unexpectedly Connection refused Host %1 not found HTTP request failed Invalid HTTP response header Invalid HTTP chunked body Host %1 found Connected to host %1 Connection to %1 closed Host found Connected to host Connection closed Proxy authentication required Authentication required HTTPS connection requested but SSL support not compiled in Connection refused (or timed out) Proxy requires authentication Host requires authentication Data corrupted Unknown protocol specified SSL handshake failed QHttpSocketEngine Authentication required QIBaseDriver Error opening database Could not start transaction Unable to commit transaction Unable to rollback transaction QIBaseResult Unable to create BLOB Unable to write BLOB Unable to open BLOB Unable to read BLOB Could not find array Could not get array data Could not get query info Could not start transaction Unable to commit transaction Could not allocate statement Could not prepare statement Could not describe input statement Could not describe statement Unable to close statement Unable to execute query Could not fetch next item Could not get statement info QIODevice Permission denied Too many open files No such file or directory No space left on device Unknown error QInputContext XIM XIM input method Windows input method Mac OS X input method QLibrary QLibrary::load_sys: Cannot load %1 (%2) QLibrary::unload_sys: Cannot unload %1 (%2) QLibrary::resolve_sys: Symbol "%1" undefined in %2 (%3) Could not mmap '%1': %2 Plugin verification data mismatch in '%1' Could not unmap '%1': %2 The plugin '%1' uses incompatible Qt library. (%2.%3.%4) [%5] The plugin '%1' uses incompatible Qt library. Expected build key "%2", got "%3" Unknown error The shared library was not found. The file '%1' is not a valid Qt plugin. The plugin '%1' uses incompatible Qt library. (Cannot mix debug and release libraries.) QLineEdit &Undo &Redo Cu&t &Copy &Paste Delete Select All QLocalServer %1: Name error %1: Permission denied %1: Address in use %1: Unknown error %2 QLocalSocket %1: Connection refused %1: Remote closed %1: Invalid name %1: Socket access error %1: Socket resource error %1: Socket operation timed out %1: Datagram too large %1: Connection error %1: The socket operation is not supported %1: Unknown error %2 QMYSQLDriver Unable to open database ' Unable to connect Unable to begin transaction Unable to commit transaction Unable to rollback transaction QMYSQLResult Unable to fetch data Unable to execute query Unable to store result Unable to prepare statement Unable to reset statement Unable to bind value Unable to execute statement Unable to bind outvalues Unable to store statement results Unable to execute next query Unable to store next result QMdiArea (Untitled) QMdiSubWindow %1 - [%2] Close Minimize Restore Down &Restore &Move &Size Mi&nimize Ma&ximize Stay on &Top &Close - [%1] Maximize Unshade Shade Restore Help Menu QMenu Close Open Execute QMessageBox Help OK About Qt <p>This program uses Qt version %1.</p> Show Details... Hide Details... <p>This program uses Qt Open Source Edition version %1.</p><p>Qt Open Source Edition is intended for the development of Open Source applications. You need a commercial Qt license for development of proprietary (closed source) applications.</p><p>Please see <a href=""></a> for an overview of Qt licensing.</p> <h3>About Qt</h3>%1<p>Qt is a C++ toolkit for cross-platform application development.</p><p>Qt provides single-source portability across MS Windows, Mac OS X, Linux, and all major commercial Unix variants. Qt is also available for embedded devices as Qt for Embedded Linux and Qt for Windows CE.</p><p>Qt is a Nokia product. See <a href=""></a> for more information.</p> QMultiInputContext Select IM QMultiInputContextPlugin Multiple input method switcher Multiple input method switcher that uses the context menu of the text widgets QNativeSocketEngine The remote host closed the connection Network operation timed out Out of resources Unsupported socket operation Protocol type not supported Invalid socket descriptor Network unreachable Permission denied Connection timed out Connection refused The bound address is already in use The address is not available The address is protected Unable to send a message Unable to receive a message Unable to write Network error Another socket is already listening on the same port Unable to initialize non-blocking socket Unable to initialize broadcast socket Attempt to use IPv6 socket on a platform with no IPv6 support Host unreachable Datagram was too large to send Operation on non-socket Unknown error The proxy type is invalid for this operation QNetworkAccessFileBackend Request for opening non-local file %1 Error opening %1: %2 Write error writing to %1: %2 Cannot open %1: Path is a directory Read error reading from %1: %2 QNetworkAccessFtpBackend Cannot open %1: is a directory Logging in to %1 failed: authentication required Error while downloading %1: %2 Error while uploading %1: %2 QNetworkReply Error downloading %1 - server replied: %2 Protocol "%1" is unknown QNetworkReplyImpl Operation canceled QOCIDriver Unable to logon Unable to initialize QOCIDriver Unable to begin transaction Unable to commit transaction Unable to rollback transaction QOCIResult Unable to bind column for batch execute Unable to execute batch statement Unable to goto next Unable to alloc statement Unable to prepare statement Unable to bind value Unable to execute select statement Unable to execute statement QODBCDriver Unable to connect Unable to connect - Driver doesn't support all needed functionality Unable to disable autocommit Unable to commit transaction Unable to rollback transaction Unable to enable autocommit QODBCResult QODBCResult::reset: Unable to set 'SQL_CURSOR_STATIC' as statement attribute. Please check your ODBC driver configuration Unable to execute statement Unable to fetch next Unable to prepare statement Unable to bind variable Unable to fetch last Unable to fetch Unable to fetch first Unable to fetch previous QObject Operation not supported on %1 Invalid URI: %1 Write error writing to %1: %2 Read error reading from %1: %2 Socket error on %1: %2 Remote host closed the connection prematurely on %1 Protocol error: packet of size 0 received QPPDOptionsModel Name Nume Value Valoare QPSQLDriver Unable to connect Could not begin transaction Could not commit transaction Could not rollback transaction Unable to subscribe Unable to unsubscribe QPSQLResult Unable to create query Unable to prepare statement QPageSetupWidget Centimeters (cm) Millimeters (mm) Inches (in) Points (pt) Form Paper Page size: Width: Height: Paper source: Orientation Portrait Landscape Reverse landscape Reverse portrait Margins top margin left margin right margin bottom margin QPluginLoader Unknown error The plugin was not loaded. QPrintDialog locally connected Aliases: %1 unknown Print To File ... File %1 is not writable. Please choose a different file name. %1 already exists. Do you want to overwrite it? %1 is a directory. Please choose a different file name. A0 A1 A2 A3 A4 A5 A6 A7 A8 A9 B0 B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 C5E DLE Executive Folio Ledger Legal Letter Tabloid US Common #10 Envelope Custom &Options >> &Print &Options << Print to File (PDF) Print to File (Postscript) Local file Write %1 file QPrintPreviewDialog Page Setup Print Preview Next page Previous page First page Last page Fit width Fit page Zoom in Zoom out Portrait Landscape Show single page Show facing pages Show overview of all pages Print Page setup Close QPrintPropertiesWidget Form Page Advanced QPrintSettingsOutput Form Copies Print range Print all Pages from to Selection Output Settings Copies: Collate Reverse Options Color Mode Color Grayscale Duplex Printing None Long side Short side QPrintWidget Form Printer &Name: P&roperties Location: Preview Type: Output &file: ... QProgressDialog Cancel QPushButton Open QRadioButton Check QRegExp no error occurred disabled feature used bad char class syntax bad lookahead syntax bad repetition syntax invalid octal value missing left delim unexpected end met internal limit QSQLite2Driver Error to open database Unable to begin transaction Unable to commit transaction Unable to rollback Transaction QSQLite2Result Unable to fetch results Unable to execute statement QSQLiteDriver Error opening database Error closing database Unable to begin transaction Unable to commit transaction Unable to rollback transaction QSQLiteResult Unable to fetch row Unable to execute statement Unable to reset statement Unable to bind parameters Parameter count mismatch No query QScrollBar Scroll here Left edge Top Right edge Bottom Page left Page up Page right Page down Scroll left Scroll up Scroll right Scroll down Line up Position Line down QSharedMemory %1: unable to set key on lock %1: create size is less then 0 %1: unable to lock %1: unable to unlock %1: permission denied %1: already exists %1: doesn't exists %1: out of resources %1: unknown error %2 %1: key is empty %1: unix key file doesn't exists %1: ftok failed %1: unable to make key %1: system-imposed size restrictions %1: not attached QShortcut Space Esc Tab Backtab Backspace Return Enter Ins Del Pause Print SysReq Home End Left Up Right Down PgUp PgDown CapsLock NumLock ScrollLock Menu Help Back Forward Stop Refresh Volume Down Volume Mute Volume Up Bass Boost Bass Up Bass Down Treble Up Treble Down Media Play Media Stop Media Previous Media Next Media Record Favorites Search Standby Open URL Launch Mail Launch Media Launch (0) Launch (1) Launch (2) Launch (3) Launch (4) Launch (5) Launch (6) Launch (7) Launch (8) Launch (9) Launch (A) Launch (B) Launch (C) Launch (D) Launch (E) Launch (F) Print Screen Page Up Page Down Caps Lock Num Lock Number Lock Scroll Lock Insert Delete Escape System Request Select Yes No Context1 Context2 Context3 Context4 Call Hangup Flip Ctrl Shift Alt Meta + F%1 Home Page QSlider Page left Page up Position Page right Page down QSocks5SocketEngine Socks5 timeout error connecting to socks server Network operation timed out QSpinBox More Less QSql Delete Delete this record? Yes No Insert Update Save edits? Cancel Confirm Cancel your edits? QSslSocket Unable to write data: %1 Error while reading: %1 Error during SSL handshake: %1 Error creating SSL context (%1) Invalid or empty cipher list (%1) Error creating SSL session, %1 Error creating SSL session: %1 Cannot provide a certificate with no key, %1 Error loading local certificate, %1 Error loading private key, %1 Private key does not certificate public key, %1 QTDSDriver Unable to open connection Unable to use database QTabBar Scroll Left Scroll Right QTextControl &Undo &Redo Cu&t &Copy Copy &Link Location &Paste Delete Select All QToolButton Press Open QUdpSocket This platform does not support IPv6 QUndoGroup Undo Redo QUndoModel <empty> QUndoStack Undo Redo QUnicodeControlCharacterMenu LRM Left-to-right mark RLM Right-to-left mark ZWJ Zero width joiner ZWNJ Zero width non-joiner ZWSP Zero width space LRE Start of left-to-right embedding RLE Start of right-to-left embedding LRO Start of left-to-right override RLO Start of right-to-left override PDF Pop directional formatting Insert Unicode control character QWebFrame Request cancelled Request blocked Cannot show URL Frame load interruped by policy change Cannot show mimetype File does not exist QWebPage Bad HTTP request Submit default label for Submit buttons in forms on web pages Submit Submit (input element) alt text for <input> elements with no alt, title, or value Reset default label for Reset buttons in forms on web pages This is a searchable index. Enter search keywords: text that appears at the start of nearly-obsolete web pages in the form of a 'searchable index' Choose File title for file button used in HTML forms No file selected text to display in file button used in HTML forms when no file is selected Open in New Window Open in New Window context menu item Save Link... Download Linked File context menu item Copy Link Copy Link context menu item Open Image Open Image in New Window context menu item Save Image Download Image context menu item Copy Image Copy Link context menu item Open Frame Open Frame in New Window context menu item Copy Copy context menu item Go Back Back context menu item Go Forward Forward context menu item Stop Stop context menu item Reload Reload context menu item Cut Cut context menu item Paste Paste context menu item No Guesses Found No Guesses Found context menu item Ignore Ignore Spelling context menu item Add To Dictionary Learn Spelling context menu item Search The Web Search The Web context menu item Look Up In Dictionary Look Up in Dictionary context menu item Open Link Open Link context menu item Ignore Ignore Grammar context menu item Spelling Spelling and Grammar context sub-menu item Show Spelling and Grammar menu item title Hide Spelling and Grammar menu item title Check Spelling Check spelling context menu item Check Spelling While Typing Check spelling while typing context menu item Check Grammar With Spelling Check grammar with spelling context menu item Fonts Font context sub-menu item Bold Bold context menu item Italic Italic context menu item Underline Underline context menu item Outline Outline context menu item Direction Writing direction context sub-menu item Default Default writing direction context menu item LTR Left to Right context menu item RTL Right to Left context menu item Inspect Inspect Element context menu item No recent searches Label for only item in menu that appears when clicking on the search field image, when no searches have been performed Recent searches label for first item in the menu that appears when clicking on the search field image, used as embedded menu title Clear recent searches menu item in Recent Searches menu that empties menu's contents Unknown Unknown filesize FTP directory listing item %1 (%2x%3 pixels) Title string for images Web Inspector - %2 QWhatsThisAction What's This? QWidget * QWizard Go Back Continue Commit Done Quit Help < &Back &Finish Cancel &Help &Next &Next > QWorkspace &Restore &Move &Size Mi&nimize Ma&ximize &Close Stay on &Top Sh&ade %1 - [%2] Minimize Restore Down Close &Unshade QXml no error occurred error triggered by consumer unexpected end of file more than one document type definition error occurred while parsing element tag mismatch error occurred while parsing content unexpected character invalid name for processing instruction version expected while reading the XML declaration wrong value for standalone declaration encoding declaration or standalone declaration expected while reading the XML declaration standalone declaration expected while reading the XML declaration error occurred while parsing document type definition letter is expected error occurred while parsing comment error occurred while parsing reference internal general entity reference not allowed in DTD external parsed general entity reference not allowed in attribute value external parsed general entity reference not allowed in DTD unparsed entity reference in wrong context recursive entities error in the text declaration of an external entity QXmlStream Extra content at end of document. Invalid entity value. Invalid XML character. Sequence ']]>' not allowed in content. Namespace prefix '%1' not declared Attribute redefined. Unexpected character '%1' in public id literal. Invalid XML version string. Unsupported XML version. %1 is an invalid encoding name. Encoding %1 is unsupported Standalone accepts only yes or no. Invalid attribute in XML declaration. Premature end of document. Invalid document. Expected , but got ' Unexpected ' Expected character data. Recursive entity detected. Start tag expected. XML declaration not at start of document. NDATA in parameter entity declaration. %1 is an invalid processing instruction name. Invalid processing instruction name. Illegal namespace declaration. Invalid XML name. Opening and ending tag mismatch. Reference to unparsed entity '%1'. Entity '%1' not declared. Reference to external entity '%1' in attribute value. Invalid character reference. Encountered incorrectly encoded content. The standalone pseudo attribute must appear after the encoding. %1 is an invalid PUBLIC identifier. QtXmlPatterns An %1-attribute with value %2 has already been declared. An %1-attribute must have a valid %2 as value, which %3 isn't. Network timeout. Element %1 can't be serialized because it appears outside the document element. Attribute %1 can't be serialized because it appears at the top level. Year %1 is invalid because it begins with %2. Day %1 is outside the range %2..%3. Month %1 is outside the range %2..%3. Overflow: Can't represent date %1. Day %1 is invalid for month %2. Time 24:%1:%2.%3 is invalid. Hour is 24, but minutes, seconds, and milliseconds are not all 0; Time %1:%2:%3.%4 is invalid. Overflow: Date can't be represented. At least one component must be present. At least one time component must appear after the %1-delimiter. No operand in an integer division, %1, can be %2. The first operand in an integer division, %1, cannot be infinity (%2). The second operand in a division, %1, cannot be zero (%2). %1 is not a valid value of type %2. When casting to %1 from %2, the source value cannot be %3. Integer division (%1) by zero (%2) is undefined. Division (%1) by zero (%2) is undefined. Modulus division (%1) by zero (%2) is undefined. Dividing a value of type %1 by %2 (not-a-number) is not allowed. Dividing a value of type %1 by %2 or %3 (plus or minus zero) is not allowed. Multiplication of a value of type %1 by %2 or %3 (plus or minus infinity) is not allowed. A value of type %1 cannot have an Effective Boolean Value. Effective Boolean Value cannot be calculated for a sequence containing two or more atomic values. Value %1 of type %2 exceeds maximum (%3). Value %1 of type %2 is below minimum (%3). A value of type %1 must contain an even number of digits. The value %2 does not. %1 is not valid as a value of type %2. Operator %1 cannot be used on type %2. Operator %1 cannot be used on atomic values of type %2 and %3. The namespace URI in the name for a computed attribute cannot be %1. The name for a computed attribute cannot have the namespace URI %1 with the local name %2. Type error in cast, expected %1, received %2. When casting to %1 or types derived from it, the source value must be of the same type, or it must be a string literal. Type %2 is not allowed. No casting is possible with %1 as the target type. It is not possible to cast from %1 to %2. Casting to %1 is not possible because it is an abstract type, and can therefore never be instantiated. It's not possible to cast the value %1 of type %2 to %3 Failure when casting from %1 to %2: %3 A comment cannot contain %1 A comment cannot end with a %1. No comparisons can be done involving the type %1. Operator %1 is not available between atomic values of type %2 and %3. An attribute node cannot be a child of a document node. Therefore, the attribute %1 is out of place. A library module cannot be evaluated directly. It must be imported from a main module. A value of type %1 cannot be a predicate. A predicate must have either a numeric type or an Effective Boolean Value type. A positional predicate must evaluate to a single numeric value. The target name in a processing instruction cannot be %1 in any combination of upper and lower case. Therefore, is %2 invalid. %1 is not a valid target name in a processing instruction. It must be a %2 value, e.g. %3. The last step in a path must contain either nodes or atomic values. It cannot be a mixture between the two. The data of a processing instruction cannot contain the string %1 No namespace binding exists for the prefix %1 No namespace binding exists for the prefix %1 in %2 %1 is an invalid %2 %1 takes at most %n argument(s). %2 is therefore invalid. %1 requires at least %n argument(s). %2 is therefore invalid. The first argument to %1 cannot be of type %2. It must be a numeric type, xs:yearMonthDuration or xs:dayTimeDuration. The first argument to %1 cannot be of type %2. It must be of type %3, %4, or %5. The second argument to %1 cannot be of type %2. It must be of type %3, %4, or %5. %1 is not a valid XML 1.0 character. The first argument to %1 cannot be of type %2. If both values have zone offsets, they must have the same zone offset. %1 and %2 are not the same. %1 was called. %1 must be followed by %2 or %3, not at the end of the replacement string. In the replacement string, %1 must be followed by at least one digit when not escaped. In the replacement string, %1 can only be used to escape itself or %2, not %3 %1 matches newline characters %1 and %2 match the start and end of a line. Matches are case insensitive Whitespace characters are removed, except when they appear in character classes %1 is an invalid regular expression pattern: %2 %1 is an invalid flag for regular expressions. Valid flags are: If the first argument is the empty sequence or a zero-length string (no namespace), a prefix cannot be specified. Prefix %1 was specified. It will not be possible to retrieve %1. The root node of the second argument to function %1 must be a document node. %2 is not a document node. The default collection is undefined %1 cannot be retrieved The normalization form %1 is unsupported. The supported forms are %2, %3, %4, and %5, and none, i.e. the empty string (no normalization). A zone offset must be in the range %1..%2 inclusive. %3 is out of range. %1 is not a whole number of minutes. Required cardinality is %1; got cardinality %2. The item %1 did not match the required type %2. %1 is an unknown schema type. Only one %1 declaration can occur in the query prolog. The initialization of variable %1 depends on itself No variable by name %1 exists The variable %1 is unused Version %1 is not supported. The supported XQuery version is 1.0. The encoding %1 is invalid. It must contain Latin characters only, must not contain whitespace, and must match the regular expression %2. No function with signature %1 is available A default namespace declaration must occur before function, variable, and option declarations. Namespace declarations must occur before function, variable, and option declarations. Module imports must occur before function, variable, and option declarations. It is not possible to redeclare prefix %1. Only the prefix %1 can be declared to bind the namespace %2. By default, it is already bound to the prefix %1. Prefix %1 is already declared in the prolog. The name of an option must have a prefix. There is no default namespace for options. The Schema Import feature is not supported, and therefore %1 declarations cannot occur. The target namespace of a %1 cannot be empty. The module import feature is not supported A variable by name %1 has already been declared in the prolog. No value is available for the external variable by name %1. The namespace for a user defined function cannot be empty (try the predefined prefix %1 which exists for cases like this) The namespace %1 is reserved; therefore user defined functions may not use it. Try the predefined prefix %2, which exists for these cases. The namespace of a user defined function in a library module must be equivalent to the module namespace. In other words, it should be %1 instead of %2 A function already exists with the signature %1. No external functions are supported. All supported functions can be used directly, without first declaring them as external An argument by name %1 has already been declared. Every argument name must be unique. The name of a variable bound in a for-expression must be different from the positional variable. Hence, the two variables named %1 collide. The Schema Validation Feature is not supported. Hence, %1-expressions may not be used. None of the pragma expressions are supported. Therefore, a fallback expression must be present The %1-axis is unsupported in XQuery %1 is not a valid numeric literal. No function by name %1 is available. The namespace URI cannot be the empty string when binding to a prefix, %1. %1 is an invalid namespace URI. It is not possible to bind to the prefix %1 Namespace %1 can only be bound to %2 (and it is, in either case, pre-declared). Prefix %1 can only be bound to %2 (and it is, in either case, pre-declared). Two namespace declaration attributes have the same name: %1. The namespace URI must be a constant and cannot use enclosed expressions. An attribute by name %1 has already appeared on this element. A direct element constructor is not well-formed. %1 is ended with %2. The name %1 does not refer to any schema type. %1 is an complex type. Casting to complex types is not possible. However, casting to atomic types such as %2 works. %1 is not an atomic type. Casting is only possible to atomic types. %1 is not a valid name for a processing-instruction. Therefore this name test will never match. %1 is not in the in-scope attribute declarations. Note that the schema import feature is not supported. The name of an extension expression must be in a namespace. empty zero or one exactly one one or more zero or more Required type is %1, but %2 was found. Promoting %1 to %2 may cause loss of precision. The focus is undefined. It's not possible to add attributes after any other kind of node. An attribute by name %1 has already been created. Only the Unicode Codepoint Collation is supported(%1). %2 is unsupported. VolumeSlider Muted Volume: %1% WebCore::PlatformScrollbar Scroll here Left edge Top Right edge Bottom Page left Page up Page right Page down Scroll left Scroll up Scroll right Scroll down | https://www.virtualbox.org/download/testcase/nls/4.2/qt_ro.ts | CC-MAIN-2015-27 | refinedweb | 5,360 | 50.46 |
Last week we looked at creating our first Ruby on Rails Model.
Rails Models are backed by a database table. So for example, if you had an
Article Model you would also need an
articles database table.
In order to control our database schema and to make deployment and changes easier we can use migration files.
A migration file is a set of instructions that will be run against your database.
This means if you need to make a change to your database, you can capture that change within a migration file so it can be repeated by other installations of your project.
This will be extremely useful when it comes to deploying your application or working with other developers.
In today’s tutorial we will be taking a look at Rails migrations.
The purpose of Migrations
In order to fully understand migrations and why you need them, first it’s important to understand their purpose.
The majority of all web applications will need a database. A database is used for storing the data of the web application. So for example, that might be blog posts or your user’s details.
The database is made up of tables that store your data. Typically you would run SQL statements to create or modify the tables and columns of a database.
Rails introduces a specific Domain-specific language for writing instructions for how a database should be created. This saves you from writing SQL statements.
A migration is a file that contains a specific set of instructions for the database. For example, last week we created a migration file to create the
articles table with columns for
title and
body.
When this migration file is run, Rails will be able to make the changes to the database and automatically create the table for us.
Over time as the database evolves, the migration files will act as a versioned history of how the database has changed. This means you will be able to recreate the database from the set of instruction files.
What are the benefits of using Migrations?
There are a number of benefits to using Migrations as part of your Rails project.
Firstly, your application is going to be pretty useless without a database. When someone grabs a copy of the code they need to set up a local version of the database. Instead of passing around a file of SQL statements, the Rails project will automatically have everything it needs to recreate the database.
Secondly, when working with other developers, you will likely face a situation where one developer needs to modify the database in order to implement the feature she is working on. When that developer pushes her code and you pull down the changes, your application will be broke without also modifying the database. The migration file will be able to automatically make the change so you don’t have to try and reverse engineer what has changed.
And thirdly, when you deploy your application to a production server you need a way for the production database to be created or updated. Updating anything in production manually can be a bit hairy as any human error can potentially be disastrous. By using database migrations we can completely remove the human element of updating a production database.
What do Migration files look like?
Last week we ran the Rails generator to create a new
Article Model.
bin/rails g model Article title:string body:text [/bash] As part of this generation process, Rails also created a migration file: ```ruby class CreateArticles < ActiveRecord::Migration def change create_table :articles do |t| t.string :title t.text :body t.timestamps null: false end end end
This migration will create a new database table called
articles. We’ve also specified that the table should have a
title column of type
string and a
body column of type
text.
Rails will also add a primary key column of
id as well as
created_at and
updated_at timestamps by default.
Understanding the Migration file
The migration file is essentially just a regular Ruby class that is run during the migration process:
class CreateArticles < ActiveRecord::Migration end
The
CreateArticles class inherits from the
ActiveRecord::Migration class.
In the
change method we have the instructions of what the migration should do:
def change create_table :articles do |t| t.string :title t.text :body t.timestamps null: false end end
In this example we are calling the
create_table method with an argument of
:articles for the table name.
We also pass a block to the method that allows us to specify the names and types of the columns.
Creating Migration files
In the previous example we saw how we can automatically generate a migration file using the model generator command.
However, you will need to generate a migration file whenever you want to alter an existing table or create a join table between two tables.
So how do you generate a standalone migration file?
Rails provides a migration generator command that looks something like this:
bin/rails g migration AddSlugToArticles [/bash] This should create the following migration file ```ruby class AddSlugToArticles < ActiveRecord::Migration def change end end
You can also add the columns you want to add to the table to the generator command:
bin/rails g migration AddSlugToArticles slug:string [/bash] This will create the following migration file: ```ruby class AddSlugToArticles < ActiveRecord::Migration def change add_column :articles, :slug, :string end end
As you can see in this example, we are calling the
add_column method and passing the table name
:articles, the column name
:slug and the type
:string.
The migration will automatically created with the
add_column method if your migration begins with
Add.
If you wanted to remove a column from a table, you would generate a migration that begins with
Remove:
bin/rails g migration RemoveSlugFromArticles slug:string [/bash] This would generate the following migration: ```ruby class RemoveSlugFromArticles < ActiveRecord::Migration def change remove_column :articles, :slug, :string end end
If you would like to create a new table you should run a generator command where the migration begins with
Create. For example:
bin/rails g migration CreateComments name:string comment:text [/bash] Running the command above would generate the following migration: ```ruby class CreateComments < ActiveRecord::Migration def change create_table :comments do |t| t.string :name t.text :comment end end end
Running Migrations
Once you have generated your migration files and you are happy with the structure of the tables that you have defined, it’s time to run the migrations to update the database.
To run the migration files you can use the following command:
bin/rake db:migrate [/bash] This will work through each of your migration files in the order of which they were generated and perform that action on the database. If everything goes smoothly you should see the output from the command console telling you which migrations were run. If you try to run the `db:migrate` command again you will notice that nothing will happen. Rails keeps track of which migrations have already been run. When you run the `db:migrate` command, Rails will only run the migrations that have yet to be run. ## Rolling back migrations If you have made a mistake and you want to roll back the previous migration you can do so using the following command: ```bash bin/rake db:rollback [/bash] For simple migrations Rails can automatically revert the changes from the `change` method. If you need to rollback through multiple migrations you can add the `STEP` parameter to the command: ```bash bin/rake db:rollback STEP=2 [/bash] This will rollback through the previous 2 migrations. ## Conclusion Migrations are an important part of a modern web application framework. Nearly all web applications have a database, and so it’s pretty important that you have the right tooling. Migrations make it easy to create database tables. You will no longer have to remember the syntax for creating tables or defining column. If you were to switch databases, you will also find that the migrations are agnostic to the type of database you are using. However, really the biggest benefit of migrations is how easy it is to create a new database, or modify an existing one. When working on a team of developers this is going to be essential! | https://www.culttt.com/2015/10/07/understanding-ruby-on-rails-migrations/ | CC-MAIN-2018-51 | refinedweb | 1,382 | 51.78 |
balancer
This post assumes that you have followed the instructions in my previous post and run Lab14, so that you are now proud owner of a working OpenStack installation including Octavia. If you have not done this yet, here are the instructions to do so.
git clone cd openstack-labs/Lab14 wget vagrant up ansible-playbook -i hosts.ini site.yaml
Now, let us bring up a test environment. As part of the Lab, I have provided a playbook which will create two test instances, called web-1 and web-2. To run this playbook, enter
ansible-playbook -i hosts.ini demo.yaml
In addition to the test instances, the playbook is creating a demo user and a role called load-balancer_admin. The default policy distributed with Octavia will grant all users to which this role is assigned the right to read and write load balancer configurations, so we assign this role to the demo user as well. The playbook will also set up an internal network to which the instances are attached plus a router, and will assign floating IP addresses to the instances, creating the following setup.
Once the playbook completes, we can log into the network node and inspect the running servers.
vagrant ssh network source demo-openrc openstack server list
Now its time to create our first load balancer. The load balancer will listen for incoming traffic on an address which is traditionally called virtual IP address (VIP). This terminology originates from a typical HA setup, in which you would have several load balancer instances in an active-passive configuration and use a protocol like VRRP to switch the IP address over to a new instance if the currently active instance fails. In our case, we do not do this, but still the term VIP is commonly used. When we create a load balancer, Octavia will assign a VIP for us and attach the load balancer to this network, but we need to pass the name of this network to Octavia as a parameter. With this, our command to start the load balancer and to monitor the Octavia log file to see how the provisioning process progresses is as follows.
openstack loadbalancer create \ --name demo-loadbalancer\ --vip-subnet external-subnet sudo tail -f /var/log/octavia/octavia-worker.log
In the log file output, we can nicely see that Octavia is creating and signing a new certificate for use by the amphora. It then brings up the amphora and tries to establish a connection to its port 9443 (on which the agent will be listening) until the connection succeeds. If this happens, the instance is supposed to be ready. So let us wait until we see a line like “Mark ACTIVE in DB…” in the log file, hit ctrl-c and then display the load balancer.
openstack loadbalancer list openstack loadbalancer amphora list
You should see that your new load balancer is in status ACTIVE and that an amphora has been created. Let us get the IP address of this amphora and SSH into it.
amphora_ip=$(openstack loadbalancer amphora list \ -c lb_network_ip -f value) ssh -i amphora-key ubuntu@$amphora_ip
Nice. You should now be inside the amphora instance, which is running a stripped down version of Ubuntu 16.04. Now let us see what is running inside the amphora.
ifconfig -a sudo ps axwf sudo netstat -a -n -p -t -u
We find that in addition to the usual basic processes that you would expect in every Ubuntu Linux, there is an instance of the Gunicorn WSGI server, which runs the amphora agent, listening on port 9443. We also see that the amphora agent holds a UDP socket, this is the socket that the agent uses to send health messages to the control plane. We also see that our amphora has received an IP address on the load balancer management network. It is also instructive to display the configuration file that Octavia has generated for the agent – here we find, for instance, the address of the health manager to which the agent should send heartbeats.
This is nice, but where is the actual proxy? Based on our discussion of the architecture, we would have expected to see a HAProxy somewhere – where is it? The answer is that Octavia puts this HAProxy into a separate namespace, to avoid potential IP address range conflicts between the subnets on which HAProxy needs to listen (i.e. the subnet on which the VIP lives, which is specified by the user) and the subnet to which the agent needs to be attached (the load balancer management network, specified by the administrator). So let us look for this namespace.
ip netns list
In fact, there is a namespace called amphora-haproxy. We can use the ip netns exec command and nsenter to take a look at the configuration inside this namespace.
sudo ip netns exec amphora-haproxy ifconfig -a
We see that there is a virtual device eth1 inside the namespace, with an IP address on the external network. In addition, we see an alias on this device, i.e. essentially a second IP address. This the VIP, which could be detached from this instance and attached to another instance in an HA setup (the first IP is often called the VRRP address, again a term originating from its meaning in a HA setup as being the IP address of the interface across which the VRRP protocol is run).
At this point, no proxy is running yet (we will see later that the proxy is started only when we create listeners), so the configuration that we find is as displayed below.
Monitoring load balancers
We have mentioned several times that the agent is send hearbeats to the health manager, so let us take a few minutes to dig into this. During the installation, we have created a virtual device called lb_port on our network node, which is attached to the integration bridge to establish connectivity to the load balancer management network. So let us log out of the amphora to get back to the network node and dump the UDP traffic crossing this interface.
sudo tcpdump -n -p udp -i lb_port
If we look at the output for a few seconds, then we find that every 10 seconds, a UDP packet arrives, coming from the UDP port of the amphora agent that we have seen earlier and the IP address of the amphora on the load balancer management network, and targeted towards port 5555. If you add -A to the tcpdump command, you find that the output is unreadable – this is because the heartbeat message is encrypted using the heartbeat key that we have defined in the Octavia configuration file. The format of the status message that is actually sent can be found here in the Octavia source code. We find that the agent will transmit the following information as part of the status messages:
- The ID of the amphora
- A list of listeners configured for this load balancer, with status information and statistics for each of them
- Similarly, a list of pools with information on the members that the pool contains
We can display the current status information using the OpenStack CLI as follows.
openstack loadbalancer status show demo-loadbalancer openstack loadbalancer stats show demo-loadbalancer
At the moment, this is not yet too exciting, as we did not yet set up any listeners, pools and members for our load balancer, i.e. our load balancer is not yet accepting and forwarding any traffic at all. In the next post, we will look in more detail into how this is done.
One thought on “OpenStack Octavia – creating and monitoring a load balancer” | https://leftasexercise.com/2020/05/04/openstack-octavia-creating-and-monitoring-a-load-balancer/ | CC-MAIN-2020-50 | refinedweb | 1,282 | 55.98 |
pyConditions 0.0.1
Guava Like Preconditions in Python.
Guava like precondition enforcing for Python.
Has been tested against:
- 2.6
- 2.7
- 3.2
- 3.3
- pypy
Decorate functions with preconditions so that your code documents itself and at the same time removes the boilerplate code that is typically required when checking parameters.
An Example:
def divideAby1or10( a, b ): if not ( 1 <= b <= 10 ): <raise some error> else: return a / b
Simply becomes the following:
from pyconditions.pre import Pre pre = Pre() @pre.between( "b", 1, 10 ) def divideAbyB( a, b ) return a / b
In the above example the precondition pre.between ensures that the b variable is between (1, 10) inclusive. If it is not then a PyCondition exception is thrown with the error message detailing what went wrong.
More Examples:
from pyconditions.pre import Pre pre = Pre() @pre.notNone( "a" ) @pre.between( "a", "a", "n" ) @pre.notNone( "b" ) @pre.between( "b", "n", "z" ) def concat( a, b ): return a + b
The above ensures that the variables a and b are never None and that a is between ( ?a?, ?n? ) inclusively and b is between ( ?n?, ?z? ) inclusively.
from pyconditions.pre import Pre pre = Pre() BASES = [ 2, 3, 4 ] @pre.custom( a, lambda x: x in BASES ) @pre.custom( b, lambda x: x % 2 == 0 ) def weirdMethod( a, b ): return a ** b
Using the custom precondition you are able to pass in any function that receives a single parameter and perform whatever condition checking you need.
- Downloads (All Versions):
- 162 downloads in the last day
- 495 downloads in the last week
- 2754 downloads in the last month
- Author: Sean Reed
- Maintainer: Sean Reed
- Keywords: preconditions conditions assertion decorators
- License: LICENSE.txt
- Categories
- Package Index Owner: streed
- DOAP record: pyConditions-0.0.1.xml | https://pypi.python.org/pypi/pyConditions/0.0.1 | CC-MAIN-2016-07 | refinedweb | 295 | 58.28 |
Hi,
I try to create a managed folder in pyspark recipe, the folder is created, and then attached as an output to the recipe, but a few lines later when I try to access it it fails:
Managed folder LWVG21ww cannot be used : declare it as input or output of your recipe
Maybe the DSS is having the old version of recipe outputs in cache and does not realize that it has been changed during the execution. Is there a way to to change it?
Thanks
Not sure what you are trying to accomplish but I made a recipe that dynamically creates sub-folders of a managed folder as needed. I use the OS module to make the directories.
import dataiku
import os
# OUTPUT_FOLDER is the name of the managed folder
folder_handle = dataiku.Folder(OUTPUT_FOLDER)
# NEW_FOLDER_NAME is the name of the folder you want to create
os.mkdir(folder_handle.get_path() + "/" + NEW_FOLDER_NAME)
From there you can save files to the new folder like you normally would. The OUTPUT_FOLDER needs to be created before running this code. Use the name of a managed folder that you create in the flow of your project.
©Dataiku 2012-2018 - Privacy Policy | https://answers.dataiku.com/2474/creating-a-managed-folder-in-recipe | CC-MAIN-2019-22 | refinedweb | 196 | 69.21 |
XPath
Since Camel 1.1
Camel supports XPath to allow an Expression or Predicate to be used in the DSL or Xml Configuration. For example you could use XPath to create an Predicate in a Message Filter or as an Expression for a Recipient List.
Streams
If the message body is stream based, which means the input it receives is submitted to Camel as a stream. That means you will only be able to read the content of the stream once. So often when you use XPath as Message Filter or Content Based Router then you need to access the data multiple times, and you should use Stream Caching or convert the message body to a
String prior which is safe to be re-read multiple times.
from("queue:foo"). filter().xpath("//foo")). to("queue:bar")
from("queue:foo"). choice().xpath("//foo")).to("queue:bar"). otherwise().to("queue:others"); if there is a header with the given key
from exchange.properties if there is a property with the given key
Functions
Camel adds the following XPath functions that can be used to access the exchange:
Here’s an example showing some of these functions in use.
And the new functions introduced in Camel 2
If you have a standard set of namespaces you wish to work with and wish to share them across many different XPath expressions you can use the NamespaceBuilder as shown in this example
In this sample we have a:
And the spring XML equivalent of the route:.
i.e. cut and paste upper code to your own project in a different package and/or annotation name then add whatever namespace prefix/uris you want in scope when you use your annotation on a method parameter. Then when you use your annotation on a method parameter all the namespaces you want will be available for use in your XPath expression.
For example
public class Foo { @MessageDriven(uri = "activemq:my.queue") public void doSomething(@MyXPath("/ns1:foo/ns2:bar/text()") String correlationID, @Body String body) { // process the inbound message here } }
Using XPathBuilder without an Exchange
Since: accesible:
[me: {prefix -> namespace}, {prefix -> namespace}], [parent: [me: {prefix -> namespace}, {prefix -> namespace}], [parent: [me: {prefix -> namespace}]]]
Any of these options can be used to activate this logging:
Enable TRACE logging INFO level keyword:=[,]}path("resource:classpath:myxpath.txt", String.class) | https://camel.apache.org/components/2.x/languages/xpath-language.html | CC-MAIN-2020-40 | refinedweb | 388 | 52.63 |
This is an exploration of the limits of C# - how much it could be pushed to create Fluent APIs or Internal Domain Specific Languages (DSLs) that have different kinds of flow than the traditional use of the language. On the way we look at indexers, implicit cast, delegates and operator overloading, and use them - sometimes in a bit unexpected ways.
The example that I use is an API for generation of XML / (X)HTML with more Fluent syntax. Let's call it Fluent.Xml.Linq.
I have included some code as screenshots, because styling is so important for the purposes of this article, but you can find all the source in the accompanied ZIP-file.
I love C#, I am a huge fan of Anders Hejlsberg, but there is one thing I am envious of Visual Basic developers of: Quite often you need to produce XML or (X)HTML within code, and they've got this great language feature called XML literals. You are probably familiar with that, but here is an example just in case:
Now, if the software that you are writing does a lot of XML production, you should really consider template technologies - my favorite is StringTemplate - but often, it is just some XML blocks here and there, and it is not really worth the extra DLL references. String concatenation is one option, but let's say we want to stay somewhat strongly typed...
I was thinking to first show here the above VB-example using System.Xml.XmlDocument. It would probably be dozens of lines of code. But then, I thought it is waste of space: most of you - if not all - are already familiar with System.Xml.Linq, so this is what the above example would look using LINQ to XML:
System.Xml.XmlDocument
System.Xml.Linq
Well, it is not bad compared to System.Xml, but there are some problems:
System.Xml
new
XElement
XAttribute
Really, what I would like to write is something like this:
It would be easy to create a DSL like that - using Oslo/MGrammar tools, or my favorite Antlr tools. But everyone who has tried creating DSLs knows that while there are quite good tools for creating the actual parsing of DSLs and using DSLs as stand-alone programs, the deployment, maintenance, and IDE integration are still very hard.
Hopefully, in the future, Microsoft will get their DSL story straight so that it would be easy to mix, e.g., C# and DSLs within Visual Studio. But for now, what I would really like to do is to be able to write this kind of syntax in C#.
Well, this article is an exploration of how you could do that.
The first thing that we should tackle is the easiest part - the attributes. It is easy to create an indexer to get the attrName["attrValue"] syntax. You could just create a class like this:
attrName["attrValue"]
public class FXAttrDefinition
{
private XName name;
public FXAttrDefinition(XName name)
{
this.name = name;
}
public XAttribute this[object value]
{
get
{
return new XAttribute(this.name, value);
}
}
}
... which would allow you to create the attributes like this:
FXAttrDefinition type = new FXAttrDefinition("type"), value = new FXAttrDefinition("value");
XElement input = new XElement("input", type["text"], value["Default value"]);
but the constructors are really long to write like that - they take too much space. You could easily have dozens of different attributes even for simple HTML. Couldn't we replace them with something nicer?
What we can do is make an implicit cast so that our FXAttrDefinition class can be automatically converted from string the same way as double is converted from integer or XName is automatically converted from a string. You do it like this:
FXAttrDefinition
string
double
XName
public static implicit operator FXAttrDefinition(string name)
{
return new FXAttrDefinition(name);
}
public static implicit operator FXAttrDefinition(XName name)
{
return new FXAttrDefinition(name);
}
I decided to also allow the cast from XName, so that you can use the XNamespace + string => XName construct, which is the way of LINQ to XML to create namespaces. So, now you could use the code like this:
FXAttrDefinition tpe = "type", val = "value";
XElement input = new XElement("input", tpe["text"], val["Default value"]);
Now, that starts to be quite succinct as far as the attributes are concerned.
Next, we should take care of the XElement. You could of course use the exact same structure as with attributes - there is an example of that as the FXElemDefinition2 class in the sample files. However, it is not quite what I am looking for. Just look at the result:
FXElemDefinition2
Firstly, we are using an object array (marked "params") as the indexer! Yes - to my surprise - it is allowed, and it works, but it is starting to be a bit strange if you think about the intended purpose of indexer as a language feature.
object
params
More importantly, to my quest for that perfect XML generation syntax, readability is better in my first example where we have round brackets () around element children and square brackets are reserved for attribute value.
So, how can I get those round brackets?
Of course, you could just create static methods to your class for each of the elements you are going to use. But then, you would end up having each of the XML generating classes cluttered with dozens of methods like form(), input() etc. Unacceptable!
form()
input()
Or, if there was a separate class, you would need to first instantiate it" var x = new FunctionCollection();. Then, you would use syntax like x.form(). OK, best option so far, but I really want to get rid of that x..
var x = new FunctionCollection();
x.form()
x.
Delegates are an important language feature introduced in C# 2.0. They are basically functions assigned to a variable. These functions can be static, from instantiated objects, or even anonymous. We can use them like this to get our nice function-syntax on elements:
delegate XElement FXElemDefinition(params object[] children);
XElement pDel(params object[] children)
{
return new XElement("p", children);
}
XElement inputDel(params object[] children)
{
return new XElement("input", children);
}
private void getElemExample()
{
// Element definitions
FXElemDefinition p = pDel, ipt = inputDel; ;
// Creating the <p><input type="text" /></p> with nice syntax
XElement elem = p(ipt(new XAttribute("type", "text")));
}
We define a delegate named FXElemDefinition and we define functions for paragraph and input. OK, so now our element syntax is the way I want it, but how can we get rid of that boilerplate code of defining functions?
FXElemDefinition
Let's take a step back and think what we are doing: We need to call a constructor (= function) new XElement() with two parameters: name and children. But, we assign the variable p to be another function that already knows the element name "p", and thus we can call it just with the children parameter. So essentially, we want to reduce the function from having two parameters to only one parameter. This is called currying, and it is a much used mechanism in functional languages. In C#, it is a bit limited. For example, it does not play well together with the "params" keyword, and I had real trouble implementing it first. But fortunately, there is one syntax that we can use in our case:
new XElement()
name
children
p
FXElemDefinition p = x => new XElement("p", x);
FXElemDefinition ipt = x => new XElement("input", x);
But that's really messy, the syntax is not intuitive unless you have worked with lambdas a lot. And again, any real-life XML generation scenario would have a lot of different elements. Can we use implicit cast again to make it shorter?
Unfortunately: not! The delegate class is marked sealed so we cannot touch that (probably for a good reason . We cannot do casts or operator overloading unless we are able to extend the class. We can, however, create a helper class like this:
delegate
sealed
public delegate XElement FXElemDefinition(params object[] children);
public class FXElementHelper
{
public FXElemDefinition this[XName elementName]
{
get
{
return remainingChildrenParam =>
new XElement(elementName, remainingChildrenParam);
}
}
}
and then we can call it like this:
// Element definitions
var h = new FXElementHelper();
FXElemDefinition p = h["p"], ipt = h["input"];
// Creating the <p><input type="text" /></p>
XElement elem = p(ipt(new XAttribute("type", "text")));
So now, we have achieved our goal and can create XML/XHTML, with quite nice syntax like this (FluentSample.cs in the source code):
If I knew how, I would get rid of the FXElementHelper and h[""], but I can live with this syntax. It is nice to read - you get the structure of the HTML immediately: wlement and attribute names are first defined, and after that, they are strongly typed. It is quite fast to write, and supports intellisense nicely.
FXElementHelper
h[""]
I actually started writing this article thinking that the DSL that I would demonstrate would use operator overloading. You know, the XML element would be created like an expression:
There is an unfinished - but somewhat working - example of this in the "Old" folder of the source code if you are interested. But I just couldn't get it "flowing fluently" enough using operators. The problem is the hierarchical structure of XML: the Nested Functions approach is much more natural.
But I am sure there are good uses for operator overloading in Internal DSLs that have other purposes than XML. I also think that this method could be used more than it is now to produce more intuitive and naturally flowing syntax. One good example I already mentioned above is the XNamespace, which overloads the plus operator when the second operand is string:
XNamespace
XNamespace ns = "MyNameSpace";
XName qualifiedName = ns + "tagName";
Console.WriteLine(new XElement(qualifiedName));
// produces <tagName xmlns="MyNameSpace" />
But you could get so much further with this: overloading different operators (*, /, -, ==) and so on to build complete expressions. Perhaps, I will figure out a good example and write about it someday.
The term of Fluent Interfaces (or Internal DSLs) was introduced by Martin Fowler as he, in 2005, described the then new style of interfaces that were often characterized by the use of method chaining. I decided not to cover method chaining here as it is so well known these days. Just put "Fluent" in the CodeProject search box or "Fluent API" in Google, and you will find several examples. I am sure you will recognize the style - even if you didn't recognize the term.
I would say LINQ to XML is already an example of a Fluent interface. Sure, it does not use method chaining using dots, but it uses the most natural way of constructing XML trees: hierarchical function structure (or Nested Functions, as Fowler describes in this article). In that sense, it satisfies Fowler's description of "intent is to do something along the lines of an internal Domain Specific Language" and "the API is primarily designed to be readable and to flow".
What I tried to demonstrate above is ways to take that "fluency" one step further using other language features that are available to us in this great language of C#.
To me, the pros of these kinds of approaches are clear: APIs like this are easier to read and write:
The main downside is as Fowler wrote in the original article:
"One of the problems of methods in a fluent interface is that they don't make much sense on their own. Looking at a method browser of a method by method documentation doesn't show much sense to [function named] with. Indeed sitting there on its own, I'd argue that it's a badly named method that doesn't communicate its intent at all well. It's only in the context of the fluent action that it shows its strengths."
Well said. I would only like to add to this a few aspects: Often in Fluent APIs, certainly in the examples above, you are "bending" the original intent and conventions of the language (say, for example, using indexers when there is really no collection . For someone who is not familiar with your DSL and the intention of it, this might be confusing.
Another thing is that making a good Fluent API is an effort. It is much faster to create a traditional API. In fact, Fowler suggests - and I agree - it makes sense to first create a traditional API and then create the Fluent API on top of it.
So, there are serious pros and cons in Fluent interfaces. This is still kind of a new approach - experience is limited. Whether you use this stuff or not in your projects is in your discretion - well, like any other pattern I suppose. It's just an. | http://www.codeproject.com/Articles/58716/Fluent-Xml-Linq-Exploring-the-limits-of-C-syntax?msg=3371851 | CC-MAIN-2014-52 | refinedweb | 2,108 | 57.91 |
Instructions on how to write a new DDE plugin.
When to write a plugin
Only write a plugin if you want to create an DDE export for databases like UDD or apt-xapian-index.
If you just periodically generate a small amount of static data, you can publish it by saving it in a format like yaml, json or pickle and put it in ~/.dde (see DDE/StaticData).
The plugin interface
Plugins are instances of an octofuss.Tree, which represent a subtree in the DDE information space; an octofuss.Tree is defined as an object which implements the following 4 methods:
lhas, that tells if a subtree exists or not
lget, that gets the value of a node in the subtree
llist, that list the child nodes of a node in the subtree
ldoc, that gives documentation about a node in the subtree
Plugins get one parameter called path which the path of the requested subtree split into a list. [] means the node itself.
Some optional keyword arguments can be passed as hints for optimization purposes, and you should normally ignore them: this explains why every method also has a **kw parameter, but besides remembering to put it there, you should not worry about it.
Here is some annotated example plugin implementing a node with a value and no subtrees:
import octofuss # Plugin interface class Plugin(octofuss.Tree): def __init__(self): # The first parameter is the name of this node in the tree # The second parameter is the documentation for this node to use # if the 'ldoc' method is missing super(Plugin, self).__init__("plugin", "Example plugin") def lhas(self, path, **kw): """ Return True if the path exists, False if not. """ # We are empty, so no subtrees exist: if path: return False # But we exist, so if path is [] return True: return True def lget(self, path, **kw): """ Return the value of the node at path, or None if the path does not exist """ # We are empty, so no subtrees exist: if path: return None # If path is [], return our value return "I am an example plugin" def llist(self, path, **kw): """ Return a list with the names of the child nodes. If there are no child nodes or the path is invalid, return [] """ return [] def ldoc(self, path, **kw): """ Return the documentation for the given node. Return None if the node does not exist. """ if path: return None return "Example plugin"
To allow DDE to instantiate the plugin, add to your module an 'init' function. It is passed optional information as keyword parameters (currently not really used, but planned for the future), and generates dictionaries with the octofuss.Tree objects and their mount point in the DDE tree:
def init(**kw): if os.environ.get("DDE_AVOID_CRUFT", None) is not None: return yield dict( tree = Plugin(), # Mount as /test/plugin root = "/test" )
The init function should only generate those plugins that are in a condition to be run (for example, ensuring that their data files exist, or that connections to their databases can be established).
Tips & Tricks
Do not start to write a plugin from scratch: have a look at the existing plugins at the DDE git repository for inspiration and for stealing ideas.
Get in touch with Enrico if you need help
- Only implement views corresponding to common use cases: DDE should be simple to query. You should not reimplment SQL or LDAP using URLs: for special needs, people can craft a SQL or LDAP query, and if the need becomes more general, the query can be turned into a DDE plugin.
See also: | https://wiki.debian.org/DDE/WritePlugin | CC-MAIN-2019-47 | refinedweb | 595 | 53.44 |
Tải bản đầy đủ
Công Nghệ Thông Tin
Kỹ thuật lập trình
C++ quick syntax reference xiii
About the Technical Reviewer��������������������������������������������������������� xv
Introduction����������������������������������������������������������������������������������� xvii
■■Chapter 1: Hello World�������������������������������������������������������������������� 1
■■Chapter 2: Compile and Run����������������������������������������������������������� 3
■■Chapter 3: Variables����������������������������������������������������������������������� 5
■■Chapter 4: Operators�������������������������������������������������������������������� 11
■■Chapter 5: Pointers����������������������������������������������������������������������� 15
■■Chapter 6: References������������������������������������������������������������������ 19
■■Chapter 7: Arrays������������������������������������������������������������������������� 21
■■Chapter 8: String�������������������������������������������������������������������������� 23
■■Chapter 9: Conditionals���������������������������������������������������������������� 27
■■Chapter 10: Loops������������������������������������������������������������������������� 29
■■Chapter 11: Functions������������������������������������������������������������������ 31
■■Chapter 12: Class������������������������������������������������������������������������� 37
■■Chapter 13: Constructor��������������������������������������������������������������� 41
■■Chapter 14: Inheritance���������������������������������������������������������������� 45
■■Chapter 15: Overriding����������������������������������������������������������������� 47
■■Chapter 16: Access Levels������������������������������������������������������������ 51
iii
■ Contents at a Glance
■■Chapter 17: Static������������������������������������������������������������������������� 55
■■Chapter 18: Enum������������������������������������������������������������������������� 57
■■Chapter 19: Struct and Union������������������������������������������������������� 59
■■Chapter 20: Operator Overloading������������������������������������������������ 63
■■Chapter 21: Custom Conversions������������������������������������������������� 67
■■Chapter 22: Namespaces�������������������������������������������������������������� 69
■■Chapter 23: Constants������������������������������������������������������������������ 73
■■Chapter 24: Preprocessor������������������������������������������������������������� 77
■■Chapter 25: Exception Handling��������������������������������������������������� 83
■■Chapter 26: Type Conversions������������������������������������������������������ 87
■■Chapter 27: Templates������������������������������������������������������������������ 93
■■Chapter 28: Headers��������������������������������������������������������������������� 99
Index���������������������������������������������������������������������������������������������� 103
iv book.
xvii
Chapter 1
Hello World
Choosing an IDE
To begin developing in C++ you should download and install an Integrated Development
Environment (IDE) that supports C++. A good choice is Microsoft’s own Visual Studio.1
If you do not have Visual Studio but would like to try out the examples in this book in
a similar environment you can download Visual Studio Express2 ➤ New ➤ Project in Visual Studio, or File ➤ New ➤ Solution
Explorer) you can see that the project consists of three empty folders: Header Files,
Resource Files and Source Files. Right click on the Source Files folder and select
Add ➤ New Item. From the Add New Item dialog box choose the C++ File (.cpp) template.
Give this source file the name “MyApp” and click the Add button. An empty cpp file will
now be added to your project and also opened for you.
1
2
1
CHAPTER 1 ■ Hello World
int main()
{
std::cout << "Hello World";
}
Using namespace
To make things a bit easier you can add a line specifying that the code file uses the
standard namespace. You then no longer have to prefix cout with the namespace (std::)
since it is now used by default.
#include
using namespace std;
int main()
{
cout << "Hello World";
}
2
Chapter 2
Compile and Run
Visual Studio compilation.
using namespace std;
int main()
{
cout << "Hello World";
cin.get();
}
Console compilation
As an alternative to using an IDE you can also compile source files from the command
line as long as you have a C++ compiler.1
1
3
CHAPTER 2 ■ Compile and Run
C++ has two kinds of comment notations – single-line and multi-line. These are used to
insert notes into the source code and will have no effect on the end program.
// single-line comment
/* multi-line
comment */
4
Chapter 3.
Data Type
Size (byte)
Description
char
1
Integer or character
short
2
int
4
long
4
float
4
double
8
long double
8
bool
1
Integer
Floating-point number
Boolean.
5
CHAPTER 3 ■ Variables equal)
6
CHAPTER 3 ■ Variables
7
CHAPTER 3 ■ Variables five 32767
8
CHAPTER 3 ■ Variables
9
Chapter 4
Operators
The numerical operators in C++ can be grouped into five types: arithmetic, assignment,
comparison, logical and bitwise operators.
Arithmetic operators
There are the four basic arithmetic operators, as well as the modulus operator (%) which is
used to obtain the division remainder.
x = 3 + 2; // 5 // addition
x = 3 - 2; // 1 // subtraction
x = 3 * 2; // 6 // multiplication
x = 3 / 2; // 1 // division
x = 3 % 2; // 1 // modulus (division remainder)
Notice that the division sign gives an incorrect result. This is because it operates on
two integer values and will therefore truncate the result and return an integer. To get the
correct value one of the numbers must be explicitly converted to a floating-point number.
x = 3 / (float.
11
CHAPTER 4 ■ Operators
x
x
x
x
x
+=
-=
*=
/=
%=
5;
5;
5;
5;
5;
//
//
//
//
//
x
x
x
x
x
=
=
=
=
=
x+5;
x-5;
x*5;
x/5;
x%5;
Increment and decrement operators
Another common operation is to increment or decrement a variable by one. This can be
simplified with the increment (++) and decrement (--) operators.
x++; // x = x+1;
x--; // x = x-1;
Both of these++; // y=5, x=6
x = 5; y = ++x; // y=6, x=6
Comparison operators
The comparison operators compare two values and return either true or false. They
are mainly used to specify conditions, which are expressions that evaluate to either true
or false.
bool x = (2 == 3); // false // equal to
x = (2 != 3); // true // not equal to
x = (2 > 3); // false // greater than
x = (2 < 3); // true // less than
x = (2 >= 3); // false // greater than or equal to
x = (2 <= 3); // true // less than or equal to
Logical operators
The logical operators are often used together with the comparison operators. Logical and
(&&) evaluates to true if both the left and right sides are true, and logical or (||) is true if
either the left or right side is true. For inverting a Boolean result there is the logical not (!)
12
CHAPTER 4 ■ Operators
operator. Note that for both “logical and” and “logical or” the right-hand side will not be
evaluated if the result is already determined by the left-hand side.
bool x = (true && false); // false // logical and
x = (true || false); // true // logical or
x = !(true);
// false // logical not
Bitwise operators
The bitwise operators can manipulate individual bits inside an integer. For example,
the “bitwise or” operator (|) makes the resulting bit 1 if the bits are set on either side
of the operator.
int x = 5 & 4; // 101 & 100 = 100 (4)
// and
x = 5 | 4; // 101 | 100 = 101 (5)
// or
x = 5 ^ 4; // 101 ^ 100 = 001 (1)
// xor
x = 4 << 1; // 100 << 1 =1000 (8)
// left shift
x = 4 >> 1; // 100 >> 1 = 10 (2)
// right shift
x = ~4;
// ~00000100 = 11111011 (-5) // invert
The bitwise operators also have combined assignment operators.
int precedence
In C++, expressions are normally evaluated from left to right. However, when an
expression contains multiple operators, the precedence of those operators decides the
order that they are evaluated in. The order of precedence can be seen in the table below.
This same order also applies to many other languages, such as Java and C#.
Pre
Operator
Pre
Operator
1
++ -- ! ~
7
&
2
*/%
8
^
3
+-
9
|
4
<< >>
10
&&
5
< <= > >=
11
||
6
== !=
12
= op=
13
CHAPTER 4 ■ Operators
For example, logical and (&&) binds weaker than relational operators, which in turn
bind weaker than arithmetic operators.
bool x = 2+3 > 1*4 && 5/5 == 1; // true
To make things clearer, parentheses can be used to specify which part of the
expression will be evaluated first. Parentheses have the highest precedence of all
operators.
bool x = ((2+3) > (1*4)) && ((5/5) == 1); // true
14
Chapter 5
15
CHAPTER 5 ■ Pointers
16
CHAPTER 5 ■ Pointers.
delete d;
// ...
if (d != NULL) { *d = 10; } // check for null pointer
17
Chapter 6 x = 5;
int& r = x; // r is an alias to x
int &s = x; // alternative syntax
Once the reference has been assigned, or seated, it can never be reseated to another
variable. The reference has in effect become an alias for the variable and can be used
exactly as though it was the original variable.
r = 10; // assigns value to r/x
References and pointers
A reference is similar to a pointer that always points to the same thing. However, while
a pointer is a variable that points to another variable, a reference is only an alias and does
not have an address of its own.
19
CHAPTER 6 ■ References)
20
Chapter 7
21
CHAPTER 7 ■ Arrays
22
Tài liệu liên quan
c# 3.0 the complete reference (3rd edition)
Quick Reference
A Quick Tour of the C++CLI Language Features
Appendix A Quick Reference
Excel 2007 for Dummies Quick Reference P1
Chapter 6 Quick Reference
Chapter 7 Quick Reference
Quick reference
Excel 2007 for Dummies Quick Reference P2
Chapter 8 Quick Reference
Tài liệu bạn tìm kiếm đã sẵn sàng tải về
(1.95 MB) - C++ quick syntax reference
Tải bản đầy đủ ngay
× | https://text.123doc.org/document/4791697-c-quick-syntax-reference.htm | CC-MAIN-2018-13 | refinedweb | 1,335 | 55.47 |
import lotus notes into MS Outlook 2000
With Lotus iConnect for Outlook, O2000 can be used as a front-end to Domino/Notes R5. But that might not solve all of your problems:
- If you're talking about a 'legacy' Notes, that's probably Notes R4
- Even with iConnect, Outlooks datastore is not used automatically. You'd probably have to cut and paste between both storages, but it should work.
Hope this helps,
<Erik> - The Netherlands
import lotus notes into MS Outlook 2000
There isd a company called Binarytree that wrote all the exchange to notes migration tools. I'm sure that do stuff to go the other way. I'd advise you to check then out. Their url is if they don't have the product I'm somebody in their forums will be able to point you in the right direction.
import lotus notes into MS Outlook 2000
I'd like to import Lotus Notes files into my Outlook 2000. I am not referring to the Notes Organizer - this is not the program that I used in the past. I used Lotus Notes. I have search the Internet for some kind of import utility but couldn't find one. Any help on this matter is greatly appreciated.
Thanks.
tk
This conversation is currently closed to new comments. | https://www.techrepublic.com/forums/discussions/import-lotus-notes-into-ms-outlook-2000/ | CC-MAIN-2018-26 | refinedweb | 219 | 72.16 |
Introduction and requirements
We will be talking about Nerves in this lesson. The Nerves project is a framework for using Elixir in embedded software development. As the website for Nerves says, it allows you to “craft and deploy bulletproof embedded software in Elixir”. This lesson will be a bit different from other Elixir School lessons. Nerves is a bit more difficult to get into as it requires both some advanced system setup and additional hardware, so may not be suitable for beginners.
To write embedded code using Nerves, you will need one of the supported targets, a card reader with a memory card supported by the hardware of your choice, as well as wired networking connection to access this device by the network.
However, we would suggest using a Raspberry Pi, due to it having controllable LED onboard. It is also advisable to have a screen connected to your target device as this will simplify debugging using IEx.
Setup
The Nerves project itself has an excellent Getting started guide, but the amount of detail there may be overwhelming for some users. Instead, this tutorial will try and present “fewer words, more code”.
Firstly, you will need an environment set up. You can find the guide in the Installation part of Nerves wiki. Please make sure that you have the same version of both OTP and Elixir mentioned in the guide. Not using the right version can cause trouble as you progress. At the time of writing, any Elixir (compiled with Erlang/OTP 21) should work.
After getting set up, you should be ready to build your first Nerves project!
Our goal will be getting to the “Hello world” of embedded development: a blinking LED controlled by calling a simple HTTP API.
Creating a project
To generate a new project, run
mix nerves.new network_led and answer
Y when prompted whether to fetch and install dependencies.
You should get the following output:
Your Nerves project was created successfully. You should now pick a target. See for supported targets. If your target is on the list, set `MIX_TARGET` to its tag name: For example, for the Raspberry Pi 3 you can either $ export MIX_TARGET=rpi3 Or prefix `mix` commands like the following: $ MIX_TARGET=rpi3 mix firmware If you will be using a custom system, update the `mix.exs` dependencies to point to desired system's package. Now download the dependencies and build a firmware archive: $ cd network_led $ mix deps.get $ mix firmware If your target boots up using an SDCard (like the Raspberry Pi 3), then insert an SDCard into a reader on your computer and run: $ mix firmware.burn Plug the SDCard into the target and power it up. See target documentation above for more information and other targets.
Our project has been generated and is ready to be flashed to our test device! Let’s try it now!
In the case of a Raspberry Pi 3, you set
MIX_TARGET=rpi3, but you can change this to suit the hardware you have depending on the target hardware (see the list in the Nerves documentation).
Let’s set up our dependencies first:
$ export MIX_TARGET=rpi3 $ cd network_led $ mix deps.get .... Nerves environment MIX_TARGET: rpi3 MIX_ENV: dev Resolving Nerves artifacts... Resolving nerves_system_rpi3 => Trying |==================================================| 100% (133 / 133) MB => Success Resolving nerves_toolchain_arm_unknown_linux_gnueabihf => Trying |==================================================| 100% (50 / 50) MB => Success
Note: be sure you have set the environment variable specifying the target platform before running
mix deps.get, as it will download the appropriate system image and toolchain for the specified platform.
Burning the firmware
Now we can proceed to flashing the drive. Put the card into the reader, and if you set up everything correctly in previous steps, after running
mix firmware.burn and confirming the device to use you should get this prompt:
Building ......../network_led/_build/rpi_dev/nerves/images/network_led.fw... Use 7.42 GiB memory card found at /dev/rdisk2? [Yn]
If you are sure this is the card you want to burn - pick
Y and after some time the memory card is ready:
Use 7.42 GiB memory card found at /dev/rdisk2? [Yn] |====================================| 100% (32.51 / 32.51) MB Success! Elapsed time: 8.022 s
Now it is time to put the memory card into your device and verify whether it works.
If you have a screen connected - you should see a Linux boot sequence on it after powering up the device with this memory card inserted.
Setting up networking
The next step is to set up the network. The Nerves ecosystem provides a variety of packages, and nerves_network is what we will need to connect the device to the network over the wired Ethernet port.
It is already present in your project as a dependency of
nerves_init_gadget. However, by default, it uses DHCP (see the configuration for it in
config/config.exs after running
config :nerves_init_gadget). It is easier to have a static IP address.
To set up static networking, you need to add the following lines to
config/config.exs:
# Statically assign an address config :nerves_network, :default, eth0: [ ipv4_address_method: :static, ipv4_address: "192.168.88.2", ipv4_subnet_mask: "255.255.255.0", nameservers: ["8.8.8.8", "8.8.4.4"] ]
Please note that this configuration is for a wired connection. If you want to use wireless connection - take a look at the Nerves network documentation.
Note that you need to use your local network parameters here - in my network there is an unallocated IP
192.168.88.2, which I am going to use. However, in your case, it may differ.
After changing this, we will need to burn the changed version of the firmware via
mix firmware.burn, then start up the device with the new card.
When you power up the device, you can use
ping to see it coming online.
Request timeout for icmp_seq 206 Request timeout for icmp_seq 207 64 bytes from 192.168.88.2: icmp_seq=208 ttl=64 time=2.247 ms 64 bytes from 192.168.88.2: icmp_seq=209 ttl=64 time=2.658 ms
This output means that the device now is reachable from the network.
Network firmware burning
So far, we have been burning SD cards and physically load them into our hardware. While this is fine to start with, it is more straightforward to push our updates over the network. The
nerves_firmware_ssh package does just that. It is already present in your project by default and is configured to auto-detect and find SSH keys in your directory.
To use the network firmware update functionality, you will need to generate an upload script via
mix firmware.gen.script. This command will generate a new
upload.sh script which we can run to update the firmware.
If the network is functional after the previous step, you are good to go.
To update your setup, the simplest way is to use
mix firmware && ./upload.sh 192.168.88.2: the first comand creates the updated firmware, and the second one pushes it over the network and reboots the device. You can finally stop having to swap SD cards in and out of the device!
_Hint:
ssh 192.168.88.2 gives you an IEx shell on the device in the context of the app. _
Troubleshooting: If you don’t have an existing ssh key in your home folder, you will have an error
No SSH public keys found in ~/.ssh.. In this case, you will need to run
ssh-keygen and re-burn the firmware to use the network update feature.
Setting up the LED control
To interact with LEDs, you need nerves_leds package installed which is done by adding
{:nerves_leds, "~> 0.8", targets: @all_targets}, to
mix.exs file.
After setting up the dependency, you need to configure the LED list for the given device. For example, for all Raspberry Pi models, there is only one LED onboard:
led0. Let’s use it by adding a
config :nerves_leds, names: [green: "led0"] line to the
config/config.exs.
For other devices, you can take a look at the corresponding part of the nerves_examples project.
After configuring the LED itself, we surely need to control it somehow. To do that, we will add a GenServer (see details about GenServers in OTP Concurrency lesson) in
lib/network_led/blinker.ex with these contents:
defmodule NetworkLed.Blinker do use GenServer @moduledoc """ Simple GenServer to control GPIO #18. """ require Logger alias Nerves.Leds def start_link(state \\ []) do GenServer.start_link(__MODULE__, state, name: __MODULE__) end def init(state) do enable() {:ok, state} end def handle_cast(:enable, state) do Logger.info("Enabling LED") Leds.set(green: true) {:noreply, state} end def handle_cast(:disable, state) do Logger.info("Disabling LED") Leds.set(green: false) {:noreply, state} end def enable() do GenServer.cast(__MODULE__, :enable) end def disable() do GenServer.cast(__MODULE__, :disable) end end
To enable this, you also need to add it to the supervision tree in
lib/network_led/application.ex: add
{NetworkLed.Blinker, name: NetworkLed.Blinker} under the
def children(_target) do group.
Notice that Nerves has two different supervision trees in application - one for the host machine and one for actual devices.
After this - that’s it! You actually can upload the firmware and via running IEx through ssh on target device check that
NetworkLed.Blinker.disable() turns the LED off (which is enabled by default in code), and
NetworkLed.Blinker.enable() turns it on.
We have control over the LED from the command prompt!
Now the only missing piece of the puzzle left is to control the LED via the web interface.
Adding the web server
In this step, we will be using
Plug.Router. If you need a reminder - feel free to skim through the Plug lesson.
First, we will add
{:plug_cowboy, "~> 2.0"}, to
mix.exs and install the dependencies.
Then, add the actual process to process those requests in
lib/network_led/http.ex :
defmodule NetworkLed.Http do use Plug.Router plug(:match) plug(:dispatch) get("/", do: send_resp(conn, 200, "Feel free to use API endpoints!")) get "/enable" do NetworkLed.Blinker.enable() send_resp(conn, 200, "LED enabled") end get "/disable" do NetworkLed.Blinker.disable() send_resp(conn, 200, "LED disabled") end match(_, do: send_resp(conn, 404, "Oops!")) end
And, the final step - add
{Plug.Cowboy, scheme: :http, plug: NetworkLed.Http, options: [port: 80]} to the application supervision tree.
After the firmware update, you can try it! is returning plain text response, and with disable and enable that LED!
You can even pack Phoenix-powered user interfaces into your Nerves app, however, it will require some tweaking.
Caught a mistake or want to contribute to the lesson? Edit this page on GitHub! | https://elixirschool.com/en/lessons/advanced/nerves/ | CC-MAIN-2020-05 | refinedweb | 1,767 | 67.45 |
I've always liked the idea of ray-tracing to render 3D images with crazy accuracy. On Saturday night (being a huge nerd) I decided I'd try to write one from scratch for the hell of it. By from scratch I mean I started with this C++ code in a text file:
#include <stdlib.h>
#include <stdio.h>
int main()
{
return 0;
}
If you would like to receive an email when updates are made to this post, please register here
RSS
stdio.h and stdlib.h shouldn't appear in a true C++ project. You should be using the C++ iostream library instead.
Any chance you can post some metrics? eg. number of polys/spheres/boxes, resolution and time taken to render?
Nice effort! Well i mean great. I had always wanted to produce something like this but i am still grasping the initial understanding of c++ let alone raytracing.
Keep the efforst going i would love to see how far you get with this ;)
That is looking good, what will you be using it in, in the future?
Very cool!! It is excellent to see someone still writing cool and efficient code in C/C++!
Nice one. Seems "not that hard", though, even if I wouldn't have known where to start anything like that.
Raytracing is something I'd very briefly visited in the past, back when I was in C++ on DOS.
Your article has once again sparked my interest in it - so off to fool around with your code samples!
Hi, Can you publish some sort of app or publish the code please, it looks great! I cant wait to mess around
Cool Project. Thanks a ton for sharing. I can't wait to see your code. Would like to see more articles from you in future.
That's amazing! With a supercomputer, would this be able to render in realtime? Is this the future of computer graphics?
Wow!
Just shows what can be accomplished will a little coffee... (I'm assuming you used the power of coffee to keep you coding).
Dan
This is SO cool! Thanks for the nice crash-course in raytracing!
Very nice. Although I'm nowhere near that level, I like the idea. Also, if you felt like it, instead of making it just a screenshot, make it the renderer for your entire game... it would be very interesting to say the least. Only 480k pixels to go in a 600x400 game, but whats the fun in that :D I must say the image is spectacular for a bmp, which i am assuming they are. Just curious as to the resolution of the images. Anywho very nice.
Nice program!!!! I downloaded the source code and started to play with it and it's great. I am trying to understand the source code but it is very difficult cuz there are not much comments. Do you have a version of this program with detail comments!? If you do it would be great to have!!!
I was astounded at this article. I am a heavy C++ programmer with strong roots into things like pov-ray and ArtOfIllusion. This just blew my mind away.
A great article, 5/5
I especially like the, "Looking up at the underside of a high-in-the-sky sphere."
the download link is on the top of the page
Is the version in the pics different than the one you can download?? I ran the code to see what it can do but get is a window that pops up with nothing but #'s and &'s accrost it.
Compiles and runs just fine on MacOS X 10.4.11 (PowerPC). Very cool stuff.
I think this is an awesome idea. Kind of reminds me of icon forge though. Very cool
If you are using this with Visual C++ 2008 express, I would suggest downloading the most recent version of the source code (see download link above) and letting the program "update" the source code (which is from VS 2005).
To update, unzip the source files, then open pixelmachine_20070220/VisualStudio2005 and select the "pixelmachine" file that is a VC++ Project. Express should then launch a conversion wizard.
PS Fun Project!!
I am do not write code, and I am looking for an off the shelf software that I can learn and apply towards creating "the shape of shadows." My goal is to see what shadows are cast onto an array of solar cells from nearby structures ,such as office buildings, homes, etc. I do not need anything more than just the area and dimensions of black shadows. I will provide the 3D volumes of the nearby structures by exporting CAD models. Any suggestions?
@David, I suggest trying some off the shelf 3D rendering software. There are some open source free things that may be able to do what you're asking.
Do you know the seed that you used for that? I really like that location.
What you've done - excellent program, but no bibliography at end — not easy to understand the concepts of it.
Wow.. cool journey :)
It sure looks fun | http://blogs.msdn.com/coding4fun/archive/2007/07/16/3820716.aspx | crawl-002 | refinedweb | 854 | 83.36 |
make generate-plist
cd /usr/ports/www/p5-RT-Extension-SLA/ && make install clean
pkg install p5-RT-Extension-SLA
Number of commits found: 46 error, which should fix the index too...
Add option to allow compiling for www/rt44, and make that the default.
(Except for www/p5-RT-Extension-SLA, as the SLA functionality is now
an integral part of rt44)
Refactor the option handling code.
Add a CONFLICTS_INSTALL on www/rt44 -- the SLA module has been made a
standard feature there rather than an add-on.
Remove a duplicate LICENSE setting.
Update to 1.04
Drop support for rt38, which is long gone.
Update to 1.03
Change Log:
Update to 1.02:
Change
I'm not exactly sure why or what, but these ports do something at install time,
and it only works as root, so, mark them as NEED_ROOT.
Sponsored by: Absolight
Update to 1.01
ChangeLog:
Update to 0.08
Changes:
-
Trim remaining untrimmed headers on my ports
Where BUILD_DEPENDS and RUN_DEPENDS have the same value, initialise
RUN_DEPENDS from BUILD_DEPENDS
- Update to 0.07
- ChangeLog:
- png to 1.5.10
Update maintainer address to matthew@FreeBSD.org
Approved by: shaun (mentor)
- Update to version 0.05
- Add support for www/rt40
- Add license
- Pet portlint
PR: ports/160955 0.04
- Add LICENSE
Changes:
PR: ports/157772
Submitted by: Matthew Seaman <m.seaman@infracaninophile.co.uk> (maintainer)
- Get Rid MD5 support
Fix WWW in pkg-descr to<MODULE> for unification.
No functional changes.
Sponsored by: p5 namespace
- Chase security/libksba shlib version bump
Requested by: kwm
Pointyhat to: glarkin138
Submitted by: Matthew Seaman <m.seaman@infracaninophile.co.uk> (maintainer)
- bump all port that indirectly depends on libjpeg and have not yet been bumped
or updated
Requested by: edwin
- Add pkg-message
PR: ports/135881
Submitted by: Matthew Seaman <m.seaman@infracaninophile.co.uk> (maintainer)
-.03
PR: 133990
Submitted by: Matthew Seaman <m.seaman@infracaninophile.co.uk> (Maintainer)
- Fix plist
- Bump portrevision
Reported by: pointyhat (via pav)
- Add OPTION to switch between dependency on www/rt36 or www/rt38
(www/rt38 is now the default)
- Rename pkg-plist to pkg-plist.rt36
- Add new pkg-plist.rt38
PR: ports/130083
Submitted by: Matthew Seaman <m.seaman@infracaninophile.co.uk> (maintainer)
- Fix build
Submitted by: maintainer via private mail
Reported by: ionbot
RT's extension that allows you to automate due dates using service levels.
WWW:
PR: ports/126779 | https://www.freshports.org/www/p5-RT-Extension-SLA/ | CC-MAIN-2019-43 | refinedweb | 402 | 59.4 |
I have a fairly simple usecase for an enhancement request from which a lot of users can benefit.
Use case:
* Apache HTTPd (2.4.33) <====> Tomcat (8.5.30) via mod_proxy
* Apache logs with CustomLog ... common
* VirtualHost does not only proxy Tomcat, also hosts other unrelated apps (.e.g, Subversion), so changing the log format is not an option
* Tomcat performs authentication
* Apache logs the requests, but remote_user column is empty. This is ugly and I do not really want duplicate logging, i.e., on both sides or if both need to be consistent.
Thanks to rjung@ and jim@ I worked out a solution which does a nice job.
httpd-tomcat.conf:
> <Location "/app">
> ProxyPreserveHost On
> ProxyPass ..
> ProxyPassReverse ..
> RequestHeader set X-Forwarded-Proto "https"
> Header note X-Remote-User REMOTE_USER
> LuaHookLog /usr/local/etc/apache24/register_remote_user.lua register_remote_user
> </Location>
register_remote_user.lua:
> require 'apache2'
>
> function register_remote_user(r)
> local remote_user = r.notes["REMOTE_USER"]
> if remote_user ~= nil then
> r.user = remote_user
> -- not implemented in mod_lua
> -- r. end
> return apache2.OK
> end
On the Tomcat side I have added:
> public class ResponseRemoteUserValve extends ValveBase {
>
> @Override
> public void invoke(Request request, Response response) throws IOException, ServletException {
> String remoteUser = request.getRemoteUser();
>
> if (remoteUser != null) {
> response.setHeader("X-Remote-User", remoteUser);
> }
>
> getNext().invoke(request, response);
> }
>
> }
Ideally for request#getAuthType() to X-Remote-AuthType too. I think this is suitable for either AuthenticatorBase or RemoteIPValve.
One glitch: "Header unset X-Remote-User" is missing from the config.
Seems reasonable.
Care to prepare a patch, including javadoc + XML/HTML documentation?
(In reply to Christopher Schultz from comment #2)
> Seems reasonable.
>
> Care to prepare a patch, including javadoc + XML/HTML documentation?
The patch isn't an issue. I'd like to assess where (classwise) it fits best.
Well, there doesn't seem to be a need to implement this as a Valve (unless I'm missing something important), so let's implement it as a Filter.
The other filters Tomcat provides are all in the org.apache.catalina.filters package. The class name you have now seems awkward, but I don't have a better idea for it.
It would be nice to be able to set the header field-names, and enable either/or X-Remote-User and X-Remote-AuthType.
I wonder how hard it would be to rewrite something like mod_headers in Java. Similar to our RewriteValve that mimics mod_rewrite. That would be more flexible, but we would need to find a good place to put the config for the headers valve or filter. The RewriteValve uses it's own rewrite.config due to the goal of config compatibility with httpd, but mod_headers config syntax is much simpler, so maybe it can be transformed to xml style without getting to ugly.
Just an idea...
I should have added, how such a headers filter or valve would then be used:
Header set X-Remote-User %{REMOTE_USER}
That would be httpd syntax, it could be adjusted for our uses. Also %{REMOTE_USER} is httpd syntax and also used by our own RewriteValve, but we could instead use something else.
(In reply to Christopher Schultz from comment #4)
> Well, there doesn't seem to be a need to implement this as a Valve (unless
> I'm missing something important), so let's implement it as a Filter.
That is true, but opted for Valve because I can phyically register it *after* my authenticator (in context.xml) guaranteeing that auth has actually happened. I had it at Host level and it did not work of course.
Why do we need that actually separately? Why not add it to AuthenticatorBase? That seems to be perfect.
> It would be nice to be able to set the header field-names, and enable
> either/or X-Remote-User and X-Remote-AuthType.
Agreed.
(In reply to Christopher Schultz from comment #4)
Oh, I forgot. Here is the code I have put on the server now:
Fixed in:
- master for 9.0.23 onwards
- 8.5.x for 8.5.44 onwards
- 7.0.x for 7.0.97 onwards | https://bz.apache.org/bugzilla/show_bug.cgi?id=62496 | CC-MAIN-2021-17 | refinedweb | 673 | 60.11 |
Opened 8 years ago
Closed 8 years ago
Last modified 8 years ago
#10017 closed (fixed)
auth.forms.PasswordResetForm.clean_email returns None
Description
Classes inheriting from
PasswordResetForm will be unable to access
self.cleaned_data['email'] without a stub
clean_email() to return a value:
def clean_email(self): super(MyPasswordResetForm, self).clean_email() return self.cleaned_data['email']
Attachments (2)
Change History (9)
Changed 8 years ago by
comment:1 Changed 8 years ago by
comment:2 Changed 8 years ago by
comment:3 Changed 8 years ago by
Not ready for checkin - requires a test (django/contrib/auth/tests/forms.py). There aren't any tests for the PasswordResetForm at the moment (which is probably why this bug slipped through the cracks), but that's no excuse for not adding them when we have the chance.
comment:4 Changed 8 years ago by
Note that there are tests for the password reset views, and for the PasswordResetTokenGenerator, so in the tests for the PasswordResetForm the scope can be kept nice and narrow (i.e. I wouldn't worry about testing the 'save' method).
Changed 8 years ago by
comment:5 Changed 8 years ago by
I've attached several (possibly gratuitous) tests for PasswordResetForm, including one for this bug.
This would be fantastic to get checked in.
Seems an obvious bug; it just caused us problems - setting to Accepted, I'd suggest this is ready for checkin? | https://code.djangoproject.com/ticket/10017 | CC-MAIN-2016-40 | refinedweb | 234 | 60.75 |
hello,
why do i get the error/message "int cannot be dereferenced" in bluej?
import java.util.Random;
public class KassaSimulator
{
Type: Posts; User: Wolverine89
hello,
why do i get the error/message "int cannot be dereferenced" in bluej?
import java.util.Random;
public class KassaSimulator
{
whats a profiler?
Hi guys,
What's wrong with my code? I have a Link class and a Linked List class. The Link class contains one random number. The Linked List class creates a new list. The size of that list is up to...
Hi guys,
The code below works. The code does what it should do. But is there a way to make the code more compact. Writing "beautifull code". Someone got some suggestions?
public void...
this was the solution:
item.setAccelerator(KeyStroke.getKeyStroke(
java.awt.event.KeyEvent.VK_S,
java.awt.event.InputEvent.SHIFT_MASK |...
and do you know how i can make an desktop executable? i have my code in eclipse with a main method in the controller class but i want something with a pictogram so i starts like a normal program...
hello any idea how i can set a accelerator for a menu item with shift, mac, letter
item.setAccelerator(KeyStroke.getKeyStroke(
java.awt.event.KeyEvent.VK_S,
java.awt.Event.META_MASK));
this...
hello any idea how i can set a accelerator for a menu item with shift, mac, letter
item.setAccelerator(KeyStroke.getKeyStroke(
java.awt.event.KeyEvent.VK_S,
...
i have a map in the package assets. this map contains the chesspieces .gif files. Next i made a new package in the src folder and imported the assets map in the view package
fucked up the program
the imageicons in chesspiecesview class are null pointers tried to add them but doesnt work. I made a folder in eclipse with the name assets copied the .gif files in this map...
i understand sorry i am slow witted
--- Update ---
one question after i split them how can i add them to the different .setFen() and .setSubscription() methods
this piece of code is working. It reads the lines i have saved in a file but this is the result and puts both lines in the subscription bar. Any idea how i can separate them?
Default...
yes but you have to code this in java right? so the program knows we want to save it that way
--- Update ---
its more like how can i code that? I access the data with infopanel.getFen() and...
Yes i read your posts! But i dont know how to apply that.
I have to use the bufferedwriter right? and if i write something to the file i need infopanel.getFen and infopanel.getDescription to get...
i use a scanner to read next lines in the file and add it to a string but how can i split the subscription from the fen
--- Update ---
private void openFile(File file){
...
Description
Fen
whats this?
2266
first jtextfield is description right under it jtextfield fen-notation
and my question is how can i design a file layout to handle that can you give code examples need to finish this deadline is tomorrow
and thanks for all you help last week! learned a lot!
Each chessboard is associated with one FEN String, if the file has more boards it also contains more fens. so you say i have to save the FEN? what about the description
The fen string with description! So i fill in a fen-string with description white-to-win.. that needs to be saved i think. What you suggest is to save the fen and description and when its opened the...
another question.
How can i save the current diagram? All ready have some save en open code with extension cmd but how can i save the chessboard and fen to a file is in a container with...
I have to program this in the controller class wright? the controller class makes the frame and the jpanel with different classes from the view which contains jpanel
i asked but its not necessary to define how many objects are allowed to be made.
Another question. This is the program single window with fen, board en some buttons. this represents one FEN...
thanks, another question is there a way to define how many instances of a specific object are allowed to be made.
Like there may only be one WHITE KING K object. Only 8 BLACK PAWN P objects are...
To check if two Strings are the same, you can use .equals(); Is there a similair methode for type char (Character) to check if two char are the same | http://www.javaprogrammingforums.com/search.php?s=e15108ad0bfcb241fadea939dae9efe4&searchid=1627796 | CC-MAIN-2015-27 | refinedweb | 767 | 83.96 |
Web Sockets with server side logic (2)
A few days ago I posted an update to my websocket chat demo that talked about associating a CFC with the web socket to perform server side operations. While testing the chat, a user (hope he reads this and chimes in to take credit) noted another security issue with the code. I had blogged on this topic already, specifically how my chat handler was escaping HTML but could be bypassed easily enough. The user found another hole though. Let’s examine it, and then I’ll demonstrate the fix.
When the chat button is pressed, the following code is run:
$("#sendmessagebutton").click(function() {
var txt = $.trim($("#newmessage").val());
if(txt == "") return;
msg = {
type: "chat",
username: username,
chat: txt
};
chatWS.publish("chat",msg);
$("#newmessage").val("");
});
I've removed my HTML escaping code since the server handles it. But pay attention to the message payload. It contains a type, a username, and a chat. The username value is set after you sign in. It's a simple global JavaScript variable. It's also trivial to modify. Just create your own structure and pass it to the web socket object:
chatWS.publish("chat", {type:"chat",username:"Bob Newhart", chat:"Howdy"});
The server will gladly accept that and pass it along to others. Not good. Luckily there is a simple enough fix for this. My first change was to remove the username from the packet completely.
$("#sendmessagebutton").click(function() {
var txt = $.trim($("#newmessage").val());
if(txt == "") return;
msg = {
type: "chat",
chat: txt
};
chatWS.publish("chat",msg);
$("#newmessage").val("");
});
If you remember, we had a CFC associated with our web socket that was handling a variety of tasks. One of them supported stripping HTML. Here is the original method:
public any function beforeSendMessage(any message, Struct subscriberInfo) {
if(structKeyExists(message, "type") && message.type == "chat") message.chat=rereplace(message.chat, "<.*?>","","all");
return message;
}
Notice the second argument we didn't use? This a structure of data associated with the client. We modified this a bit on our initial subscription to include our username. That means we can make use of it again:
message["username"]=arguments.subscriberInfo.userinfo.username;
This will now get returned in our packet. Check it our yourself below. I've included a zip of the code. (And this is my last chat demo. Honest.)
OOPS!
Turns out I have a critical mistake in my fix, but it's one of those awesome screwups that lead to learning. As soon as I posted my demo, a user noted that his chats were being marked as coming from me. I had no idea why. I then modified my CFC to do a bit of logging:
var myfile = fileOpen(expandPath("./log.txt"),"append");
fileWriteLine(myfile,serializejson(message) & "----" & serializejson(subscriberInfo));
I saw this in the log file:
{"chat":"TestAlphaOne","type":"chat"}----{"userinfo":{"username":"Ray"},"connectioninfo":{"connectiontime":"March, 02 2012 09:56:18","clientid":1511146919,"authenticated":false},"channelname":"chat"}
{"chat":"TestAlphaOne","type":"chat"}----{"userinfo":{"username":"chk"},"connectioninfo":{"connectiontime":"March, 02 2012 09:57:49","clientid":542107549,"authenticated":false},"channelname":"chat"}
At the time of this test, there were two users. My one message was sent out 2 times. So this is interesting. To me, I thought beforeSendMessage was called once, but it's actually called N times, one for each listener. That's kind of cool. It means you could - possibly - stop a message from going to one user. Of course, it totally breaks my code.
One possible fix would be to simply see if USERNAME existed in the message packet. But if I did that, an enterprising hacker would simply supply it.
When I figure this out, I'll post again.
SECOND EDIT Woot! I figured it out. Turns out I should have been using beforePublish. It makes total sense once you remember it exists. It also makes more sense to have my HTML "clean" there too. Things got a bit complex though.
beforePublish is sent a structure too, but it only contains the clientinfo packet. It does not contain the custom information added by the front-end code. I'm thinking this is for security. But, we have the clientid value and we have a server-side function, wsGetSubscribers. If we combine the two, we can create a way to get the proper username:
public any function beforePublish(any message, Struct publisherInfo) {
if(structKeyExists(message, "type") && message.type == "chat") {
//gets the user list, this is an array of names only
var users = getUserList();
var myclientid = publisherInfo.connectioninfo.clientid;
var me = users[arrayFind(wsGetSubscribers('chat'), function(i) {
return (i.clientid == myclientid);
})];
message.chat=rereplace(message.chat, "<.*?>","","all");
message["username"]=me;
}
return message;
}
Does that logic make sense? Basically we are just comparing clientids. I've restored the demo so have at it! | https://www.raymondcamden.com/2012/03/02/Web-Sockets-with-server-side-logic-2 | CC-MAIN-2018-22 | refinedweb | 792 | 60.51 |
Ok, in this assignment I have to write the code for a game that involves picking up and putting down stones in the right combination in order to unlock the treasure chest.
We've been advised that it only needs 4 classes i.e Player, Game, TreasureChest and Stones
My problem comes when I try to get the Game class to run a toString method that when it's finished should return something like this "Jack(0) #(1) @(2) %(3) $(4) Treasure/6\(5)"
This is what my code looks like for the Game class so far:
Code Java:
import java.util.Scanner; public class Game { private int maxMoves; private int combination; private String name; private TreasureChest chest; private Player player; public Game() { Game game1 = new Game(); Scanner keyboard = new Scanner(System.in); System.out.print("Enter combination for the treasure chest (5-10): "); int combination = keyboard.nextInt(); keyboard.nextLine(); chest = new TreasureChest(combination); System.out.print("Enter maximum allowed moves: "); int maxMoves = keyboard.nextInt(); keyboard.nextLine(); System.out.print("Enter player name: "); String name = keyboard.next(); keyboard.nextLine(); game1.toString(); } public String toString() { return name + ""; } }
I still need to add the other information that the toString method is supposed to return, but I can't get it to return even the name. When I try to initialise the Game class it is supposed to ask for the combination, max moves and player name. Instead I get the error "java.lang.StackOverflowError:null"
Any pointers? | http://www.javaprogrammingforums.com/%20whats-wrong-my-code/11041-troubles-tostring-static-non-static-printingthethread.html | CC-MAIN-2014-15 | refinedweb | 245 | 61.87 |
I wanted to know how I can make my own custom 2 dimensional enum array class. I found a similar article online and here's what I've come up with:
public enum myEnum { value1, value2, value3, value4, value5 };
[System.Serializable]
public class myClass {
public int columns = 9;
public int rows = 9;
public myEnum[] myArray = new myEnum [ columns * rows ];
public myEnum this [ int x, int y ] {
get {
return myArray [y * columns + x];
} set {
myArray [y * columns + x] = value;
}
}
}
Now I can call my array like this:
myClass myNewClass = new myClass ();
myNewClass [2,5] = myEnum.value3 // Set value
Instead of using myNewClass [2,5] I wanted to use myNewClass.myArray [2,5]. How do I go about doing that? I don't want to use Unity's original multidimensional method since it doesn't serialize in the inspector.
myNewClass [2,5]
myNewClass.myArray [2,5]
Answer by Bunny83
·
Mar 31, 2018 at 08:31 PM
That's not possible unless your "myArray" is actually a custom class / struct which provides an indexer. Indexer can only be defined in classes / structs.
You can create a class like this:
[System.Serializable]
public class Serializable2DimArray<T>
{
private int m_Columns = 1;
[SerializeField]
private T[] m_Data;
public T this [ int aCol, int aRow ]
{
get { return m_Data[aRow * m_Columns + aCol]; }
set { m_Data[aRow * m_Columns + aCol] = value; }
}
public int Columns {get { return m_Columns; }}
public int Rows { get { return m_Data.Length / m_Columns; }}
public Serializable2DimArray(int aCols, int aRows)
{
m_Columns = aCols;
m_Data = new T[aCols * aRows];
}
}
With this struct you can simply declare a serializable 2-dimensional array for any type. Something like this:
[System.Serializable]
public class MyEnum2DimArray : Serializable2DimArray<myEnum> { }
This "type" can be used in any other serializable class just like a normal 2 dim array. The only exception is that you would create an array with "method brackets" instead of index brackets:
public MyEnum2DimArray myArray = new MyEnum2DimArray(2, 3);
Of course when you want to edit this array in the inspector you most likely need a propertydrawer for this class. The array length always has to be a multiple of the column count. Though it's not clear for what purpose you actually need this 2dim array. It's difficult to actually display a 2dim array if you have a large column count
That's good knowledge. I'm just curious to see how Unity sets up their 2D array and how I can customize my own. I've already set up my own table in the inspector for such a class. I'm just wondering how unity does1 People are following this question.
Multiple Cars not working
1
Answer
Distribute terrain in zones
3
Answers
Serialize custom multidimensional array from Inspector
3
Answers
C# how to setup a Binary Serialization
3
Answers
How do you use serialized values at creation?
1
Answer | https://answers.unity.com/questions/1487794/how-to-make-custom-2-dimensional-enum-array-class.html | CC-MAIN-2019-39 | refinedweb | 466 | 51.48 |
C++ Tutorial - Functors(Function Objects) - 2017
Functors (Function Objects or Functionals) are simply put object + ().
In other words, a functor is any object that can be used with () in the manner of a function.
This includes normal functions, pointers to functions, and class objects for which the () operator (function call operator) is overloaded, i.e., classes for which the function operator()() is defined.
Sometimes we can use a function object when an ordinary function won't work. The STL often uses function objects and provides several function objects that are very helpful.
Function objects are another example of the power of generic programming and the concept of pure abstraction. We could say that anything that behaves like a function is a function. So, if we define an object that behaves as a function, it can be used as a function.
For example, we could define a struct names absValue that encapsulates the operation of converting a value of type float to its absolute valie:
#include <iostream> struct absValue { float operator()(float f) { return f > 0 ? f : -f; } }; int main( ) { using namespace std; float f = -123.45; absValue aObj; float abs_f = aObj(f); cout << "f = " << f << " abs_f = " << abs_f << endl; return 0; }
Output is:
f = -123.45 abs_f = 123.45
As we see from the definition, even though aObj is an object and not a function, we were able to make a call on that object. The effect is to run the overloaded call operator defined by the object absValue. The operator takes a float value and returns its absolute value. Note that the function-call operator must be declared as a member function.
So, Objects of class types, like the absValue object, that define the call operator are referred to as function object.
Let's look at another example simulating a line. The class object is working as a functor taking x for a given y-intercept(b) and slope(a) giving us the corresponding y coordinate.
y = ax + b
Here is the code:
#include <iostream> using namespace std; class Line { double a; // slope double b; // y-intercept public: Line(double slope = 1, double yintercept = 1): a(slope),b(yintercept){} double operator()(double x){ return a*x + b; } }; int main () { Line fa; // y = 1*x + 1 Line fb(5.0,10.0); // y = 5*x + 10 double y1 = fa(20.0); // y1 = 20 + 1 double y2 = fb(3.0); // y2 = 5*3 + 10 cout << "y1 = " << y1 << " y2 = " << y2 << endl; return 0; }
Here y1 is calculated using the expression 1 * 20 + 1 and y2 is calculated using the expression 5 * 3 + 10. In the expression a *x + b, the values for b and a come from the constructor for the object, and the value of x comes from the argument to operator()().
Now that we have a little bit of taste for functor, let's step back and think. So, what's the behavior of a function? A functional behavior is something that we can call by using parentheses and passing arguments. For example:
func(arg1, arg2);
If we want objects to behave this way, we have to make it possible to call them by using parentheses and passing arguments. It's not that difficult. All we have to do is to define operator () with the appropriate parameter types:
Class X { public: // define "function call" operator return-value operator() (arguments) const; ... };
Then we can use object of this class to behave as a function that we can call:
X fn; ... fn(arg1, arg2); // call operator () for function object fn
This call is equivalent to:
fn.operator()(arg1,arg2); // call operator () for function object fn
Here is a function object example.
#include <iostream> #include <vector> #include <algorithm> using namespace std; class Print { public: void operator()(int elem) const { cout << elem << " "; } }; int main () { vector<int> vect; for (int i=1; i<10; ++i) { vect.push_back(i); } Print print_it; for_each (vect.begin(), vect.end(), print_it); cout << endl; return 0; }
The for_each function applied a specific function to each member of a range:
for_each (vect.begin(), vect.end(), print_it);
In general, the 3rd argument could be a functor, not just a regular function. Actually, this raises a question. How do we declare the third argument? We can't declare it as a function pointer because a function pointer specifies the argument type. Because a container can contain just about any type, we don't know in advance what particular type should be used. The STL solves that problem by using template.
The class Print defines object for which we can call operator () with an int argument. The expression
print_itin the statement
for_each (vect.begin(), vect.end(), print_it);creates a temporary object of this class, which is passed to the for_each() algorithm as an argument. The for_each algorithm looks like this:
template<class Iterator, class Function> Function for_each(Iterator first, Iterator last, Function f) { while (first != last) { f(*first); ++first; } return f; }for_each uses the temporary function f to call f(*first) for each element first. If the third parameter is an ordinary function, it simply calls it with *first as an argument. If the third parameter is a function object, it calls operator() for the function object f with *first as an argument. Thus, in this example for_each() calls:
print_it::operator()(*first)The output is:
1 2 3 4 5 6 7 8 9
Here are some advantages of function object listed in "The C++ Standard Library" by Nicolai M. Josuttis.
- Function object. ....
-.
A template nontype parameter is a constant value inside the template definition. We can use a nontype parameter when we need constant expressions. In the example, we need that to specify the size of an array. So, when arrayInit is called, the compiler figures out the value of the nontype parameter from the array argument.
Here is an example using integer as a non-type parameters. In the code below, we want to assign values for all the elements of a vector. So, we set the template parameter with an integer value as non-type, and as we traverse each element of the vector container, we set the value we want by calling setValue().
#include <iostream> #include <vector> #include <algorithm> using namespace std; template <int val> void setValue(int& elem) { elem = val; } template <typename T> class PrintElements { public: void operator()(T& elm) const {cout << elm << ' ';} }; int main() { int size = 5; vector<int> v(size); PrintElements<int> print_it; for_each(v.begin(), v.end(), print_it); for_each(v.begin(), v.end(), setValue<10>); for_each(v.begin(), v.end(), print_it); return 0; }
Output:
0 0 0 0 0 10 10 10 10 10
Here is a similar example but this time it's adding some value to each element of the vector at runtime:
#include <iostream> #include <vector> #include <algorithm> using namespace std; template <typename T> class Add { T x; public: Add(T xx):x(xx){} void operator()(T& e) const { e += x; } }; template <typename T> class PrintElements { public: void operator()(T& elm) const { cout << elm << ' ';} }; int main() { int size = 5; vector<int> v; for(int i = 0; i < size; i++) v.push_back(i); PrintElements<int> print_it; for_each(v.begin(), v.end(), print_it); cout << endl; for_each(v.begin(), v.end(), Add<int>(10)); for_each(v.begin(), v.end(), print_it); cout << endl; for_each(v.begin(), v.end(), Add<int>(*v.begin())); for_each(v.begin(), v.end(), print_it); return 0; }
Output:
0 1 2 3 4 10 11 12 13 14 20 21 22 23 24
The first call to Add(10) creates a Add type temporary object which is initialized with x = 10. Then, it adds its member value 10 to each of the element of the vector v while traversing. The second call to Add(*v.begin()) sets the member of the temporary object to v.begin() element, and then adds that value to each of the element of the vector. applied to one container element at a time.
A special auxiliary function for algorithm is a predicate. Predicates are functions that return a Boolean value (or something that can be implicitly converted to bool). In other words, a predicate class is a functor class whose operator() function is a predicate, i.e., its operator() returns true or false.
Predicates are widely used in the STL. The comparison functions for the standard associative containers are predicates, and predicate functions are commonly passed as parameters to algorithms like find_if. Depending on their purpose, predicates are unary or binary.
- A unary function that returns a bool value is a predicate.
- A binary function that returns a bool value is a binary predicate.
A typical example of predefined function object in STL is set. It sorts element in increasing order because by default it's using less(<) sorting criteria. So, when we do:
set<int> mySet;
internal code looks like this:
set<int, less<int> > mySet;
So, we want to store our elements in decreasing order in a set, we do:
set<int, greater<int> > mySet;
In the example below, the negate<int>() creates a function object from the predefined template class negate. It simply returns the negated int element.
The second transform() algorithm combines the elements from two vector containers by using multiplies<int>() operation. Then it writes the results into the 3rd vector container.
#include <iostream> #include <vector> #include <algorithm> using namespace std; template <typename T> class PrintElements { public: void operator()(T& elm) const { cout << elm << ' ';} }; int main() { PrintElements<int> print_it; int size = 5; vector<int> v; for(int i = 0; i < size; i++) v.push_back(i); for_each(v.begin(), v.end(), print_it); cout << endl; transform(v.begin(), v.end(), // source v.begin(), // destination negate<int>()); // operation for_each(v.begin(), v.end(), print_it); cout << endl; transform(v.begin(), v.end(), // source v.begin(), // second source v.begin(), // destination multiplies<int>()); // operation for_each(v.begin(), v.end(), print_it); cout << endl; return 0; }
Output:
0 1 2 3 4 0 -1 -2 -3 -4 0 1 4 9 16
Here are the two types of transform() algorithms:
- Transforming elements
OutputIterator transform ( InputIterator source.begin(), InputIterator source.end(), OutputIterator destination.begin(), UnaryFunc op )
- Combining elements of two sequences
OutputIterator transform ( InputIterator1 source1.begin(), InputIterator1 source1.end(), InputIterator2 source2.begin() OutputIterator destination.begin(), BinaryFunc op )
Function adapter converts some other interface to an interface used by STL.
Here is the declaration of bind2nd():
template <class Operation, class T> binder2nd<Operation> bind2nd (const Operation& op, const T& x);
It returns function object with second parameter bound. This function constructs an unary function object from the binary function object op by binding its second parameter to the fixed value x.
The function object returned by bind2nd() has its operator() defined such that it takes only one argument. This argument is used to call binary function object op with x as the fixed value for the second argument.
#include <iostream> #include <vector> #include <algorithm> #include <iterator> using namespace std; struct PrintElm { void operator()(int & elm) const { cout << elm << ' ';} }; int main() { int size = 10; vector<int> v; for(int i = 0; i < size; i++) v.push_back(i); for_each(v.begin(), v.end(), PrintElm()); cout << endl; replace_if(v.begin(), v.end(), bind2nd(equal_to<int>(),0), 101); for_each(v.begin(), v.end(), PrintElm()); cout << endl; v.erase( remove_if(v.begin(), v.end(), bind2nd(less<int>(), 3)), v.end() ); for_each(v.begin(), v.end(), PrintElm()); cout << endl; transform(v.begin(), v.end(), ostream_iterator<int>(cout, " "), negate<int>()); return 0; }
Output:
0 1 2 3 4 5 6 7 8 9 101 1 2 3 4 5 6 7 8 9 101 3 4 5 6 7 8 9 -101 -3 -4 -5 -6 -7 -8 -9
Here are additional examples of bind2nd():
- To count all the elements within a vector that are less than or equal to 100, we can use count_if():
count_if(vec.begin(), vec.end(), bind2nd(less_equal<int>(), 100));The 3rd argument uses bind2nd() function adaptor. This adaptor returns a function object that applies the <= operator using 100 as the right hand operand. So, the call to count_if() counts the number of elements in the input range (from vec.begin() to vec.end()) that are less than or equal to 100.
- We can negate the binding of less_equal:
count_if(vec.begin(), vec.end(), not1(bind2nd(less_equal<int>(), 100)));As in the previous sample, we first bind the second operand of the less_equal object to 100, transforming that binary operation into a unary operation. Then, we negate the return from the operation using not1. So, what it does is that each element will be tested to see if it is <= 100. Then, the truth value of the result will be negated. Actually, the call counts those elements that are not <= 100.
Here is another sample of using bind2nd() with transform(). The code multiplies 10 to each element of an array and outputs them in Print() function:
#include <iostream> #include <iterator> #include <algorithm> using namespace std; template <typename ForwardIter> void Print(ForwardIter first, ForwardIter last, const char* status) { cout << status << endl; while(first != last) cout << *first++ << " "; cout << endl; } int main() { int arr[] = {1, 2, 3, 4, 5}; Print(arr, arr+5, "Initial values"); transform(arr, arr+5, arr, bind2nd(multiplies<int>(), 10)); Print(arr, arr+5, "New values:"); return 0; }
Output:
$ g++ -o bind bind.cpp $ ./bind Initial values 1 2 3 4 5 New values: 10 20 30 40 50 | http://www.bogotobogo.com/cplusplus/functors.php | CC-MAIN-2017-34 | refinedweb | 2,212 | 55.74 |
This class safely opens a temporary file in TEMP_DIR (usually /tmp). More...
#include <l_stdio_wrap.h>
This class safely opens a temporary file in TEMP_DIR (usually /tmp).
This class is a wrapper around Unix system call mkstemp which securely generates a unique temporary filename and opens the file too, so that there is no race condition.
This object deletes its underlying file only when the object is destroyed. If the program aborts, the temporary file might not be deleted. In that case, you might want to clean up files called /tmp/libkjb-XXXXXX manually, where XXXXXX denotes any arbitrary combination of six printable characters.
Example usage:
close file before destruction; rarely needed; safe to do twice.
Reimplemented from kjb::File_Ptr. | http://kobus.ca/research/resources/doc/doxygen/classkjb_1_1Temporary__File.html | CC-MAIN-2022-21 | refinedweb | 119 | 65.32 |
Odoo Help
This community is for beginners and experts willing to share their Odoo knowledge. It's not a forum to discuss ideas, but a knowledge base of questions and their answers.
How to pre-fill the tax in sale order lines before adding a product?
As a service company we do not use products and therefore we cannot predefine the tax in order lines which is usually defined by the product. I wrote a simple extension to set the tax on sale order line on creation:
class SaleOrderLine(osv.osv):
_inherit = 'sale.order.line'
def create(self, cr, uid, values, context=None):
tax = self.pool.get('account.tax').browse(cr, uid, 12, context=context)
values['tax_id'] = [[6, 0, [tax.id]]]
return super(SaleOrderLine, self).create(cr, uid, values, context=context)
My questions are:
1.) How can I obtain the default sales tax from account configuration within the SaleOrderLine extension?
2.) At the moment the tax will be set after saving the sale order. Is there a way to pre-fill the tax field on creatrion of an order line in the frontend (Link: Add Entry)? I presume this must be done with JavaScript but I do not know how to obtain the default sales tax there and how to add a default value to the tax field.
Thanks for your help!
Great! that answers my second question.
But how can I access the default sales tax from account configurations?
Got the answer myself:
ir_values = self.pool.get('ir.values')
taxes_id = ir_values.get_default(cr, uid, 'product.product', 'taxes_id', company_id=company_id)! | https://www.odoo.com/forum/help-1/question/how-to-pre-fill-the-tax-in-sale-order-lines-before-adding-a-product-87591 | CC-MAIN-2016-50 | refinedweb | 260 | 60.31 |
Here we will see how to get the differences between first and last X digits of a number N. The number and X are given. To solve this problem, we have to find the length of the number, then cut the last x digits using modulus operator. After that cut all digits from the number except first x digits. Then get the difference, and return the result. Let the number is N = 568424. The X is 2 so first two digits are 56, and last two digits are 24. The difference is (56 - 24) = 32.
begin p := 10^X last := N mod p len := length of the number N while len is not same as X, do N := N / 10 len := len -1 done first := len return |first - last| end
#include <iostream> #include <cmath> using namespace std; int lengthCount(int n){ return floor(log10(n) + 1); } int diffFirstLastDigits(int n, int x) { int first, last, p, len; p = pow(10, x); last = n % p; len = lengthCount(n); while(len != x){ n /= 10; len--; } first = n; return abs(first - last); } main() { int n, x; cout << "Enter number and number of digits from first and last: "; cin >> n >> x; cout << "Difference: " << diffFirstLastDigits(n,x); }
Enter number and number of digits from first and last: 568424 2 Difference: 32 | https://www.tutorialspoint.com/absolute-difference-between-the-first-x-and-last-x-digits-of-n | CC-MAIN-2020-10 | refinedweb | 216 | 76.05 |
How would I code a reversible shuffle algorithm in C# ArrayList which uses a key to shuffle and can be reversed to the original state?
private class myItem { public Int32 ID { get; set; } public Int32 Rand { get; set; } public String Value { get; set; } } private void Main(object sender, EventArgs e) { string[] array = new string[] { "alpha", "beta", "gamma", "delta" }; List<myItem> myArray = addKeys(array);//adds keys to array required for randomization and reversing string[] randomarray = randomize(myArray); string[] reversedarray = reverse(myArray); } private List<myItem> addKeys(string[] array) { Random random = new Random(); List<myItem> myArr = new List<myItem>(); for (int i = 0; i < array.Length; i++) { myArr.Add(new myItem { ID = i, Rand = random.Next(), Value = array[i] }); } return myArr; } private string[] randomize(List<myItem> myArray) { return (from item in myArray orderby item.Rand select item.Value).ToArray(); } private string[] reverse(List<myItem> myArray) { return (from item in myArray orderby item.ID select item.Value).ToArray(); }
You can use KeyValuePair. You can get a detailed explanation here on the following site:
However as you want to reverse it, I suggest using a DataTable. In first column you can have original sort order (integers may be), in second column you can have random numbers created. In third column you can have the values you need to sort. Then use the sorting by second column to shuffle and then sort it by first column to revert it.
Shuffling an array is done by swapping the elements inside the array several times.
If you take an array, swap the elements, then swap them again, but in the reverse order you swapped them initially, you get the original array back.
These shuffling positions can be generated from
PRNG (Pseudorandom number generator). For a given
seed, a
PNRG will always generate the same numbers : that will be your "key".
Here is some code using that idea (Fisher Yates is used for the shuffle):
public static void Shuffle(ArrayList list, int seed, bool reverse = false) { Random rng = new Random(seed); List<Tuple<int, int>> swaps = new List<Tuple<int, int>>(); int n = list.Count; //prepare swapping positions while (n > 1) { n--; int k = rng.Next(n + 1); swaps.Add(new Tuple<int, int>(k, n)); } //reverse if needed if(reverse) swaps.Reverse(); //swap the items foreach(Tuple<int, int> swap in swaps) { object value = list[swap.Item1]; list[swap.Item1] = list[swap.Item2]; list[swap.Item2] = value; } }
Usage :
ArrayList a = new ArrayList(); int key = 2742; Shuffle(a, key); Shuffle(a, key, true);
Note that if you don't want the key to be an integer, but a string, you can use exact same method but the integer using
"somekey".GetHashCode()
Here is a simple way that only requires you to store your "key", as well as not change the shuffling algorithm.
It does not matter which shuffle algorithm you use as long as you can reproduce it, meaning that if you use the same key or seed, and run the shuffle algorithm twice, it will shuffle the exact same way both times.
If you can do that, using a key, or a seed to your shuffle algorithm, then you're good to go.
All you have to do in order to reverse it is to shuffle an array of integers, where the values in the array before the shuffle corresponds to the index of each element. After shuffling, with the same key or seed as above, you now have a shuffled copy where each value is the index of the original placement in the unshuffled array.
Let me give you an example.
Let's assume an array of 4 integers, values 0 through 3 (element at index 0 has a value of 0, element at index 1 has a value of 1, and so on).
Shuffle this, to obtain this array: 3, 1, 2, 0.
This now means that the first element of the shuffled array used to be at index 3, the second element was at index 1, the third at index 2, and the fourth at index 0. | http://www.dlxedu.com/askdetail/3/75d16ad81d9a6d303293d09b01355458.html | CC-MAIN-2019-13 | refinedweb | 672 | 60.35 |
POSIX-compatibile (sorta) threading support. More...
#include <unistd.h>
#include <sys/_pthread.h>
#include <sys/types.h>
#include <time.h>
#include <sys/sched.h>
Go to the source code of this file.
POSIX-compatibile (sorta) threading support.
This file was imported (with a few changes) from Newlib. If you really want to know about the functions in here, you should probably consult the Single Unix Specification and the POSIX specification. Here's a link to that:
The rest of this file will remain undocumented, as it isn't really a part of KOS proper... Also, doxygen tends to mangle this whole thing anyway... | http://cadcdev.sourceforge.net/docs/kos-2.0.0/pthread_8h.html | CC-MAIN-2018-05 | refinedweb | 103 | 63.05 |
23 April 2012 01:45 [Source: ICIS news]
LONDON (ICIS)--One worker was killed and 21 workers and nearby residents were injured in an explosion and fire at a resorcinol plant at Mitsui Chemicals’ Iwakuni-Ohtake site in ?xml:namespace>
Mitsui Chemicals said the explosion occurred at around 02:15 hours
The fire, which spread to a cymene plant at the site, was brought under control at 17:15 hours, the company said.
The accident caused extensive damage to windows and roofs at 14 plants at the site, the company added.
Mitsui Chemicals said that the 21 injured included seven Mitsui employees, two workers from a company that works for Mitsui, and two workers at an adjacent JX Nippon Oil refinery, as well as 10 community residents. Two of the Mitsui workers suffered serious injuries, the company added.
The cause of the accident is not yet known. | http://www.icis.com/Articles/2012/04/23/9552541/blast-at-mitsui-chemicals-plant-in-japan-kills-one-worker.html | CC-MAIN-2014-42 | refinedweb | 147 | 56.08 |
Let's face it, forms are everywhere across the web, and they often take significant time to build depending on the requirements.
In this tutorial we will dynamically build pages with forms using Next.js, and GraphQL.
Chapters:
- Define a solid content model
- Create the content model in GraphCMS
- Create an example Page and Form with Fields as a content editor
- Reordering form fields
- Query our Page, Form and Fields with GraphQL
- Configure public API access
- Setup Next.js project with dependencies
- Build Pages programatically with Next.js
- Build our Form Field components
- Render our Form to our individual Pages
- Managing form state and submissions
- Submitting our Form to GraphCMS with GraphQL Mutations
- Deploy to Vercel
TLDR;
1. Define a solid content model
Before we dive into creating our schema, let's first think about what we're going to need to enable our marketing team to spin up landing page forms from just using the CMS.
It all starts with a Page. Pages must have a
slug field so we can easily look up content from the params of any request.
Next, for simplicity, each page will have an associated
Form model. For the sake of this tutorial, we'll pick 4 form field types;
- Input
- Textarea
- Checkbox
Form Fields
If we think of a traditional form, let's try and replace all of the data points we need to recreate a simple contact form like the following:
<form> <div> <label for="name">Name</label> <input type="text" id="name" placeholder="Your name" required /> </div> <div> <label for="email">Email</label> <input type="email" id="email" placeholder="Your email" required /> </div> <div> <label for="tel">Tel</label> <input type="tel" id="tel" placeholder="Your contact no." /> </div> <div> <label for="favFramework">What's your favorite framework?</label> <select id="favFramework"> <option value="react">React</option> <option value="vue">Vue</option> <option value="angular">Angular</option> <option value="svelte">Svelte</option> </select> </div> <div> <label for="message">Message</label> <textarea id="message" placeholder="Leave a message" /> </div> <div> <label for="terms"> <input id="terms" type="checkbox" /> I agree to the terms and privacy policy. </label> </div> <div> <button type="submit">Submit</button> </div> </form>
In the above form, we have some
<input />'s that are required, some which are of type
tel and
text, while the
<select /> has no placeholder or is required.
GraphCMS has support for GraphQL Union Types. This means we can define models for each of our form field types, and associate them to our
Form model as one "has many" field.
Our schema will end up looking a little something like the following...
Models
Page
- Title, String, Single line text, Required, and used as a Title
- Slug, String, Single line text, Required
- Form, Reference to
Form
Form
- Page, Reference, Accepts multiple
Pagevalues
- Fields, Reference, Accepts multiple
FormInput,
FormTextarea,
FormSelectand
FormCheckboxvalues
FormInput
- Name, String, Single line text, and used as a Title
- Type, Enum,
FormInputTypedropdown
- Label, String, Single line text
- Placeholder, Single line text
- Required, Boolean
- Form, Reference to
Form
FormTextarea
- Name, String, Single line text, and used as a Title
- Label, String Single line text
- Placeholder, String, Single line text
- Required, Boolean
- Form, Reference to
Form
FormSelect
- Name, String, Single line text, and used as a Title
- Label, String, Single line text
- Required, Boolean
- Choices, Reference, Accepts multiple
FormOptionvalues
- Form, Reference to
Form
FormOption
- Value, String, Single line text, Required, and used as a Title
- Option, String, Single line text
- FormSelect, Reference, Belongs to
FormSelect
FormCheckbox
Name, String, Single line text, and used as a Title
Label, String, Single line text, Required
Required, Boolean
Form, Reference to
Form
Enumerations
FormInputType values
TEXT
TEL
🖐 You could add more, but it's not required for this tutorial.
2. Create the models in GraphCMS
Now we have an idea of how our content model looks like. Let's create the models and their associations with eachother inside GraphCMS.
You'll need an account to continue. Sign up or head to the Dashboard.
Once logged in, head to the Schema editor by selecting Schema from the side.
Click + Add in the sidebar above default system
Assetmodel.
Go ahead and create the 7 models above. Don't worry about creating relations just yet, you can do them all at once after creating the other fields.
3. Create an example Page and Form with Fields as a content editor
So that we are able to query, and build our forms, we're going to need some content inside our models.
- Inside the Dashboard, head to the Content editor by selecting Content from the side.
- Select the Page model and click + Create New from the top right.
- Give your page a
titleand
slug. I'll call use
contact, respectively.
- Now underneath
Form, click Create and add a new form.
- Inside the inline
Formcontent editor, click on Create and add a new document.
- From the dropdown, select FormInput.
- Inside the inline
FormInputcontent editor, enter a
name,
type,
labeland
placeholderfor your form field. I'll add the values
Name,
TEXT,
Your name,
Nameand set required to
true.
- Now click Save and publish.
Repeat steps 5-8 to add additional fields.
🖐 To follow along with the rest of this tutorial, I will be using the following values for my fields...
3 x
FormInput's
Name
- Name:
name
- Type:
TEXT
- Label:
Name
- Placeholder:
Your name
- Required:
true
- Name:
- Type:
- Label:
- Placeholder:
Your email
- Required:
true
Tel
- Name:
tel
- Type:
TEL
- Label:
Tel
- Placeholder:
Your contact no.
- Required:
false
1 x
FormTextarea
- Name:
- Label:
- Placeholder:
Leave a message
- Required:
true
1 x
FormCheckbox
- Terms
- Name:
terms
- Label:
I agree to the terms and privacy policy.
- Required:
true
1 x
FormSelect
The
FormSelect is a little special because it also references another model
FormSelect.
First, create your
FormSelect document as usual, entering the following.
- Favourite Framework
- Name:
favFramework
- Label:
What's your favorite frontend framework?
- Required:
false
- Next below Options, click on Create and add a new formOption.
Now for each of our choices below, repeat the steps to "Create and add a new formOption", and provide the
value/
option for each:
react/
React
vue/
Vue
angular/
Angular
svelte/
Svelte
Finally, click Save and publish on this and close each of the inline editors, making sure to publish any unsaved changes along the way.
4. Reordering form fields
Now we have created our fields, we can now reorder them using the content editor. This may be useful if you decide to add or remove some fields later, you can order the fields exactly the way you want them to appear.
✨ Simply drag each of the Field rows into the order you want. ✨
5. Query our Page, Form and Fields with GraphQL
We have two pages, with two separate forms:
- Contact Form
- Request a Demo
Let's start by querying for all pages and their forms using the API Playground available from the sidebar within your project Dashboard.
Query pages, form and field
__typename
{ pages { title slug form { id fields { __typename } } } }
Union Type Query
As we're using Union Types for our form
fields, we must use the
... on TypeName notation to query each of our models.
Let's go ahead and query
on all of our models we created earlier.
{ pages { title slug form { id fields { __typename ... on FormInput { name type inputLabel: label placeholder required } ... on FormTextarea { name textareaLabel: label placeholder required } ... on FormCheckbox { name checkboxLabel: label required } ... on FormSelect { name selectLabel: label options { value option } required } } } } }
The response should look a little something like the following:
{ "data": { "pages": [ { "title": "Contact us", "slug": "contact", "form": { "id": "ckb9j9y3k004i0149ypzxop4r", "fields": [ { "__typename": "FormInput", "name": "Name", "type": "TEXT", "inputLabel": "Name", "placeholder": "Your name", "required": true }, { "__typename": "FormInput", "name": "Email", "type": "EMAIL", "inputLabel": "Email address", "placeholder": "you@example.com", "required": true }, { "__typename": "FormInput", "name": "Tel", "type": "TEL", "inputLabel": "Phone no.", "placeholder": "Your phone number", "required": false }, { "__typename": "FormSelect", "name": "favFramework", "selectLabel": "What's your favorite frontend framework?", "options": [ { "value": "React", "option": "React" }, { "value": "Vue", "option": "Vue" }, { "value": "Angular", "option": "Angular" }, { "value": "Svelte", "option": "Svelte" } ], "required": false }, { "__typename": "FormTextarea", "name": "Message", "textareaLabel": "Message", "placeholder": "How can we help?", "required": true }, { "__typename": "FormCheckbox", "name": "Terms", "checkboxLabel": "I agree to the terms and privacy policy.", "required": true } ] } } ] } }
6. Configure public API access
GraphCMS has a flexible permissions system, which includes enabling certain user groups to do actions, and most importantly restrict who can query what data.
For the purposes of querying data to build our pages and forms, we'll enable public API queries.
To do this, go to your project Settings.
- Open the API Access page
- Enable Content from stage Published under Public API permissions
- Save ✨
That's it! You can test this works using the API Playground and selecting
Environment: master Public from the dropdown in the section above your query/result.
🖐 Make sure to copy your API Endpoint to the clipboard. We'll need it in step 8.
7. Setup Next.js project with dependencies
Now we have our schema, and content, let's begin creating a new Next.js project with all of the dependencies we'll need to build our pages and forms.
Inside the Terminal, run the following to create a new Next.js project.
npm init next-app dynamic-graphcms-forms
When prompted, select
Default starter app from the template choices.
cd dynamic-graphcms-forms
This template will scaffold a rough folder structure following Next.js best practices.
Next, we'll install
graphql-request for making GraphQL queries via fetch.
yarn add -E graphql-request # or npm install ...
Now, if you run the project, you should see the default Next.js welcome page at.
yarn dev # or npm run dev
8. Build Pages programatically with Next.js
This comes in two significant parts. First we create the routes (or "paths") and then query for the data for each page with those path params.
8.1 Create programmatic page routes
First up is to add some code to our Next.js application that will automatically generate pages for us. For this we will be exporting the
getStaticPaths function from a new file called
[slug].js in our
pages directory.
touch pages/[slug].js
Having a filename with square brackets may look like a typo, but rest assured this is a Next.js convention.
Inside
pages/[slug].js add the following code to get going:
export default function Index(props) { return ( <pre>{JSON.stringify(props, null, 2)}</pre> ) }
If you're familiar with React already, you'll notice we are destructuring
props from the
Index function. We'll be updating this later to destructure our individual page data, but for now, we'll show the
props data on each of our pages.
Inside
pages/[slug].js, let's import
graphql-request and initialize a new
GraphQLClient client.
🖐 You'll need your API Endpoint from Step 6 to continue.
import { GraphQLClient } from "graphql-request"; const graphcms = new GraphQLClient("YOUR_GRAPHCMS_ENDOINT_FROM_STEP_6");
Now the
graphcms instance, we can use the
request function to send queries (with variables) to GraphCMS.
Let's start by querying for all pages, and get their slugs, inside a new exported function called
getStaticPaths.
export async function getStaticPaths() { const { pages } = await graphcms.request(`{ pages { slug } }`) return { paths: pages.map(({ slug }) => ({ params: { slug } })), fallback: false } }
There's quite a bit going on above, so let's break it down...
const { pages } = await graphcms.request(`{ pages { slug } }`)
Here we are making a query and destructuring the response
pages from the request. This will be similar to the results we got back in step 5.
return { paths: pages.map(({ slug }) => ({ params: { slug } })), fallback: false }
Finally inside
getStaticPaths we are returning
paths for our pages, and a
fallback. These build the dynamic paths inside the root
pages directory, and each of the slugs will become
pages/[slug].js.
The
fallback is
false in this example, but you can read more about using that here.
🖐
getStaticPaths alone does nothing, we need to next query data for each of the pages.
8.2 Query page data
Now we have programmatic paths being generated for our pages, it's now time to query the same data we did in step 5, but this time, send that data to our page.
Inside
pages/[slug].js, export the following function:
export async function getStaticProps({ params: variables }) { const { page } = await graphcms.request( `query page($slug: String!) { page(where: {slug: $slug}) { title slug form { fields { __typename ... on FormInput { name type inputLabel: label placeholder required } ... on FormTextarea { name textareaLabel: label placeholder required } ... on FormCheckbox { name checkboxLabel: label required } ... on FormSelect { name selectLabel: label options { value option } required } } } } } `, variables ); return { props: { page, }, }; }
Now just like before, there's a lot going on, so let's break it down...
export async function getStaticProps({ params: variables }) { // ... }
Here we are destructuring the
params object from the request sent to our page. The params here will be what we sent in
getStaticPaths, so we'd expect to see
slug here.
🖐 As well as destructuring, we are also renaming (or reassigning) the variable
params to
variables.
const { page } = await graphcms.request(`...`, variables); return { props: { page, }, };
Next we're sending the same query we did in step 5, but this time we've given the query a name
page which expects the
String variable
slug.
Once we send on our renamed
params as
variables, we return an object with our
page inside of
props.
Now all that's left to do is run our Next development server and see our response JSON on the page!
yarn dev # or npm run dev
Now you should see at the data from GraphCMS for our Page.
9. Build our Form Field components
We are now ready to dynamically build our form using the data from GraphCMS.
The
__typename value will come in handy when rendering our form, as this will decide which component gets renderered.
Inside a new directory
components, add a
Form.js file.
mkdir components touch components/Form.js
In this this file, we will create the structure of our basic form, and
map through each of our
fields to return the appropreciate field.
Add the following code to
components/Form.js
import * as Fields from "./FormFields"; export default function Form({ fields }) { if (!fields) return null; return ( <form> {fields.map(({ __typename, ...field }, index) => { const Field = Fields[__typename]; if (!Field) return null; return <Field key={index} {...field} />; })} <button type="submit">Submit</button> </form> ); }
Once you have this component setup, now create the file
components/FormFields/index.js and add the following:
export { default as FormCheckbox } from "./FormCheckbox"; export { default as FormInput } from "./FormInput"; export { default as FormSelect } from "./FormSelect"; export { default as FormTextarea } from "./FormTextarea";
All we're doing in this file is importing each of our different form fields and exporting them.
The reason we do this is that when we import using
import * as Fields, we can grab any of the named exports by doing
Fields['FormCheckbox'], or
Fields['FormInput'] like you see in
components/Form.js.
Now that that we are importing these new fields, we next need to create each of them!
For each of the imports above, create new files inside
components/FormFields for:
FormCheckbox.js
FormInput.js
FormSelect.js
FormTextarea.js
Once these are created, let's export each of the components as default, and write a minimum amount of code to make them work.
The code in the below files isn't too important. What's key about this tutorial is how we can very easily construct forms, and in fact any component or layout, using just data from the CMS. Magic! ✨
FormCheckbox.js
export default function FormCheckbox({ checkboxLabel, ...rest }) { const { name } = rest; return ( <div> <label htmlFor={name}> <input id={name} type="checkbox" {...rest} /> {checkboxLabel || name} </label> </div> ); }
FormInput.js
Since this component acts as a generic
<input />, we will need to lowercase the
type enumeration to pass to the input.
export default function FormInput({ inputLabel, type: enumType, ...rest }) { const { name } = rest; const type = enumType.toLowerCase(); return ( <div> {inputLabel && <label htmlFor={name}>{inputLabel || name}</label>} <input id={name} type={type} {...rest} /> </div> ); }
FormSelect.js
export default function FormSelect({ selectLabel, options, ...rest }) { const { name } = rest; if (!options) return null; return ( <div> <label htmlFor={name}>{selectLabel || name}</label> <select id={name} {...rest}> {options.map(({ option, ...opt }, index) => ( <option key={index} {...opt}> {option} </option> ))} </select> </div> ); }
FormTextarea.js
export default function FormTextarea({ textareaLabel, ...rest }) { const { name } = rest; return ( <div> <label htmlFor={name}>{textareaLabel || name}</label> <textarea id={name} {...rest} /> </div> ); }
We're done on the form components, for now...!
10. Render our Form to our individual Pages
Let's recap...
- We have our form model and content coming from GraphCMS
- We have our form fields created
- We have our form pages automatically created
Let's now render the form we created in step 9 to our page.
Inside
pages/[slug].js, we'll need to import our Form component and return that inside of the default export.
Below your current import (
graphql-request), import our Form component:
import Form from "../components/Form";
Lastly, update the default export to return the
<Form />.
export default function Index({ page }) { const { form } = page; return <Form {...form} />; }
Next run the Next.js development server:
yarn dev # or npm run dev
Once the server has started, head to (or a
slug you defined in the CMS) to see your form!
I'll leave the design and UI aesthetics up to you!
As far as creating dynamic forms with React, Next.js and GraphQL goes, this is it! Next we'll move onto enhancing the form to be accept submissions.
11. Managing form state and submissions
In this step we will install a library to handle our form state, and submissions, as well as create an
onSubmit that'll we'll use in Step 12 to forward onto GraphCMS.
Inside the terminal, let's install a new dependency:
yarn add -E react-hook-form # or npm install ...
Now it's not essential we use
react-hook-form for managing our form, I wanted to provide a little closer to real world scenario than your typical
setState example that are used in tutorials.
After we complete this tutorial, you should be in a position to return to each of your form fields, add some CSS, error handling, and more, made easy with
react-hook-form!
Inside
components/Form.js, add the following import to the top of the file:
import { useForm, FormContext } from "react-hook-form";
Then inside your
Form function after you
return null if there are no
fields, add the following:
const { handleSubmit, ...methods } = useForm(); const onSubmit = (values) => console.log(values);
Finally, you'll need to wrap the current
<form> with
<FormContext {...methods}>, and add a
onSubmit prop to the
<form> that is
onSubmit={handleSubmit(onSubmit)}.
Your final
components/Form.js should look like this:
import { useForm, FormContext } from "react-hook-form"; import * as Fields from "./FormFields"; export default function Form({ fields }) { if (!fields) return null; const { handleSubmit, ...methods } = useForm(); const onSubmit = (values) => console.log(values); return ( <FormContext {...methods}> <form onSubmit={handleSubmit(onSubmit)}> {fields.map(({ __typename, ...field }, index) => { const Field = Fields[__typename]; if (!Field) return null; return <Field key={index} {...field} />; })} <button type="submit">Submit</button> </form> </FormContext> ); }
Now all that's happening here is we're initializing a new
react-hook-form instance, and adding a
FormContext provider around our form + fields.
Next we'll need to update each of our
FormFields/*.js and
register them with the
react-hook-form context.
First update
components/FormFields/FormInput.js to include the hook
useFormContext from
react-hook-form.
At the top of the file add the following import:
import { useFormContext } from 'react-hook-form'
Then inside the
FormInput function, add the following before the
return:
const { register } = useFormContext();
Now all that's left to do add
register as a
ref to our
<input /> and pass in the
required value.
<input ref={register({ required: rest.required })} id={name} type={type} {...rest} />
The final
FormInput should look like:
import { useFormContext } from "react-hook-form"; export default function FormInput({ inputLabel, type: enumType, ...rest }) { const { register } = useFormContext(); const { name } = rest; const type = enumType.toLowerCase(); return ( <div> {inputLabel && <label htmlFor={name}>{inputLabel || name}</label>} <input ref={register({ required: rest.required })} id={name} type={type} {...rest} /> </div> ); }
Great! Now let's do the same for the other 3 field components:
FormCheckbox.js
import { useFormContext } from "react-hook-form"; export default function FormCheckbox({ checkboxLabel, ...rest }) { const { register } = useFormContext(); const { name } = rest; return ( <div> <label htmlFor={name}> <input ref={register({ required: rest.required })} id={name} type="checkbox" {...rest} /> {checkboxLabel || name} </label> </div> ); }
FormSelect.js
import { useFormContext } from "react-hook-form"; export default function FormSelect({ selectLabel, options, ...rest }) { if (!options) return null; const { register } = useFormContext(); const { name } = rest; return ( <div> <label htmlFor={name}>{selectLabel || name}</label> <select ref={register({ required: rest.required })} id={name} {...rest}> {options.map(({ option, ...opt }, index) => ( <option key={index} {...opt}> {option} </option> ))} </select> </div> ); }
FormTextarea.js
import { useFormContext } from "react-hook-form"; export default function FormTextarea({ textareaLabel, ...rest }) { const { register } = useFormContext(); const { name } = rest; return ( <div> <label>{textareaLabel || name}</label> <textarea ref={register({ required: rest.required })} htmlFor={name} id={name} {...rest} /> </div> ); }
🖐 Let's start the Next.js development server, and view the console when we submit the form!
yarn dev # or npm run dev
Once the server has started, head to (or a
slug you defined in the CMS) to see your form!
Open the browser developer tools console, and then fill out the form and click submit!
You should now see the form values submitted!
12. Submitting our Form to GraphCMS with GraphQL Mutations
It's now time to take our form to the next level. We are going to update our GraphCMS schema with a new
Submission model that will be used to store submissions.
Inside the GraphCMS Schema Editor, click + Add to create a new model.
- Give the model a name of
Submission,
- Add a new JSON Editor field with the Display Name
Form Data, and, API ID as
formData,
- Add a new Reference field with the Display Name/API ID
Form/
form, and select
Formas the Model that can be referenced,
- Configure the reverse field to Allow multiple values and set the default Display Name/API ID to (
Submissions/
submissions) respectively.
Things should look a little something like the following:
And the
Form model should now have a new field
submisson:
Since we want full control via the CMS what appears on our form, we'll just save all of that data inside
formData JSON field.
🖐 Using something like webhooks would enable you to forward
formData onto a service like Zapier, and do what you need to with the data, all without writing a single line of code! ✨
In order to use the Mutations API, we'll need to configure our API access to permit mutations and create a dedicated Permanent Auth Token. Don't enable Mutations for the Public API, as anybody will be able to query/mutate your data!
Head to
Settings > API Access > Permanent Auth Tokens and create a token with the following setup:
Copy the token to the clipboard once created.
Inside of the root of your Next.js project, create the file
.env and, add the following, replacing
YOUR_TOKEN_HERE with your token:
GRAPHCMS_MUTATION_TOKEN=YOUR_TOKEN_HERE
With this token added, let's also do some housekeeping. Replace the API Endpoint you created in
/pages/[slug].js with a the
.env variable
GRAPHCMS_ENDPOINT and assign the value inside
.env:
// pages/[slug].js // ... const graphcms = new GraphQLClient(process.env.GRAPHCMS_ENDPOINT); // ...
Now before we can use the
GRAPHCMS_MUTATION_TOKEN, we'll need to update our
components/Form/index.js to
POST the values to a Next.js API route.
Inside the form, let's do a few things:
- import
useStatefrom React,
- Invoke
useStateinside your
Formfunction,
- Replace the
onSubmitfunction,
- Render
errorafter the submit
<button />
import { useState } from 'react' // ... export default function Form({ fields }) { if (!fields) return null; const [success, setSuccess] = useState(null); const [error, setError] = useState(null); // ... const onSubmit = async (values) => { try { const response = await fetch("/api/submit", { method: "POST", body: JSON.stringify(values), }); if (!response.ok) throw new Error(`Something went wrong submitting the form.`); setSuccess(true); } catch (err) { setError(err.message); } }; if (success) return <p>Form submitted. We'll be in touch!</p>; return ( // ... <button type="submit">Submit</button> {error && <span>{error}</span>}} ) }
Finally we'll create the API route
/api/submit that forwards requests to GraphCMS securely. We need to do this to prevent exposing our Mutation Token to the public.
One of the best ways to scaffold your mutation is to use the API Playground inside your GraphCMS project. It contains all of the documentation and types associated with your project/models.
If you've followed along so far, the following mutation is all we need to create + connect form submissions.
mutation createSubmission($formData: Json!, $formId: ID!) { createSubmission(data: {formData: $formData, form: {connect: {id: $formId}}}) { id } }
The
createSubmission mutation takes in 2 arguments;
formData and
formId.
In the
onSubmit function above, we're passing along
values which will be our
formData. All we need to do now is pass along the form ID!
We are already querying for the form
id inside
pages/[slug].js, so we can use this
id passed down to the
Form component.
Inside
components/Form.js, destructure
id when declaring the function:
export default function Form({ id, fields }) { // ... }
.... and then pass that
id into the
onSubmit
body:
const response = await fetch("/api/submit", { method: "POST", body: JSON.stringify({ id, ...values }), });
Then, inside the
pages directory, create the directory/file
api/submit.js, and add the following code:
import { GraphQLClient } from "graphql-request"; export default async ({ body }, res) => { const { id, ...data } = JSON.parse(body); const graphcms = new GraphQLClient(process.env.GRAPHCMS_ENDPOINT, { headers: { authorization: `Bearer ${process.env.GRAPHCMS_MUTATION_TOKEN}`, }, }); try { const { createSubmission } = await graphcms.request(` mutation createSubmission($data: Json!, $id: ID!) { createSubmission(data: {formData: $data, form: {connect: {id: $id}}}) { id } }`, { data, id, } ); res.status(201).json(createSubmission); } catch ({ message }) { res.status(400).json({ message }); } };
That's it! ✨
Now go ahead and submit the form, open the content editor and navigate to the
Submission content.
You should see your new entry!
You could use GraphCMS webhooks to listen for new submissions, and using another API route forward that onto a service of your choice, such as email, Slack or Zapier.
13. Deploy to Vercel
Now all that's left to do is deploy our Next.js site to Vercel. Next.js is buil, and managed by the Vercel team and the community.
To deploy to Vercel, you'll need to install the CLI.
npm i -g vercel # or yarn global add vercel
Once installed, all it takes to deploy is one command!
vercel # or vc
You'll next be asked to confirm whether you wish to deploy the current directory, and what the project is named, etc. The defaults should be enough to get you going! 😅
Once deployed, you'll get a URL to your site. Open the deployment URL and append
/contact to see your form!
Discussion (1)
Excellent! Clear and concise. Bravo! I didn't know that vercel had a CLI. The way I've been deploying is to commit to github then Vercel automatically triggers a build. Thanks for mentioning the vercel cli; I'll look into it. | https://practicaldev-herokuapp-com.global.ssl.fastly.net/graphcms/programmatically-create-forms-and-capture-submissions-with-next-js-and-graphql-3pn5 | CC-MAIN-2021-49 | refinedweb | 4,509 | 56.96 |
And update comm-central in sync, of course. /configure.in { 6319 if test -n "$MOZ_PLACES"; then 6321 if test -z "$MOZ_MAIL_NEWS"; then 6322 MOZ_MORK= 6323 fi 6325 fi } The current m-c defaults are MOZ_PLACES=1 and MOZ_MORK=1: this seems a little odd wrt this code. The values and the code should be reworked to be more explicit. /extensions/cookie/nsCookiePermission.cpp { 93 #ifdef MOZ_MAIL_NEWS 94 // returns PR_TRUE if URI appears to be the URI of a mailnews protocol 95 // XXXbz this should be a protocol flag, not a scheme list, dammit! 96 static PRBool 97 IsFromMailNews(nsIURI *aURI) 98 { 107 } 108 #endif 197 nsCookiePermission::CanAccess(nsIURI *aURI, 200 { 201 #ifdef MOZ_MAIL_NEWS 204 if (IsFromMailNews(aURI)) { 208 #endif // MOZ_MAIL_NEWS } At least, rename it to "DENY_MAILNEWS_COOKIES" or the like. Ideally, bz seems to suggest that each protocol should have a flag to allow/deny cookies ... which should let us get rid of these #ifdef :-)
(In reply to comment #0) > /configure.in Dupe of bug 556253 which I just haven't checked in yet. > /extensions/cookie/nsCookiePermission.cpp ... > At least, rename it to "DENY_MAILNEWS_COOKIES" or the like. > Ideally, bz seems to suggest that each protocol should have a flag to > allow/deny cookies ... which should let us get rid of these #ifdef :-) IMHO renaming is a waste of time and effort and it would be better to consider a long term solution for cookies (for which there are various bugs filed that would cover it iirc).
Created attachment 556001 [details] [diff] [review] WIP mozilla-central patch The basic idea here is to provide an opt-out flag for the URI to not allow it to access cookies. I'll attach the comm-central patch in a moment. The patches work together for Thunderbird and pass tests, but I still need to fix (separate out) the unit test for TestCookie.cpp so that it keeps passing for Firefox, as Firefox doesn't have the mailnews protocols defined.
Created attachment 556002 [details] [diff] [review] WIP comm-central patch
Created attachment 567849 [details] [diff] [review] Proposed fix (mozilla-central) So the mailnews cookie tests will be covered by the comm-central patch, hence we shouldn't really need them in the mozilla-central one. I'm going to push this to try and see if it comes back green.
Try run for 019ee71450ce is complete. Detailed breakdown of the results available here: Results (out of 4 total builds): exception: 3 failure: 1 Builds available at
Created attachment 567880 [details] [diff] [review] Proposed fix v2 (mozilla-central) The previous patch had forgotten to take account of the PRBool rename.
Try run for d8022c1e9bf7 is complete. Detailed breakdown of the results available here: Results (out of 65 total builds): success: 54 warnings: 4 failure: 7 Builds available at
Created attachment 568406 [details] [diff] [review] Proposed fix (comm-central) This is the comm-central part of the fix, with unit test.
Comment on attachment 567880 [details] [diff] [review] Proposed fix v2 (mozilla-central) Try server passed, the only failures where random orange. This gets the mailnews specific cookie stuff out of gecko and allows protocols to forbid cookies individually.
Note for the build to work for Thunderbird in comm-central, the patch from bug 686278 must be applied first.
Comment on attachment 568406 [details] [diff] [review] Proposed fix (comm-central) for some reason, my build failed in ldap with these two patches applied. I'll try a clobber build next.
Comment on attachment 568406 [details] [diff] [review] Proposed fix (comm-central) ugh, apologies, but I've tried this several times, and my Windows build fails here: nsLDAPService.cpp c:/builds/tbirdhq/objdir-tb/ldap/xpcom/src/../../../../ldap/xpcom/src/nsLDAPURL. cpp(48) : fatal error C1083: Cannot open include file: 'nsMsgUtils.h': No such f ile or directory c:\builds\tbirdhq\config\rules.mk:1308:0: command 'c:/mozilla-build/python/pytho n2.6.exe -O c:/builds/tbirdhq/mozilla/build/cl.py cl -FonsLDAPURL.obj -c -D_HAS_ If I back out the patches, the build is fine. Clobbering didn't help.
oh, I should say, I'm doing a pymake; not sure if that matters or not.
Comment on attachment 568406 [details] [diff] [review] Proposed fix (comm-central) Did you apply the patch from the blocking bug?
ah, sorry, I read that as having to apply the moz-central patch. With the other patch applied, this does build. The protocol tests (imap, mailbox, nntp) are failing, however. Perhaps I need to do a clobber build.
clobber build didn't help - still seeing the protocol tests failing: TEST-UNEXPECTED-FAIL | c:/builds/tbirdhq/objdir-tb/mozilla/_tests/xpcshell/mailn ews/local/test/unit/test_mailboxProtocol.js | 32928 == 160 - See following stack : JS frame :: c:\builds\tbirdhq\mozilla\testing\xpcshell\head.js :: do_throw :: li ne 453 JS frame :: c:\builds\tbirdhq\mozilla\testing\xpcshell\head.js :: _do_check_eq : : line 547 JS frame :: c:\builds\tbirdhq\mozilla\testing\xpcshell\head.js :: do_check_eq :: line 568 JS frame :: c:/builds/tbirdhq/objdir-tb/mozilla/_tests/xpcshell/mailnews/local/t est/unit/test_mailboxProtocol.js :: run_test :: line 30 JS frame :: c:\builds\tbirdhq\mozilla\testing\xpcshell\head.js :: _execute_test I'll try without the patches, I guess.
Oh, they are because we're testing the default protocol flags for those protocols and I've now changed the default flags. I'll fix in a bit once my compile's finished.
Created attachment 568900 [details] [diff] [review] Proposed fix v2 (comm-central) Fixed the unit tests.
Comment on attachment 568900 [details] [diff] [review] Proposed fix v2 (comm-central) thx, that works. Since this is just test code that's being moved, I won't complain about the egregious code duplication :-)
Comment on attachment 567880 [details] [diff] [review] Proposed fix v2 (mozilla-central) Not heard anything from dwitte :-( So trying some other cookie peers...
Comment on attachment 567880 [details] [diff] [review] Proposed fix v2 (mozilla-central) Review of attachment 567880 [details] [diff] [review]: ----------------------------------------------------------------- r=sdwilsh
Checked in both patches:
This changeset has been pointed out by the email regression detector in dev.tree-management. Could you please have a look at it? Talos Regression :( Dromaeo (SunSpider) decrease 4.03% on Win7 Firefox-Non-PGO Talos Regression :( Dromaeo (DOM) decrease 3.61% on Linux x64 Firefox-Non-PGO This could have been blamed incorrectly. If so, if you mention it on the thread we can work on narrowing this down to other changes. | https://bugzilla.mozilla.org/show_bug.cgi?id=557047 | CC-MAIN-2017-30 | refinedweb | 1,064 | 55.54 |
This to automatically scale up or down to meet resource demands. One of the keys to building scalable cloud solutions is understanding queue-centric patterns. Recognizing the opportunity to provide multiple points to independently scale an architecture ultimately protects against failures while it enables high availability.
In my previous post, I discussed Autoscaling Azure Virtual Machines where I showed how to leverage a Service Bus queue to determine if a solution requires more resources to keep up with demand. In that post, we had to pre-provision resources, manually deploy our code solution to each machine, and then use PowerShell to enable a startup task. This post shows how to create an Azure cloud service that contains a worker role, and we will see how to automatically scale the worker role based on the number of messages in a queue.
The previous post showed that we had to pre-provision virtual machines, and that the autoscale service simply turned them on and off. This post will demonstrate that autoscaling a cloud service means creating and destroying the backing virtual machines. Because data is not persisted on the local machine, we use Azure Storage to export diagnostics information, allowing persistent storage that survive instances.
Create the Cloud Service
In Visual Studio, create a new Cloud Service. I named mine “Processor”.
The next screen enables you to choose from several template types. I chose a Worker Role with Service Bus Queue as it will generate most of the code that I need. I name the new role “ProcessorRole”.
Two projects are added to my solution. The first project, ProcessorRole, is a class library that contains the implementation for my worker role and contains the Service Bus boilerplate code. The second project, Processor, contains the information required to deploy my cloud service.
Some Code
The code for my cloud service is very straightforward, but definitely does not follow best practices. When the worker role starts, we output Trace messages and then listen for incoming messages from the Service Bus queue. As a message is received, we output a Trace message then wait for 3 seconds.
Note: This code is different than the code generated by Visual Studio that uses a ManualResetEvent to prevent a crash. Do not take the code below as a best practice, but rather as an admittedly lazy example used to demonstrate autoscaling based on the messages in a queue backing up.
- using Microsoft.ServiceBus.Messaging;
- using Microsoft.WindowsAzure;
- using Microsoft.WindowsAzure.ServiceRuntime;
- using System;
- using System.Diagnostics;
- using System.Net;
-
- namespace ProcessorRole
- {
- public class WorkerRole : RoleEntryPoint
- {
- // The name of your queue
- const string _queueName = "myqueue";
-
- // QueueClient is thread-safe. Recommended that you cache
- // rather than recreating it on every request
- QueueClient _client;
-
- public override void Run()
- {
- Trace.WriteLine("Starting processing of messages");
-
-
- while (true)
- {
- // Not a best practice to use Receive synchronously.
- // Done here as an easy way to pause the thread,
- // in production you'd use _client.OnMessage or
- // _client.ReceiveAsync.
- var message = _client.Receive();
- if (null != message)
- {
- Trace.WriteLine("Received " + message.MessageId + " : " + message.GetBody<string>());
- message.Complete();
- }
-
- // Also a terrible practice… use the ManualResetEvent
- // instead. This is shown only to control the time
- // between receive operations
- System.Threading.Thread.Sleep(TimeSpan.FromSeconds(3));
- }
- }
-
- public override bool OnStart()
- {
- // Set the maximum number of concurrent connections
- ServicePointManager.DefaultConnectionLimit = 12;
-
- // Initialize the connection to Service Bus Queue
- var connectionString = CloudConfigurationManager.GetSetting("Microsoft.ServiceBus.ConnectionString");
- _client = QueueClient.CreateFromConnectionString(connectionString, _queueName);
-
- return base.OnStart();
- }
-
- public override void OnStop()
- {
- // Close the connection to Service Bus Queue
- _client.Close();
- base.OnStop();
- }
- }
- }
Again, I apologize for the code sample that does not follow best practices.
The Configuration
Contrary to how we pre-provisioned virtual machines and copied the code to each virtual machine, Cloud Services will provision new virtual machine instances and then deploy the code from the Azure fabric controller to the cloud service. If the role is upgraded or maintenance performed on the host, the underlying virtual machine is destroyed. This is a fundamental difference between roles and persistent VMs. This means we no longer RDP into each instance and make changes manually, we incorporate any desired changes into the deployment package itself. The package we are creating uses desired state configuration to tell the Azure fabric controller how to provision the new role instance.
As an example, you might add the Azure Role instance to a virtual network as I showed in the post Deploy Azure Roles Joined to a VNet Using Eclipse where I edited the .cscfg file to indicate the virtual network and subnet to add the role to. Using Visual Studio, you can affect various settings such as the number of instances, the size of each virtual machine, and the diagnostics settings:
For a more detailed look at handling multiple configurations, debugging locally, and managing connection strings, see Developing and Deploying Microsoft Azure Cloud Services Using Visual Studio.
Deploying the Cloud Service
Right-click on the deployment project and choose Publish. You are prompted to create a cloud service and an automatically created storage account.
The next screen provides more settings regarding whether it the code is being deployed to Staging or Production, whether it is a Debug or Release build, and which configuration to use. You can also configure the ability to enable Remote Desktop to each of the roles. I enable that, and provide the username and password to log into each role.
Click Publish and you can watch the status in the Microsoft Azure Activity Log pane. Notice that the output shows we are uploading a package:
Once deployed, we can see the services are running in Visual Studio:
We can also see the deployed services in the management portal:
Managing Autoscale
Just like we did in the article Autoscaling Azure Virtual Machines, we will use the autoscale service to scale our cloud service based on queue length. Go to the management portal, open the cloud service that you just created, and click the Dashboard tab. Then click the “Configure Autoscale” link:
On that screen you will see that we have a minimum of 2 instances because we specified 2 instances in the deployment package.
Click on the Queue option, and we can now scale between 1 and 350 instances!
OK, that’s a little much… let’s go with a minimum of 2 instances, maximum of 5 instances, and scale 1 instance at a time over 5 minutes based on messages in my Service Bus queue.
Click save, and within seconds our configuration is saved.
Testing it Out
I wrote a quick Console application that will send messages to the queue once per second. The receiver only processes messages once every 3 seconds, so we should quickly have more messages in queue than 2 instances can handle, forcing an autoscale event to occur.
-++;
- }
-
- }
- }
- }
If we let the Sender program run for awhile, it is sending messages to the queue faster than the receivers can process them. I go to the portal and look at the Service Bus queue, I can see that the queue length is now 61 after running for a short duration.
Hit refresh, and we see it is continuing to increase. Next, go to the Azure Storage Account used for deployment and look at the WADLogsTable:
Double-click and you will see that the roles are processing the messages, just not faster than the Sender program is sending them.
After a few minutes, the autoscale service sees that there are more messages in queue than we configured, our current role instances cannot keep up with demand, so a new virtual machine instance is created.
This is very different than when we used virtual machines. When using cloud services, the roles are created and destroyed as necessary. I then stop the sender program, and the number of queue messages quickly falls as our current number of instances can handle the demand:
And we wait for a few more minutes to see that the autoscale service has now destroyed the newly created virtual machine according to our autoscale rules.
This is a good thing, as the virtual machine was automatically created according to our autoscale rules as well. This should highlight the importance, then, of not simply using Remote Desktop to connect to a cloud service role instance to configure something. Those settings must be applied within the deployment package itself.
Monitoring
If we go to the operation logs we can see the deployment operation (note: some values are redacted by me):
- <SubscriptionOperation xmlns=""
- xmlns:
- <OperationId>3c8f4a65-64c4-7c6a-86db-672972e504b9</OperationId>
- <OperationObjectId>/REDACTED/services/hostedservices/kirkeautoscaledemo/deployments/REDACTED</OperationObjectId>
- <OperationName>ChangeDeploymentConfigurationBySlot<>deploymentSlot</d2p1:Name>
- <d2p1:Value>Production</d2p1:Value>
- </OperationParameter>
- <OperationParameter>
- <d2p1:Name>input</d2p1:Name>
- <d2p1:Value><?xml version="1.0" encoding="utf-16"?>
- <ChangeConfiguration xmlns:i=""
-
- <Configuration>REDACTED
- <>InProgress</Status>
- </OperationStatus>
- <OperationStartedTime>2015-02-22T13:38:12Z</OperationStartedTime>
- <OperationKind>UpdateDeploymentOperation</OperationKind>
- </SubscriptionOperation>
For More Information
Developing and Deploying Microsoft Azure Cloud Services Using Visual Studio
Autoscaling Azure Virtual Machines
Excellent Article – certainly reduces monitoring and configuring
Well written post. It worked as explained.
Thanks Kirk !
Great post, thank you.
But in keeping with true Columbo style, there just one more thing sir…
"Click on the Queue option, and we can now scale between 1 and 350 instances!"
is the bit I'm struggling with.
Given,
The basic unit of deployment and scale in Azure is the Cloud Service, consisting of a set of roles. Each role contains a set of identical role instances, each running a specialized cloud configured version of Windows Server. (best practices quote)
I understand these Vms "instances" are non persistent, and invoked/destroyed in code.
The maximum number of roles or instances per cloud service is 25.
Even assuming your subscription has sufficient storage, cores, service bus/queue services, etc;
How,
Can elastic scale break the built-in cloud service limit, and scale to 350 instances?
Note,
I have also read that practically, you can have very large number (1000+) of instances per role limited only by subscription.
Which is at odds with what I have read (in manuals) of the 25 roles, or instances max, per cloud service.
Wouldn't this have a knock effect for instance dns naming, input endpoints, internal endpoints for instances, rdp access, subnet sizes, Ip's, upgrade and fault domains, etc; ?
Any response, links, or humour would be appreciated,
Ta.
@Eoin – Good question! See azure.microsoft.com/…/azure-subscription-service-limits.
"Each Cloud Service with Web/Worker roles can have two deployments, one for production and one for staging. Also note that this limit refers to the number of distinct roles (configuration) and not the number of instances per role (scaling)."
The 25 limit doesn't refer to the number of instances per role.
Excellent article! works as given.. similar quetion to eoin, I can only see 20 as maximum number of possible instances in autoscaling block.you see 350! what does it depend on? Is it because Mine is msdn subcription meant for testing and dev? or something else | https://blogs.msdn.microsoft.com/kaevans/2015/02/23/autoscaling-azurecloud-services/ | CC-MAIN-2017-22 | refinedweb | 1,829 | 53.71 |
The following example code generates a simple plot, then saves it to 'fig1.pdf', then displays it, then saves it again to 'fig2.pdf'. The first image looks as expected, but the second one is blank (contains a white square). What's actually going on here? The line
plt.show()
import numpy as np
import matplotlib.pyplot as plt
x = np.linspace(-1, 1, 100)
y = x**2
plt.plot(x,y)
plt.savefig('fig1.pdf')
plt.show()
plt.savefig('fig2.pdf')
If you want to save the figure after displaying it, you'll need to hold on to the figure instance. The reason that
plt.savefig doesn't work after calling
show is that the current figure has been reset.
pyplot keeps track of which figures, axes, etc are "current" (i.e. have not yet been displayed with
show) behind-the-scenes.
gcf and
gca get the current figure and current axes instances, respectively.
plt.savefig (and essentially any other
pyplot method) just does
plt.gcf().savefig(...). In other words, get the current figure instance and call its
savefig method. Similarly
plt.plot basically does
plt.gca().plot(...).
After
show is called, the list of "current" figures and axes is empty.
In general, you're better off directly using the figure and axes instances to plot/save/show/etc, rather than using
plt.plot, etc, to implicitly get the current figure/axes and plot on it. There's nothing wrong with using
pyplot for everything (especially interactively), but it makes it easier to shoot yourself in the foot.
Use
pyplot for
plt.show() and to generate a figure and an axes object(s), but then use the figure or axes methods directly. (e.g.
ax.plot(x, y) instead of
plt.plot(x, y), etc) The main advantage of this is that it's explicit. You know what objects you're plotting on, and don't have to reason about what the pyplot state-machine does (though it's not that hard to understand the state-machine interface, either).
As an example of the "recommended" way of doing things, do something like:
import numpy as np import matplotlib.pyplot as plt x = np.linspace(-1, 1, 100) y = x**2 fig, ax = plt.subplots() ax.plot(x, y) fig.savefig('fig1.pdf') plt.show() fig.savefig('fig2.pdf')
If you'd rather use the
pyplot interface for everything, then just grab the figure instance before you call
show. For example:
import numpy as np import matplotlib.pyplot as plt x = np.linspace(-1, 1, 100) y = x**2 plt.plot(x, y) fig = plt.gcf() fig.savefig('fig1.pdf') plt.show() fig.savefig('fig2.pdf') | https://codedump.io/share/poZnlrbcEq5c/1/saving-a-figure-after-invoking-pyplotshow-results-in-an-empty-file | CC-MAIN-2017-17 | refinedweb | 446 | 69.28 |
Mailsync is a way of keeping a collection of mailboxes synchronized. The
mailboxes may be on the local filesystem or on an IMAP server.
WWW:
make generate-plist
To install the port: cd /usr/ports/mail/mailsync/ && make install cleanTo add the package: pkg install mailsync
cd /usr/ports/mail/mailsync/ && make install clean
pkg install mailsync
PKGNAME: mailsync
distinfo:
SHA256 (mailsync_5.2.1.orig.tar.gz) = 8a4f35eedff0003a7e17a6b06b79ad824c8a3ab80cb8351e540948ee94001e6d
SIZE (mailsync_5.2.1.orig.tar.gz) = 139967
NOTE: FreshPorts displays only information on required and default dependencies. Optional dependencies are not covered.
===> The following configuration options are available for mailsync-5.2.1_3:
DOCS=on: Build and/or install documentation
EXAMPLES=on: Build and/or install examples
===> Use 'make config' to modify these settings
gmake ssl
Number of commits found: 24
mail/mailsync: support building with any SSL base
Remove ${PORTSDIR}/ from dependencies, categories m, n, o, and p.
With hat: portmgr
Sponsored by: Absolight
- Add LICENSE
- Switch to options helpers
- Regenerate patches with `make makepatch`
- Don't install COPYING
Cleanup plist
- Support staging
- USES -> gmake
- New LIB_DEPENDS definition
- Define EXAMPLES option
- Define DOCS option
Add NO_STAGE all over the place in preparation for the staging support (cat:
mail)
- Use single space after WWW:
- remove MD5
- respect NOPORTEXAMPLES and fix plist
- use SF macro
- bump PORTREVISION
Prompted by: QA Tindy run>
Remove USE_REINPLACE from ports starting with M
- Add SHA256
- Overhaul the port, unbreak, undeprecate
- Drop maintainership (see ports/84011)
Approved by: portmgr (erwin)
This port is scheduled for deletion on 2005-09-22 if it is still broken
at that time and no PRs have been submitted to fix it.
BROKEN: Configure fails
- Update to 5.2.1
PR: ports/78241
Submitted by: maintainer
- Unset USE_GCC=2.95 and add patch to fix build with gcc 3.x
- Respect CFLAGS
- Portlint
PR: ports/69022
Submitted by: Andrey Slusar <vasallia@ukr.net>
- SIZE'ify
PR: ports/64609
Submitted by: maintainer
Bump PORTREVISION on all ports that depend on gettext to aid with upgrading.
(Part 1)
The __WORD_BIT constant in GCC's stl_bvector.h caused a namespace
conflict which kept the mailsync port from compiling. Resolve it.
Move inclusion of bsd.port.pre.mk later in the file for conditional BROKEN
tag. Early inclusion caused problems for some ports, so to be safe I'm
updating all of them.
Pointy hat to: kris
BROKEN on 5.1: bad C++
Add mailsync 4.4.4, mailsync is a way of keeping a collection of mailboxes
synchronized.
PR: ports/48601
Submitted by: Maxim Tulyuk <mt@primats.org.ua>
Servers and bandwidth provided byNew York Internet, SuperNews, and RootBSD
14 vulnerabilities affecting 77 ports have been reported in the past 14 days
* - modified, not new
All vulnerabilities
Last updated:2017-12-10 18:59:21 | https://www.freshports.org/mail/mailsync/ | CC-MAIN-2017-51 | refinedweb | 460 | 54.42 |
What's New for Developers in Windows 10 Version 1511 and the 10586 SDK
.
I need Emulator support of Windows 10 Continuum for Phones.
I need Emulator support of Windows 10 Continuum for Phones.
So the best way to implement your adaptive UI for continuum is switch on screensize not on device family.
How about monetizing? I feel that advertisement adaptive UI is left behind. I can't find much about how to implement this. I think many of your loyal UWP developers need income from this source.
"I need Emulator support of Windows 10 Continuum for Phones"
You don't really need an emulator. Instead of thinking of screen size, think window size. As you reduce the size of the app you re-organize the placement and the layout of your app. You can see this if you resize Groove or the map app (on Windows 10).
@MaulRiEEZZZ just position and resize the advertisement based on the screen/window size that is available.
Adapting the UI of an app to a small size is also very useful on a desktop where I may not want the app to take my entire screen, makes room for other apps.
The additions to app package manifest to opt out of (mobile) continuum are actually incompatible with the tooling.
Once added it renders the manifest uneditable in the editor & generates the following warning:
The element 'Extensions' in namespace '' has invalid child element 'Extension' in namespace ''. List of possible elements expected: 'ApplicationExtensionChoice' in namespace '' as well as 'Extension' in namespace '' as well as 'Extension' in namespace ''. notepad.mobile C:\Users\chrissn\Documents\Visual Studio 2015\Projects\notepad\wx-app\notepad.mobile\Package.appxmanifest 45 | https://channel9.msdn.com/events/Windows/Developers-Guide-to-Windows-10-Version-1511/Building-Apps-for-Continuum | CC-MAIN-2019-30 | refinedweb | 280 | 56.66 |
Over this column and the next one (and possibly the one after that, depending on how detailed we get), we're going to discuss kernel and module debugging using proc files. Specifically, we're going to discuss the seq_file implementation of proc files, which represents the newest and most powerful variation of the proc files we're interested in.
This first column will introduce the simpler variation of sequence files, while Part 2 (and beyond) will cover the more complete and formal usage of those files. And, not surprisingly, we're once again going to steal shamelessly from, and build on, what you can read in the classic book LDD3, found online here. In particular, we'll be working out of Chapter 4 of that book, so it would be in your best interest to have that portion of the chapter handy as you read what follows, since I plan on referring to parts of it as we go along.
Oh, and fair warning: a tiny part of this discussion is speculation, so feel free to use the comments section to correct any misinformation.
(The archive of all previous "Kernel Newbie Corner" articles can be found here.)
This is ongoing content from the Linux Foundation training program. If you want more content, please consider signing up for one of these classes.
What is the /proc Directory, Anyway?
If you've worked with Linux for even a short time, you probably already know the answer to that question. The /proc directory is an example of what is known as a "pseudo" filesystem in Linux; that is, something that isn't really a filesystem in that it doesn't take up any space on disk.
Rather, files under the /proc directory act as interfaces to internal kernel data structures, such that accessing entries under /proc magically accesses the underlying contents in kernel space. Put another way, the "files" you see under /proc aren't really "files"; instead, their contents are typically generated dynamically whenever you access them. And it's that dynamic content generation that's the theme of this and next week's columns because that's how you're going to debug the kernel and your loadable modules--by writing loadable modules that create simple proc files that allow you to list the contents of specific kernel data whenever you want.
Examples. We Need Examples.
Of course you do, so let's examine some of the proc files that already typically exist as a standard part of the Linux kernel. Consider listing the "version" of the running kernel via /proc/version:
$ cat /proc/version
Linux version 2.6.31-rc5 (rpjday@localhost.localdomain)
(gcc version 4.4.0 20090506 (Red Hat 4.4.0-4) (GCC) ) #3
SMP Mon Aug 3 11:24:19 EDT 2009
$
Where did all that information come from? It's certainly not stored in that file, since asking for the long listing of that file produces:
$ ls -l /proc/versionIt's a file with zero size. But that's because, again, it's not a real file, it's a pseudo file, whose "contents" are generated by some underlying code that implements that file whenever it's accessed. Put another way, the "contents" of numerous proc files is whatever the kernel programmer decided to generate as "output" for those proc files, using whatever combination of appropriate "print" statments that seemed appropriate at the time.
-r--r--r--. 1 root root 0 2009-08-07 11:22 /proc/version
Let's list some other proc files, all of which contain potentially useful information, all of which are zero size, and all of which generate their content based on some underlying code we'll examine a bit later:
$ cat /proc/cpuinfo
$ cat /proc/cmdline
$ cat /proc/modules
$ cat /proc/meminfo
$ cat /proc/interrupts
... and so on and so on, check it out ...
And note well that the contents of various proc files are generated new each time, which is why you'll probably see slightly different output every time you list the contents of, say, /proc/meminfo. Our mission in this week's column is to show you how to design and create your own proc files, so you can, whenever you want, list them to examine whatever kernel data you choose to associate with them.
(Note that, while the contents of some proc files should change constantly as the system runs, others should remain static. For instance, you don't expect the contents of /proc/version or /proc/cpuinfo to change no matter how many times you list them. You get the idea, right?)
As an aside, you can also create writable proc files, so that writing data to a proc file can be used to modify kernel data structures. But since this is a column on simple debugging, we'll restrict ourselves to just readable proc files. Anyone wanting to get more ambitious is welcome to read the appropriate docs.
Exercise for the reader: Take some time and examine some of the other files under /proc that look like they might contain useful information. Don't be scared to go into some of those subdirectories. If you have the time, read the kernel documentation file Documentation/filesystems/proc.txt.
So What's a "Sequence" File, Then?
And here's where things get a bit tricky. Technically, a "proc file" is nothing more than the file you can see under the /proc directory--it has a name, an owner and group, a size of (typically) zero and some permissions that dictate who is allowed to perform what operations on it. And that's all.
A proc file by itself does absolutely nothing. What's necessary is to then implement some read and/or write code behind it that defines what it means to read from (or write to) that file. And a sequence file is simply one of the possible implementations you can use to define those operations. So what's so special about a sequence file?
At this point, for the sake of brevity, I'm going to refer you to the proper sections of LDD3 that discuss the rationale for sequence files, but I'll at least summarize it here. The "older" and current implementations of proc files had an awkward limitation of not being able to "print" more than a single page of output (a page being defined by the definition of the kernel PAGE_SIZE macro). Sequence files solve this problem by generating the "output" of a proc file as a sequence of writes, each of which can be up to a page in size, with no limit on the number of writes, effectively solving the problem and allowing unlimited output from a single proc file.
In fact, it's probably safe to say that, while you'll still find a lot of the old implementation in the current kernel tree, sequence files are easily the preferred way to implement output-only proc files, even when the output is very brief.
To emphasize the difference between proc files and sequence files, two important points:
- You can create proc files that don't use the underlying seq_file implementation, and
- you can create files based on the seq_file implementation in places other than under /proc.
An Example, Please
At this point, we definitely need a live example, so let's whip up a trivial proc file that will display the current value of jiffies (the tick counter) whenever we list it. Consider the loadable module jif.c (in this case, for a 64-bit system):
#include <linux/module.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/fs.h> // for basic filesystem
#include <linux/proc_fs.h> // for the proc filesystem
#include <linux/seq_file.h> // for sequence files
#include <linux/jiffies.h> // for jiffies
static struct proc_dir_entry* jif_file;
static int
jif_show(struct seq_file *m, void *v)
{
seq_printf(m, "%llu\n",
(unsigned long long) get_jiffies_64());
return 0;
}
static int
jif_open(struct inode *inode, struct file *file)
{
return single_open(file, jif_show, NULL);
}
static const struct file_operations jif_fops = {
.owner = THIS_MODULE,
.open = jif_open,
.read = seq_read,
.llseek = seq_lseek,
.release = single_release,
};
static int __init
jif_init(void)
{
jif_file = proc_create("jif", 0, NULL, &jif_fops);
if (!jif_file) {
return -ENOMEM;
}
return 0;
}
static void __exit
jif_exit(void)
{
remove_proc_entry("jif", NULL);
}
module_init(jif_init);
module_exit(jif_exit);
MODULE_LICENSE("GPL");
At this point (using what you know from previous columns in this series), create the corresponding Makefile, and compile the module and load it. As soon as you load that module, verify that the following proc file now exists:
# ls -l /proc/jifat which point, you should (as a regular user given the read permissions), be able to list that file as much as you want to display the current value of the appropriate kernel jiffies variable:
-r--r--r--. 1 root root 0 2009-08-10 19:48 /proc/jif
#
$ cat /proc/jifafter which you can (as root) unload the module, at which point the proc file is removed. Yes, it really is that simple. And now, to work.
4329225958
$ cat /proc/jif
4329226854
$ cat /proc/jif
4329227174
$ cat /proc/jif
4329227486
$ cat /proc/jif
4329227798
$ cat /proc/jif
4329228078
$
So What Just Happened There?
Let's summarize just the critical features of the above example, so you can start implementing your own (simple) debugging proc files; we'll leave the more complicated features for Part 2. So, about the above:
- The necessary header files should be self-explanatory--just use that list.
- The pointer of type proc_dir_entry (defined in the kernel header file include/linux/proc_fs.h, if you're interested) is used to keep track of the proc file that is created but, in fact, if your proc file code is simple enough, you might not even need to keep track of that. (You'll see why shortly.)
- Your "show" routine (in our case, jif_show()) is what defines what is displayed when someone lists your proc file. That routine always takes the same two argument types but, in simple cases like ours, you can ignore the second argument--it's only relevant when you're actually "sequencing" through the output, as you'll see in Part 2.)
In short, you can define whatever output you want to print based on any kernel routines or data structures you can think of, but keep in mind that, if this is a loadable module, you will have access to only that kernel data that's been "exported."
- The next two portions of our example--the jif_open routine and the jif_fops structure--should be self-explanatory as well. Just substitute your own "open" routine name in there, and leave the rest as is. The fact that this is a trivial sequence file that doesn't even need sequencing is represented by the use of the "single_" variation of the operations.
- The entry and exit routines should also be reasonably self-explanatory. The module entry routine creates the proc file with a name of "jif", the "0" argument represents the default file permission of 0444, the "NULL" means that the file should be created directly under the /proc directory, and the final argument identifies the file operations structure to associate with that file. In other words, the creation of the proc file and its association with its open and I/O routines is all done in a single call.
The exit routine is much simpler--it simply deletes the file by name.
Simple, no? But there is one more point worth making.
Can We Make This Any Simpler?
In fact, we can if we want to cut some corners. As you can see, our entry routine does the proper error checking on whether or not we could even create our proc file and, if that failed, we return a negative error code which, as always, causes the module to fail to load:
static int __init
jif_init(void)
{
jif_file = proc_create("jif", 0, NULL, &jif_fops);
if (!jif_file) {
return -ENOMEM;
}
return 0;
}
However, if you tighten up the code, you don't even need to save the pointer to the proc_dir_entry structure:
static int __init
jif_init(void)
{
if (!proc_create("jif", 0, NULL, &jif_fops)) {
return -ENOMEM;
}
return 0;
}
You can do that since, if you look carefully, you don't really need that pointer anywhere else in the module. In an example this trivial, the exit routine simply has to delete the proc file by its name--it has no need for that pointer so, really, there's no need to hang onto it. At least for now.
And if you really wanted to tighten things up, you could bypass the error-checking altogether:
static int __init
jif_init(void)
{
proc_create("jif", 0, NULL, &jif_fops);
return 0;
}
The above simply assumes that there's no possible way for that file creation step to fail. That's probably not wise for production-level code, but the chance of that step failing is typically small so it's probably acceptable for informal testing.
What About Some Kernel Examples?
And since you've seen the very basics of creating a short, output-only sequence file, it's worth seeing the code that's responsible for printing a number of those short files you saw under /proc earlier, such as /proc/version. A number of those files can be found in the kernel source tree, in the fs/proc directory, so let's examine version.c
#include <linux/fs.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/utsname.h>
static int version_proc_show(struct seq_file *m, void *v)
{
seq_printf(m, linux_proc_banner,);
All of the above should look reasonably similar to your loadable module code, but keep two distinctions in mind:
- Notice that many of the kernel proc files in that directory have no module exit routine. That's because they're built into the kernel proper, and have no option to be built as loadable modules. Hence, they have no need for an exit routine. (For the same reason, they don't need to set a file operations "owner" field because they will never have a module owner.)
- As we've already noted, the proc files that are built into the kernel have access to all kernel symbols with external linkage, while your modules can only access what is exported.
Exercises for the reader: Take a look at some of the other simple sequence files in the kernel fs/proc directory, to see how they match up with their corresponding /proc files that we listed earlier. Some of them should be simple enough that you can see how they work.
In addition, if you have the time, write a loadable module that, when loaded, creates an output-only proc file (say, /proc/hz) that, when read, displays the kernel "HZ" value--that is, the configured kernel tick rate. It's up to you to figure out where to get that value and how to print it out.
Next week: More sequence files, of course. | http://www.linux.com/learn/linux-career-center/37985-the-kernel-newbie-corner-kernel-debugging-using-proc-qsequenceq-files-part-1 | CC-MAIN-2013-20 | refinedweb | 2,495 | 58.82 |
# # This file is part of Audio-MPD # # This software is copyright (c) 2007 by Jerome Quelin. # # This is free software; you can redistribute it and/or modify it under # the same terms as the Perl 5 programming language system itself. # use 5.008; use warnings; use strict; package Audio::MPD; # ABSTRACT: class to talk to MPD (Music Player Daemon) servers $Audio::MPD::VERSION = '2.004'; use Audio::MPD::Common::Item; use Audio::MPD::Common::Stats; use Audio::MPD::Common::Status; use Audio::MPD::Common::Output; use Encode; use IO::Socket::IP; use Moose; use MooseX::Has::Sugar; use MooseX::SemiAffordanceAccessor; use Audio::MPD::Collection; use Audio::MPD::Playlist; use Audio::MPD::Types; has conntype => ( ro, isa=>'CONNTYPE', default=>'once' ); has host => ( ro, lazy_build ); has password => ( rw, lazy_build, trigger=>sub { $_[0]->ping } ); has port => ( ro, lazy_build ); has collection => ( ro, lazy_build, isa=>'Audio::MPD::Collection' ); has playlist => ( ro, lazy_build, isa=>'Audio::MPD::Playlist' ); has version => ( rw ); has _socket => ( rw, isa=>'IO::Socket' ); #-- # initializer & lazy builders sub BUILD { my $self = shift; # create the connection if conntype is set to $REUSE $self->_connect_to_mpd_server if $self->conntype eq 'reuse'; # try to issue a ping to test connection - this can die. $self->ping; } # # my ($passwd, $host, $port) = _parse_env_var(); # # parse MPD_HOST environment variable, and extract its components. the # canonical format of MPD_HOST is passwd@host:port. # sub _parse_env_var { return (undef, undef, undef) unless defined $ENV{MPD_HOST}; return ($1, $2, $3) if $ENV{MPD_HOST} =~ /^([^@]+)\@([^:@]+):(\d+)$/; # passwd@host:port return ($1, $2, undef) if $ENV{MPD_HOST} =~ /^([^@]+)\@([^:@]+)$/; # passwd@host return (undef, $1, $2) if $ENV{MPD_HOST} =~ /^([^:@]+):(\d+)$/; # host:port return (undef, $ENV{MPD_HOST}, undef); } sub _build_host { return ( _parse_env_var() )[1] || 'localhost'; } sub _build_port { return $ENV{MPD_PORT} || ( _parse_env_var() )[2] || 6600; } sub _build_password { return $ENV{MPD_PASSWORD} || ( _parse_env_var() )[0] || ''; } sub _build_collection { Audio::MPD::Collection->new( _mpd => $_[0] ); } sub _build_playlist { Audio::MPD::Playlist ->new( _mpd => $_[0] ); } #-- # Private methods # # $mpd->_connect_to_mpd_server; # # This method connects to the mpd server. It can die on several conditions: # - if the server cannot be reached, # - if it's not an mpd server, # - or if the password is incorrect, # sub _connect_to_mpd_server { my ($self) = @_; # try to connect to mpd. my $socket; if ($self->host =~ m{^/}) { eval q{use IO::Socket::UNIX qw(); 1} or die "Could not load IO::Socket::UNIX: $@\n"; $socket = IO::Socket::UNIX->new($self->host) or die "Could not create socket: $!\n"; } else { $socket = IO::Socket::IP->new( PeerAddr => $self->host, PeerPort => $self->port, ) or die "Could not create socket: $!\n"; } # parse version information. my $line = $socket->getline; chomp $line; die "Not a mpd server - welcome string was: [$line]\n" if $line !~ /^OK MPD (.+)$/; $self->set_version($1); # send password. if ( $self->password ) { $socket->print( 'password ' . encode('utf-8', $self->password) . "\n" ); $line = $socket->getline; die $line if $line =~ s/^ACK //; } # save socket $self->_set_socket($socket); } # # my @result = $mpd->_send_command( $command ); # # This method is central to the module. It is responsible for interacting with # mpd by sending the $command and reading output - that will be returned as an # array of chomped lines (status line will not be returned). # # This method can die on several conditions: # - if the server cannot be reached, # - if it's not an mpd server, # - if the password is incorrect, # - or if the command is an invalid mpd command. # In the latter case, the mpd error message will be returned. # sub _send_command { my ($self, $command) = @_; $self->_connect_to_mpd_server if $self->conntype eq 'once'; my $socket = $self->_socket; # ok, now we're connected - let's issue the command. $socket->print( encode('utf-8', $command) ); my @output; while (defined ( my $line = $socket->getline ) ) { chomp $line; die $line if $line =~ s/^ACK //; # oops - error. last if $line =~ /^OK/; # end of output. push @output, decode('utf-8', $line); } # close the socket. $socket->close if $self->conntype eq 'once'; return @output; } # # my @items = $mpd->_cooked_command_as_items( $command ); # # Lots of Audio::MPD methods are using _send_command() and then parse the # output as a collection of AMC::Item. This method is meant to factorize # this code, and will parse the raw output of _send_command() in a cooked # list of items. # sub _cooked_command_as_items { my ($self, $command) = @_; my @lines = $self->_send_command($command); my (@items, %param); # parse lines in reverse order since "file:" or "directory:" lines # come first. therefore, let's first store every other parameter, # and the last line will trigger the object creation. # of course, since we want to preserve the playlist order, this means # that we're going to unshift the objects instead of push. foreach my $line (reverse @lines) { my ($k,$v) = split /:\s/, $line, 2; $param{$k} = $v; next unless $k eq 'file' || $k eq 'directory' || $k eq 'playlist'; # last param of item unshift @items, Audio::MPD::Common::Item->new(%param); %param = (); } return @items; } # # my %hash = $mpd->_cooked_command_as_kv( $command ); # # Lots of Audio::MPD methods are using _send_command() and then parse the # output to get a list of key / value (with the colon ":" acting as separator). # This method is meant to factorize this code, and will parse the raw output # of _send_command() in a cooked hash. # sub _cooked_command_as_kv { my ($self, $command) = @_; my %hash = map { split(/:\s/, $_, 2) } $self->_send_command($command); return %hash; } # # my @list = $mpd->_cooked_command_strip_first_field( $command ); # # Lots of Audio::MPD methods are using _send_command() and then parse the # output to remove the first field (with the colon ":" acting as separator). # This method is meant to factorize this code, and will parse the raw output # of _send_command() in a cooked list of strings. # sub _cooked_command_strip_first_field { my ($self, $command) = @_; my @list = map { ( split(/:\s+/, $_, 2) )[1] } $self->_send_command($command); return @list; } #-- # Public methods # -- MPD interaction: general commands sub ping { my ($self) = @_; $self->_send_command( "ping\n" ); } # sub version {} # implemented as an accessor. sub kill { my ($self) = @_; $self->_send_command("kill\n"); } # implemented by password trigger (from moose) sub updatedb { my ($self, $path) = @_; $path ||= ''; $self->_send_command("update $path\n"); } sub urlhandlers { my ($self) = @_; return $self->_cooked_command_strip_first_field("urlhandlers\n"); } # -- MPD interaction: handling volume & output sub volume { my ($self, $volume) = @_; if ($volume =~ /^(-|\+)(\d+)/ ) { my $current = $self->status->volume; $volume = $1 eq '+' ? $current + $2 : $current - $2; } $self->_send_command("setvol $volume\n"); } sub outputs { my ($self) = @_; my @lines = $self->_send_command("outputs\n"); my (@outputs, %param); # parse lines in reverse order since "id" lines come first foreach my $line (reverse @lines) { my ($k,$v) = split /:\s/, $line, 2; $k =~ s/^output//; $param{$k} = $v; next unless $k eq 'id'; # last output param unshift @outputs, Audio::MPD::Common::Output->new(%param); %param = (); } return @outputs; } sub output_enable { my ($self, $output) = @_; $self->_send_command("enableoutput $output\n"); } sub output_disable { my ($self, $output) = @_; $self->_send_command("disableoutput $output\n"); } # -- MPD interaction: retrieving info from current state sub stats { my ($self) = @_; my %kv = $self->_cooked_command_as_kv( "stats\n" ); return Audio::MPD::Common::Stats->new(\%kv); } sub status { my ($self) = @_; my %kv = $self->_cooked_command_as_kv( "status\n" ); my $status = Audio::MPD::Common::Status->new( \%kv ); return $status; } sub current { my ($self) = @_; my ($item) = $self->_cooked_command_as_items("currentsong\n"); return $item; } sub song { my ($self, $song) = @_; return $self->current unless defined $song; my ($item) = $self->_cooked_command_as_items("playlistinfo $song\n"); return $item; } sub songid { my ($self, $songid) = @_; return $self->current unless defined $songid; my ($item) = $self->_cooked_command_as_items("playlistid $songid\n"); return $item; } # -- MPD interaction: altering settings sub repeat { my ($self, $mode) = @_; $mode = not $self->status->repeat unless defined $mode; # toggle if no param $mode = $mode ? 1 : 0; # force integer $self->_send_command("repeat $mode\n"); } sub random { my ($self, $mode) = @_; $mode = not $self->status->random unless defined $mode; # toggle if no param $mode = $mode ? 1 : 0; # force integer $self->_send_command("random $mode\n"); } sub fade { my ($self, $value) = @_; $value ||= 0; $self->_send_command("crossfade $value\n"); } # -- MPD interaction: controlling playback sub play { my ($self, $number) = @_; $number = '' unless defined $number; $self->_send_command("play $number\n"); } sub playid { my ($self, $number) = @_; $number ||= ''; $self->_send_command("playid $number\n"); } sub pause { my ($self, $state) = @_; $state ||= ''; # default is to toggle $self->_send_command("pause $state\n"); } sub stop { my ($self) = @_; $self->_send_command("stop\n"); } sub next { my ($self) = @_; $self->_send_command("next\n"); } sub prev { my($self) = shift; $self->_send_command("previous\n"); } sub seek { my ($self, $time, $song) = @_; $time ||= 0; $time = int $time; $song = $self->status->song if not defined $song; # seek in current song $self->_send_command( "seek $song $time\n" ); } sub seekid { my ($self, $time, $song) = @_; $time ||= 0; $time = int $time; $song = $self->status->songid if not defined $song; # seek in current song $self->_send_command( "seekid $song $time\n" ); } no Moose; __PACKAGE__->meta->make_immutable; 1; __END__ =pod =head1 NAME Audio::MPD - class to talk to MPD (Music Player Daemon) servers =head1 VERSION version 2.004 =head1 SYNOPSIS use Audio::MPD; my $mpd = Audio::MPD->new; $mpd->play; sleep 10; $mpd->next; =head1 DESCRIPTION L<Audio::MPD> gives a clear object-oriented interface for talking to and controlling MPD (Music Player Daemon) servers. A connection to the MPD server is established as soon as a new L<Audio::MPD> object is created. Since mpd is still in 0.x versions, L<Audio::MPD> sticks to latest mpd (0.15 as time of writing) protocol & behaviour, and does B C<idle> command (new in mpd 0.14) is B<not> (and will not) be supported in L<Audio::MPD>. This will be implemented in L<POE::Component::Client::MPD>. B</!\> Note that L<Audio::MPD> is using high-level, blocking sockets. This means that if the mpd server is slow, or hangs for whatever reason, or even crash abruptly, the program will be hung forever in this sub. The L<POE::Component::Client::MPD> module is way safer - you're advised to use it instead of L<Audio::MPD>. Or you can try to set C<conntype> to C<$REUSE> (see L<Audio::MPD> constructor for more details), but you would be then on your own to deal with disconnections. =head2 Searching the collection To search the collection, use the C<collection()> accessor, returning the associated L<Audio::MPD::Collection> object. You will then be able to call: $mpd->collection->all_songs; See L<Audio::MPD::Collection> documentation for more details on available methods. =head2 Handling the playlist To update the playlist, use the C<playlist()> accessor, returning the associated L<Audio::MPD::Playlist> object. You will then be able to call: $mpd->playlist->clear; See L<Audio::MPD::Playlist> documentation for more details on available methods. =head1 ATTRIBUTES =head2 host The hostname where MPD is running. Defaults to environment var C<MPD_HOST>, then to 'localhost'. Note that C<MPD_HOST> can be of the form C<password@host:port> (each of C<password@> or C<:port> can be omitted). =head2 port The port that MPD server listens to. Defaults to environment var C<MPD_PORT>, then to parsed C<MPD_HOST> (cf above), then to 6600. =head2 password The password to access special MPD functions. Defaults to environment var C<MPD_PASSWORD>, then to parsed C<MPD_HOST> (cf above), then to empty string. =head2 conntype Change how the connection to mpd server is handled. It should be of a C<CONNTYPE> type (cf L<Audio::MPD::Types>). Use either the C<reuse> string to reuse the same connection or C<once> to open a new connection per command (default). =head1 METHODS =head2 new my $mpd = Audio::MPD->new( \%opts ); This is the constructor for L<Audio::MPD>. One can specify any of the attributes (cf above). =head1 CONTROLLING THE SERVER =head2 ping $mpd->ping; Sends a ping command to the mpd server. =head2 version my $version = $mpd->version; Return mpd's version number as advertised during connection. Note that mpd returns B<protocol> version when connected. This protocol version can differ from the real mpd version. eg, mpd version 0.13.2 is "speaking" and thus advertising version 0.13.0. =head2 kill $mpd->kill; Send a message to the MPD server telling it to shut down. =head2 set_password $mpd->set_password( [$password] ); Change password used to communicate with MPD server to C<$password>. Empty string is assumed if C<$password> is not supplied. =head2 updatedb $mpd->updatedb( [$path] ); Force mpd to recan its collection. If C<$path> (relative to MPD's music directory) is supplied, MPD will only scan it - otherwise, MPD will rescan its whole collection. =head2 urlhandlers my @handlers = $mpd->urlhandlers; Return an array of supported URL schemes. =head1 HANDLING VOLUME & OUTPUT =head2 volume $mpd->volume( [+][-]$volume ); Sets the audio output volume percentage to absolute C<$volume>. If C<$volume> is prefixed by '+' or '-' then the volume is changed relatively by that value. =head2 outputs my @outputs = $mpd->outputs( ); Return a list of C<Audio::MPD::Common::Outputs> with all outputs available within MPD. =head2 output_enable $mpd->output_enable( $output ); Enable the specified audio output. C<$output> is the ID of the audio output. =head2 output_disable $mpd->output_disable( $output ); Disable the specified audio output. C<$output> is the ID of the audio output. =head1 RETRIEVING INFO FROM CURRENT STATE =head2 stats my $stats = $mpd->stats; Return an L<Audio::MPD::Common::Stats> object with the current statistics of MPD. See the associated pod for more information. =head2 status my $status = $mpd->status; Return an L<Audio::MPD::Common::Status> object with various information on current MPD server settings. See the associated pod for more information. =head2 current my $song = $mpd->current; Return an L<Audio::MPD::Common::Item::Song> representing the song currently playing. =head2 song my $song = $mpd->song( [$song] ); Return an L<Audio::MPD::Common::Item::Song> representing the song number C<$song>. If C<$song> is not supplied, returns the current song. =head2 songid my $song = $mpd->songid( [$songid] ); Return an L<Audio::MPD::Common::Item::Song> representing the song with id C<$songid>. If C<$songid> is not supplied, returns the current song. =head1 ALTERING MPD SETTINGS =head2 repeat $mpd->repeat( [$repeat] ); Set the repeat mode to C<$repeat> (1 or 0). If C<$repeat> is not specified then the repeat mode is toggled. =head2 random $mpd->random( [$random] ); Set the random mode to C<$random> (1 or 0). If C<$random> is not specified then the random mode is toggled. =head2 fade $mpd->fade( [$seconds] ); Enable crossfading and set the duration of crossfade between songs. If C<$seconds> is not specified or $seconds is 0, then crossfading is disabled. =head1 CONTROLLING PLAYBACK =head2 play $mpd->play( [$song] ); Begin playing playlist at song number C<$song>. If no argument supplied, resume playing. =head2 playid $mpd->playid( [$songid] ); Begin playing playlist at song ID C<$songid>. If no argument supplied, resume playing. =head2 pause $mpd->pause( [$state] ); Pause playback. If C<$state> is 0 then the current track is unpaused, if C<$state> is 1 then the current track is paused. Note that if C<$state> is not given, pause state will be toggled. =head2 stop $mpd->stop; Stop playback. =head2 next $mpd->next; Play next song in playlist. =head2 prev $mpd->prev; Play previous song in playlist. =head2 seek $mpd->seek( $time, [$song]); Seek to C<$time> seconds in song number C<$song>. If C<$song> number is not specified then the perl module will try and seek to C<$time> in the current song. =head2 seekid $mpd->seekid( $time, $songid ); Seek to C<$time> seconds in song ID C<$songid>. If C<$song> number is not specified then the perl module will try and seek to C<$time> in the current song. =for Pod::Coverage BUILD =head1 SEE ALSO You can find more information on the mpd project on its homepage at L<>.wikia.com>. Original code (2005) by Tue Abrahamsen C<< <tue.abrahamsen@gmail.com> >>, documented in 2006 by Nicholas J. Humfrey C<< <njh@aelius.com> >>. You can look for information on this module at: =over 4 =item * Search CPAN L<> =item * See open / report bugs L<> =item * Mailing-list L<> =item * Git repository L<> =item * AnnoCPAN: Annotated CPAN documentation L<> =item * CPAN Ratings L<> =back =head1 AUTHOR Jerome Quelin =head1 COPYRIGHT AND LICENSE This software is copyright (c) 2007 by Jerome Quelin. This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself. =cut | http://web-stage.metacpan.org/release/Audio-MPD/source/lib/Audio/MPD.pm | CC-MAIN-2019-51 | refinedweb | 2,667 | 51.89 |
#################### Bazaar Release Notes #################### .. toctree:: :maxdepth: 1 bzr 2.5.2 ######### :2.5.2: NOT RELEASED YET External Compatibility Breaks ***************************** .. These may require users to change the way they use Bazaar. New Features ************ .. New commands, options, etc that users may wish to try out. Improvements ************ .. Improvements to existing commands, especially improved performance or memory usage, or better results. Bug Fixes ********* .. Fixes for situations where bzr would previously crash or give incorrect or undesirable results. * ``bzr config`` properly handles aliases and references in the ``--directory`` parameter (Vincent Ladeuil, Wouter van Heyst, #947049) * Empty arguments in EDITOR are now properly preserved. (Ross Lagerwall, #1089792) * Fix a traceback when trying to checkout a tree that also has an entry with file-id `TREE_ROOT` somewhere other than at the root directory. (John Arbash Meinel, #830947) * Lightweight checkouts of remote repositories had a bug with how they extracted texts from the repository. (Just an ordering constraint on how they consumed the stream.) (John Arbash Meinel, #1046284) * ``osutils.send_all`` now detects if we get a series of zero bytes sent, and fails with a ECONNRESET. It seems if paramiko gets disconnected, it will get into a state where it returns 0 bytes sent, but doesn't raise an error. This change allows us to get a couple hiccups of no content sent, but if it is consistent, we will consider it to be a failure. (John Arbash Meinel, #1047309) * Revert use of --no-tty when gpg signing commits. (Jelmer Vernooij, #1014570) * Some filesystems give ``EOPNOTSUPP`` when trying to call ``fdatasync``. This shouldn't be treated as a fatal error. (John Arbash Meinel, #1075108) * Some small bug fixes wrt lightweight checkouts and remote repositories. A test permutation was added that runs all working tree tests against a lightweight checkout. (John Arbash Meinel, #1046697) Documentation ************* .. Improved or updated documentation. API Changes *********** .. Changes that may require updates in plugins or other code that uses bzrlib. Internals ********* .. Major internal changes, unlikely to be visible to users or plugin developers, but interesting for bzr developers. Testing ******* .. Fixes and changes that are only relevant to bzr's test framework and suite. This can include new facilities for writing tests, fixes to spurious test failures and changes to the way things should be tested. bzr 2.5.1 ######### :2.5.1: 2012-05-22 This is a bugfix release. Most of the bugs dealt with https and colocated branches glitches. Upgrading is recommended for all users of earlier 2.5 releases. External Compatibility Breaks ***************************** None. New Features ************ None. Improvements ************ * ``bzr rmbranch`` now supports removing colocated branches. (Jelmer Vernooij, #920653) * ``bzr rmbranch`` no longer removes active branches unless ``--force`` is specified. (Jelmer Vernooij, #922953) Bug Fixes ********* * Connecting with HTTPS via HTTP now correctly uses the host name of the destination rather than the proxy when checking certificates. (Martin Packman, #944696) * Fixed merge tool availability checking and invocation to search the Windows App Path registry in addition to the PATH. (Gordon Tyler, #939605) * Fixed problem with getting errors about failing to open /dev/tty when using Bazaar Explorer to sign commits. (Mark Grandi, #847388) * Fix UnicodeEncodeError when translated progress task messages contain non-ascii text. (Martin Packman, #966934) * Make sure configuration options can provide their own help topic. (Jelmer Vernooij, #941672) Documentation ************* * The alpha-quality texinfo sphinx builder has been deprecated. Sphinx >= 1.1.2 now provides a better one. Most of the documentation can now be generated to the texinfo format with ``make texinfo-sphinx``. This will generate both the ``.texi`` files and the ``.info`` ones. (Vincent Ladeuil, #940164) API Changes *********** None. Testing ******* * Add support for pyftpdlib >= 0.7.0 and drop support for previous pyftpdlib versions. (Vincent Ladeuil, #956027) * Run smoketest for setup.py isolated in a tempdir. (Martin Packman, #140874) bzr 2.5.0 ######### :Codename: Phillip :2.5.0: 2012-02-24 This release marks the start of a new long-term-stable series. From here, we will only make bugfix releases on the 2.5 series (2.5.1, etc, and support it until April 2017), while 2.6 will become our new development series. This is a bugfix and polish release over the 2 ***************************** None. New Features ************ None. Improvements ************ * The names of colocated branches are used as branch nicks if no nick is specified. (Aaron Bentley) Bug Fixes ********* * Show locks in ``bzr info`` on control directories without a repository. (Jelmer Vernooij, #936767) * Disable ssl certificate verification on osx and windows until a native access to the the root certificates is provided there. (Vincent Ladeuil, #929179) Testing ******* * Stop depending on the particular CPython ordering of dictionary keys when testing the result of BzrDir.get_branches. (Wouter van Heyst) bzr 2.5b6 ######### :2.5b6: 2012-02-02 This is the sixth (and last (really)) beta of the 2.5 series, leading to a 2.5.0 release in March 2012. Beta releases are suitable for everyday use but may cause some incompatibilities with plugins. This introduces the support for colocated branches into the '2a' format in a backward compatible way, fix more glitches in the colocated UI, verify https certificates for the urllib https client implementation, fix some more unicode issues and more. All bugs fixed in previous series known at the time of this release are included. External Compatibility Breaks ***************************** None. New Features ************ * Support for colocated branches is now available in the default format ("2a"). (Jelmer Vernooij) Improvements ************ * ``bzr switch -b`` in a standalone tree will now create a colocated branch. (Jelmer Vernooij, #918197) * ``bzr info`` now reports when there are present (but unused) colocated branches. (Jelmer Vernooij, #891646) * Checkouts can now be into target directories that already have a control directory (but no branch or working tree). (Jelmer Vernooij, #913980) * Colocated branches can now have names including forward slashes, to allow for namespaces. (Jelmer Vernooij, #907980) * New HPSS call for ``BzrDir.get_branches``. (Jelmer Vernooij, #894460) * Checkouts of colocated branches are now always lightweight. (Jelmer Vernooij, #918828) Bug Fixes ********* * ``bzr branch`` now fetches revisions when branching into an empty control directory. (Jelmer Vernooij, #905594) * A sane default is provided for ``ssl.ca_certs`` which should points to the Certificate Authority bundle for supported platforms. (Vincent Ladeuil, #920455) * ``bzr branch`` generates correct target branch locations again if not specified. (Jelmer Vernooij, #919218) * ``bzr send`` works on treeless branches again. (Jelmer Vernooij, #921591) * ``bzr version`` no longer throws a UnicodeDecodeError if the .bzr.log path contains non-ascii characters. (Martin Packman, #312841) * Support scripts that don't call bzrlib.initialize() but still call run_bzr(). (Vincent Ladeuil, #917733) * Test for equality instead of object identity where ROOT_PARENT is concerned. (Wouter van Heyst, #881142) * urllib-based HTTPS client connections now verify the server certificate validity as well as the hostname. (Jelmer Vernooij, Vincent Ladeuil, #651161) API Changes *********** * ``config.config_dir`` and related functions now always return paths as unicode. (Martin Packman, #825826) * ``ControlDir`` now has a new method ``set_branch_reference`` which can be used for setting branch references. (Jelmer Vernooij) * ``ControlDir.destroy_branch`` now raises ``NotBranchError`` rather than ``NoSuchFile`` if the branch didn't exist. (Jelmer Vernooij, #921693) Internals ********* * A new matcher ``RevisionHistoryMatches`` has been added. (Jelmer Vernooij) * Add new module ``bzrlib.url_policy_open``. (Jelmer Vernooij, #850843) * ``MutableTree`` has two new hooks ``pre_transform`` and ``post_transform`` that are called for tree transform operations. (Jelmer Vernooij, #912084) Testing ******* * Be more careful about closing open files for pypy interoperability. (Wouter van Heyst) bzr 2.5b5 ######### :2.5b5: 2012-01-12 This is the fifth (and last), enhancements to the config framework and more internal uses, bug fixes related to unicode and locale support and more. All bug fixed in previous series known at the time of this release are included.) bzr 2.5b4 ######### :2.5b4: 2011-12-08 This is the fourth, optimizations for revision specifiers to avoid history sized operations, enhancements to the config framework, bug fixes related to unicode paths and more. All bug fixed in previous series known at the time of this release are included. External Compatibility Breaks ***************************** None. New Features ************ * Provides a ``po_merge`` plugin to automatically merge ``.po`` files with ``msgmerge``. See ``bzr help po_merge`` for details. (Vincent Ladeuil, #884270) Improvements ************ * ``bzr branch --stacked`` now only makes a single connection to the remote server rather than three. (Jelmer Vernooij, #444293) * ``bzr export --uncommitted`` will export the uncommitted tree. (Jelmer Vernooij, #555613) * ``bzr rmbranch`` can now remove colocated branches. (Jelmer Vernooij, #831464) * ``bzr status`` no longer shows shelves if files are specified. (Francis Devereux) * ``bzr switch`` now accepts colocated branch names to switch to. (Jelmer Vernooij, #826814) * Plugins can now register additional "location aliases". (Jelmer Vernooij) * Revision specifiers will now only browse as much history as they need to, rather than grabbing the whole history unnecessarily in some cases. (Jelmer Vernooij) * When using ``bzr switch`` to switch to a sibling of the current branch, the relative branch name should no longer be url-encoded. (Jelmer Vernooij) Bug Fixes ********* * A new section local option ``basename`` is available to help support some ``bzr-pipeline`` workflows and more generally help mapping local paths to remote ones. See ``bzr help configuration`` for more details. (Vincent Ladeuil, #843211) * Add HPSS call for looking up revision numbers from revision ids on remote repositories. (Jelmer Vernooij, #640253) * Add HPSS call for retrieving file contents from remote repositories. Should improve performance for lightweight checkouts and exports of from remote repositories. (Jelmer Vernooij, #368717, #762330, #608640) * Allow lazy compiled patterns from ``bzrlib.lazy_regex`` to be pickled. (Jelmer Vernooij, #893149) * ``bzr info`` no longer shows empty output if only a control directory is present. (Jelmer Vernooij, #159098) * Cope with missing revision ids being specified to ``Repository.gather_stats`` HPSS call. (Jelmer Vernooij, #411290) * Fix test failures on windows related to locations.conf handling. (Vincent Ladeuil, #892992) * Fixed parsing of the timestamp given to ``commit --commit-time``. Now prohibits several invalid strings, reads the correct number of seconds, and gives a better error message if the time zone offset is not given. (Matt Giuca, #892657) * Give meaningful file/line references when reporting deprecation warnings for _CompatabilityThunkFeature based test features. (Vincent Ladeuil, #897718) * Make reporting of mistakes involving unversioned files with non-ascii filenames work again without 'Unprintable exception' being shown. (Martin Packman, #898408) * Provide names for lazily registered hooks. (Neil Martinsen-Burrell, #894609) * Raise BadIndexKey exception in btree_index when a key is too large, fixing an infinite recursion issue. (Shannon Weyrick, #720853) * Resolve regression from colocated branch path handling, by ensuring that unreserved characters are unquoted in URLs. (Martin Packman, #842223) * Split segments from URLs for colocated branches without assuming the combined form is valid. (Martin Packman, #842233) * Support looking up revision numbers by revision id in empty branches. (Jelmer Vernooij, #535031) * Support verifying signatures on remote repositories. (Jelmer Vernooij, #889694) * Teach the bzr client how to reconnect if we get ``ConnectionReset`` while making an RPC request. This doesn't handle all possible network disconnects, but it should at least handle when the server is asked to shutdown gracefully. (John Arbash Meinel, #819604) * When a remote format is unknown, bzr will now print a single-line error message rather than a backtrace. (Jelmer Vernooij, #687226) API Changes *********** * ``BzrDir.open_branch`` and ``BranchFormat.open`` now take an optional ``possible_transports`` argument. (Jelmer Vernooij) * New method ``Transport.set_segment_parameter``. (Jelmer Vernooij) * ``Repository.verify_revision`` has been renamed to ``Repository.verify_revision_signature``. (Jelmer Vernooij) * ``RevisionSpec.wants_revision_history`` now defaults to ``False`` and is deprecated. The ``revs`` argument of ``RevisionInfo.from_revision_id`` is now deprecated. (Jelmer Vernooij) * ``Tree.get_file_by_path`` is now deprecated. Use ``Tree.get_file`` instead. (Jelmer Vernooij, #666897) * Some global options for use with commands have been removed, construct an ``Option`` with the name instead. (Martin Packman) * The unused exception ``HistoryMissing`` has been removed. (Jelmer Vernooij) Internals ********* * Add HPSS call for ``Repository.pack``. (Jelmer Vernooij, #894461) * ``bzr config`` uses the new configuration implementation. (Vincent Ladeuil) * Custom HPSS error handlers can now be installed in the smart server client using the ``error_translators`` and ``no_context_error_translators`` registries. (Jelmer Vernooij) * New HPSS calls ``Repository.has_signature_for_revision_id``, ``Repository.make_working_trees``, ``BzrDir.destroy_repository``, ``BzrDir.has_workingtree``, ``Repository.get_physical_lock_status``, ``Branch.get_physical_lock_status``, ``Branch.put_config_file``, ``Branch.break_lock``, ``BzrDir.destroy_branch``, ``Repository.break_lock``, ``VersionedFileRepository.get_serializer_format``, ``Repository.all_revision_ids``, ``Repository.start_write_group``, ``Repository.commit_write_group``, ``Repository.abort_write_group`` ``Repository.check_write_group``, ``Repository.iter_revisions``, ``Repository.add_signature_revision_text`` and ``Repository.get_revision_signature_text``. (Jelmer Vernooij) * ``RemoteBranch.get_config_stack`` and ``RemoteBzrDir.get_config_stack`` will now use HPSS calls where possible. (Jelmer Vernooij) * The registry of merge types has been moved to ``merge`` from ``option`` but ``merge.get_merge_type_registry`` remains as an accessor. (Martin Packman) Testing ******* * Avoid failures in test_transform when OS error messages are localised. (Martin Packman, #891582) * Tests are now subject to a time limit: by default 300s, and 120s when run from 'make check', controlled by the `selftest.timeout` configuration option. This is currently not supported on Windows. (Martin Pool) bzr 2.5b3 ######### :2.5b3: 2011-11-10 This is the third beta of the 2.5 series, leading to a 2.5.0 release in February 2012. Beta releases are suitable for everyday use but may cause some incompatibilities with plugins. This release includes log options for ``push`` and ``pull``, more UI polish for colocated branches, a better and more coherent implementation for UI dialogs, enhancements to the config framework and more. This release includes all bug fixed in previous series known at the time of this release. External Compatibility Breaks ***************************** None New Features ************ * The ``log_format`` configuration can be used with ``-Olog_format=line`` to change the format ``push`` and ``pull`` use to display the revisions. I.e.: ``bzr pull -v -Olog_format=short`` will use the ``short`` format instead of the default ``long`` one. (Vincent Ladeuil, #861472) * The new config scheme allows an alternative syntax for the 'appenpath' policy relying on option expansion and defining a new 'relpath' option local to a section. Instead of using ' | http://doc.bazaar.canonical.com/latest/en/_sources/release-notes/bzr-2.5.txt | CC-MAIN-2018-22 | refinedweb | 2,246 | 50.73 |
Credit: Alex Martelli
You need to perform frequent tests for
membership in a sequence. The
O(N) behavior of repeated
in operators hurts performance, but you
can’t switch to using just a dictionary, as you also
need the sequence’s order.
Say you need to append items to a list only if they’re not already in the list. The simple, naive solution is excellent but may be slow:
def addUnique1(baseList, otherList): for item in otherList: if item not in baseList: baseList.append(item)
If
otherList is large, it may be faster to build
an auxiliary dictionary:
def addUnique2(baseList, otherList): auxDict = {} for item in baseList: auxDict[item] = None for item in otherList: if not auxDict.has_key(item): baseList.append(item) auxDict[item] = None
For a list on which you must often perform membership tests, it may
be best to wrap the list, together with its auxiliary dictionary,
into a class. You can then define a special
_ _contains_ _ method to
speed the
in operator. The dictionary must be
carefully maintained to stay in sync with the sequence.
Here’s a version that does the syncing just in time,
when a membership test is required and the dictionary is out of sync,
and works with Python 2.1 or later:
from _ _future_ _ import nested_scopes import UserList try: list._ _getitem_ _ except: Base = UserList.UserList else: Base = list class FunkyList(Base): def _ _init_ _(self, initlist=None): Base._ _init_ _(self, initlist) self._dict_ok = 0 def _ _contains_ _(self, item): if not self._dict_ok: self._dict = {} for item in self: self._dict[item] = 1 self._dict_ok = 1 return self._dict.has_key(item) def _wrapMethod(methname): _method = getattr(Base, methname) def wrapper(self, *args): # Reset 'dictionary OK' flag, then delegate self._dict_ok = 0 return _method(self, *args) setattr(FunkyList, methname, wrapper) for meth in 'setitem delitem setslice delslice iadd'.split( ): _wrapMethod('_ _%s_ _'%meth) for meth in 'append insert pop remove extend'.split( ): _wrapMethod(meth) del _wrapMethod
Python’s
in operator is extremely handy, but
it’s O(N)
when applied to an N-item sequence. If a
sequence is subject to frequent
in tests, and the
items are hashable, an auxiliary dictionary at the
sequence’s side can provide a signficant performance
boost. A membership check (using the
in operator)
on a sequence of N items is
O(N); if
M such tests are performed, the overall time is
O(M x
N). Preparing an auxiliary dictionary whose keys
are the sequence’s items is also roughly
O(N), but the
M tests are roughly
O(M), so overall we have
roughly
O(N+M).
This is rather less than O(N
x M) and can thus offer a
very substantial performance boost when M and
N are large.
Even better overall performance can often be obtained by permanently placing the auxiliary dictionary alongside the sequence, encapsulating both into one object. However, in this case, the dictionary must be maintained as the sequence is modified, so that it stays in sync with the actual membership of the sequence.
The
FunkyList class in this recipe, for example,
extends
list (
UserList in
Python 2.1) and delegates every method to it. However, each method
that can modify list membership is wrapped in a closure that resets a
flag asserting that the auxiliary dictionary is in sync. The
in operator calls the
_ _contains_ _ method when it is applied to an instance that has such a
method. The
_ _contains_ _ method rebuilds the
auxiliary dictionary, unless the flag is set, proving that the
rebuilding is unnecessary.
If our program needs to run only on Python 2.2 and later versions, we
can rewrite the
_ _contains_ _ method in a much
better way:
def __contains__(self, item): if not self.dict_ok: self._dict = dict(zip(self,self)) self.dict_ok = 1 return item in self._dict
The built-in type
dict, new in Python 2.2, lets us
build the auxiliary dictionary faster and more concisely.
Furthermore, the ability to test for membership in a dictionary
directly with the
in operator, also new in Python
2.2, has similar advantages in speed, clarity, and conciseness.
Instead of building and installing the wrapping closures for all the
mutating methods of the list into the
FunkyList
class with the auxiliary function
_wrapMethod, we
could simply write all the needed
defs for the
wrapper methods in the body of
FunkyList, with the
advantage of extending backward portability to Python versions even
older than 2.1. Indeed, this is how I tackled the problem in the
first version of this recipe that I posted to the online Python
cookbook. However, the current version of the recipe has the
important advantage of minimizing boilerplate (repetitious plumbing
code that is boring and voluminous and thus a likely home for bugs).
Python’s advanced abilities for introspection and
dynamic modification give you a choice: you can build method
wrappers, as this recipe does, in a smart and concise way, or you can
choose to use the boilerplate approach anyway, if you
don’t mind repetitious code and prefer to avoid what
some would call the “black magic”
of advanced introspection and dynamic modification of class objects.
Performance characteristics depend on the actual pattern of membership tests versus membership modifications, and some careful profiling may be required to find the right approach for a given use. This recipe, however, caters well to a rather common pattern of use, where sequence-modifying operations tend to happen in bunches, followed by a period in which no sequence modification is performed, but several membership tests may be performed.
Rebuilding the dictionary when needed is far simpler than
incrementally maintaining it at each sequence-modifying step.
Incremental maintenance requires careful analysis of what is being
removed and of what is inserted, particularly upon such operations as
slice assignment. If that strategy is desired, the values in the
dictionary should probably be a
count of the
number of occurrences of each key’s value in the
sequence. A list of the indexes in which the value is present is
another possibility, but that takes even more work to maintain.
Depending on usage patterns, the strategy of incremental maintenance
can be substantially faster or slower.
Of course, all of this is necessary only if the sequence itself is
needed (i.e., if the order of items in the sequence is significant).
Otherwise, keeping just the dictionary is obviously simpler and more
effective. Again, the dictionary can map values to
counts, if you the need the data structure to be,
in mathematical terms, a bag rather than a set.
An important requisite for any of these membership-test optimizations
is that the values in the sequence must be hashable (otherwise, of
course, they cannot be keys in a dictionary). For example, a list of
tuples might be subjected to this recipe’s
treatment, but for a list of lists the recipe as it stands is not
applicable. You can sometimes use
cPickle.dumps to
create dictionary keys—or, for somewhat different application
needs, the object’s
id—but
neither workaround is always fully applicable. In the case of
cPickle.dumps, even when it is applicable, the
overhead may negate some or most of the optimization.
No credit card required | https://www.oreilly.com/library/view/python-cookbook/0596001673/ch02s10.html | CC-MAIN-2019-18 | refinedweb | 1,222 | 52.9 |
Write a program in C# that asks the user for two numbers and one operation (+, -, x, /) then calculate the operation and display the result on the screen.
Show the text Unrecognized character if the operation symbol is different from the previous ones.
You should use the if block.
4
x
4
4x4= 16
using System;
public class BasicCalculatorIf
{
public static void Main(string[] args)
{
int a = Convert.ToInt32(Console.ReadLine());
char operation = Convert.ToChar(Console.ReadLine());
int b = Convert.ToInt32(Console.ReadLine());
if (operation == '+')
Console.WriteLine("{0}+{1}= {2}", a, b, a + b);
else if (operation == '-')
Console.WriteLine("{0}-{1}= {2}", a, b, a - b);
else if ((operation == 'x') || (operation == '*'))
Console.WriteLine("{0}x{1}= {2}", a, b, a * b);
else if (operation == '/')
Console.WriteLine("{0}/{1}= {2}", a, b, a / b);
else
Console.WriteLine("Unrecognized character");
}
}
Practice C# anywhere with the free app for Android devices.
Learn C# at your own pace, the exercises are ordered by difficulty.
Own and third party cookies to improve our services. If you go on surfing, we will consider you accepting its use. | https://www.exercisescsharp.com/flow-controls-a/basic-calculator-using-if/ | CC-MAIN-2022-05 | refinedweb | 180 | 52.56 |
Introduction
Python
asyncio is a library for efficient single-thread concurrent applications. In my last blog post “Python AsyncIO Event Loop”, we have understood what an event loop is in Python
asyncio by looking at the Python source code. This seems to be effective to understand how Python
asyncio works.
In this blog post, I would like to take one step further and discuss the mechanisms of the three key
asyncio awaitables, including
Coroutine,
Future, and
Task, by looking at the Python source code again.
Coroutine
Starting from Python 3.5,
coroutine functions are defined using
async def and
Coroutine objects are created by calling
coroutine functions. The abstracted class of
Coroutine is just as follows. It does not have method overloading because the derived class and method overload is generated by Python interpreter for the
coroutine functions defined using
async def. The key method for
Coroutineclass is
send. It is trying to mimic the behavior of trampoline.
class Coroutine(Awaitable): __slots__ = () @abstractmethod def send(self, value): """Send a value into the coroutine. Return next yielded value or raise StopIteration. """ raise StopIteration @abstractmethod def throw(self, typ, val=None, tb=None): """Raise an exception in the coroutine. Return next yielded value or raise StopIteration. """ if val is None: if tb is None: raise typ val = typ() if tb is not None: val = val.with_traceback(tb) raise val def close(self): """Raise GeneratorExit inside coroutine. """ try: self.throw(GeneratorExit) except (GeneratorExit, StopIteration): pass else: raise RuntimeError("coroutine ignored GeneratorExit") @classmethod def __subclasshook__(cls, C): if cls is Coroutine: return _check_methods(C, '__await__', 'send', 'throw', 'close') return NotImplemented
“Fortunately”, Python
asyncio
coroutine was once implemented using a
@asyncio.coroutine decorator on a Python generator in Python 3.4. Hopefully the logic of the
coroutine in Python 3.5+ is similar to the
coroutine in Python 3.4 that it yields sub
coroutine upon calling.
A typical
coroutine could be implemented using a Python generator just like the follows.()
The
@asyncio.coroutine decorator implementation is as follows.
def coroutine(func): """Decorator to mark coroutines. If the coroutine is not yielded from before it is destroyed, an error message is logged. """ warnings.warn('"@coroutine" decorator is deprecated since Python 3.8, use "async def" instead', DeprecationWarning, stacklevel=2) if inspect.iscoroutinefunction(func): # In Python 3.5 that's all we need to do for coroutines # defined with "async def". return func if inspect.isgeneratorfunction(func): coro = func else: @functools.wraps(func) def coro(*args, **kw): res = func(*args, **kw) if (base_futures.isfuture(res) or inspect.isgenerator(res) or isinstance(res, CoroWrapper)): res = yield from res else: # If 'res' is an awaitable, run it. try: await_meth = res.__await__ except AttributeError: pass else: if isinstance(res, collections.abc.Awaitable): res = yield from await_meth() return res coro = types.coroutine(coro) if not _DEBUG: wrapper = coro else: @functools.wraps(func) def wrapper(*args, **kwds): w = CoroWrapper(coro(*args, **kwds), func=func) if w._source_traceback: del w._source_traceback[-1] # Python < 3.5 does not implement __qualname__ # on generator objects, so we set it manually. # We use getattr as some callables (such as # functools.partial may lack __qualname__). w.__name__ = getattr(func, '__name__', None) w.__qualname__ = getattr(func, '__qualname__', None) return w wrapper._is_coroutine = _is_coroutine # For iscoroutinefunction(). return wrapper
Without looking into the details, this
@asyncio.coroutine decorator almost does not change the generator at all, since most likely
wrapper $\approx$
coro.
When we tried to run
coroutine with loop.run_until_complete, we see from the comment that if the argument is a
coroutine then it would be converted to a
Task in the first place, and
loop.run_until_complete is actually scheduling
Tasks. So we would look into
Task shortly.
Future
Future has closed relationship with
Task, so let’s look at
Future first.
Future use has an event loop. By default, it is the event loop in the main thread.
class Future: """This class is *almost* compatible with concurrent.futures.Future. Differences: - This class is not thread-safe. - result() and exception() do not take a timeout argument and raise an exception when the future isn't done yet. - Callbacks registered with add_done_callback() are always called via the event loop's call_soon(). - This class is not compatible with the wait() and as_completed() methods in the concurrent.futures package. (In Python 3.4 or later we may be able to unify the implementations.) """ # Class variables serving as defaults for instance variables. _state = _PENDING _result = None _exception = None _loop = None _source_traceback = None # This field is used for a dual purpose: # - Its presence is a marker to declare that a class implements # the Future protocol (i.e. is intended to be duck-type compatible). # The value must also be not-None, to enable a subclass to declare # that it is not compatible by setting this to None. # - It is set by __iter__() below so that Task._step() can tell # the difference between # `await Future()` or`yield from Future()` (correct) vs. # `yield Future()` (incorrect). _asyncio_future_blocking = False __log_traceback = False = format_helpers.extract_stack( sys._getframe(1)) _repr_info = base_futures._future_repr_info
The key method of
Future is
future.set_result. Let’s check what will happen if we call
future.set_result.
def set_result(self, result): """Mark the future done and set its result. If the future is already done when this method is called, raises InvalidStateError. """ if self._state != _PENDING: raise exceptions.InvalidStateError(f'{self._state}: {self!r}') self._result = result self._state = _FINISHED self.__schedule_callbacks()
def __schedule_callbacks(self): """Internal: Ask the event loop to call all callbacks. The callbacks are scheduled to be called as soon as possible. Also clears the callback list. """ callbacks = self._callbacks[:] if not callbacks: return self._callbacks[:] = [] for callback, ctx in callbacks: self._loop.call_soon(callback, self, context=ctx)
Once
future.set_result is called, it would trigger
self.__schedule_callbacks asking the even loop to call all the
callbacks related to the
Future as soon as possible. These
Future related
callbacks are added or removed by
future.add_done_callback or
future.remove_done_callback. If no
Future related
callbacks, no more
callbacks are scheduled in the event loop.
So we have known what will happen after the
Future got result. What happens when the
Future is scheduled in the event loop?
From the last blog post “Python AsyncIO Event Loop”, we have seen the
Future was scheduled into the event loop via
loop.ensure_future. “If the argument is a Future, it is returned directly.” So when the
Future is scheduled in the event loop, there is almost no
callback scheduled, until the
future.set_result is called. (I said almost no
callback because there is a default
callback
_run_until_complete_cb added as we have seen in the last blog post.)')
Task
Because
_PyFuture = Future,
Task is just a derived class of
Future. The task of a
Task is to wrap a
coroutine in a
Future.
class Task(futures._PyFuture): # Inherit Python Task implementation # from a Python Future implementation. """A coroutine wrapped in a Future.""" # An important invariant maintained while a Task not done: # # - Either _fut_waiter is None, and _step() is scheduled; # - or _fut_waiter is some Future, and _step() is *not* scheduled. # # The only transition from the latter to the former is through # _wakeup(). When _fut_waiter is not None, one of its callbacks # must be _wakeup(). # If False, don't log a message if the task is destroyed whereas its # status is still pending _log_destroy_pending = True def __init__(self, coro, *, loop=None, name=None): super().__init__(loop=loop) if self._source_traceback: del self._source_traceback[-1] if not coroutines.iscoroutine(coro): # raise after Future.__init__(), attrs are required for __del__ # prevent logging for pending task in __del__ self._log_destroy_pending = False raise TypeError(f"a coroutine was expected, got {coro!r}") if name is None: self._name = f'Task-{_task_name_counter()}' else: self._name = str(name) self._must_cancel = False self._fut_waiter = None self._coro = coro self._context = contextvars.copy_context() self._loop.call_soon(self.__step, context=self._context) _register_task(self)
In the constructor, we see that the
Task schedules a
callback
self.__step in the event loop. The
task.__step is a long method, but we should just pay attention to the
try block and the
else block since these two are the ones mostly likely to be executed.
def __step(self, exc=None): if self.done(): raise exceptions.InvalidStateError( f'_step(): already done: {self!r}, {exc!r}') if self._must_cancel: if not isinstance(exc, exceptions.CancelledError): exc = self._make_cancelled_error() self._must_cancel = False coro = self._coro self._fut_waiter = None _enter_task(self._loop, self) # Call either coro.throw(exc) or coro.send(None). try: if exc is None: # We use the `send` method directly, because coroutines # don't have `__iter__` and `__next__` methods. result = coro.send(None) else: result = coro.throw(exc) except StopIteration as exc: if self._must_cancel: # Task is cancelled right before coro stops. self._must_cancel = False super().cancel(msg=self._cancel_message) else: super().set_result(exc.value) except exceptions.CancelledError as exc: # Save the original exception so we can chain it later. self._cancelled_exc = exc super().cancel() # I.e., Future.cancel(self). except (KeyboardInterrupt, SystemExit) as exc: super().set_exception(exc) raise except BaseException as exc: super().set_exception(exc) else: blocking = getattr(result, '_asyncio_future_blocking', None) if blocking is not None: # Yielded Future must come from Future.__iter__(). if futures._get_loop(result) is not self._loop: new_exc = RuntimeError( f'Task {self!r} got Future ' f'{result!r} attached to a different loop') self._loop.call_soon( self.__step, new_exc, context=self._context) elif blocking: if result is self: new_exc = RuntimeError( f'Task cannot await on itself: {self!r}') self._loop.call_soon( self.__step, new_exc, context=self._context) else: result._asyncio_future_blocking = False result.add_done_callback( self.__wakeup, context=self._context) self._fut_waiter = result if self._must_cancel: if self._fut_waiter.cancel( msg=self._cancel_message): self._must_cancel = False else: new_exc = RuntimeError( f'yield was used instead of yield from ' f'in task {self!r} with {result!r}') self._loop.call_soon( self.__step, new_exc, context=self._context) elif result is None: # Bare yield relinquishes control for one event loop iteration. self._loop.call_soon(self.__step, context=self._context) elif inspect.isgenerator(result): # Yielding a generator is just wrong. new_exc = RuntimeError( f'yield was used instead of yield from for ' f'generator in task {self!r} with {result!r}') self._loop.call_soon( self.__step, new_exc, context=self._context) else: # Yielding something else is an error. new_exc = RuntimeError(f'Task got bad yield: {result!r}') self._loop.call_soon( self.__step, new_exc, context=self._context) finally: _leave_task(self._loop, self) self = None # Needed to break cycles when an exception occurs.
Here we see the
coroutine.send method again. Each time we call
coroutine.send in the
try block, we get a
result. In the
else blcok, we always have another
self._loop.call_soon call. We do this in a trampoline fashion until
Coroutine runs out of results to
send.
Trampoline Function
import asyncio import time def trampoline(loop: asyncio.BaseEventLoop, name: str = "") -> None: current_time = time.time() print(current_time) loop.call_later(0.5, trampoline, loop, name) return current_time loop = asyncio.get_event_loop() loop.call_soon(trampoline, loop, "") loop.call_later(5, loop.stop) loop.run_forever()
The flavor of the wrapping of
Task to
Coroutine is somewhat similar to trampoline. Every time we call
coroutine.send, we got some returned values and scheduled another
callback.
Conclusion
The implementation of
asyncio is complicated and I don’t expect I could know all the details. But trying to understand more about the low-level design might be useful for implementing low-level
asyncio libraries and prevent stupid mistakes in high-level
asyncio applications.
The key to scheduling the key
asyncio awaitables,
Coroutine,
Future, and
Task, are that the awaitables are all wrapped into
Future in some way under the hood of
asyncio interface. | https://leimao.github.io/blog/Python-AsyncIO-Awaitable-Coroutine-Future-Task/ | CC-MAIN-2021-25 | refinedweb | 1,930 | 53.88 |
§Play Application Overview
This tutorial is implemented as a simple Play application that we can examine to start learning about Play. Let’s first look at what happens at runtime. When you enter in your browser:
- The browser requests the root
/URI from the HTTP server using the
GETmethod.
- The Play internal HTTP Server receives the request.
- Play resolves the request using the
routesfile, which maps URIs to controller action methods.
- The action method renders the
indexpage, using Twirl templates.
- The HTTP server returns the response as an HTML page.() { ok(views.html.index.render()); }
- Scala
def index = Action { Ok(views.html.index()) }
To view the route that maps the browser request to the controller method, open the
conf/routes file. A route consists of an HTTP method, a path, and an action. This control over the URL schema makes it easy to design clean, human-readable, bookmarkable URLs. The following line maps a GET request for the root URL
/ to the
index action in
HomeController:
GET / controllers.HomeController.index
Open
app/views/index.scala.html with your text editor. The main directive in this file calls the main template
main.scala.html with the string Welcome to generate the page. You can open
app/views/main.scala.html to see how a
String parameter sets the page title.
With this overview of the tutorial application, you are ready to add a “Hello World” greeting.
Next: Implementing Hello World | https://www.playframework.com/documentation/2.8.0/PlayApplicationOverview | CC-MAIN-2020-05 | refinedweb | 239 | 68.87 |
Details
Description
It would be nice to be able to update some fields on a document without having to insert the entire document.
Given the way lucene is structured, (for now) one can only modify stored fields.
While we are at it, we can support incrementing an existing value - I think this only makes sense for numbers.
for background, see:
Issue Links
- depends upon
-
- is depended upon by
-
- is related to
LUCENE-1879 Parallel incremental indexing
- Open
-
-
- relates to
-
-
Activity
- All
- Work Log
- History
- Activity
- Transitions
I am using apache-solr 4.0.
I am trying to post the following document -
curl -H "Content-Type: text/xml" --data-binary '<add commitWithin="5000"><doc boost="1.0"><field name="accessionNumber" update="set">3165297</field><field name="status" update="set">ORDERED</field><field name="account.accountName" update="set">US LABS DEMO ACCOUNT</field><field name="account.addresses.address1" update="set">2601 Campus Drive</field><field name="account.addresses.city" update="set">Irvine</field><field name="account.addresses.state" update="set">CA</field><field name="account.addresses.zip" update="set">92622</field><field name="account.externalIds.sourceSystem" update="set">10442</field><field name="orderingPhysician.lcProviderNumber" update="set">60086</field><field name="patient.lpid" update="set">5571351625769103</field><field name="patient.patientName.lastName" update="set">test</field><field name="patient.patientName.firstName" update="set">test123</field><field name="patient.patientSSN" update="set">643522342</field><field name="patient.patientDOB" update="set">1979-11-11T08:00:00.000Z</field><field name="patient.mrNs.mrn" update="set">5423</field><field name="specimens.specimenType" update="set">Bone Marrow</field><field name="specimens.specimenType" update="set">Nerve tissue</field><field name="UID">3165297USLABS2012</field></doc></add>'
This document gets successfully posted. However, the multi-valued field 'specimens.specimenType', gets stored as following in SOLR -
<arr name="specimens.specimenType"> <str>{set=Bone Marrow}</str> <str>{set=Nerve tissue}</str> </arr>
I did not expect "{set=" to be stored along with the text "Bone Marror".
My Solr schema xml definition for the field specimens.SpecimenType is -
<field indexed="true" multiValued="true" name="specimens.specimenType" omitNorms="false" omitPositions="true" omitTermFreqAndPositions="true" stored="true" termVectors="false" type="text_en"/>
Can someone help?
I believe yonik chose to implement it by using updateLog features.
i think it has to be - the "real time get" support provided by the updateLog is the only way to garuntee that the document will be available to atomicly update it.
Lukas: if the atomic update code path isn't throwing a big fat error if you try to use it w/o updateLog configured then that sounds to me like a bug – can you please file a Jira for that
we need to look a little closer as to why/whether the <updateLog> directive is really always needed for partial document update.
I believe yonik chose to implement it by using updateLog features.
Oh, yeah, that. I actually was going to mention it, but I wanted to focus on running with the stock Solr example first. Actually, we need to look a little closer as to why/whether the <updateLog> directive is really always needed for partial document update. That should probably be a separate Jira issue.
Ok, I finally figured it out by diffing every single difference from my test case to the stock Solr 4.0 example using git bisect.
The culprit was a missing <updateLog /> directive in solrconfig.xml. As soon as I configured a transaction log, atomic updates worked as expected. I added a note about this at .
Certainly, to the extent that I can.
about the lack of documentation
Since you are eating some of this pain, perhaps you could give a hand when you have it figured out and contribute to our wiki?
No apology necessary for the noise. I mean, none of us was able to offer a prompt response to earlier inquiries and this got me focused on actually trying the feature for the first time.
Thanks for your response. I thought I had the issue reduced to a simple enough test case, but apparently not. I will try again with a clean stock Solr 4.0, and file a seperate issue if necessary, or look for support on the mailing list. My choice of words ('doesn't work as advertised') might have been influenced by frustration about the lack of documentation, sorry for the noise.
I just tried it and the feature does work as advertised. If there is a bug, that should be filed as a separate issue. If there is a question or difficulty using the feature, that should be pursued on the Solr user list.
For reference, I took a fresh, stock copy of the Solr 4.0 example, no changes to schema or config, and added one document:
curl 'localhost:8983/solr/update?commit=true' -H 'Content-type:application/json' -d ' [{"id":"id-123","title":"My original Title", "content": "Initial content"}]'
I queried it and it looked fine.
I then modified only the title field:
curl 'localhost:8983/solr/update?commit=true' -H 'Content-type:application/json' -d ' [{"id":"id-123","title":{"set":"My new title"}}]'
I tried the XML equivalents and that worked fine as well, with the original content field preserved.
This feature doesn't work as advertised in Solr 4.0.0 (final).
Since it's not documented, I used the information in these blog posts (yonik.com, solr.pl) and this ticket to try to get it working, and asked in the #solr IRC channel, to no avail.
Whenever I use the 'set' command in an update message, it mangles the value to something like
<str name="Title">{set=My new title}</str>
, and drops all other fields.
I tried the JSON as well as the XML Syntax for the update message, and I tried it with both a manually defined 'version' field and without.
Relevant parts from my schema.xml:
<schema name="solr-instance" version="1.4"> <fields> <field name="Creator" type="string" indexed="true" stored="true" required="false" multiValued="false" /> <!-- ... --> <field name="Title" type="text" indexed="true" stored="true" required="false" multiValued="false" /> <field name="UID" type="string" indexed="true" stored="true" required="true" multiValued="false" /> <field name="_version_" type="long" indexed="true" stored="true" required="false" multiValued="false" /> </fields> <!-- ... --> <uniqueKey>UID</uniqueKey> </schema>
I initially created some content like this:
$ curl 'localhost:8983/solr/update?commit=true' -H 'Content-type:application/json' -d '[{"UID":"7cb8a43c","Title":"My original Title", "Creator": "John Doe"}]'
Which resulted in this document:
<doc> <str name="UID">7cb8a43c</str> <str name="Title">My original Title</str> <str name="Creator">John Doe</str> </doc>
Then I tried to update that document with this statement:
$ curl 'localhost:8983/solr/update?commit=true' -H 'Content-type:application/json' -d '[{"UID":"7cb8a43c","Title":{"set":"My new title"}}]'
Which resulted in this mangled document:
<doc> <str name="UID">7cb8a43c</str> <str name="Title">{set=My new title}</str> </doc>
(I would have expected the document to still have the value 'John Doe' for the 'Creator' field,
and have the value of its 'Title' field update to 'My new title')
I tried using the XML format for the update message as well:
<add> <doc> <field name="UID">7cb8a43c</field> <field name="Title" update="set">My new title</field> </doc> </add>
Same result as above.
Thanks!
Hi,
I'm a fan of the feature, but not really a fan of the syntax, for the following reasons:
- It is extremely verbose for batch update operations, e.g. setting a new field on all documents in the index. Surely the update modes should be specified outside of each individual record (either as URL query parameters, or in some content header/wrapper). The current approach is entirely inappropriate for extending to CSV, which might otherwise be an obvious choice of format when adding a single field to each of a set of objects.
- The distinction between an "insert" and an "update" operation (in SQL terms) is implicit, only indicated by the presence of an object in a JSON value, or by the presence of update in any one of the specified fields. Since insert and update operations are quite distinct on the server, it should select between these on a per-request basis, not per-record.
- The JSON syntax would appear as if one could extend {"set":100}
to{"set":100,"inc":2}
on the same field, which is nonsense. It uses JSON a object for inappropriate semantics, where what one actually means is{"op":"set","val":100}
, or even{"name":"price","op":"set","val":100}
.
- It may be worth reserving JSON-object-as-value for something more literal in the future.
I will be out of the office until the 29th of October.
If you need immediate assistance, please contact IT Integration (itintegration@paml.com) or my manager Jon Tolley (jtolley@paml.com).
Thanks.
PAML EMAIL DISCLAIMER:
Information contained in this message may be privileged and confidential.
If the reader of this message is not the intended recipient, be notified
that any dissemination, distribution or copying of this communication is
strictly prohibited. If this communication is received in error, please
notify the sender immediately by replying to the message and deleting
from your computer. Thank you
Can we get this on the Wiki somewhere? I've been looking around, haven't been able to find it. Not sure where to put it...
It appears that SolrJ does not yet (as of 4.0 Alpha) support updating fields in a document. Is there a separate Jira ticket for this?
Unassigned issues -> 4.1
Christopher,
Here is how I am able to update a document by posting an XML
<add>
<doc>
<field name="id">VA902B</field>
<field name="price" update="set">300</field>
</doc>
</add>
Yonik,
Do you have an example with the XML syntax? I have been trying to test this in 4.0-Beta, but am obviously not grokking the right syntax =(
Also, have you tried to use this with a join query? I can think of some interesting use cases =)
Regards
The create-if-not-exist patch was committed to both trunk and 4x branch.
Per the discussion on the mailing list, here's a patch that creates the document being updated if it doesn't exist already. The standard optimistic concurrency mechanism can be used to specify that a document must exist if desired.
Yonik,
it's hard to follow for me. Can't you clarify what's actually can be updated stored/indexed field, field cache? Where update will be searchable?
Thanks
The schema has "price" and "price_c" (copy field from price)
For this feature to work correctly, source fields (the ones you normally send in) should be stored, and copyField targets (like price_c) should be un-stored.
yonik, I see a small issue with this.
- I have added one of the example document from exampledocs folder. The schema has "price" and "price_c" (copy field from price)
1.When i add the document, it">185.0</float>
<arr name="price_c">
<str>185,USD</str>
</arr>
2. Now I want to set price field with 100 , so I sent a json for it.
[{"id":"TWINX2048-3200PRO","price":{"set":100}}]
Now the document">100.0</float>
<arr name="price_c">
<str>100,USD</str>
<str>185,USD</str>
</arr>
as you can see, the old price value still there in the "price_c". Is there a workaround/patch we can do for this?
Committed (5 years after the issue was opened!)
I'll keep this issue open and we can add follow-on patches to implement increment and other set operations.
Here's an updated patch with XML support and cloud tests.
The underlying mechanism for updating a field always change in the future (and depending on the field type), so the really important part of this is the API.
I plan on committing soon unless someone comes does come up with some API improvements.
David, please see LUCENE-3837 for a low-level partial update of inverted fields without re-indexing other fields. That is very much work in progress, and it's more complex. This issue provides a shortcut to a "retrieve stored fields, modify, delete original doc, add modified doc" sequence that users would have to execute manually.
Yonik; I don't see your patch.
Try sorting by date, or click on the "All" tab and you can see where I added it.
what is the essence of the implementation?
This is the simplest form. Original fields need to be stored. The document is retrieved and then re-indexed after modification.
Yonik; I don't see your patch. Will this support the ability to replace a text field that is indexed? If so, what is the essence of the implementation? The most common use-case I need for my employer involves essentially a complete re-indexing of all docs but only for a few particular fields. I tend to think, at least for my use-case, that it is within reach if I had enough time to work on it. The implementation concept I have involves building parallel segments with aligned docids. Segment merging would clean out older versions of a field.
Cool. Any plans for supporting modification of existing value?
Definitely!
- increment or inc (add is taken for adding additional field values). decrement not needed (just use a negative increment)
- append/prepend (and maybe allow "add" to mean "append" if a text/string field is multi-valued)
- set operations for multi-valued fields (union, intersection, remove, etc)
- we could get into conditionals, but at some point we should just punt that to a script-updator (i.e. update this document using the given script)
Cool. Any plans for supporting modification of existing value? Most useful would be add, subtract (for numeric) and append text for textual. (In FAST ESP we had this as part of the partial update APIs)
I'm working in getting an XML syntax going. The easiest/least disruptive seems to be an attribute:
Current:
<field name="foo" boost="2.5">100</foo>
Proposed:
<field name="foo" boost="2.5" update="add">100</foo>
Here's a patch for updateable docs that reuses the infrastructure we put in place around versioning and realtime-get.
Only the JSON parser is currently implemented.
Recall that a normal field value in JSON is of the form:
"myfield":10
This patch extends the JSON parser to support extended values as a Map. For example, to add an additional value to a multi-valued field, you would use
"myfield":{"add":10}
Using an existing data structure (a Map) should allow us to pass this through javabin format w/o any additional changes.
This patch depends on optimistic locking (it's currently included in this patch) and updating documents fully supports optimistic locking (i.e. you can conditionally update a document based on it's version)
Right now only "add" and "set" are supported as mutators (and setting a null value is like "remove"), but I figured it would be best to do a slice first from start to finish.
I need this feature. How much do I have to pay in order to get this issue fixed? Can I pass around a piggy bank?
This seems like a very old issue as someone suggested. Is there any update on whether it will ever be resolved? It is quite important feature and causes big problems when you have a huge index and only need to update one column. notice this is a very long lived issue and that it is marked for 1.5. Are there outstanding issues or problems with its usage if I apply it to my 1.4 source?
ParallelReader assumes you have two indexes that "line up" so the internal docids match. Maintaining something like that would currently be pretty hard or impractical.
It would make sense of adding ParallelReader functionality so a core can read from several index-dirs.
Guess it complicates things a little since you would need to have support for adding data as well to more than one index.
Suggestion:
/update/coreX/index1 - Uses schema1.xml
/update/coreX/index2 - Uses schema2.xml
/select/coreX - Uses all schemas e.g. A ParallelReader.
Seing quite a lot questions on the mailinglist about users that want to be able to update a single field while maintaining the rest of the index intact (not reindex).
Marking for 1.5
there are pros and cons w/ each approaches (am I discovering a universal truth here
)
Many approaches can to confuse users . I can propose something like
<modifiable>true</modifiable> in the manIndex .
And say
<modificationStrategy>solr.SepareteIndexStrategy</modificationStrategy>
or
<modificationStrategy>solr.SameIndexStrategy</modificationStrategy>
(I did not mention the other two because of personal preferences
)
to me the key is getting an interface that would allow for the existing fields to be stored a number of ways:
- within the index itself
- within an independent index (as you suggest)
- within SQL
- on the file system
- ...
Shall I open another issue on my idea keeping another index for just stored fields (in an UpdateRequestProcessor).
Is it a good idea to have multiple approaches for the same feature?
Or should I post the patch in this issue only?
- If your index is really big most likely you may have a master/slave deployment. In that case only master needs to store the data copy. The slaves do not have to pay the 'update tax'
Perhaps we want to write the xml to disk when it is indexed, then reload it when the file is 'updated', perhaps the content should be stored in a SQL db.
Writing the xml may not be an option if I use DIH for indexing (there is no xml). And what if I use CSV. That means multiple storage formats. RDDBMS storage is a problem because of incompatibility of Lecene/RDBMS data structures. Creating a schema will be extremely hard because of dynamic fields
I guess we can have multiple solutions. I can provide a simple DuplicateIndexUpdateProcessor (for lack of a better name) which can store all the data in a duplicate index. Let the user decide what he wants
It's possible to recover unstored fields, if the purpose of such recovery is to make a copy of the document and update other fields.
It is not wise to invest our time to do 'very clever' things because it is error prone. Unless Lucene gives us a clean API to do so
[quote.[/quote]
Splitting it out into another store is much better at scale. A distinct lucene index works relatively well.
It's possible to recover unstored fields, if the purpose of such recovery is to make a copy of the document and update other fields. The process is time-consuming, because you need to traverse all postings for all terms, so it might be impractical for larger indexes. Furthermore, such recovered content may be incomplete - tokens may have been changed or skipped/added by analyzers, positionIncrement gaps may have been introduced, etc, etc.
Most of this functionality is implemented in Luke "Restore & Edit" function. Perhaps it's possible to implement a new low-level Lucene API to do it more efficiently.
Ryan, I know of course the index can get big because one needs to store all the data for re-indexing; but due to Lucene's fundamental limitations, we can't get around that fact. Moving the data off to another place (a DB of some sort or whatever) doesn't change the fundamental problem. If one is unwilling to store a copy somewhere convenient due to data scalability issues then we simply cannot support the feature because Lucene doesn't have an underlying update capability.
If the schema's field isn't stored, then it may be useful to provide an API that can fetch un-stored fields for a given document. I don't think it'd be common to use the API and definitely wouldn't be worth providing a default implementation.
There are many approaches to make this work – i don't think there will be a one-size-fits all approach though. Storing all fields in the lucene index may be fine. Perhaps we want to write the xml to disk when it is indexed, then reload it when the file is 'updated', perhaps the content should be stored in a SQL db..
In general, I think we just need an API that will allow for a variety of storage mechanisms.
It's unclear to me how your suggestion, Paul, is better. If at the end of the day, all the fields must be stored on disk somewhere (and that is the case), then why complicate matters and split out the storage of the fields into a separate index? In my case, nearly all of my fields are stored so this would be redundant. AFAIK, having your "main" index "small" only matters as far as index storage, not stored-field storage. So I don't get the point in this.
How about maintaining a separate index at store.index and write all the documents to that index also.
In the store.index , all the fields must be stored and none will be indexed. This index will not have any copy fields . It will blindly dump the data as it is.
Updating can be done by reading data from this. Deletion must also be done on both the indices
This way the original index will be small and the users do not have to make all fields stored
the biggest reason this patch won't work is that with
SOLR-559, the DirectUpdateHandler2 does not keep track of pending updates – to get this to work again, it will need to maintain a list somewhere.
Also, we need to make sure the API lets you grab stored fields from somewhere else – as is, it forces you to store all fields for all documents.
I noticed that this bug is no longer included in the 1.3 release. Are there any outstanding issues if all the fields are stored? Requiring that all fields are stored for a document to be update-able seems like reasonable to me. This feature will simplify things for Solr users who are doing a query to get all the fields following by an add when they only want to update a very small number of fields.
For non-stored fields that need to be retained (Otis & Yoniks concern)... I wonder what Lucene exposes about the indexed data for a non-stored field. We'd just want to copy this low-level data over to a new document, basically.
Lucene maintains an inverted index, so the "indexed" part is spread over the entire index (terms point to documents). Copying an indexed field would require looping over every indexed term (all documents with that field). It would be very slow once an index got large.
I agree Lance. I don't know what the "first design" is that might be bogus you're talking about... but we definitely eventually want to handle non-stored fields. I have an index that isn't huge and I'm salivating at the prospects of doing an update. For non-stored fields... it seems that if an update always over-writes such fields with new data then we should be able to support that now easily because we don't care what the old data was.
For non-stored fields that need to be retained (Otis & Yoniks concern)... I wonder what Lucene exposes about the indexed data for a non-stored field. We'd just want to copy this low-level data over to a new document, basically.
updated patch to work with trunk
If I may comment?
Is it ok if a first implementation requires all fields to stored, and then a later iteration supports non-stored fields? This seems to be a complex problem, and you might decide later that the first design is completely bogus.
I'm looking at this issue now. I used to think "sure, this will work fine", but I'm looking at a 1B doc index (split over N servers) and am suddenly very scared of having to store more than necessary in the index. In other words, writing custom field value loaders from external storage sounds like the right thing to do. Perhaps one such loader could simply load from the index itself.
that is part of why i thought having it in an update request processor makes sense – it can easily be subclassed to pull the existing fields from whereever it needs. Even if it is directly in the UpdateHandler, there could be some interface to loadExistingFields( id ) or something similar.
I'm having second thoughts if this is a good enough approach to really put in core Solr.
Requiring that all fields be stored is a really large drawback, esp for large indicies with really large documents.
updated to work with trunk. no real changes.
The final version of this will need to move updating logic out of the processor into the UpdateHandler
A useful feature would be "update based on query", so that documents matching the query condition will all be modified in the same way on the given update fields.
This feature would correspond to SQL's UPDATE command, so that Solr would now cover all the basic commands SQL provides. (While this is a theoretic motivation, I just missed this feature for my Solr projects..)
applies with trunk...
applies to /trunk
I'm back and have some time to focus on this... Internally, I need to move the document modification to the update handler, but before I get going it would be nice to agree what we want the external interface to look like.
- - -
Should we deprecate the AddUpdateCommand and replace it with something else? Do we want one command to do Add/Update/Modify? Two?
- - -
what should happen if you "modify" a non-existent document?
a. error – makes sense in a 'tagging' context
b. treat it as an add – makes sense in a "keep these fields up to date" context (i don't want to check if the document already exists or not)
I happen to be working with context b, but I suspect 'a' makes more sense.
- - -
Should we have different xml syntax for document modification vs add? Erik suggested:
<update overwrite="title" distinct="cat">
...
</update>
since 'update' is already used in a few ways, maybe <modify>?
Should we put the id as an xml attribute? this would make it possible to change the unique key.
<modify id="ID">
<field name="id">new id</field>
</modify>
That may look weird if the uniqueKeyField is not called "id"
Assuming we put the modes as attributes, I guess multiple fields would be comma delimited?
<modify distinct="cat,keyword">
Do you like the default mode called "default" or "mode"?
<modify id="ID" default="overwrite">?
<modify id="ID" mode="overwrite">?
Updated Erik's patch to /trunk and added back the solrj modifiable document tests.
For posterity, here's the patch I'm running in Collex production right now, and its working fine thus far.
Sorry, I know this issue has a convoluted set of patches to follow at the moment. I trust that when Ryan is back in action we'll get this tidied up and somehow get Yonik-satisfying refactorings
Added DELETE support in ModifyDocumentUtils like this:
case DELETE:
if( field != null ) {
Collection<Object> collection = existing.getFieldValues(name);
if (collection != null) {
collection.remove(field.getValue());
if (collection.isEmpty())
else{ existing.setField(name, collection, field.getBoost()); }
}
}
// TODO: if field is null, should the field be deleted?
break;
> how about using <update> instead of <add>
We had previously talked about making this distinction (and configuration for each field) in the URL:
This makes it usable and consistent for different update handlers and formats (CSV, future SQL, future JSON, etc)
but perhaps if we allowed the <add> tag to optionally be called something more neutral like <docs>?
wrt patches, I think the functionality needs refactoring so that modify document logic is in the update handler. It seems like it's the only clean way from a locking perspective, and it also leaves open future optimizations (like using different indices depending on the fieldname and using a parallel reader across them).
Some thoughts on the update request, how about using <update> instead of <add>? And put the modes as attributes to <update>...
<update overwrite="title" distinct="cat">
<field name="id">ID</field>
<field name="title">new title</field>
<field name="cat">NewCategory</field>
</update>
Using <add> has the wrong implication for this operation. Thoughts?
There is a bug in the last patch that allows an update to a non-existent document to create a new document.
I've corrected this by adding this else clause in ModifyExistingDocumentProcessor:
if( existing != null ){ cmd.solrDoc = ModifyDocumentUtils.modifyDocument(existing, cmd.solrDoc, modes, schema ); }
else{ throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "Cannot update non-existent document: " + id); }
I can't generate a patch, yet, that is clean addition of just that bit of code along with the other changes.
> One thing I noticed is that all fields sent in the update must be stored, but that doesn't really need to be the case with fields being overwritten - perhaps that restriction should be lifted and only applies when the stored data is needed.
+1
> Perhaps a "text" field that could be optionally set by the client, and is also the destination of copyField's of title, author, etc.
Seems like things of this form can always be refactored by adding a field "settableText" and adding another copyField from that to "text"
A mistake I had was my copyField target ("tag") was stored. Setting it to be unstored alleviated the need to overwrite it - thanks!
One thing I noticed is that all fields sent in the update must be stored, but that doesn't really need to be the case with fields being overwritten - perhaps that restriction should be lifted and only applies when the stored data is needed.
As for sending in a field that was the target of a copyField - I'm not doing this nor can I really envision this case, but it seemed like it might be a case to consider here. Perhaps a "text" field that could be optionally set by the client, and is also the destination of copyField's of title, author, etc.
I think there are a number of restrictions for using this feature:
1) all source fields (not copyField targets) need to be stored, or included in modify commands
2) copyField targets should either be unstored, or if stored, should never be explicitly set by the user
What tags would you send in that aren't user tags? If they are some sort of global tags, then you could do global_tag=foo (reusing your dynamic user_tag field), or create a new globalTags field and an additional copyField to the tag field.
I agree in the gut feel to how copyFields and overwrite should work, but what about the situation where the client is sending in values for that field also? If it got completely regenerated during a modify operation, data would be lost. No?
Thanks for the note about field-per-user strategy. In our case, even thousands of users is on down the road. The simplicity of having fields per user is mighty alluring and is working nicely in the prototype for now.
IMO, copyField targets should always be re-generated... so no, it doesn't seem like you should have to say anything about tag if you update erik_tag.
On a side note, I'm not sure how scalable the field-per-user strategy is. There are some places in Lucene (segment merging for one) that go over each field, merging the properties. Low thousands would be OK, but millions would not be OK.
One thing to note about overwrite and copyFields is that to keep a purely copyFielded field in sync you must basically remove it (overwrite without providing a value).
For example, my schema:
<dynamicField name="*_tag" type="string" indexed="true" stored="true" multiValued="true"/>
<field name="tag" type="string" indexed="true" stored="true" multiValued="true"/>
and then this:
<copyField source="*_tag" dest="tag"/>
The client never provides a value for "tag" only ever <username>_tag values. I was seeing old values in the tag field after doing overwrites of <username>_tag expecting "tag" to get rewritten entirely. Saying mode=tag:OVERWRITE does the trick. This is understandable, but confusing, as the client then needs to know about purely copyFielded fields that it never sends directly.
I'm experimenting with this patch with tagging. I'm modeling the fields in this way, beyond general document metadata fields:
<username>_tags
usernames
And copyFielding *_tags into "tags".
usernames field allows seeing all users who have tagged documents.
Users are allowed to "uncollect" an object, which would remove the <username>_tags field and remove their name from the usernames field. Removing the <username>_tags field use case is covered with the <username>_tags:OVERWRITE mode. But removing a username from the multiValued and non-duplicating usernames field is not.
An example (from some conversations with Ryan):
id: 10
usernames: ryan, erik
You want to be able to remove 'ryan' but keep 'erik'.
Perhaps we need to add a 'REMOVE' mode to remove the first (all?)
matching values
/update?mode=OVERWRITE,username=REMOVE
<doc>
id=10,
usernames=ryan
</doc>
and make the output:
id: 10
usernames: erik
But what about duplicate values?
applies with trunk
Updated to use Yonik's 'getStoredFields'. The one change to that is to have UpdateHandler addFields() load fields as the Object (not external string) and to skip copy fields:
void addFields(Document luceneDoc, SolrDocument solrDoc) {
for (Fieldable f : (List<Fieldable>)luceneDoc.getFields()) {
SchemaField sf = schema.getField( f.name() );
if( !schema.isCopyFieldTarget( sf ) )
}
}
- - - -
This still implements modifiable documents as a RequestProcessor.
>
> So I think that we need a modifyDocument() call on updateHandler, and perhaps a ModifyUpdateCommand to go along with it.
>
> I'm not sure yet what this means for request processors. Perhaps another method that handles the reloaded storedFields?
>
Another option might be some sort of transaction or locking model. Could it block other requests while there is an open transaction/lock?
Consider the case where we need the same atomic protection for fields loaded from non-stored fields loaded from a SQL database. In this case, it may be nice to have locking/blocking happen at the processor level.
I don't know synchronized well enough to know if this works or is a bad idea, but what about something like:
class ModifyExistingDocumentProcessor {
void processAdd(AddUpdateCommand cmd) {
String id = cmd.getIndexedId(schema);
synchronized( updateHandler )
}
}
This type of approach would need to make sure everyone modifying fields was locking on the same updateHandler.
- - - -
I'm not against adding a ModifyUpdateCommand, I just like having the modify logic sit outside the UpdateHandler.
So the big issue now is that I don't think we can use getStoredFields() and do document modification outside the update handler. The biggest reason is that I think we need to be able to update documents atomically (in the sense that updates should not be lost).
Consider the usecase of adding a new tag to a multi-valued field: if two different clients tag a document at the same time, it doesn't seem acceptable that one of the tags could be lost. So I think that we need a modifyDocument() call on updateHandler, and perhaps a ModifyUpdateCommand to go along with it.
I'm not sure yet what this means for request processors. Perhaps another method that handles the reloaded storedFields?
Attaching a patch for getStoredFields that appears to work.
Darn, you're right: writer.addDocument() is outside of the synchronized block.
We could do as you suggested, downgrading to a read lock from commit. It should only reduce concurrently when the document is in pending state.
The locking logic for getStoredFields() is indeed flawed.
closing the writer inside the sync block of getStoredFields() doesn't project callers of addDoc() from concurrently using that writer. The commit lock aquire will be needed after all... no getting around it I think.
I disabled logging on all of "org.apache.solr" via a filter, and voila, OOM problems are gone.
Perhaps the logger could not keep up with the number of records and they piled up over time time (does any component of the logging framework use another thread that might be getting starved?)
Anyway, it doesn't look like Solr has a memory leak.
On to the next issue.
OOM still happens from the command line also after lucene updates to 2.2.
Looks like it's time for old-school instrumentation (printfs, etc).
Weird... I put a profiler on it to try and figure out the OOM issue, and I never saw heap usage growing over time (stayed between 20M and 30M), right up until I got an OOM (it never registered on the profiler).
Here's the latest patch, with the beginnings of a test, but I've run into some issues.
1) I'm getting occasional errors about the index writer already being closed (with lucene trunk), or null pointer exception with the lucene version we have bundled. It's the same issue I believe... somehow the writer gets closed when someone else is trying to use it. If I comment out verifyLatest() in the test (that calls getStoredFields), it no longer happens.
2) When I comment out getStoredFields, things seem to run correctly, but memory grows without bound and i eventually run out. 100 threads each limiting themselves to manipulating 16 possible documents each should not be able to cause this.
At this point, I think #2 is the most important to investigate. It may be unrelated to this latest patch, and could be related to other peoples reports of running out of memory while indexing.
Thanks Mike, I've made those changes to my local copy.
It is my fault that the DUH2 locking is so hairy to begin with, so I should at least review changes to it
With your last change, the locking looks sound. However, I noticed a few things:
This comment is now inaccurate:
+ // need to start off with the write lock because we can't aquire
+ // the write lock if we need to.
Should openSearcher() call closeSearcher() instead of doing it manually? It looks like searcherHasChanges is not being reset to false.
Found another bug just by looking at it a little longer... pset needs to be synchronized if all we have is the read lock, since addDoc() actually changes the pset while only holding the readLock (but it also synchronizes). I expanded the sync block to encompass the pset access and merged the two sync blocks. Hopefully this one should be correct.
updated getStoredFields.patch to use synchronization instead of the write lock since you can't upgrade a read lock to a write lock. I could have started out with the write lock and downgraded it to a read lock, but that would probably lead to much worse concurrency since it couldn't proceed in parallel with other operations such as adds.
It would be good if someone could review this for threading issues.
Here's a patch that adds getStoredFields to updateHandler.
Changes include leaving the searcher open as long as possible (and in conjunction with the writer, if the reader has no unflushed deletions). This would allow reuse of a single index searcher when modifying multiple documents that are not in the pending set.
No tests for getStoredFields() yet... I plan on a comprehensive update handler test that uses multiple threads to try and flush out any possible concurrency bugs.
FYI, I'm working on a loadStoredFields() for UpdateHandler now.
>
> ... avoid touching the less-modified/bigger fields ...
>
aaah, perhaps a future updateHandler getDocument() function could take a list of fields it should extract. Still problems with what to do when you add it.. maybe it checks if anything has changed in the less-modified index? I see your point.
>> What are you thinking? Adding the processor as a parameter to AddUpdateCommand?
>
> I didn't have a clear alternative... I was just pointing out the future pitfalls of assuming too much implementation knowledge.
>
I am fine either way – in the UpdateHandler or the Processors.
Request plumbing-wise, it feels the most natural in a processor. But if we rework the AddUpdateCommand it could fit there too. I don't know if it is an advantage or disadvantage to have the 'modify' parameters tied to the command or the parameters. either way has its +-, with no real winner (or loser) IMO
In the end, I want to make sure that I never need a custom UpdateHandler (80% is greek to me), but can easily change the 'modify' logic.
>> ... ParallelReader, where some fields are in one sub-index ...
> the processor would ask the updateHandler for the existing document - the updateHandler deals with
> getting it to/from the right place.
The big reason you would use ParallelReader is to avoid touching the less-modified/bigger fields in one index when changing some of the other fields in the other index.
> What are you thinking? Adding the processor as a parameter to AddUpdateCommand?
I didn't have a clear alternative... I was just pointing out the future pitfalls of assuming too much implementation knowledge.
>
> The update handler could call the processor when it was time to do the manipulation too.
>
What are you thinking? Adding the processor as a parameter to AddUpdateCommand?
> ... ParallelReader, where some fields are in one sub-index ...
the processor would ask the updateHandler for the existing document - the updateHandler deals with getting it to/from the right place.
we could add something like:
Document getDocumentFromPendingOrCommited( String indexId )
to UpdateHandler and then that is taken care of.
Other then extracting the old document, what needs to be done that cant be done in the processor?
> I like having the the actual document manipulation happening in the Processor because it is an easy
> place to put in other things like grabbing stuff from a SQL database.
The update handler could call the processor when it was time to do the manipulation too.
Consider the case of future support for ParallelReader, where some fields are in one sub-index and some fields are in another sub-index. I'm not sure if our current processor interfaces will be able to handle that scenario well (but I'm not sure if we should worry about it too much right now either).
There is another way to perhaps mitigate the problem of a few frequently modified documents causing thrashing: a document cache in the update handler.
> the udpate handler knows much more about the index than we do outside
Yes. The patch i just attached only deals with documents that are already commited. It uses req.getSearcher() to find existing documents.
Beyond finding commited or non-commited Documents, is there anything else that it can do better?
Is it enought to add something to UpdateHandler to ask for a pending or commited document by uniqueId?
I like having the the actual document manipulation happening in the Processor because it is an easy place to put in other things like grabbing stuff from a SQL database.
Updated patch to work with
SOLR-269 UpdateRequestProcessors.
One thing I think is weird about this is that it uses parameters to say the mode rather then the add command. That is, to modify a documetn you have to send:
/update?mode=OVERWRITE,count:INCREMENT
<add>
<doc>
<field name="id">1</field>
<field name="count">5</field>
</doc>
</add>
rather then:
<add mode="OVERWRITE,count:INCREMENT">
<doc>
<field name="id">1</field>
<field name="count">5</field>
</doc>
</add>
This is fine, but it makes it hard to have an example 'modify' xml document.
Some general issues w/ update processors and modifiable documents, and keeping this stuff out of the update handler is that the udpate handler knows much more about the index than we do outside, and it constrains implementation (and performance optimizations).
For example, if modifiable documents were implemented in the update handler, and the old version of the document hasn't been committed yet, the update handler could buffer the complete modify command to be done at a later time (the much slower alternative is closing the writer and opening the reader to get the latest stored fields), then closing the reader and re-opening the writer.
implementing modifiable documents in UpdateRequestProcessor.
This adds two example docs:
modify1.xml and modify2.xml
<add mode="OVERWRITE,price:INCREMENT,cat:DISTINCT">
<doc>
<field name="id">CSC</field>
<field name="name">Campbell's Soup Can</field>
<field name="manu">Andy Warhol</field>
<field name="price">23.00</field>
<field name="popularity">100</field>
<field name="cat">category1</field>
</doc>
</add>
will increment the price by 23 each time it is run.
> (as long as you don't need multiple of them).
If you had a bunch of independent things you wanted to do, having a single processor forces you to squish them together (say you even added another document type to your index and then wanted to add another mutation operation).
What if we had some sort of hybrid between the two approaches. Had a list of processors, and the last would have primary responsibility for getting the documents in the index?
> How can the UpdateHandler get access to pending documents? should it just use req.getSearcher()?
I don't think so... the document may have been added more recently than the last searcher was opened.
We need to use the IndexReader from the UpdateHandler. The reader and writer can both remain open as long as the reader is not used for deletions (which is a write operation and hence exclusive with IndexWriter).
> So you are suggesting [...]
I don't have a concrete implementation idea, I'm just going over all the things I know people will want to do (and many of these I have an immediate use for).
> Do you mean as input or output?
Input, for index-only fields. Normally field values need to be stored for an "update" to work, but we could also allow the user to get these field values from an external source.
> we would need a hook at the end.
Yes, it might make sense to have more than one callback method per UpdateRequestProcessor
Of course now that I finally look at the code, UpdateRequestProcessor isn't quite what I expected.
I was originally thinking more along the lines of DocumentMutator(s) that manipulate a document, not that actually initiate the add/delete/udpate calls. But there is a certain greater power to what you are exposing/allowing too (as long as you don't need multiple of them).
In UpdateRequestProcessor , instead of
protected final NamedList<Object> response;
Why not just expose SolrQueryRequest, SolrQueryResponse?
So you are suggesting pulling this out of the UpdateHandler and managing the document merging in the UpdateRequestProcessor? (this might makes sense - It was not an option when the patch started in feb)
How can the UpdateHandler get access to pending documents? should it just use req.getSearcher()?
>.
>
1 & 2 seem pretty straightforwad
> example3: some field values are pulled from a database when missing rather than being stored values.
>
Do you mean as input or output? The UpdateRequestProcessor could not affect if a field is stored or not, it could augment a document with more fields before it is indexed. To add fields from a database rather then storing them, we would need a hook at the end.
Another use case to keep in mind... someone might use an UpdateRequestProcessor
to add new fields to a document (something like copyField, but more complex). Some of this logic might be based on the value of other fields themselves..
example3: some field values are pulled from a database when missing rather than being stored values.
I think we need to try to find the right UpdateRequestProcessor interface to support all these with UpdateableDocuments.
- Updated for new SolrDocument implementation.
- Testing via solrj.
I think this should be committed soon. It should only affect performance if you specify a 'mode' in the add command.
- Updated to work with trunk.
- moved MODE enum to common
This could be added without affecting any call to the XmlUpdateHandler – (for now) this only works with the StaxUpdateHandler. Adding it will help clarify some of the DocumentBuilder needs/issues...
updated with trunk
An updated and cleaned up versoin. The big change is to the SolrDocument interface - rather then expose Maps directly, it hides them in an interface:
Still needs some test cases
added missing files. this should is ready to check over if anyone has some time
This is a minor change that keeps the OVERWRITE property if it is specified. The previous version ignored the update mode if everthing was OVERWRITE.
I added a new version of
SOLR-139-IndexDocumentCommand.patch that:
- gets rid of 'REMOVE' option
- uses a separate searcher to search for existing docs
- includes the XmlUpdateRequestHandler
- moves general code from XmlUpdateRequestHandler to SolrPluginUtils
- adds a few more tests
Can someone with a better lucene understanding look into re-useint the existing searcher as Yonik suggests above - I don't quite understand the other DUH2 implications.
I moved the part that parses (and validates) field mode parsing into SolrPluginUtils. This could be used by other RequestHandlers to parse the mode map.
The XmlUpdateRequestHandler in this patch should support all legacy calls except cases where overwritePending != overwriteCommitted. There are no existing tests with this case, so it is not a problem from the testing standpoint. I don't know if anyone is using this (mis?) feature.
> I added a second searcher to DirectUpdatehandler2 that is only closed when you call commit();
That's OK for right now, but longer term we should re-use the other searcher.
Previously, the searcher was only used for deletions, hence we always had to close it before we opened a writer.
If it's going to be used for doc retrieval now, we should change closeSearcher() to flushDeletes() and only really close the searcher if there had been deletes pending (from our current deleteByQuery implementation).
>
> That wouldn't work for multi-valued fields though, right?
'REMOVE' on a mult-valued field would clear the old fields before adding the new ones. It is essentially the same as OVERWRITE, but you may or may not pass in a new value on top of the old one.
> If we keep this option, perhaps we should find a better name...
how about 'IGNORE' or 'CLEAR' It is awkward because it refers to what was in in the document before, not what you are passing in.
The more i think about it, I think we should drop the 'REMOVE' option. You can get the same effect using 'OVERWRITE' and passing in a null value.
Is this what you are suggesting?
I added a second searcher to DirectUpdatehandler2 that is only closed when you call commit();
// Check if the document has not been commited yet
Integer cnt = pset.get( id.toString() );
if( cnt != null && cnt > 0 )
if( committedSearcher == null ){ committedSearcher = core.newSearcher("DirectUpdateHandler2.committed"); }
Term t = new Term( uniqueKey.getName(), uniqueKey.getType().toInternal( id.toString() ) );
int docID = committedSearcher.getFirstMatch( t );
if( docID >= 0 )
- - - - - - -
This passes a new test that adds the same doc multiple times. BUT it does commit each time.
Another alternative would be to keep a Map<String,Document> of the pending documents in memory. Then we would not have to commit each time something has changed.
If modes are in a single param, a ":" syntax might be nicer because you could cut-n-paste it into a URL w/o escaping.
> I'm using 'REMOVE' to say "remove the previous value of this field before doing anything." Essentially, this makes sure you new
> document does not start with a value for 'sku'.
That wouldn't work for multi-valued fields though, right?
If we keep this option, perhaps we should find a better name... to me "remove" suggests that the given value should be removed. Think adding / removing tag values from a document.
>
> If the field modes were parameters, they could be reused for other update handlers like SQL or CSV
> Perhaps something like:
> /update/xml?modify=true&f.price.mode=increment,f.features.mode=append
>
Yes. I think we should decide a standard 'modify modes' syntax that can be used across handlers. In this example, I am using the string:
mode=cat=DISTINCT,features=APPEND,price=INCREMENT,sku=REMOVE,OVERWRITE
and passing it to 'parseFieldModes' in XmlUpdateRequestHandler.
Personally, I think all the modes should be specified in a single param rather then a list of them. I vote for a syntax like:
<lst name="params">
<str name="mode">cat=DISTINCT,features=APPEND,price=INCREMENT,sku=REMOVE,OVERWRITE</str>
</lst>
or:
<lst name="params">
<str name="mode">cat:DISTINCT,features:APPEND,price:INCREMENT,sku:REMOVE,OVERWRITE</str>
</lst>
rather then:
<lst name="params">
<str name="f.cat.mode">DISTINCT</str>
<str name="f.features.mode">APPEND</str>
<str name="f.price.mode">INCREMENT</str>
<str name="f.sku.mode">REMOVE</str>
<str name="modify.default.mode">OVERWRITE</str>
</lst>
>> sku=REMOVE is required because sku is a stored field that is written to with copyField.
> I'm not sure I quite grok what REMOVE means yet, and how it fixes the copyField problem.
>
I'm using 'REMOVE' to say "remove the previous value of this field before doing anything." Essentially, this makes sure you new document does not start with a value for 'sku'.
> Another way to work around copyField is to only collect stored fields that aren't copyField targets.
I just implemented this. It is the most normal case, so it should be the default. It can be overridden by setting the mode for a copyField explicitly.
Oh, so we obviously need some tests that modify the same document multiple times w/o a commit in between.
I browsed the code really quick, looking for the tricky part... It's here:
+ openSearcher();
+ Term t = new Term( uniqueKey.getName(), uniqueKey.getType().toInternal( id.toString() ) );
+ int docID = searcher.getFirstMatch( t );
When you overwrite a document, it is really just adds another instance... so the index contains multiple copies. When we "commit", deletes of the older versions are performed. So you really want the last doc matching a term, not the first.
Also, to need to make sure that the searcher you are using can actually "see" the last document (once a searcher is opened, it sees documents that were added since the last IndexWriter close().
So a quick fix would be to do a commit() first that would close the writer, and then delete any old copies of docments.
Opening and closing readers and writers is very expensive though.
You can get slightly more sophisticated by checking the pset (set of pending documents), and skip the commit() if the doc you are updating isn't in there (so you know an older searcher will still have the freshest doc for that id).
We might be able to get more efficient yet in the future by leveraging NewIndexModifier:
SOLR-124
Haven't had a chance to check out any code, but a few quick comments:
If the field modes were parameters, they could be reused for other update handlers like SQL or CSV
Perhaps something like:
/update/xml?modify=true&f.price.mode=increment,f.features.mode=append
> sku=REMOVE is required because sku is a stored field that is written to with copyField.
I'm not sure I quite grok what REMOVE means yet, and how it fixes the copyField problem.
Another way to work around copyField is to only collect stored fields that aren't copyField targets. Then you run the copyField logic while indexing the document again (as you should anyway).
I think I'll have a tagging usecase that requires removing a specific field value from a multivalued field. Removing based on a regex might be nice too. though.
f.tag.mode=remove
f.tag.mode=removeMatching or removeRegex
I just attached a modified XmlUpdateRequestHandler that uses the new IndexDocumentCommand . I left this out originally because I think the discussion around syntax and functionality should be separated... BUT without some example, it is tough to get a sense how this would work, so i added this example.
Check the new file:
monitor-modifier.xml
It starts with:
<add mode="cat=DISTINCT,features=APPEND,price=INCREMENT,sku=REMOVE,OVERWRITE">
<doc>
<field name="id">3007WFP</field>
...
If you run ./post.sh monitor-modifier.xml multiple times and check: you should notice
1) the price increments by 5 each time
2) there is an additional 'feature' line each time
3) the categories are distinct even if the input is not
sku=REMOVE is required because sku is a stored field that is written to with copyField.
Although I think this syntax is reasonable, this is just an example intended to spark discussion. Other things to consider:
- rather then 'field=mode,' we could do 'field:mode,' this may look less like HTTP request parameter syntax
- The update handler could skip any stored field that is the target of a 'copyField' automatically. This is the most normal case, so it may be the most reasonable thing to do.
SOLR-139-IndexDocumentCommand.patch adds a new command to UpdateHandler and deprecates 'AddUpdateCommand'
This patch is only concerned with adding updateability to the UpdateHandler, it does not deal with how request handlers specify what should happen with each field.
I added:
public class IndexDocumentCommand
{
public enum MODE
;
public boolean overwrite = true;
public SolrDocument doc;
public Map<SchemaField,MODE> mode; // What to do for each field. null is the default
public int commitMaxTime = -1; // make sure the document is commited within this much time
}
RequestHandlers will need to fill up the 'mode' map if they want to support updateability. Setting the mode.put( null, APPEND ) sets the default mode.
Closed after release. | https://issues.apache.org/jira/browse/SOLR-139?focusedCommentId=13021007&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel | CC-MAIN-2015-40 | refinedweb | 9,681 | 64 |
I know you haven't had time to properly test the due with the shield and thus make any changes to your library; so I was hoping to do it for you...
However I've hit a roadblock I can't figure out myself: when trying to upload a sketch that uses the AFMotor library, I get the error: "_B was not declared in this scope". It makes me think that the core Arduino functions are not being loaded properly and I'm guessing it has to do with the following lines of code found in AFMotor.cpp:
- Code: Select all
#if (ARDUINO >= 100)
#include "Arduino.h"
#else
#if defined(__AVR__)
#include <avr/io.h>
#endif
#include "WProgram.h"
#endif
Even if I get rid of the if statement
- Code: Select all
#include "Arduino.h"
I still get erros with _BV.
Any quick suggestions? | http://adafruit.com/forums/viewtopic.php?f=31&t=35544 | CC-MAIN-2013-20 | refinedweb | 144 | 73.37 |
Puppet, PowerShell and Facter
Puppet uses a tool called Facter to gather system information during a Puppet run.
Facter is Puppet’s cross-platform system profiling library. It discovers and reports per-node facts, which are available in your Puppet manifests as variables.
There are some core facts which are processed on all operating systems, but two additional types of facts can be used to extend facter; External Facts and Custom Facts.
External facts
External facts provide a way to use arbitrary executables or scripts to generate facts as basic key / value pairs. If you’ve ever wanted to write a custom fact in Perl, C or PowerShell, this is how. Alternatively, external facts may contain static structured data in a JSON or YAML file.
Custom Facts
Custom facts are written in Ruby and have more advanced features, allowing for programmatic confinement to specific environments.
Most people new to Facter will write PowerShell scripts as external facts. However, there is a downside. The execution time for PowerShell scripts can be a little slow as a result of the time required to start a new PowerShell process for each fact. Another downside is that Windows will use file extensions to determine if a fact may be executed, while Unix based operating systems will look for the executable bit - it can be easy to forget these rules, especially when building cross-platform modules. While there are some tricks to help stop this, it’s easy for Windows scripts to log warnings and errors making it harder to figure out when real issues occur.
But the apparent learning curve to writing Ruby looks steep; if all you want to do is read a registry key and output the result, why should a Windows administrator have to learn Ruby? Well, this blog post should help reduce the effort it takes to write custom facts. Also, if you squint, the Ruby language looks a lot like (and in some cases operates similarly to) PowerShell.
The source code for these examples is available on my blog github repo.
Writing a registry based custom fact
The external fact
For this example we’ll convert a batch file based external fact to a Ruby external fact. This fact reads the
EditionID of the operating system from the registry and then populates a fact called
Windows_Edition.
@ECHO OFF for /f "skip=1 tokens=3" %%k in ('reg query "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion" /v EditionID') do ( set Edition=%%k ) Echo Windows_Edition_external=%Edition%
For example on my Windows 10 laptop it outputs:
Windows_Edition_external=Professional
And from within Puppet:
> puppet facts ... "kernel": "windows", "windows_edition_external": "Professional", "domain": "internal.local", "virtual": "physical", ...
The custom fact
Firstly we need to create a boiler plate custom fact in the module by creating the following file
lib/facter/windowsedition.rb
Facter.add('windows_edition_custom') do confine :osfamily => :windows setcode do 'testvalue' end end
This creates a custom fact called
windows_edition_custom which has a value of
testvalue. Running facter on my laptop we see:
> puppet facts ... "windows_edition_custom": "testvalue", "clientcert": "glennsarti.internal.local", "clientversion": "5.0.0", ...
Breaking down the custom fact code
So lets break down the boiler plate code:
Facter.add('windows_edition_custom') do ... end
This instructs Facter to create a new fact called
windows_edition_custom
confine :osfamily => :windows
The
confine statement instructs Facter to only attempt resolution of this fact on Windows operating systems.
setcode do .. end
setcode instructs Facter to run the code block to resolve the fact’s value
'testvalue'
As this is just a demonstration, we are using a static string. This is the code we’ll subsequently change to output a real value.
Reading the registry in Puppet and Ruby
You can access registry functions using the
Win32::Registry namespace. This is our new custom fact:
Facter.add('windows_edition_custom') do confine :osfamily => :windows setcode do value = nil Win32::Registry::HKEY_LOCAL_MACHINE.open('SOFTWARE\Microsoft\Windows NT\CurrentVersion') do |regkey| value = regkey['EditionID'] end value end end
So we’ve added five lines of code to read the registry. Let’s break these down too:
value = nil
First we set value of the fact to nil. We need to initialise the variable here, otherwise when its value is later set inside the codeblock, its value will be lost due to variable scoping
Next we open the registry key
SOFTWARE\Microsoft\Windows NT\CurrentVersion. Note that unlike the batch file, it doesn’t have the
HKLM at the beginning. This is because we’re using the
HKEY_LOCAL_MACHINE class, so adding that to the name is redundant. By default the registry key is opened as Read Only and for 64-bit access.
Next, once we have an open registry key, we get the registry value as a key in the
regkey object, thus
regkey['EditionID'].
Lastly, we output the value for Facter. Ruby uses the output from the last line so we don’t need an explicit
return statement like you would in langauges like C#.
When we run the updated fact we get:
> puppet facts ... "windows_edition_custom": "Professional", "clientcert": "glennsarti.internal.local", "clientversion": "5.0.0", ...
Tada!, we’ve now converted a batch file based external registry fact, to a custom Ruby fact in 10 lines. But there’s still a bit of cleaning up to do.
Final touches
If the registy key or value does not exist, Facter raises a warning. For example, if I change
value = regkey['EditionID'] to
value = regkey['EditionID_doesnotexist'] I see these errors output:
> puppet facts ... Warning: Facter: Could not retrieve fact='windows_edition_custom', resolution='<anonymous>': The system cannot find the file specified. { ...
We could write some code to test for existence of registry keys, but as this is just a fact we can simply swallow any errors and not output the fact. We can do this with a
begin /
rescue block.
Facter.add('windows_edition_custom') do confine :osfamily => :windows setcode do begin value = nil Win32::Registry::HKEY_LOCAL_MACHINE.open('SOFTWARE\Microsoft\Windows NT\CurrentVersion') do |regkey| value = regkey['EditionID'] end value rescue nil end end end
Much like the
try /
catch in PowerShell or C#,
begin /
rescue will catch the error and just output
nil for the fact value if an error occurs.
Writing a WMI based custom fact
The external fact
For this example we’ll convert a PowerShell file based external fact, to a Ruby external fact. This fact reads the
ChassisTypes property of the Win32_SystemEnclosure WMI Class. This describes the type of physical enclosure for the computer, for example a Mini Tower, or in my case, a Portable device.
$enclosure = Get-WMIObject -Class Win32_SystemEnclosure | Select-Object -First 1 Write-Output "chassis_type_external=$($enclosure.ChassisTypes)"
For example on my Windows 10 laptop it outputs:
chassis_type_external=8
And from within Puppet:
> puppet facts ... "kernel": "windows", "chassis_type_external": "8", "domain": "internal.local", "virtual": "physical", ...
The custom fact
Just like the last example, we start with a boilerplate custom fact in the module by creating the following file
lib/facter/chassistype.rb
Facter.add('chassis_type_custom') do confine :osfamily => :windows setcode do 'testvalue' end end
Accessing WMI in Puppet and ruby
We can access WMI using the
WIN32OLE Ruby class and
winmgmts:// WMI namespace. If you ever used WMI in VBScript (yes I’m that old!) this may look familiar.
Note - I’ve already added the
begin /
rescue block:
Facter.add('chassis_type_custom') do confine :osfamily => :windows setcode do begin require 'win32ole' wmi = WIN32OLE.connect("winmgmts:\\\\.\\root\\cimv2") enclosure = wmi.ExecQuery("SELECT * FROM Win32_SystemEnclosure").each.first enclosure.ChassisTypes rescue end end end
So again, let’s break this down:
require 'win32ole'
Much like in PowerShell or C#, we need to import modules (or gems for Ruby) into our code. We do this with the
require statement. This enables us to use the
WIN32OLE object on later lines.
wmi = WIN32OLE.connect("winmgmts:\\\\.\\root\\cimv2")
We then connect to the local computer (local computer is denoted by the period) WMI, inside the
root\cimv2 scope. Note that in Ruby the backslash is an escape character so each backslash must be escaped as a double backslash. Although WMI can understand using forward slashes I had some Ruby crashes in Ruby 2.3 using forward slashes.
enclosure = wmi.ExecQuery("SELECT * FROM Win32_SystemEnclosure").each.first
Now that we have a WMI connection we can send it a standard WQL query for all Win32_SystemEnclosure objects. As this returns an array, and there is only a single enclosure, we get the first element (
.each.first) and discard anything else
enclosure.ChassisTypes
And now we simply output the
ChassisTypes parameter as the fact value.
This gives the following output:
> puppet facts ... "chassis_type_custom": [ 8 ], "clientcert": "glennsarti.internal.local", "clientversion": "5.0.0", ...
Huh. So the output is slightly different. In external executable facts all output is considered a string. However as we are now using WMI and custom ruby facts, we can properly understand data types. Looking at the MSDN documentation
ChassisTypes is indeed an array type.
If this was ok for any dependent Puppet code, we could leave the code as is.
However if you wanted just the first element we could use:
enclosure.ChassisTypes.first
and this would output a single number, instead of a string:
> puppet facts ... "chassis_type_custom": 8, "clientcert": "glennsarti.internal.local", "clientversion": "5.0.0", ...
If you wanted it to be exactly like the external fact, we could then convert the integer into a string using
to_s
enclosure.ChassisTypes.first.to_s
and this would output a single string , instead of a number:
> puppet facts ... "chassis_type_custom": "8", "clientcert": "glennsarti.internal.local", "clientversion": "5.0.0", ...
Final notes
Structured Facts
Structured facts allow people to send more data than just a simple text string. This is usually as encoded JSON or YAML data. External facts have been able to provide structured facts, for instance using a batch file to output pre-formatted JSON text, but this is not yet enabled for PowerShell ().
puppet facts vs
facter
In my examples above I was using the command
puppet facts whereas most people would probably use
facter. This is mostly because I’m lazy. By default just running Facter (
facter) won’t evaluate custom facts in modules. External facts are fine due to pluginsync. By running
puppet facts, Puppet automatically runs Facter with all of the custom facts paths loaded. Note,
facter -p works but is deprecated in favour of
puppet facts
One other reason is debugging. In most modern Puppet installations Facter is running as native Facter which can make debugging native Ruby code trickier (though not impossible). However, when using Puppet as gem instead of installing the puppet-agent package, common during module development, it uses the Facter gem. The Facter gem allows for using standard Ruby debugging tools to help me out.
Conclusion
I hope this blog post helps you see that writing simple custom facts isn’t too daunting. In fact, the hardest part is setting up a Ruby development environment. I came across a blog post which explains setting up Ruby, very similar to my environment. Even though it’s for Chef, it still works for Puppet too.
The source code for these examples is available on my blog github repo.
Thanks to Ethan Brown (@ethanjbrown) for editing. | https://glennsarti.github.io/blog/puppet-ruby-facts/ | CC-MAIN-2018-17 | refinedweb | 1,845 | 56.45 |
To answer that question, I collected the world record progression for running events of various distances and plotted grows exponentially.
Let's look at each of those pieces in detail:
Physiological factors that determine running potential include VO2 max, anaerobic capacity, height, body type, muscle mass and composition (fast and slow twitch), conformation, bone strength, tendon elasticity, healing rate and probably more. Psychological factors include competitiveness, persistence, tolerance of pain and boredom, and focus.
Most of these factors have a large component that is inherent, they are mostly independent of each other, and any one of them can be a limiting factor. That is, if you are good at all of them, and bad at one, you will not be a world-class runner. There is only one way to be fast, but there are a lot of ways to be slow.
As a simple model of these factors, we can generate a random person by picking N random numbers, where each number is normally-distributed under a logistic transform. This yields a bell-shaped distribution bounded between 0 and 1, where 0 represents the worst possible value (at least for purposes of running speed) and 1 represents the best possible value.
Then to represent the running potential of each person, we compute the minimum of these factors. Here's what the code looks like:
def GeneratePerson(n=10):
factors = [random.normalvariate(0.0, 1.0) for i in range(n)]
logs = [Logistic(x) for x in factors]
return min(logs)
Yes, that's right, I just reduced a person to a single number. Bring in the humanities majors lamenting the blindness and arrogance of scientists. Then explain to them that this is supposed to be an explanatory model, so simplicity is a virtue. A model that is as rich and complex as the world is not a model.
Here's what the distribution of potential looks like for different values of N:
When N=1, there are many people near the maximum value. If we choose 100,000 people at random, we are likely to see someone near 98% of the limit. But as N increases, the probability of large values drops fast. For N=5, the fastest runner out of 100,000 is near 85%. For N=10, he is at 65%, and for N=50 only 33%.
In this kind of genetic lottery, it takes a long time to hit the jackpot. And that's important, because it suggests that even after 6 billion people, we might not have seen anyone near the theoretical limit.
Let's see what effect this model has on the progression of world records. Imagine that we choose a million people and test them one at a time for running potential (and suppose that we have a perfect test). As we perform tests, we keep track of the fastest runner we have seen, and plot the "world record" as a function of the number of tests.
Here's the code:
def WorldRecord(m=100000, n=10):
data = []
best = 0.0
for i in xrange(m):
person = GeneratePerson(n)
if person > best:
best = person
data.append(i/m, best))
return data
And here are the results with M=100,000 people and the number of factors N=10:
The x-axis is the fraction of people we have tested. The y-axis is the potential of the best person we have seen so far. As expected, the world record increases quickly at first and then slows down.
In fact, the time between new records grows geometrically. To see why, consider this: if it took 100 people to set the current record, we expect it to take 100 more to exceed the record. Then after 200 people, it should take 200 more. After 400 people, 400 more, and so on. Since the time between new records grows geometrically, this curve is logarithmic.
So if we test the same number of people each year, the progression of world records is logarithmic, not linear. But if the number of tests per year grows exponentially, that's the same as plotting the previous results on a log scale. Here's what you get:
That's right: a log curve on a log scale is a straight line. And I believe that that's why world records behave the way they do.
This model is unrealistic in some obvious ways. We don't have a perfect test for running potential and we don't apply it to everyone. Not everyone with the potential to be a runner has the opportunity to develop that potential, and some with the opportunity choose not to..
So let's get back to the original question: when will a marathoner break the 2-hour barrier? Before 1970, marathon times improved quickly; since then the rate has slowed. But progress has been consistent for the last 40 years, with no sign of an impending asymptote. So I think it is reasonable to fit a line to recent data and extrapolate. Here are the results:
The red line is the target: 13.1 mph. The blue line is a least squares fit to the data. So here's my prediction: there will be a 2-hour marathon in 2045. I'll be 78, so my chances of seeing it are pretty good. But maybe that's a topic for another article.
Related posts: | https://allendowney.blogspot.com/2011/04/ | CC-MAIN-2019-43 | refinedweb | 898 | 71.24 |
Crash - Testing Model/View Tutorial
Hello,
I am very new to Qt.
I wanted to create a simple Model/View functionality as in "": .
However, when I want to add the QTimer as in chapter 2.3, it crashes.
If I uncomment the QTimer and instead define an int x = 0 in the constructor of MyModel, it also crashes.
That's weird.
Any ideas what could be the problem?
Thanks in advance!
like that it is difficult to said (for me). Could you paste your code (use gist github maybe for paste it) ? the .cpp file of the model and his .h file, and this one with the view.
@
#ifndef GAMEMODEL_H
#define GAMEMODEL_H
#include <QAbstractTableModel>
#include <QTimer>
class GameModel : public QAbstractTableModel
{
Q_OBJECT
int x;
public:
GameModel(QObject *parent);
int rowCount(const QModelIndex &parent = QModelIndex()) const ;
int columnCount(const QModelIndex &parent = QModelIndex()) const;
QVariant data(const QModelIndex &index, int role = Qt::DisplayRole) const;
//QTimer *timer;
private slots:
//void timerHit();
};
#endif // GAMEMODEL_H
@
why define "int x" here ?
where is your class destructor ?
do you run a QMake after put the Q_OBJECT macro ? (because if you want to use slots with Q_OBJECT, you also need to call a qmake after write the Q_OBJECT in your h file)
could you show also the cpp file (you not need to write it here, because the place for write is stretch (except if it is really short code), you can use "gist github" for show the code and paste a link adress to this code.
Then the source:
@
GameModel::GameModel(QObject *parent)
:QAbstractTableModel(parent)
{
x = 0;
//selectedCell = 0;
//timer = new QTimer(this);
// timer->setInterval(1000);
// connect(timer, SIGNAL(timeout()) , this, SLOT(timerHit()));
// timer->start();
}
@
It doesn't crash without the x = 0.
- dheerendra
Hope you are allocating the QTimer object. Also try to re-arrange QTImer and Int vars in Private section separately. Clean, Run QMAKe and build it. It should work. author="Dheerendra" date="1420685575"]Hope you are allocating the QTimer object. Also try to re-arrange QTImer and Int vars in Private section separately. Clean, Run QMAKe and build it. It should work.[/quote]
Yeah, I cleaned it and 're-qmade' it. This worked. Thanks!
All the other stuff you guys adviced,... I dunno. int x is per se private since this is a class and not a struct, but it shouldn't alter the program's runtime behavior either way, etc...
for a struct (when you will need it), you have to define it first out of the class definition, and then define a link inside the class at public or private.
like that: // this is a classeur.h file
@
struct classeur_infos {
int id;
int id_parent;
QString name;
QString comment;
};
class Classeur : public QObject
{
Q_OBJECT
public:
explicit Classeur(QObject *parent = 0, int id_parent = NULL);
~Classeur(); // this is your destructor part
static QMap<int, int> Classeur_order; // this is a static QMap
classeur_infos Classeur_Info; // this is the public struct
.
.
.
@
hope this would help for clean code in C++
[quote author="jerome_isAviable?" date="1420685945"]
Sorry, but no.
The declaration of int x is fine where it is. The default for class is private, so anything above the public: line is private. I am used to the style to declare private variables as the very last section in the declaration under an explicit private: keyword, but the code as is is fine.
Creating an explicit destructor also is not required. Your compiler will create one if needed. There are few cases where you need to make one explicitly, even if it is empty, but not here.
Last: even that struct of your last post can be declared inside the scope of your class, if you so wish.
thanks Andre, i didn't know this. (you not have to be sorry also)
i learned to do like that, but i'm also happy to learn more how this could be done to. | https://forum.qt.io/topic/49893/crash-testing-model-view-tutorial | CC-MAIN-2018-09 | refinedweb | 643 | 73.37 |
Simple machine learning question. Probably numerous ways to solve this:
There is an infinite stream of 4 possible events:
'event_1', 'event_2', 'event_4', 'event_4'
The events do not come in in completely random order. We will assume that there are some complex patterns to the order that most events come in, and the rest of the events are just random. We do not know the patterns ahead of time though.
After each event is received, I want to predict what the next event will be based on the order that events have come in in the past. So my question is: What machine learning algorithm should I use for this predictor?
The predictor will then be told what the next event actually was:
Predictor=new_predictor()prev_event=Falsewhile True: event=get_event() if prev_event is not False: Predictor.last_event_was(prev_event) predicted_event=Predictor.predict_next_event(event)
Predictor=new_predictor()
prev_event=False
while True:
event=get_event()
if prev_event is not False:
Predictor.last_event_was(prev_event)
predicted_event=Predictor.predict_next_event(event)
The question arises of how long of a history that the predictor should maintain, since maintaining infinite history will not be possible. I'll leave this up to you to answer. The answer can't be infinte though for practicality.
So I believe that the predictions will have to be done with some kind of rolling history. Adding a new event and expiring an old event should therefore be rather efficient, and not require rebuilding the entire predictor model, for example.
Specific code, instead of research papers, would add for me immense value to your responses. Python or C libraries are nice, but anything will do.
It seems like a sequence prediction problem, that needs the Recurrent neural networks or hidden Markov models.
Your predictor can’t look back in time longer than the size of your window. RNNs and HMMs can do that after proper implementation of the model.
Here is a pybrain code for your problem:
from pybrain.datasets import SequentialDataSetfrom pybrain.supervised.trainers import BackpropTrainerfrom pybrain.tools.shortcuts import buildNetworkfrom pybrain.structure import SigmoidLayerINPUTS = 4HIDDEN = 10OUTPUTS = 4net =()
from pybrain.datasets import SequentialDataSet
from pybrain.supervised.trainers import BackpropTrainer
from pybrain.tools.shortcuts import buildNetwork
from pybrain.structure import SigmoidLayer
INPUTS = 4
HIDDEN = 10
OUTPUTS = 4
net =()
The code will train the recurrent network for 1000 epochs and print out the error after every epoch. You can check for correct predictions like this:
net.reset()for i in sequence: next_item = net.activate(i) > 0.5 print next_item
net.reset()
for i in sequence:
next_item = net.activate(i) > 0.5
print next_item
This will print an array of booleans for every event.
The above code is not tested on the current version, there might be some issues with the updated version.
Hope this answer helps. | https://intellipaat.com/community/7369/machine-learning-algorithm-for-predicting-order-of-events | CC-MAIN-2021-04 | refinedweb | 457 | 59.09 |
slack-webhook is a python client library for slack api Incoming Webhooks on Python 3.6 and above.
Project description
slack-webhook
slack-webhook is a python client library for slack api Incoming Webhooks on Python 3.6 and above.
Installation
$ pip install slack-webhook
Usage
basic
from slack_webhook import Slack slack = Slack(url='') slack.post(text="Hello, world.")
advanced
from slack_webhook import Slack slack = Slack(url='') slack.post(text=" } ] } ] }] )
Getting started
For help getting started with Incoming Webhooks, view our online documentation.
Contributing
- Fork it
- Create your feature branch (
git checkout -b my-new-feature)
- Commit your changes (
git commit -am 'Add some feature')
- Push to the branch (
git push origin my-new-feature)
- Create new Pull Request
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
slack-webhook-1.0.6.tar.gz (2.8 kB view hashes) | https://pypi.org/project/slack-webhook/1.0.6/ | CC-MAIN-2022-40 | refinedweb | 156 | 59.4 |
Semantics of @Name, @DataModelSelectiondjfjboss Oct 1, 2007 12:37 PM
In an attempt to learn Seam I am experimenting with a simple CRUD example, partly based on the Yuan/Heute book.
The integration example shows an edit link as:
<a href="person.seam?pid=#{fan.id}">Edit</a>
With the Person class having:
@Entity @Name("person") public class Person { private long id; @Id @GeneratedValue public long getId() { return id;} public void setId(long id) { this.id = id; }
and person.xhtml having:
<h2>Edit #{person.name}</h2> <h:form> <input type="hidden" name="pid" value="#{person.id}"/>
and the ManagerAction class having:
@Stateful @Name("manager") public class ManagerAction implements Manager { @In (required=false) @Out (required=false) private Person person; // @RequestParameter Long pid; @DataModel private List <Person> fans; @DataModelSelection private Person selectedFan; public void setPid (Long pid) { this.pid = pid; if (pid != null) { person = (Person) em.find(Person.class, pid); } else { person = new Person (); } } public Long getPid () { return pid; }
Apologies for taking so long to get to the point but the issue I have is that I keep getting property X not found on type Y. I had assumed that a Person instance would be maintained in the session state and injected/outjected as appropriate..
To make this work, up to a point, I have had to add @Scope(SESSION) to the entity and session beans but I still find that I am faced with a newly constructed Person instance rather than the one I thought I was operating on and that had been outjected into the shared context.
In desperation, I've also tried using the @DataModelSelection but without success - property not found again.
I suspect I'm missing something fundamental and have read about bijection and studied the example which look straightforward enough but something is missing and I can't figure out what it is!
1. Re: Semantics of @Name, @DataModelSelectionJacob Orshalick Oct 1, 2007 2:36 PM (in response to djfjboss)
"djfjboss" wrote:.
The pid attribute will not be set unless you uncomment the @RequestParameter annotation (which will directly inject the value) or add an entry in your pages.xml which will call your set method such as,
<pages> <page view- <param name="pid" value="#{manager.id}"/> </page> ... </pages>
I also do not see where fan is being outjected to the page so I don't know where your id would be coming from. Are you providing a list of persons and then the user can edit a person from the list? If you are looking to achieve a RESTful URL here (which is the only reason I could think of for doing it this way), an easier way to accomplish this would be...
@Stateful @Name("manager") public class ManagerAction implements Manager { @RequestParameter Long pid; @In (required=false) @Out (required=false) private Person person; @DataModel private List <Person> fans; @Factory(value="person") public void loadPerson(Long pid) { if (pid != null) { person = (Person) em.find(Person.class, pid); } else { person = new Person (); } } ...
Also note that you will not be in the same conversation when the Person is loaded because you are using a general link <a href ...>. Any conversation state will not be propagated. Hope that helps.
2. Re: Semantics of @Name, @DataModelSelectionJacob Orshalick Oct 1, 2007 2:38 PM (in response to djfjboss)
Sorry, the factory method in the above snippet should read,
... @Factory(value="person") public void loadPerson() { if (pid != null) { person = (Person) em.find(Person.class, pid); } else { person = new Person (); } } ...
No value should be passed into the loadPerson method.
3. Re: Semantics of @Name, @DataModelSelectiondjfjboss Oct 2, 2007 10:52 AM (in response to djfjboss)
Many thanks once again for your help.
I'm finding progress with Seam frustratingly slow. It sounds like a very powerful technology that should be a great productivity aid but, although the examples look straightforward, when I come to try them there are hidden subtleties and inconsistencies that continue to trip me up (perhaps they're not so straightforward!).
In addition there are oddities such as the commented out @RequestParameter annotation that's not even explained in the book. I'm reading Yuan/Heute and Nusairat in parallel, neither is the best text book I've ever encountered - the RoR books put them both to shame.
Do you have any suggestions for better sources of understanding? A clear explanation of the underlying concepts and the semantics of the annotations would hopefully go a long way to dispelling some of my confusion and frustration.
4. Re: Semantics of @Name, @DataModelSelectionJacob Orshalick Oct 2, 2007 11:50 AM (in response to djfjboss)
There is a new release of the Yuan/Heute book coming out (the examples in the current release are a little out of date for the latest release of Seam), but likely not until early next year some time. I will be posting an article soon that may help understanding of the semantics of conversations (it covers conversations and nested conversations) along with an example, but the best documentation out there that I've seen is the Seam reference manual.
I read the Nusairat book when I was getting started but past that I've kept the Seam source in my IDE at all times. I've heard that Java Persistence with Hibernate (Bauer, King) provides a good introduction to Seam, but I have not yet read this text. The forum here is great and I would recommend posting JIRA issues as you have difficulties with the documentation (the Seam team is very good about resolving these) as it will help everybody out. Sorry I can't be of more help.
5. Re: Semantics of @Name, @DataModelSelectionJacob Orshalick Oct 2, 2007 11:58 AM (in response to djfjboss)
By the way, to demonstrate responsiveness to documentation issues based on your conversationList difficulties the other day:
Resolved in a matter of days. Thanks Seam team!
6. Re: Semantics of @Name, @DataModelSelectiondjfjboss Oct 2, 2007 12:14 PM (in response to djfjboss)
I've read the Seam chapter in the 800 page Hibernate tome, and all the rest of it bar one chapter. The book is generally quite dense but is very readable in the most part. The chapter on Seam was what got me interested in Seam. Given his writing quality, a book on Seam by Gavin King would probably be very useful, but that's probably asking too much given all the other demands on his time!
Many thanks once again for your support and tolerance of my stumblings and suggestions for further reading. It certainly is encouraging to see that the JBoss team have already acted on your documentation observations. | https://developer.jboss.org/thread/138972 | CC-MAIN-2017-39 | refinedweb | 1,101 | 51.38 |
Hexbin Plots for Geospatial Data
tessellation, resolution and you:
This is a visualization I made back in May of 2013’s yearly-averaged sulfur dioxide readings. When you’re working with latitude & longitude or other coordinate systems, there’s no visualization more powerful than the hexgrid.
Today we’ll be looking further into two options that are somewhat easier to get into than ArcGIS: Matplotlib’s
hexbin() and
Basemap.
Hexplotlib
I always figured the ‘mat’ in matplotlib stands for ‘matrix’ following MATLAB’s naming convention. But do hexagons translate neatly into matrices?
That’s an interesting question that’ll surely have me writing another article.
For now, check out the source code for their hexplotting example:
Any continuous data can technically be hexgraphed, in the same way that any numeric dataset can fit into a simple point graph or kernel density estimate plot.
We start by creating some X and Y data centered around a standard distribution with
np.random.standard_normal(). Notice the use of max & min values to set a valid axis.
The standard matploglib
fig, axs = plt.subplots() is what creates the figure, but to populate it we call
ax.hexbin()which comes with a host of interesting parameters to explore.
Geographic Pandemic
NASA’s website is currently down (?) so rather than wait to re-download the GEODISC dataset, I browsed Kaggle for a more immediately relevant source of geospatial data. Today we’ll work with the John Hopkins COVID dataset that details the all of their data on the pandemic since 2019. It’s most recent update was October 3rd 2020, so the data should be quite interesting.
I loaded the 44MB CSV with pandas and drew a quick hexplot.
def draw_hexbin(df, grid=(20,20)):
x = df['Longitude']
y = df['Latitude']
C = df['Deaths']
plt.hexbin(x,y,C,gridsize=grid,)
plt.show()
plt.figure(figsize=(22,12))
draw_hexbin(df)
I laughed — not just exhaled air out of my nose, laughed — at the result of this first basic graph. A flattened beehive or some unfortunately asymmetric Rorschach, perhaps.
What we need to do is add nuance, namely by editing
gridsize=(more,hexagons) and refining the color map & bin distribution scale.
def draw_better_hexbin(df, grid=(120,120)):
x = df['Longitude']
y = df['Latitude']
C = df['Deaths']
hx = plt.hexbin(x,y,C,gridsize=grid, cmap='plasma', bins='log')
cb = plt.colorbar(hx)
plt.show()
plt.figure(figsize=(22,12))
draw_hexbin(df)
This is more tolerable. Specifying a logarithmic mapping scale with
bins='log' allows us to make better use of our ‘plasma’ colormap, which we detail by including
plt.colorbar(hexbin).
But it’s still missing something important — the world map we’re trying to measure. Data is useless without context.
Basemapping Hexes
Matplotlib’s Basemap library is designed around the needs of various earth scientists, and aids immensely in fields such as weather forecasting and oceanography.
We can use it to easily draw 2D maps of the earth over our coordinate data. It invokes PROJ & Matplotlib to handle a huge amount of figure fitting and drawing behind the scenes with just a few lines of code.
from mpl_toolkits.basemap import Basemap
x = df['Longitude']
y = df['Latitude']
C = df['Confirmed']fig = plt.figure(figsize=(22,12))
m = Basemap() # create Basemap object
m.drawcoastlines() # draw coastlines
m.drawcountries() # draw political boundaries# hexbin with prior parameters
m.hexbin(x,y,C=C, gridsize=(250), cmap='inferno', bins='log')
plt.show()
Borders & coastlines make this all much easier to read. We’ve gone from hex-points floating in space to an annotated map.
But of course, when it comes to aesthetics we can always do better. Let’s invoke
Basemap.shadedrelief():
fig = plt.figure(figsize=(22,12))
m = Basemap()
m.drawcoastlines()
m.drawcountries()m.shadedrelief() # graphic overlay
hx = m.hexbin(x,y,C=C, gridsize=(220), cmap='inferno', bins='log')
# the colorbar messed up the ratio scaling
plt.show()
Basemap draws us a beautifully blended terrain-shaded overlay to make our hexmap feel more like home.
My favorite so far is calling
Basemap.bluemarble():
This is quite close, I daresay, to being useful to look at. The hexagons are small enough to look like dots, but we can easily control that resolution at a very granular level with
gridsize=(x,y).
Every pandemic map begins at some point to resemble a population density graph. However, we can start to see some of the nuances of COVID-19’s overall impact throughout the world. As brighter colors indicate more cases, darker indicates fewer and none means no cases, we can get a clearer picture of how the world ended up over the past year.
Of course, this begets a host of other questions — are regions with no colored hexes free of infection? Only if they’re free of people, I’d reckon — more likely that many areas lack sufficient testing infrastructure. Hopkins can’t be reasonably expected to accurately aggregate the distributed (and often siloed) data of various countries.
Here in the US, we can see a visual representation of the spikes in coastal states such as California, Florida & NJ. A more detailed hexmap of the US may be of analytical use, considering the country’s troubled history with lockdown procedures.
In any case, there’s plenty of mapping resources to explore in the Basemap examples library, and you can start plotting your data with only the few lines printed above. I wish you all well in your own hexagonal journeys. | https://mark-s-cleverley.medium.com/hexbin-plots-for-geospatial-data-502db3c8348d?responsesOpen=true&source=user_profile---------8---------------------------- | CC-MAIN-2021-39 | refinedweb | 911 | 57.37 |
468B
Description
We have n (1 \leq n \leq 10^5) distinct integers q_1~q_n, a and b. There are two sets A and B. If x belongs to A, A must also contains a-x. It is the same with B and b. Output how qs can be divided into the two sets. Each q belongs and only belongs to one set.
Tutorial
If we have number x and a-x, they should be in the same set. If x belongs to A, it is obvious that a-x belongs to A. If x is not in A, then a-x cannot find its partner in A, so they it cannot be in A any more. Therefore, they can only all be in B. It is the same as the number x, b - x.
In additon, we should also know that if a-x does not exist, x can only belong to B. It is the same as A.
So we can use Disjoint Sets to solve this problem. Join the qs that must belongs to one set. Join those who must belong to A with a special node. Join those who must belong to B with another special node. Finally, if the two special nodes are in joined, there is no solution. Otherwise, solution exists.
Use STL map to get the positions of a-x and b-x.
Solution
#include <cstdio> #include <map> #include <algorithm> using namespace std; #define D(x) #define MAX_N 100005 int sum_a, sum_b; int f[MAX_N]; int n; struct Disjoint_sets { int father[MAX_N]; Disjoint_sets() {} Disjoint_sets(int n) { for (int i = 0; i < n; i++) { father[i] = i; } } int root(int a) { int ret = a; while (father[ret] != ret) ret = father[ret]; while (father[a] != a) { int b = a; a = father[a]; father[b] = ret; } return ret; } void join(int a, int b) { father[root(a)] = father[root(b)]; } }d_set; void input() { scanf("%d%d%d", &n, &sum_a, &sum_b); for (int i = 0; i < n; i++) { scanf("%d", &f[i]); } } bool work() { d_set = Disjoint_sets(n + 2); map<int, int> pos; for (int i = 0; i < n; i++) { pos[f[i]] = i; } for (int i = 0; i < n; i++) { if (pos.find(sum_a - f[i]) != pos.end()) { d_set.join(i, pos[sum_a - f[i]]); }else { d_set.join(i, n); } if (pos.find(sum_b - f[i]) != pos.end()) { d_set.join(i, pos[sum_b - f[i]]); }else { d_set.join(i, n + 1); } } return d_set.root(n) != d_set.root(n + 1); } void output() { puts("YES"); for (int i = 0; i < n; i++) { if (i != 0) { putchar(' '); } if (d_set.root(i) == d_set.root(n)) { putchar('1'); }else { putchar('0'); } } } int main() { input(); if (!work()) { puts("NO"); }else { output(); } return 0; } | https://notes.haifengjin.com/competitive_programming/codeforces/468B/ | CC-MAIN-2022-27 | refinedweb | 450 | 84.27 |
Vibrational modes of the H2O molecule¶.
- Read the script below and try to understand what it does.
from math import cos, sin, pi from ase import Atoms from ase.optimize import QuasiNewton from ase.vibrations import Vibrations from gpaw import GPAW # Water molecule: d = 0.9575 t = pi / 180 * 104.51 H2O = Atoms('H2O', positions=[(0, 0, 0), (d, 0, 0), (d * cos(t), d * sin(t), 0)]) H2O.center(vacuum=3.5) calc = GPAW(h=0.2, txt='h2o.txt', mode='lcao', basis='dzp') H2O.set_calculator(calc) QuasiNewton(H2O).run(fmax=0.05) """Calculate the vibrational modes of a H2O molecule.""" # Create vibration calculator vib = Vibrations(H2O) vib.run() vib.summary(method='frederiksen') # Make trajectory files to visualize normal modes: for mode in range(9): vib.write_mode(mode)
- Run the script and look at the output frequencies. Compare them to literature values, which are 1595cm-1 for the bending mode, 3657cm-1 for the symmetric stretching mode and 3756cm-1 for the anti-symmetric stretching mode. How good is the accuracy and what are possible error sources?
- Now we want to look at the modes to see how the atoms move. For this we use the files
vib.?.trajwhere
?is the number of the mode counted in the order they are printed out. You can look at these trajectories with the ase gui command - click Play to play the movie. Do they look like you expected and what would you have expected (you may have learned something about symmetry groups at one point)? Did you assign the modes correctly in the previous question? | https://wiki.fysik.dtu.dk/gpaw/exercises/vibrations/vibrations.html | CC-MAIN-2019-09 | refinedweb | 264 | 69.07 |
description of the many commands that may be placed
into a submit description file.
In addition, the index lists entries for each command under the
heading of Submit Commands.
Note that job ClassAd attributes can be set directly in a submit
file using the +<attribute> = <value> syntax (see
for details.)
In addition to the examples of submit description files given
here, there are more in the condor_submit manual page (see
). the file inputfile, as specified by the input command, and standard output for this job will go to the file outputfile, as specified by the output command. HTCondor expects to find inputfile in the current working directory when this job is submitted, and the system will take care of getting the input file to where it needs to be when the job is executed, as well as bringing back the output results (to the current working directory)iB of physical memory, and the rank command expresses a preference to run each instance of the program on machines with more than 64 MiB. It also advises HTCondor that this standard universe job will use up to 28000 KiB 4: Show off some fancy features including # the use of pre-defined macros. # #################### Executable = foo Universe = standard requirements = OpSys == "LINUX" && Arch =="INTEL" rank = Memory >= 64 image_size = 28000 request_memory = 32 error = err.$(Process) input = in.$(Process) output = out.$(Process) log = foo.log queue 150
A wide variety of job submissions can be specified with extra information to the queue submit command. This flexibility eliminates the need for a job wrapper or Perl script for many submissions.
The form of the queue command defines variables and expands values, identifying a set of jobs. Square brackets identify an optional item.
queue [<int expr>]
queue [<int expr>] [<varname>] in [slice] <list of items>
queue [<int expr>] [<varname>] matching [files | dirs] [slice] <list of items with file globbing>
queue [<int expr>] [<list of varnames>] from [slice] <file name> | <list of items>
All optional items have defaults:
The list of items uses syntax in one of two forms. One form is a comma and/or space separated list; the items are placed on the same line as the queue command. The second form separates items by placing each list item on its own line, and delimits the list with parentheses. The opening parenthesis goes on the same line as the queue command. The closing parenthesis goes on its own line. The queue command specified with the key word from will always use the second form of this syntax. Example 3 below uses this second form of syntax.
The optional slice specifies a subset of the list of items using the Python syntax for a slice. Negative step values are not permitted.
Here are a set of examples.
transfer_input_files = $(filename) arguments = -infile $(filename) queue filename matching files *.datThe use of file globbing expands the list of items to be all files in the current directory that end in .dat. Only files, and not directories are considered due to the specification of files. One job is queued for each file in the list of items. For this example, assume that the three files initial.dat, middle.dat, and ending.dat form the list of items after expansion; macro filename is assigned the value of one of these file names for each job queued. That macro value is then substituted into the arguments and transfer_input_files commands. The queue command expands to
transfer_input_files = initial.dat arguments = -infile initial.dat queue transfer_input_files = middle.dat arguments = -infile middle.dat queue transfer_input_files = ending.dat arguments = -infile ending.dat queue
queue 1 input in A, B, CVariable input is set to each of the 3 items in the list, and one job is queued for each. For this example the queue command expands to
input = A queue input = B queue input = C queue
queue input,arguments from ( file1, -a -b 26 file2, -c -d 92 )Using the from form of the options, each of the two variables specified is given a value from the list of items. For this example the queue command expands to
input = file1 arguments = -a -b 26 queue input = file2 arguments = -c -d 92 queue
Here is an example of a queue command for which the values of these automatic variables are identified.
queue 3 in (A, B)
Externally defined submit commands can be incorporated into the submit description file using the syntax
include : <what-to-include>
The
<what-to-include> specification may specify a single file,
where the contents of the file will be incorporated
into the submit description file at the point within the file
where the include is.
Or,
<what-to-include> may cause a program to be executed,
where the output of the program is incorporated
into the submit description file.
The specification of
<what-to-include> has the bar character
(|) following the name of the program to be executed.
The include key word is case insensitive. There are no requirements for white space characters surrounding the colon character.
Included submit commands may contain further nested include specifications, which are also parsed, evaluated, and incorporated. Levels of nesting on included files are limited, such that infinite nesting is discovered and thwarted, while still permitting nesting.
Consider the example
include : list-infiles.sh |In this example, the bar character at the end of the line causes the script list-infiles.sh to be invoked, and the output of the script is parsed and incorporated into the submit description file. If this bash script contains
echo "transfer_input_files = `ls -m infiles/*.dat`"then the output of this script has specified the set of input files to transfer to the execute host. For example, if directory infiles contains the three files A.dat, B.dat, and C.dat, then the submit command
transfer_input_files = infiles/A.dat, infiles/B.dat, infiles/C.datis incorporated into the submit description file.
Conditional
if/
else semantics
are available in a limited form.
The syntax:
if <simple condition> <statement> . . . <statement> else <statement> . . . <statement> endif
An
else key word and statements are not required,
such that simple
if semantics are implemented.
The
<simple condition> does not permit compound conditions.
It optionally contains the exclamation point character (
!)
to represent the not operation,
followed by
definedkeyword followed by the name of a variable. If the variable is defined, the statement(s) are incorporated into the expanded input. If the variable is not defined, the statement(s) are not incorporated into the expanded input. As an example,
if defined MY_UNDEFINED_VARIABLE X = 12 else X = -1 endifresults in X = -1, when MY_UNDEFINED_VARIABLE is not yet defined.
versionkeyword, representing the version number of of the daemon or tool currently reading this conditional. This keyword is followed by an HTCondor version number. That version number can be of the form
x.y.zor
x.y. The version of the daemon or tool is compared to the specified version number. The comparison operators are
==for equality. Current version 8.2.3 is equal to 8.2.
>=to see if the current version number is greater than or equal to. Current version 8.2.3 is greater than 8.2.2, and current version 8.2.3 is greater than or equal to 8.2.
<=to see if the current version number is less than or equal to. Current version 8.2.0 is less than 8.2.2, and current version 8.2.3 is less than or equal to 8.2.
if version >= 8.1.6 DO_X = True else DO_Y = True endifresults in defining DO_X as True if the current version of the daemon or tool reading this if statement is 8.1.6 or a more recent version.
Trueor
yesor the value 1. The statement(s) are incorporated.
Falseor
noor the value 0 The statement(s) are not incorporated.
$(<variable>)may be used where the immediately evaluated value is a simple boolean value. A value that evaluates to the empty string is considered
False, otherwise a value that does not evaluate to a simple boolean value is a syntax error.
The syntax
if <simple condition> <statement> . . . <statement> elif <simple condition> <statement> . . . <statement> endifis the same as syntax
if <simple condition> <statement> . . . <statement> else if <simple condition> <statement> . . . <statement> endif endif
Here is an example use of a conditional in the submit description file. A portion of the sample.sub submit description file uses the if/else syntax to define command line arguments in one of two ways:
if defined X arguments = -n $(X) else arguments = -n 1 -debug endif
Submit variable X is defined on the condor_submit command line with
condor_submit X=3 sample.subThis command line incorporates the submit command X = 3 into the submission before parsing the submit description file. For this submission, the command line arguments of the submitted job become
-n 3
If the job were instead submitted with the command line
condor_submit sample.subthen the command line arguments of the submitted job become
-n 1 -debug
A set of predefined functions increase flexibility. Both submit description files and configuration files are read using the same parser, so these functions may be used in both submit description files and configuration files.
Case is significant in the function's name, so use the same letter case as given in these definitions.
:default-valueis used; in which case it evaluates to default-value. For example,
A = $ENV(HOME)binds A to the value of the HOME environment variable.
"%d"is used as the format specifier.
$RANDOM_CHOICE(0,1,2,3,4,5,6,7,8)
minand
max, inclusive, is selected. The optional
stepparameter controls the stride within the range, and it defaults to the value 1. For example, to randomly chose an even integer in the range 0-8 (inclusive):
$RANDOM_INTEGER(0, 8, 2)
"%16G"is used as a format specifier.
nameand returns a substring of it. The first character of the string is at index 0. The first character of the substring is at index
start-index. If the optional
lengthis not specified, then the substring includes characters up to the end of the string. A negative value of
start-indexworks back from the end of the string. A negative value of
lengtheliminates use of characters from the end of the string. Here are some examples that all assume
Name = abcdef
Here are example uses of the function macros in a submit description file. Note that these are not complete submit description files, but only the portions that promote understanding of use cases of the function macros.
$(Process)are desired.
MyIndex = $(Process) + 1 initial_dir = run-$INT(MyIndex, %04d)Assuming that there are three jobs queued, such that
$(Process)becomes 0, 1, and 2, initial_dir will evaluate to the directories run-0001, run-0002, and run-0003.
Values = $(Process) * 10 Extension = $INT(Values, %03d) input = X.$(Extension)Assuming that there are four jobs queued, such that
$(Process)becomes 0, 1, 2, and 3, Extension will evaluate to 000, 010, 020, and 030, leading to files defined for input of X.000, X.010, X.020, and X.030.
arguments = $Fnx(FILE) transfer_input_files = $(FILE) queue FILE MATCHING ( samplerun/*.dat )Assume that two files that end in .dat, A.dat and B.dat, are within the directory samplerun. Macro FILE expands to samplerun/A.dat and samplerun/B.dat for the two jobs queued. The input files transferred are samplerun/A.dat and samplerun/B.dat on the submit host. The $Fnx() function macro expands to the complete file name with any leading directory specification stripped, such that the command line argument for one of the jobs will be A.dat and the command line argument for the other job will be B.dat.. When transfer_output_files is specified, its list governs which are transferred back at eviction time..
Setting transfer_output_files to the empty string (
"")
means no files are to be transferred..14 11 for more details about this command.
The submitter's entire environment can be copied into the job ClassAd for the job at job submission. The getenv command within the submit description file does this, as described at section 11..7.1Ver ==
request_GPUs = <n>where <n> is replaced by the integer quantity of GPUs required for the job. For example, a job that needs 1 GPU uses
request_GPUs = 1
Because there are different capabilities among GPUs, the job might need to further qualify which GPU of available ones is required. Do this by specifying or adding a clause to an existing Requirements submit command. As an example, assume that the job needs a speed and capacity of a CUDA GPU that meets or exceeds the value 1.2. In the submit description file, place
request_GPUs = 1 requirements = (CUDACapability >= 1.2) && $(requirements:True)
Access to GPU resources by an HTCondor job needs special configuration of the machines that offer GPUs. Details of how to set up the configuration are in section 3.7.1.. | http://research.cs.wisc.edu/htcondor/manual/current/2_5Submitting_Job.html | CC-MAIN-2018-34 | refinedweb | 2,141 | 57.57 |
Heads up! To view this whole video, sign in with your Courses account or enroll in your free 7-day trial. Sign In Enroll
Preview
Catching Errors with Error Boundaries5:12 with Guil Hernandez
With
componentDidCatch() comes a new concept of an error boundary. Error boundaries are wrapper components that use
componentDidCatch() to capture errors anywhere in their child component tree and display a fallback UI.
Resources
componentDidCatch()– React docs
- Better error handling – React docs
- Error Handling in React 16
- Introducing Error Boundaries
Note: Error boundaries only catch errors in the components below them in the tree. An error boundary can’t catch an error within itself.
With the componentDidCatch lifecycle method 0:00
comes a new concept of an error boundary. 0:03
Error boundaries are wrapper components that use componentDidCatch 0:06
to capture errors anywhere in their child component tree and 0:09
display a fallback UI in its place. 0:12
They provide an easier way to keep the error catching and conditional rendering 0:15
logic we wrote in the previous video reusable and maintainable in your app. 0:18
You create an error boundary as a class component. 0:23
So in the source components folder, I'll create a new file named ErrorBoundary.js. 0:26
Here, I'll import React, Component from react. 0:35
And to create the ErrorBoundary class, 0:44
I'll write export default class ErrorBoundary extends Component. 0:48
Now, I'll go ahead and move the state and 1:01
the componentDidCatch method over to the new ErrorBoundary component. 1:04
Then in the render function, 1:14
I'll write an if statement similar to the one we wrote earlier. 1:16
So if this component catches an error anywhere in its 1:21
child component tree, it's going to return an h2, 1:27
With the text, no, something went wrong. 1:35
Otherwise, it's going to return its children via this.props.children. 1:44
So now I have a reusable ErrorBoundary component I can wrap around my 1:53
entire app or specific components. 1:58
For example, in app.js, I'll import the ErrorBoundary component. 2:01
Then in the render method, I'll get rid of the if else statement. 2:14
And I'll wrap the ErrorBoundary around the StudentForm component only. 2:25
Back in the app, I'll click the Reset button to produce the error. 2:35
And we see the fallback content, but the rest of the app continues to render 2:42
because the component causing the error is inside the ErrorBoundary. 2:47
So as you can see, this is much better than having your entire app crash or 2:52
unmount any time an error occurs. 2:55
You can isolate errors, making them easier to understand and fix. 2:58
Any errors caught in the Component tree 3:02
get reported up to the ErrorBoundary's componentDidCatch method. 3:05
This provides a handy way to send error reports to an error tracking or 3:09
monitoring service. 3:13
For example, in my project, 3:14
I've already set up the config to an error reporting service I like to use. 3:15
So I'll import the config at the top of ErrorBoundry.js with 3:19
import sendError, from error-config. 3:24
The componenetDidCatch method takes two arguments to help you track and 3:34
investigate errors, error, the error instance itself, and info, 3:38
which contains the component stack trace or 3:42
the path in the component tree leading up to the offending component. 3:45
There are lots of JavaScript error tracking services out there. 3:50
And the tracking service you use is probably different than mine. 3:53
So in the componentDidCatch method, 3:57
I'll call my error reporting APIs captureException method. 3:59
And pass it the error instance. 4:09
The component stack trace contains helpful information for sorting and resolving 4:14
errors, so I'm also going to send it to my error tracking service as extra data. 4:20
So now, any time my ErrorBoundary catches an error, 4:32
information about the error gets sent over to my error tracking 4:36
services dashboard, where I can see what caused the error, 4:41
when it happened, the type of users it's affecting, and more. 4:45
So as you've learned, the componentDidCatch method and 4:53
ErrorBoundary work like try catch statements for your React components. 4:57
They provide a more consistent and dependable way to catch and 5:01
deal with errors. 5:05
You can learn more about error handling in React by reading the resources posted in 5:06
the teacher's notes. 5:10 | https://teamtreehouse.com/library/catching-errors-with-error-boundaries | CC-MAIN-2022-33 | refinedweb | 813 | 61.56 |
I'm using Archlinux with both Thrift 0.9.3 and Apache installed. In my Netbeans project, when I
import org.apache.thrift.*;
/lib/java
You need the
libthrift JAR file in order to use java code generated by the Thrift compiler.
If your project is set up to be able to use Maven repositories, you can add this artifact to your project:
<dependency> <groupId>org.apache.thrift</groupId> <artifactId>libthrift</artifactId> <version>0.9.3</version> </dependency>
Alternatively you could just download the JAR file from Maven central and add it to your project:
Also important to note is that the version of the JAR you use should match the version of the Thrift compiler that you use for code generation; so if you upgrade the Thrift compiler used for your project, you should upgrade the version of the JAR file as well. | https://codedump.io/share/5STEpKHO8Gz2/1/apache-and-thrift-installed-but-netbeans-can39t-see-import-orgapachethrift | CC-MAIN-2017-26 | refinedweb | 144 | 61.36 |
Question
Imagine you wish to estimate the betas for 2 investments, A and B. You have gathered the following return data for the market and for each of the investments over the past 10 years, 2005–2014.
b. Use the characteristic lines from part a to estimate the betas for investments A and B.
c. Use the betas found in part b to comment on the relative risks of investments A and B.
a. On a set of market return (x-axis)–investment return (y-axis) axes, use the data to draw the characteristic lines for investments A and B on the same graph.
.png)
.png)
b. Use the characteristic lines from part a to estimate the betas for investments A and B.
c. Use the betas found in part b to comment on the relative risks of investments A and B.
Answer to relevant QuestionsA security has a beta of 1.2. Is this security more or less risky than the market? Explain. Assess the impact on the required return of this security in each of the following cases. a. The market return increases by 15%. b. ...Use the capital asset pricing model to find the required return for each of the following securities in light of the data given. Jeanne Lewis is attempting to evaluate 2 possible portfolios consisting of the same 5 assets but held in different proportions. She is particularly interested in using beta to compare the risk of the portfolios and, in this ...Assume you wish to evaluate the risk and return behaviors associated with various combinations of assets V and W under 3 assumed degrees of correlation: perfect positive, uncorrelated, and perfect negative. The following ...Define and briefly discuss the investment merits of each of the following. a. Blue chips b. Income stocks c. Mid-cap stocks d. American depositary receipts e. IPOs f. Tech stocks
Post your question | http://www.solutioninn.com/imagine-you-wish-to-estimate-the-betas-for-2-investments | CC-MAIN-2016-50 | refinedweb | 315 | 56.55 |
-------------------------------------------------------------------------------- Fedora Update Notification FEDORA-2008-10794 None -------------------------------------------------------------------------------- Name : perl-namespace-clean Product : Fedora 9 Version : 0.09 Release : 1.fc9 URL : Summary : Keep your namespace tidy Description :. -------------------------------------------------------------------------------- ChangeLog: * Tue Dec 2 2008 Chris Weyl <cweyl alumni drew edu> 0.09-1 - update to 0.09 - note BR change from Scope::Guard to B::Hooks::EndOfScope -------------------------------------------------------------------------------- This update can be installed with the "yum" update program. Use su -c 'yum update perl-namespace-clean' at the command line. For more information, refer to "Managing Software with yum", available at. All packages are signed with the Fedora Project GPG key. More details on the GPG keys used by the Fedora Project can be found at -------------------------------------------------------------------------------- | https://www.redhat.com/archives/fedora-package-announce/2008-December/msg00829.html | CC-MAIN-2015-22 | refinedweb | 113 | 69.89 |
Interfacing Servo Motor with Arduino Uno
Contents
In this tutorial we will learn how to interface servo motor with Arduino Uno. Servo Motor is an electrical linear or rotary actuator which enables precise control of linear or angular position, acceleration or velocity. Usually servo motor is a simple motor controlled by a servo mechanism, which consists of a positional sensor and a control circuit. Servo motor is commonly used in applications like robotics, testing automation, manufacturing automation, CNC machine etc. The main characteristics of servo motor are low speed, medium torque, and accurate position.
Components Required
Hobby Servo Motor
Specifications
- Servo Motor consists of three pins: VCC, GROUND, and PWM signal pin.
- The VCC and GND is for providing power supply, the PWM input pin is used for controlling the position of the shaft by varying pulse width.
- Minimum Rotational Angle : 0 degrees
- Maximum Rotational Angle : 180 degrees
- Operating Voltage : +5V
- Torque : 2.5 kg/cm
- Operating Speed : 0.1 s/60 degree
- DC Supply Voltage : 4.8V to 6V
Working
The hobby servo motor which we use here consists of 4 different parts as below.
- Simple DC Motor
- Potentiometer
- Control Circuit Board
- Gear Assembly
Potentiometer is connected to output shaft of the servo motor helps to detect position. Based on the potentiometer value, the control circuit will rotate the motor to position the shaft to a certain angle. The position of the shaft can be controlled by varying the pulse width provided in the PWM input pin. Gear assembly is used to improve the torque of the motor by reducing speed.
Circuit Diagram
Description
- The VCC pin (Red Color) of the Servo Motor is connected to the 5V output of the Arduino Board.
- The GND pin (Brown Color) of the Servo Motor is connected to the GND pin of the Arduino Board.
- The PWM input pin (Yellow Color) of the Servo Motor is connected to the PWM output pin of the Arduino board.
Program
#include <Servo.h> Servo servo; int angle = 10; void setup() { servo.attach(3); servo.write(angle); } void loop() { // rotate from 0 to 180 degrees for(angle = 10; angle < 180; angle++) { servo.write(angle); delay(15); } // now rotate back from 180 to 0 degrees for(angle = 180; angle > 10; angle--) { servo.write(angle); delay(15); } }
Code Explanation
- The position of the shaft is kept at 10 degrees by default and the Servo PWM input is connected to the 3rd pin of the Arduino Uno.
- In the first for loop, motor shaft is rotated from 10 degrees to 180 degrees step by step with a time delay of 15 milliseconds.
- Once it reaches 180 degree, it is programmed to rotate back to 10 degree step by step with a delay of 15 milliseconds in the second for loop.
Functioning
Wire the circuit properly as per the circuit diagram and upload the program. You can see that the servo motor is rotating as per the program. Please see the video below for more details.
Output
Conclusion
Hope you understood about the functioning of the servo motor with the help of Arduino programming. You can modify this code for your robotic applications and have fun. Please feel free to comment below if you have any doubts. | https://electrosome.com/interfacing-servo-motor-arduino-uno/ | CC-MAIN-2021-04 | refinedweb | 537 | 54.52 |
Welcome to the Week of WPF. During the next 7 days, I’ll help show you how you can use WPF and PowerShell together.
PowerShell could always script almost everything in .NET, but, prior to the recent CTP2 you could not script Windows Presentation Foundation (WPF) in PowerShell.
Now you can script everything that WPF can do within PowerShell.
This means you can write pretty sleek looking UIs in nice little PowerShell scripts.
There are a lot of things you can do with this capability, but let’s start things off simple.
Windows Presentation Foundation (WPF) is a set of .NET libraries for making next-generation user interfaces. In order to script WPF classes, you have to start them in a Single Threaded Apartment (STA). Luckily, starting in CTP2, you can run code in an STA a few ways in PowerShell.
In order to script WPF, you have to do one of three things:
· Run the Script in Graphical PowerShell, where WPF scripts will run without any changes because Graphical PowerShell runspaces are STA.
· Run the script in PowerShell.exe and add the –STA script
· Create a background runspace that is STA, and run the script in the background runspace
In this post, I’ll show you a “Hello World” script, and how to run it in each of the three modes.
First, let’s write our “Hello World” script. Open up Graphical PowerShell (gPowerShell.exe) and create a new script file (CTRL + N).
Here’s the Hello World file:
$window = New-Object Windows.Window
$window.Title = $window.Content = “Hello World. Check out PowerShell and WPF Together.”
$window.SizeToContent = “WidthAndHeight”
$null = $window.ShowDialog()
Now in Graphical PowerShell, you should be able to just run it and be done.
You can run the current script in Graphical PowerShell with F5 or the Green Run button in the upper left of the editor.
Now let’s run the script in STA Mode PowerShell (Run PowerShell –sta). Save it as “HelloWorld.ps1”
Since PowerShell.exe –sta doesn’t load up WPF’s assemblies, lets run these three lines to add the references:
Add-Type –assemblyName PresentationFramework
Add-Type –assemblyName PresentationCore
Add-Type –assemblyName WindowsBase
.\HelloWorld.ps1
Now you have a Hello World in WPF from the console version of PowerShell.
Finally, you can embed scripts to generate WPF inside of an application or a background runspace. This has some advantages over the first two approaches… namely, without a background
The code below will create a background runspace and set the two properties required to make WPF work in a PowerShell runspace, ApartmentState and ThreadOptions. ApartmentState determines if the runspace PowerShell is single or multithreaded (WPF requires single threaded), and if the same thread is used every time a command is run, or if a new thread is created each time. For WPF scripting to work in a runspace, you have to set ApartmentState to “STA” and ThreadOptions to “ReuseThread”. In the code below, the first 3 lines set up the runspace, the next 10 lines ensure that runspace is able to load WPF classes, and the final 4 lines run our original HelloWorld.ps1
# Create a runspace to run Hello World
$rs = [Management.Automation.Runspaces.RunspaceFactory]::CreateRunspace()
$rs.ApartmentState, $rs.ThreadOptions = “STA”, “ReuseThread”
$rs.Open()
# Reference the WPF assemblies
$psCmd = {Add-Type}.GetPowerShell()
$psCmd.SetRunspace($rs)
$psCmd.AddParameter(“AssemblyName”, “PresentationCore”).Invoke()
$psCmd.Command.Clear()
$psCmd = $psCmd.AddCommand(“Add-Type”)
$psCmd.AddParameter(“AssemblyName”, “PresentationFramework”).Invoke()
$psCmd()
This is just a taste of some of the fun that PowerShell and WPF can have together, and how to get it working. You can now use everything in WPF in a PowerShell script.
During the Week of WPF, I will post one script each day that will take us a little further down the rabbit hole of what PowerShell and WPF can do together. After the week of WPF is done, keep your eyes on this space, because every Wednesday, I’ll try to post a “WPF Wednesday Widget” that uses showcase using PowerShell and WPF together.
Hope this Helps,
James Brundage [MSFT]
Join the conversationAdd Comment ‘obvious’ ‘miracle ‘ /.
Andrew,?
The questions is that I can’t execute a powershell script without the black console window. For example when I run a vbs script, we only need to run the command: wscript.exe example.vbs and this command do not open a cmd console
Another example. Look my example.ps1 script:
[void] [System.Reflection.Assembly]::LoadWithPartialName(“System.Drawing”)
[void] [System.Reflection.Assembly]::LoadWithPartialName(“System.Windows.Forms”)
$objForm = New-Object System.Windows.Forms.Form
$objForm.Text = “Example”
$objForm.Size = New-Object System.Drawing.Size(100,100)
$objForm.StartPosition = “CenterScreen”
$objForm.Add_Shown({$objForm.Activate()})
[void] $objForm.ShowDialog()
when I run this script: powershell.exe .example.ps1 the windows form appears correctly but I can´t hide the powershell console. How can I do?
Pablo,
Ah yes, the annoying black widow… 😉
I’ve tried posting URLs in the comments before and they seem to always be rejected. I’d go to the microsoft.public.windows.powershell newsgroup and ask…
I’ll include this in a later post in the series, but I’ll post it here for Pablo. You can Minimize /Maximize Powershell by calling Win32 Functions through C interop:
$showWindowAsync = Add-Type –memberDefintion @”
[DllImport(“user32.dll”)]
public static extern bool ShowWindowAsync(IntPtr hWnd, int nCmdShow);
“@ -name “Win32ShowWindowAsync” -namespace Win32Functions –passThru
# Minimize PowerShell
$showWindowAsync::ShowWindowAsync((Get-Process –id $pid).MainWindowHandle, 2)
sleep 2
# Then Restore it
$showWindowAsync::ShowWindowAsync((Get-Process –id $pid).MainWindowHandle, 4)
Hope this Helps,
James Brundage[MSFT]
Great! It is what I surely hoped.
But I hope -Minimize and -NoWindow start options for powershell.exe, too.
Are they shipping on next milestone? Or will ship?
@PowerShellTeam
Thanks. Interesting workaround
I’ll try it. It will be very interesting for IT Administrators that we can hide the console completely (if the black window is minimized the final user can close it). Thanks again for the help.
I`ve just tried your code on PS V2 CTP Graphical and I`ve received an exception:
Exception calling ".ctor" with "0" argument(s): "The calling thread must be STA, because many UI components require this"
So probably there is something to tune in runspace.
The only change I`ve made to standard PS Runspace is
Set-ExecutionPolicy RemoteSigned
Regards,
Gianluca
Gianluca,
You must be running the v2 CTP2 gPowerShell. A v2 CTP was release also, but it doesn’t support STA. I’m assuming the problem is that you’re running an old CTP version.
You are right!
I`ve missed the CTP2 release announce, sorry.
Just downloaded, thanks.
Lee and Suzanne here makes a good point. Shame on all of us for not at least acknowledging that this
Hi all,
I installed the latest PS-version for XP, SP3. When I run the example code, I also get a…
"
Exception calling ".ctor" with "0" argument(s): "The calling thread must be STA, because many UI components require this"
"
What’s wrong? -A version dump shows this:
> $PSVersionTable
Name Value
—- —–
CLRVersion 2.0.50727.3603
BuildVersion 6.0.6002.18111
PSVersion 2.0
WSManStackVersion 2.0
PSCompatibleVersions {1.0, 2.0}
SerializationVersion 1.1.0.1
PSRemotingProtocolVersion 2.1
TNX! | https://blogs.msdn.microsoft.com/powershell/2008/05/22/wpf-powershell-part-1-hello-world-welcome-to-the-week-of-wpf/ | CC-MAIN-2016-30 | refinedweb | 1,199 | 59.3 |
I am stuck at the end of 7. Plan your trip! Although my code works perfectly fine for a single result, I would like the console to ask the user for the destination, duration and planned spendings of the trip and return the exact cost of the trip.
Howerver, I keep getting this error message
"Traceback (most recent call last):
File "python", line 32, in module
File "python", line 26, in trip_cost
TypeError: unsupported operand type(s) for +: 'int' and 'unicode'" (or 'int' and 'str')
I have tried to change the type of my "spending_money" variable to number so that it can be normally added to the other variables but it fails every time. How could I correct it?
def hotel_cost(nights): return 140*nights def plane_ride_cost(city): if city=="charlotte": return 183 elif city=="tampa": return 220 elif city=="los angeles": return 475 elif city=="pittsburgh": return 222 def rental_car_cost(days): cost=40*days if days >=7: cost=cost-50 return cost elif days>=3 and days<7: cost=cost-20 return cost else: return cost def trip_cost(city, days, spending_money): return plane_ride_cost(city) + hotel_cost(days) + rental_car_cost(days) + spending_money city=raw_input('Where do you want to go?').lower() days=raw_input('How long do you want to go?') spending_money=int(raw_input('How much do you want to spend?')) print trip_cost(city, days, spending_money)
Thank you very much for your help and your time !
Greg | https://discuss.codecademy.com/t/creating-a-small-user-interface-to-ask-for-the-details-of-the-trip/46809 | CC-MAIN-2018-34 | refinedweb | 234 | 56.89 |
With the timer delay to achieve the mouse double click event events, click and double-click events independently of each other!
import java.awt.event.MouseAdapter; import java.awt.event.MouseEvent; import java.util.Date; import java.util.Timer; /** * Use the timer delay click event implements mouse double-click event , Click and double-click event from interfering with each other ! */ public class UserMouseAdapter extends MouseAdapter { private boolean flag = false;// Use to determine whether has been executed the double-click event private int clickNum = 0;// Use to determine whether the execution double-click event public void mouseClicked(MouseEvent e) { final MouseEvent me = e;// The event source this.flag = false;// Each click of the mouse initialization double-click event flag to false if (this.clickNum == 1) {// When clickNum ==1 When you perform a double-click event this.mouseDoubleClicked(me);// Perform a double-click event this.clickNum = 0;// Initializes a double-click event flag to 0 this.flag = true;// Double-click the event have already been performed , Event flag to true return; } // Define the timer Timer timer = new Timer(); // Timer start , Delay-0 .2 Seconds after the click event timer.schedule(new java.util.TimerTask() { private int n = 0;// Record timer executions public void run() { if (flag) {// If you double-click event has executed , Then click the perform directly cancel n = 0; clickNum = 0; this.cancel(); return; } if (n == 1) {// Timer wait 0 .2 Seconds , Double-click the event has not yet happened , Perform the click event mouseSingleClicked(me);// Perform the click event flag = true; clickNum = 0; n = 0; this.cancel(); return; } clickNum++; n++; } }, new Date(), 200); // To set the delay time } /** */ /** * The mouse click event * * @param e * The event source parameters */ public void mouseSingleClicked(MouseEvent e) { // System.out.println("Single Clicked!"); } /** */ /** * Mouse double-click event * * @param e * The event source parameters */ public void mouseDoubleClicked(MouseEvent e) { // System.out.println("Doublc Clicked!"); } }
Related Posts of With the timer delay to achieve the mouse double click event events, click and double-click events independently of each other! | http://www.codeweblog.com/with-the-timer-delay-to-achieve-the-mouse-double-click-event-events-click-and-double-click-events-independently-of-each-other/ | CC-MAIN-2014-41 | refinedweb | 333 | 63.8 |
hey guys~
I'm an ibus-table user, filing a bug the first time in my gentoo career.
After I updated gnome-light on my laptop from 3.4 to 3.6 today(12-29-2012),
i found that the latest app-i18n/ibus-table-1.3.9 was unable to run properly, which was built upon app-i18n/ibus-1.4.99 that is a part of dependencies of gnome-light-3.6.
When I was launching ibus-setup, the error showed like this:
~ $ ibus-setup
Traceback (most recent call last):
File "/usr/share/ibus-table/engine/main.py", line 25, in <module>
import ibus
ImportError: No module named ibus
The problem is reproducible. Probably there's version mismatch between ibus-1.4.99 and ibus-table-1.3.9, i think.
I've noticed that there is the latest version ibus-table-1.4.99, and it should help.
If someone kindly bump the latest ibus-table-1.4.99 , that would be nice!
Happy New Year to you all~
Hello all ~
I've worked it out!
After testing the packages time and time again, finally i noticed that the old app-i18n/ibus-table-1.3.9 only works properly when app-i18n/ibus-1.4.99 ' s "deprecated" USE flag enabled.
The "deprecated" USE flag triggers ibus-1.4.99's "--python-library" compiling option, which is OFF by default.
But surprisingly, the flag isn't in the dependency tree of ibus-table-1.3.9, probably because that it's newly added for the new ibus-1.4.99 ,while not existing in the older versions. However, the brand new ibus-1.4.99 is required by gnome 3.6.
So tired after two-days' test... Fortunately, it has been resolved..
That's all.
Please reopen, as the issue persists.
There is an ibus-table-1.4.99.20121112.tar.gz package available on the ibus page. Probably that version should be added to portage (in addition to fixing the useflags as wanglvyun described).
(In reply to comment #4)
>
Thanks for your reply, but a bit confused...
Why can't we enable both "deprecated" and "gtk3 introspection" of ibus? As ibus-table only work properly with ibus "deprecated" enabled.
What i've observed is that each of them, the three USE flags, doesn't conflict with one another.
My opinion, though maybe thoughtless, is:
submit a tiny modification of the ibus-table ebuild, in which ibus-table depends ibus[deprecated], thus the problem should be solved.
We already have ibus-1.5.1 and ibus-table-1.5.0 in the tree.
It seems this bug is no longer problem?
(In reply to comment #6)
> We already have ibus-1.5.1 and ibus-table-1.5.0 in the tree.
> It seems this bug is no longer problem?
yes, test completed and all have been proved to be correct | https://bugs.gentoo.org/show_bug.cgi?format=multiple&id=449166 | CC-MAIN-2020-29 | refinedweb | 485 | 80.07 |
22 March 2012 08:28 [Source: ICIS news]
SINGAPORE (ICIS)--?xml:namespace>
Market sources said there was a gas leakage at the cracker, but the company source declined comment on the matter.
“We do not know when the cracker will restart,” the company source said.
Showa Denko had shut one of two lines at the cracker on 4 March for furnace repairs and maintenance. The No 1 line was initially slated to resume operations on 31 March.
The No 2 line was also shut for the same reason on 13 March and was originally supposed to come back on line on 23 March.
Ethylene supply in Asia is expected to tighten if Showa Denko keeps | http://www.icis.com/Articles/2012/03/22/9543875/japans-showa-denko-likely-to-prolong-oita-cracker-shutdown.html | CC-MAIN-2014-41 | refinedweb | 115 | 80.62 |
How can we simple calculate the GPU memory model (nn.Module) use? Just a single GPU unit.
To calculate the memory requirement for all parameters and buffers, you could simply sum the number of these and multiply by the element size:
mem_params = sum([param.nelement()*param.element_size() for param in model.parameters()]) mem_bufs = sum([buf.nelement()*buf.element_size() for buf in model.buffers()]) mem = mem_params + mem_bufs # in bytes
However, this will not include the peak memory usage for the forward and backward pass (if that’s what you are looking for).
What’s the peak memory usage?
During training you are using intermediate tensors needed to backpropagate and calculate the gradients. These intermediate tensors will be freed once the gradients were calculated (and you haven’t used
retain_graph=True), so you’ll see more memory usage during training than the initial model parameters and buffers would use.
Is peak memory usage equivalent to forward/backward pass size here?
---------------------------------------------------------------- ----------------------------------------------------------------
It might be, but I’m not sure which utility you are using and how it estimates the memory usage.
@ptrblck, How can we measure the peak memory usage? This sounds like the most important question not to break into CUDA out of memory errors. Should I ask the separate question for this?
torch.Cuda.max_memory_allocated() should give you the max value. I’m not sure, if your currently used logging library gives a matching number, but it would be interesting to see.
Thanks, it was:
import torch torch.cuda.max_memory_allocated()
This can help me figure out the max batch size I can use on a model, hopefully. But I wonder if something similar is present in PyTorch already.
However, I am not sure if this thing will also count the memory in the garbage collector that can be free after
gc.collect().
Maybe this is called cache. | https://discuss.pytorch.org/t/gpu-memory-that-model-uses/56822 | CC-MAIN-2022-27 | refinedweb | 306 | 58.48 |
Dig into DNS: Part 3
dig-utility.jpg
In the first and second articles in this series, I introduced the powerful dig utility and its uses in performing DNS lookups along with a few time-saving examples of to put it into practice. In this part, I’ll look at my favorite dig utility feature -- which is definitely the “trace” option. For those unfamiliar with this process, let’s slowly run through it.
Without a Trace).
dns-fig1.png
Figure 1 shows the delegation from the root servers down into the key servers responsible for the delegation of the Top Level Domain (TLD) .org. Underneath those authoritative “afilias-nst” servers, we see (at the end we’ve been waiting for, and you might be surprised how many servers we had to ask in order to receive it.
Now that we've come to grips with DNS delegation and traversed the lengthy chain required for a ".org" namespace lookup, let's compare that output with a ".co.uk" DNS lookup. Figure 2 shouldn’t cause too many headaches to decipher.
dns-fig2.png
The 10 “.nic.uk” name servers (geographically disparate for resilience, which you could confirm categorically with a “WHOIS” lookup on each of their IP addresses) shows that the .UK namespace is well protected from issues. For reference, Network Information Centre (or NIC) is common parlance throughout the name server world, and if you’re ever unsure which authoritative body is responsible for a country’s domain name space, trying something like, in the case of the United Kingdom, might save the day.
Now, let’s compare that with a much more concise output using the +short option as follows. In Figure 3, you see that the NIC's inclusion is not displayed but instead just the other salient parts of the chain. The command that used was:
# dig +short +trace @8.8.4.4 sample.co.uk
dns-fig3.png pleased.
If you see very slow response times anywhere in the chain (you can see the lookup completion results at the foot of each section in Figures 1 and 2), my advice is to turn to one of the many online tools that of stems from the Internet of old -- that you need to figure out who to contact about a domain name issue or discover certain parameters that 4..
dns-fig4.png backward as seen in our example.
Apparently (but this may not be the case), that RFC 1035 holds fort for SOA recommendations, which can be found here if you would like some enlightening, bedtime reading.
Other name servers simply increment a long number such as 1000003 each time there’s a change. This process admittedly doesn’t help with when an update was made, but at least it lets you know which version of a zone file you are looking at. This is important because when sharing updates between many name servers, it’s imperative that they answer with the most up-to-date version of the DNS and do not respond with stale, potentially damaging information.
In case you’re wondering, we’re seeing the “root-servers.net” and “verisign-grs.com” domain names being mentioned because the domain name “chrisbinnie.tld” doesn’t exist (I’m just using it as an example). So Figure 4 that isn’t a root server, you would expect this to be 7200 seconds (two hours).
The “expire” field lets secondary name servers know when the answers it has been serving should be considered stale (or, in other words, how long the retrieved information remains valid).
Finally, the “minimum” field shows secondaries (also called “slaves”) how long the received information should be cached before checking again. This has been called the setting with the greatest importance thanks to the fact that, with frequent changes to your DNS, you need to keep secondaries frequently updated.
The tricky balance, however, is that if you don’t make frequent DNS updates, keeping this entry set 4 is that “a.root-servers.net” is the authoritative -- to avoid the use of the @ symbol I assume -- this field is actually translated as “nstld@verisign-grs.com” into a functioning email address.
Finally, the “1799” at the start of that line is the Time To Live -- which means how many seconds (30 minutes in this case, minus a second) connecting client software should hold onto the retrieved data before asking for it again, in the event that it’s changed in the meantime.
In the fourth and final part of this series, I’ll take a quick look look at some security options and wrap up with more examples that I have found very useful.
Chris Binnie is a Technical Consultant with 20 years of Linux experience and a writer for Linux Magazine and Admin Magazine. His new book Linux Server Security: Hack and Defend teaches you how to launch sophisticated attacks, make your servers invisible and crack complex passwords.
Learn more about network and system management in the Essentials of System Administration training course from The Linux Foundation.
-
- Log in or register to post comments
- Print This
- Like (1 like) | https://www.linux.com/news/dig-dns-part-3 | CC-MAIN-2016-40 | refinedweb | 857 | 60.04 |
liquid - Shopify: Order Print App is not picking up variables
I am trying to display a Final Sale message on order receipts for products that are 60% off. This bit of code does display the message on the individual product page, but when I insert in the template I'm using in the Order Printer app, the message does not seem to display.
I've contacted various Shopify support people, but they have not been able to identify the problem. Here's the code I'm inserting:
<!--if item is 60% off, it displays message: --> {% if product.compare_at_price %} {% assign sixtyPercentOff = product.compare_at_price | minus: product.price | times: 100.0 | divided_by: product.compare_at_price | round %} {% if sixtyPercentOff == 60 %} <p style="color: #B21F1F;"> This item is final sale -- no returns or exchanges are accepted. </p> {% endif %} {% endif %}
Is it because Order Printer does not recognize variables such as "compare_at_price"?
Answer
Solution:
product.compare_at_price doesn't exist.
For Do you mean to use something like: Or if you have variants you can use directly For first available variant of product:
product you have:
{% if product.compare_at_price_min > 0 %}
{% if product.variants[0].compare_at_price > 0 %}
Do you mean to use something like:
Or if you have variants you can use directly
For first available variant of. | https://e1commerce.com/items/shopify-order-print-app-is-not-picking-up-variables | CC-MAIN-2022-40 | refinedweb | 209 | 59.19 |
Scala supports the array data structure. An array is a fixed size data structure that stores elements of the same data type. The index of first element of an array is zero and the last element is the total number of elements minus one.
Table of Contents
Scala Array Declaration
The syntax for declaring an array variable is
Copyvar arrayname = new Array[datatype](size)
var indicates variable and arrayname is the name of the array, new is the keyword, datatype indicates the type of data such as integer, string and size is the number of elements in an array.
For example,
Copyvar student = new Array[String](5) Or var student:Array[String] = new Array[String](5)
Here student is a string array that holds five elements. Values can be assigned to an array as below.
Copyvar student = Array("John","Adam","Rob","Reena","Harry") or student(0) = "Martin" ; student(1) = "Jack" ; student(2) = "Jill" ; student(3) = "Paul" ; student(4) = "Kevin"
Scala Arrays Processing
While processing arrays we use for loop as the elements are of same data type.
Consider an example below.
Copyobject Student { def main(args: Array[String]) { var marks = Array(75,80,92,76,54) println("Array elements are : ") for ( m1 <- marks ) { println(m1 ) } var gtot = 0.0 for ( a <- 0 to (marks.length - 1)) { gtot += marks(a); } println("Grand Total : " + gtot); var average = 0.0 average = gtot/5; println("Average : " + average); } }
Here we are creating marks array of Integer data type. We are printing the elements of an array using for loop. We are calculating the total marks by adding all the elements and calculate average by dividing the total by the number of subjects.
Below image shows the execution of above program.
Scala Multi Dimensional Arrays
Multidimensional arrays can be defined as an Array whose elements are arrays.
Consider an example below.
Copyobject Multidim { def main(args:Array[String]) { val rows = 2 val cols = 3 val multidim = Array.ofDim[String](rows,cols) multidim(0)(0) = "Reena" multidim(0)(1) = "John" multidim(0)(2) = "Adam" multidim(1)(0) = "Michael" multidim(1)(1) = "Smith" multidim(1)(2) = "Steve" for { i <- 0 until rows j <- 0 until cols } println(s"($i)($j) = ${multidim(i)(j)}") } }
We are creating a two dimensional array with 2 rows and 3 columns using the
ofDim method which accepts rows and columns as arguments. Add the elements to the array as rows and columns accepting string arguments. We are using for loop retrieve the inserted elements.
Save the above code in
Multidim.scala and run as shown in below image.
Scala Concatenate Arrays
Two arrays can be concatenated using the
concat method. Array names can be passed as an argument to the concat method.
Below is an example showing how to concatenate two arrays.
Copyimport Array._ object Student { def main(args: Array[String]) { var sname = Array("John","Adam","Rob","Reena","Harry") var sname1 = Array("Jack","Jill","Henry","Mary","Rohan") var names = concat( sname, sname1) println("Student name array elements are : "); for ( n1 <- names ) { println( n1 ) } } }
We have to use
import Array._ since the array methods concat is defined in the package. We have declared two arrays sname and sname1 having student names as the elements. We are concatenating these two arrays using concat method and passing sname and sname1 as arguments and storing the resultant elements in names array.
Save the code in
Student.scala and then compile and run as shown in below image.
Scala Array with range
The range() method generates an array containing sequence of integers. The final argument is used as a step to generate the sequence of integers. If the argument is not specified then the default value assumed is 1.
Let us understand this range method through an example.
Copyimport Array._ object Student { def main(args: Array[String]) { var id = range(7, 23, 3) var age = range(15,20) for ( s <- id ) { print( " " + s) } println() for ( a <- age ) { print( " " + a ) } } }
We are declaring arrays id and age and generating the elements using range method. The elements start from 7 to 23 incrementing by 3. For the age the increment value by 1 starting from 15 until 20.
Run the above code by typing
Student.main(null) and you will see below output.
Copy7 10 13 16 19 22 15 16 17 18 19
Before I conclude the post, below are some useful Array methods:
- def concat[T]( xss: Array[T]* ): Array[T]: Concatenates all arrays into a single array
- def empty[T]: Array[T]: Returns an array of length 0
- def ofDim[T]( n1: Int, n2: Int ): Array[Array[T]]: Creates a 2-dimensional array
- def range( start: Int, end: Int, step: Int ): Array[Int]: Returns an array containing equally spaced values in some integer interval
- def copy( src: AnyRef, srcPos: Int, dest: AnyRef, destPos: Int, length: Int ): Unit: Copy one array to another
- def ofDim[T]( n1: Int, n2: Int, n3: Int ): Array[Array[Array[T]]]: Creates a 3-dimensional array
That’s all for arrays in Scala programming, we will look into other scala features in future. | https://www.journaldev.com/7915/scala-arrays-example | CC-MAIN-2019-13 | refinedweb | 843 | 53.51 |
We can automate WhatsApp to send messages by running a python script. In this tutorial, we will learn the simplest way of doing so using the pywhatkit module which utilizes the web.whatsapp.com webpage to automate message sending to any number on WhatsApp.
Now let's setup pywhatkit module and write the code to send WhatsApp message automatically.
To install the pywhatkit module, we can use the pip command:
pip install pywhatkit
This command will download the pywhatkit module. It will take some time as it will download some related modules too.
To use this python module to send message automatically on WhatsApp at a set time, we need chrome browser and you must have your WhatsApp logged into web.whatsapp.com website.
If you do not have chrome browser, then you can follow the following steps:
Download and extract the current stable release of chrome driver from
Open the downloaded file and search for an application named chrome drive, copy its path, for windows, it should look like this - C:/Users/.../chromedriver.exe.
Then call
pywhatkit.add_driver_path(path) and pass the copied path as an argument, if the path is valid, a black window along with chrome will open and close.
Now call
pywhatkit.load_QRcode() function and scan the QR code.
After following the above steps, you do not have to do anything, just run the final script to send whatsapp message.
To see the setup steps, you can use the
pywhatkit.manual() method in your python script.
The code is super simple,
import pywhatkit as kit kit.sendwhatmsg("+919*********", "I love studytonight.com!", 18, 21)
In the above code, we have specified the mobile number on which we want to send the message, then the message, and then the time at which the message has to be sent. This module follows the 24 hrs time format, hence the time 18:21 is 06:21 PM.
In 702 seconds web.whatsapp.com will open and after 60 seconds message will be delivered
Also, you should provide atleast 4-5 minutes future time from the current time while running the script, because if you will set the time 1-2 minute from current time, then the module will give error.
So that's it. You can use this script to automate WhatsApp to send Birthday wishes to your friends and family, to send daily morning message to your parents or use it for some business idea. As developers, we should look for ways to minimize our efforts and maximize the output.
If you face any issue while running this script, do share it with us in the comment section below. | https://www.studytonight.com/post/whatsapp-automation-to-send-message-using-python | CC-MAIN-2022-21 | refinedweb | 441 | 71.34 |
Albacore is software from Oxford Nanopore Technologies to perform basecalling of the reads obtained by their sequencers. Since the wrapper script is written in Python I can adapt it, and I wanted to try to add a progress bar to the script and that turned out to be surprisingly easy. Obviously, there was no need to reinvent the wheel, and a suitable Python library (progressbar2) already exists. The result is shown in the screenshot below:
This module can conveniently be installed using pip:
pip install progressbar2
If you have multiple python versions installed, make sure to use the right (Python3) pip executable, preferably using the following:
python3 -m pip install progressbar2
Since Albacore is proprietary software I cannot disclose the full code here below, but it shouldn’t be a problem to replicate my changes. I would suggest to copy the original script (called
read_fast5_basecaller.py) and make changes in the copy. Adding the progress bar involves adding just a few extra lines. Note that whenever a new version of the tool becomes available you’ll have to make the same changes again.
Important: I don’t have rights to the software and all changes you make are your own responsibility. I don’t claim this will work on each system and/or produce desirable results. Use at your own risk.
I hope my explanation on what you have to change will be understandable for both Pythonistas and novice programmers. If something is unclear, please leave a comment and I’ll clarify further.
First, add the import statement at the top of the script with the other import statements, on a separate line. The order of imports doesn’t matter.
import progressbar
Next, in the process_pipeline() function you have to add the progress bar as a “context wrapper”. Just above the
while loop you add
with progressbar.ProgressBar(max_value=num_files) as bar:
Pay attention to the indentation, which is crucial in Python. The rest of the lines of this function below our newly added line needs to be indented with another tab. This is about 65 lines which need another tab in front of them. As such you indicate that those lines ‘belong to the context manager’ we added.
Finally, just below the line
file_index += 1 you add another line (on the same indentation level):
bar.update(file_index)
That’s all!
3 thoughts on “Adding a progress bar to albacore”
In my case (python 3.5.2 on Ubuntu 16.04.2) the last line should be bar.update(file_index) instead of bar_update(file_index) as I got an error: NameError: name ‘bar_update’ is not defined
That’s actually a typo I made, I’ll correct it. Thanks for reporting! | https://gigabaseorgigabyte.wordpress.com/2017/05/02/adding-a-progress-bar-to-albacore/ | CC-MAIN-2019-22 | refinedweb | 449 | 72.97 |
I'm using model-view style coding for the first time. I've a combobox in which I've applied a model. Added some items in the model. I've connected PRESSED signal of view of combox to a function like this.
cmb.view().pressed.connect(self.udimCheck)
self.udimCheck() is like this
def udimCheck(self, index)
widget = self.sender()
This widget which I'm getting is an AbstractItemView object, But I need to get the combobox widget not the view. I looked the documentation. I didn't find a way to get the combobox when I've the view.
Typically, you simply set the model on the combobox and connect the signals to the combobox itself. I don't think you need to access it's view... unless you want to do something rather fancy.
Try this:
cmb.activated.connect(self.uimCheck) | http://tech-artists.org/t/get-widget-when-item-in-a-combobox-is-clicked-model-view/9247 | CC-MAIN-2017-51 | refinedweb | 143 | 69.68 |
Journey to Java: Episode 3 “Operators, Expressions, and Conditionals”
Learning Java so far has been an interesting process after learning other programming languages. I have studied the contents of episodes 1 and 2 for hours and only have down data types. I was impressed and shocked how many notes could be taken for every data type as opposed to learning Ruby and Javascript were pretty simple. Continuing my third week of this it is time to learn about Javas take on operators and related topics.
Operators
Operators are special symbols that perform specific actions on operands then return a result. Operands are any object being manipulated by the operator. Expressions are the entire operation together.
3 + 7;
// 3 and 7 are operands
// + is the operator
3 + 7 is the expression
List of Java Operators
+ Addition
- Subtraction
* Multiplication
/ Division Divides
% Modulus
++ Increment
-- Decrement
Conditional Logic
== equal to
!= not equal to
> greater than
>= greater than or equal to
< less than
<= less than or equal to
If-then looks at a condition to see if it is true.
boolean isOverTwentyOne = true;if (isOverTwentyOne == true){
System.out.println("over twenty one");
}// checking to see if isOverTwentyOne is true then performing a task if it is.
&& and operator
The and operator can be used in a conditional to check multiple operands being true.
int high = 70;
int low = 20;
int testNum = 55;if (testNum > low && testNum < high) {
System.out.println("test number is in the middle");
}
|| Or operator
The or operator is basically as it sounds and is the contrast to and operator. It only requires one operand to be true.
char a = "A";
char b = "B";
char newGrade = "A";if (NewGrade == a || newGrade == b) {
System.out.println("this is a good grade");
}
Shorthand for == and !=
If you want to check the truthiness of something you can just say if followed by the variable name. The bang operator will be the opposite value you want.
boolean happy = true;if (happy){
System.out.println("you are happy");
{if (!happy){
System.out.println("you are not happy");
}
Ternary Operator
Another way to check if then is to use the ternary. It checks the first condition to see its boolean value. If it is true the left operand happens and if it is false the right operand happens.
boolean isInvited = trueisInvited ? System.out.println("invited") : System.out.println("not invited");
Review Test
public class Practice { public static void main(String[] args) { //conditional test double twenty = 20.00; double eighty = 80.00; double solution = (twenty + eighty) * 100.00 % 40.00; boolean divisable = (solution == 0) ? true : false;
if (!divisable) { System.out.println("got some remainder"); } }}
Final Thoughts
I will probably have to revisit some of these topics in the future because some things have not been covered yet. I will make a follow up containing if then else statements and more complex expressions. Thankfully these topics are exactly the same as other languages I have learned. It is just going to take some repetition getting used to declaring variables with the data type it requires. I am happy to build on my foundational knowledge of programming while learning a new language that is so in demand. I am excited to start learning about java exclusive topics that are different than things I have learned in the past. | https://adamadolfo8.medium.com/journey-to-java-episode-3-operators-expressions-and-conditionals-c7d6d3183e84?source=post_internal_links---------3---------------------------- | CC-MAIN-2021-43 | refinedweb | 547 | 56.96 |
Ready to learn Artificial Intelligence? Browse courses like Uncertain Knowledge and Reasoning in Artificial Intelligence developed by industry thought leaders and Experfy in Harvard Innovation Lab.
Welcome to part five of Learning AI if You Suck at Math. If you missed part 1, part 2, part3 and part4 be sure to check them out.
Today, we’re going to write our own Python image recognition program.
To do that, we’ll explore a powerful deep learning architecture called a deep convolutional neural network (DCNN).
Convnets are the workhorses of computer vision. They power everything from self-driving cars to Google’s image search. At TensorFlow Summit 2017, a researcher showed how they’re using a convnet to detect skin cancer as well as a dermatologist with a smart phone!
So why are neural networks so powerful? One key reason:
They do automatic pattern recognition.
So what’s pattern recognition and why do we care if it’s automatic?
Patterns come in many forms but let’s take two critical examples:
- The features that define a physical form
- The steps it takes to do a task
Computer Vision
In image processing pattern recognition is known as feature extraction.
When you look at a photo or something in the real world you’re selectively picking out the key features that allow you to make sense of it. This is something you do unconsciously.
When you see the picture of my cat Dove you think “cat” or “awwwwww” but you don’t really know how you do that. You just do it.
You don’t know how you do it because it’s happening automatically and unconsciously.
My beautiful cat Dove. Your built in neural network knows this is a cat.
It seems simple to you because you do it every day, but that’s because the complexity is hidden away from you.
Your brain is a black box. You come with no instruction manual.
Yet if you really stop to think about it, what you just did in a fraction of second involved a massive number of steps. On the surface it’s deceptively simple but it’s actually incredibly complex.
- You moved your eyes.
- You took in light and you processed that light into component parts which sent signals to your brain.
- Then your brain went to work, doing its magic, converting that light to electro-chemical signals.
- Those signals fired through your built in neural network, activating different parts of it, including memories, associations and feelings.
- At the most “basic” level your brain highlighted low level patterns (ears, whiskers, tail) that it combined into higher order patterns (animal).
- Lastly, you made a classification, which means you turned it into a word, which is a symbolic representation of the real life thing, in this case a “cat.”
All of that happened in the blink of an eye.
If you tried to teach a computer to do that, where would you even begin?
- Could you tell it how to detect ears?
- What are ears?
- How do you describe them?
- Why are cat ears different than human ears or bat ears (or Batman)?
- What do ears look like from various angles?
- Are all cat ears the same (Nope, check out a Scottish Fold)?
The problems go on and on.
If you couldn’t come up with a good answer on how to teach a computer all those steps with some C++ or Python, don’t feel bad, because it stumped computer scientists for 50 years!
What you do naturally is one of the key uses for a deep learning neural network, which is a “classifier”, in this case an image classifier.
In the beginning, AI researchers tried to do the exercise we just went through. They attempted to define all the steps manually. For example, when it comes to natural language processing or NLP, they assembled the best linguists and said “write down all the ‘rules’ for languages.” They called these early AI’s “expert systems.”
The linguists sat down and puzzled out a dizzying array of if, then, unless, except statements:
- Does a bird fly?
Yes
Unless it’s:
- Dead
- Injured
- A flightless bird like a Penguin
- Missing a wing
These lists of rules and exceptions are endless. Unfortunately they’re also terribly brittle and prone to all kinds of errors. They’re time consuming to create, subject to debate and bias, hard to figure out, etc.
Deep neural networks represent a real breakthrough because instead of you having to figure out all the steps, you can let the machine extract thekey features of a cat automatically.
“Automatically” is essential because we bypass the impossible problem of trying to figure out all those thousands or millions of hidden steps we take to do any complex action.
We can let the computer figure it out for itself!
The Endless Steps of Everything
Let’s look at the second example: Figuring out the steps to do a task.
Today we do this manually and define the steps for a computer. It’s called programming. Let’s say you want to find all the image files on your hard drive and move them to a new folder.
For most tasks the programmer is the neural network. He’s the intelligence. He studies the task, decomposes it into steps and then defines each step for the computer one by one. He describes it to the computer with a symbolic representation known as a computer programming language.
Here’s an example in Python, from “Jolly Jumper” on Stack Exchange:
importshutil
import os
dst_dir = “your/destination/dir”
shutil.move(jpgfile, dst_dir)
Jolly Jumper figured out all the steps and translated them for the computer, such as:
- We need to know the source directory
- Also, we need a destination
- We need a way of classifying the types of files we want, in this case a “jpg” file
- Lastly we go into the directory, search it for any jpgs and move them from the source to the destination directory
This works well for simple and even moderately complex problems. Operating systems are some of the most complex software on Earth, composed of 100's of millions of lines of code. Each line is an explicit instruction for how computers do tasks ( like draw things on the screen, store and update information ) as well as how people do tasks ( copy files, input text, send email, view photos, chat with others, etc. ).
But as we evolve to try and solve more challenging problems we’re running into the limits of our ability to manually define the steps of the problem.
For example, how do you define driving a car?
There are hundreds of millions of tiny steps that we take to do this mind-numbingly complex task. We have to:
- Stay in the lines
- Know what a line is and be able to recognize it
- Navigate from one place to another
- Recognize obstructions like walls, people, debris
- Classify objects as helpful (street sign) or threat (pedestrian crossing a green light)
- Assess where all the drivers around us are constantly
- Make split second decisions
In machine learning this is known as a decision making problem. Examples of complex decision making problems are:
- Robot navigation and perception
- Language translation systems
- Self driving cars
- Stock trading systems
The Secret Inner Life of Neural Networks
Let’s see how deep learning helps us solve the insane complexity of the real world by doing automatic feature extraction!
If you’ve ever read the excellent book Think Like a Programmer, by V. Anton Spraul (and you should), you know that programming is about problem solving. The programmer decomposes a problem down into smaller problems, creates an action plan to solve it and then writes code to make it happen.
Deep Learning solves problems for us, but AI still needs humans at this point (thank God) to design and test AI architectures (at least for now.) So let’s decompose a neural net into its parts and build a program to recognize that the picture of my Dove is a cat.
The Deep in Deep Learning
Deep learning is subfield of machine learning. It’s name comes from the idea that we stack together a bunch of different layers to learn increasingly meaningful representations of data.
Each of those layers are neural networks, which consist of linked connections between artificial neurons.
Before we had powerful GPUs to do the math for us we could only build very small “toy” neural nets. They couldn’t do very much. Today we can stack many layers together hence the “deep” in deep learning.
Neural nets were inspired by biological research into the human brain in the 1950s. Researchers created a mathematical representation of a neuron, which you can see below (courtesy of the awesome open courseware on Convolutional Neural Nets from Stanford and Wikimedia Commons):
Biological neuron
Math model of a neuron.
Forget about all the more complex math symbols for now, because you don’t need them.
The basics are super simple. Data, represented by x0, travels through the connections between the neurons. The strength of the connections are represented by their weights (w0x0, w1x1, etc). If the signal is strong enough, it fires the neuron via its “activation function” and makes the neuron “active.”
Here is an example of a three layer deep neural net:
By activating some neurons and not others and by strengthening the connections between neurons, the system learns what’s important about the world and what’s not.
Building and Training a Neural Network
Let’s take a deeper look at deep learning and write some code as we go. All the code is available on my Github here.
The essential characteristics of the system are:
- Training
- Input data
- Layers
- Weights
- Targets
- Loss function
- Optimizer function
- Predictions
Training
Training is how we teach a neural network what we want it to learn. It follows a simple five step process:
- Create a training data set, which we will call x and load its labels as targets y
- Feed the x data forward through the network with the result being predictions y’
- Figure out the “loss” of the network, which is the difference between the predictions y’ and the correct targets y
- Compute the “gradient” of the loss (l) and which tells us how fast we’re moving towards or away from the correct targets
- Adjust the weights of the network in the opposite direction of the gradient and go back to step two to try again
Input Data
In this case the input data to a DCNN is a bunch of images. The more images the better. Unlike people, computers need a lot of examples to learn how to classify them. AI researchers are working on ways to learn with a lot less data but that’s still a cutting edge problem.
A famous example is the ImageNet data set. It consists of lots of hand labeled images. In other words, they crowd sourced the humans to use their built in neural nets to look at all the images and provide meaning to the data. People uploaded their photos and labeled it with tags, like “dog”, or a specific type of dog like a “Beagle.”
Those labels represent accurate predictions for the network. The closer the network gets to matching the hand labeled data (y) with their predictions (y’) the more accurate the network grows.
The data is broken into two pieces, a training set and testing set. The training set is the input that we feed to our neural network. It learns the key features of various kinds of objects and then we test whether it can accurately find those objects on random data in the test image set.
In our program we’ll use the well known CIFAR-10 dataset which was developed by the Canadian Institute for Advanced Research.
CIFAR-10 has 60000 32x32 color images in 10 classes, with 6000 images per class. We get 50000 training images and 10000 test images.
When I first started working with CIFAR I mistakenly assumed it would be an easier challenge than working with the larger images of the ImageNet challenge. It turns out CIFAR10 is more challenging because the images are so tiny and there are a lot less of them, so they have less identifiable characteristics for our neural network to lock in on.
While some of the biggest and baddest DCNN architectures like ResNet can hit 97% accuracy on ImageNet, it can only hit about 87% on CIFAR 10, in my experience. The current state of the art on CIFAR 10 is DenseNet, which can hit around 95% with a monstrous 250 layers and 15 million parameters! I link to those frameworks at the bottom of the article for further exploration. But it’s best to start with something simpler before diving into those complex systems.
Enough theory! Let’s write code.
If you’re not comfortable with Python, I highly, highly, highly recommend Learning Python by Fabrizio Romano. This book explains everything so well. I’ve never found a better Python book and I have a bunch of them that failed to teach me much.
The code for our DCNN is based on the Keras example code on Github.
You can find my modifications here.
I’ve adjusted the architecture and parameters, as well as added TensorBoard to help us visualize the network.
Let’s initialize our Python program, import the dataset and the various classes we’ll need to build our DCNN. Luckily, Keras already knows how to get this dataset automatically so we don’t have too much work to do.
importnumpy as np
fromkeras.callbacks import TensorBoard
fromkeras.models import Sequential
fromkeras.layers import Dense, Dropout, Activation, Flatten
fromkeras.layers import Convolution2D, MaxPooling2D
fromkeras.utils import np_utils
fromkeras import backend as K
Our neural net starts off with a random configuration. It’s as good a starting place as any but we shouldn’t expect it to start off very smart. Then again, it’s possible that some random configuration gives us amazing results completely by accident, so we seed the random weights to make sure that we don’t end up with state of the art results by sheer dumb luck!
np.random.seed(1337) # Very l33t
Layers.
Let’s see what the Stanford course on computer vision has to say about convnet scaling:
“In CIFAR-10, the image are merely respectible.”
Overfitting is when you train the network so well that it kicks ass on the training data but sucks when you show it images it’s never seen. In other words it’s not much use in the real world.
It’s as if you played the same game of chess over and over and over again until you had it perfectly memorized. Then someone makes a different move in a real game and you have no idea what to do. We’ll look at overfitting more later.
Here’s how data flows through a DCNN. It looks at only a small subset of the data, hunting for patterns. It then builds those observations up into higher order understandings.
A visual representation of a convolutional neural net from the mNeuron plugin created for MIT’s computer vision courses/teams.
Notice how the first few layers are simple patterns like edges and colors and basic shapes.
As the information flows through the layers, the system finds more and more complex patterns, like textures, and eventually it deduces various object classes.
The ideas were based on experiments on cat vision that showed that different cells responded to only certain kinds of stimuli such as an edge or a particular color.
Slides from the excellent Deep Learning open course at Oxford.
The same is true for humans. Our visual cells respond only to very specific features.
Here is a typical DCNN architecture diagram:
You’ll notice a third kind of layer in there, a pooling layer. You can find all kinds of detail in the Oxford lectures and the Standford lectures. However, I’m going to skip a lot of the granular detail because most people just find it confusing. I know I did when I first tried to make sense of it.
Here’s what you need to know about pooling layers. Their goal is simple. They do subsampling. In other words they shrink the input image, which reduces the computational load and memory usage. With less information to crunch we can work with the images more easily.
They also help reduce a second kind of overfitting where the network zeros in on anomalies in the training set that really have nothing to do with picking out dogs or birds or cats. For example, there may be some garbled pixels or some lens flares on a bunch of the images. The network may then decide that lens flare and dog go together, when they’re about as closely related as an asteroid and a baby rattle.
Lastly, most DCNNs add a few densely connected, aka fully connected layers to process out all the features maps detected in earlier layers and make predictions.
So let’s add a few layers to our convnet.
First we add some variables that we will pull into our layers.
batch_size = 128
nb_classes = 10
nb_epoch = 45
img_rows, img_cols = 32, 32
nb_filters = 32
pool_size = (2, 2)
# convolution kernel size
kernel_size = (3, 3)
The kernel and pooling size define how the convolutional network passes over the image looking for features. The smallest kernel size would be 1x1, which means we think key features are only 1 pixel wide. Typical kernel sizes check for useful features over 3 pixels at a time and then pool those features down to a 2x2 grid.
The 2x2 grid pulls the features out of the image and stacks them up like trading cards. This disconnects them from a specific spot on the image and allows the system to look for straight lines or swirls anywhere, not just in the spot it found them in the first place.
Most tutorials describe this as dealing with “translation invariance.”
What the heck does that mean? Good question.
Take a look at this image again:
Without yanking the features out, like you see in layer 1 or layer 2, the system might decide that the circle of a cat’s nose was only important right smack in the center of the image where it found it.
Let’s see how that works with my Dove. If the system originally finds a circle in her eye then it might mistakenly assume that the position of the circle in an image is relevant to detecting cats.
Instead the system should look for circles wherever they may roam, as we see below.
Before we can add the layers we need to load and process the data.
X_train = X_train.reshape(X_train.shape[0], 3, img_rows, img_cols)
X_test = X_test.reshape(X_test.shape[0], 3, img_rows, img_cols)
input_shape = (1, img_rows, img_cols)
else:
X_train = X_train.reshape(X_train.shape[0], img_rows, img_cols, 3)
X_test = X_test.reshape(X_test.shape[0], img_rows, img_cols, 3)
input_shape = (img_rows, img_cols, 3)
X_test = X_test.astype('float32')
X_train /= 255
X_test /= 255
print('X_train shape:', X_train.shape)
print(X_train.shape[0], 'train samples')
print(X_test.shape[0], 'test samples')
Y_test = np_utils.to_categorical(y_test, nb_classes)
border_mode='valid',
input_shape=input_shape))
model.add(Activation('relu'))
model.add(Convolution2D(nb_filters, kernel_size[0], kernel_size[1]))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=pool_size))
model.add(Dropout(0.25))
The layers are stacked as follows:
- Convolution
- Activation
- Convolution
- Activation
- Pooling
- Dropout
We’ve already discussed most of these layer types except for two of them, dropout and activation.
Dropout is the easiest to understand. Basically it’s a percentage of how much of the model to randomly kill off. This is similar to how Netflix uses Chaos Monkey. They have scripts that turn off random servers in their network to ensure the network can survive with its built in resilience and redundancy. The same is true here. We want to make sure the network is not too dependent on any one feature.
The activation layer is a way to decide if the neuron “fires” or gets “activated.” There are dozens of activation functions at this point. RELU is the one of the most successful because of its computational efficiency. Here is a list of all the different kinds of activation functions available in Keras.
We’ll also add a second stack of convolutional layers that mirror the first one. If we were rewriting this program for efficiency we would create a model generator and do a for loop to create however many stacks we want. But in this case we will just cut and paste the layers from above, violating the zen rules of Python for expediency sake.
model.add(Activation('relu'))
model.add(Convolution2D(nb_filters, kernel_size[0], kernel_size[1]))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=pool_size))
model.add(Dropout(0.25))
Lastly, we add the dense layers, some more drop out layers and we flatten all the features maps.
model.add(Dense(256))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(nb_classes))
model.add(Activation('softmax'))
We use a different kind of activation called softmax on the last layer, because it defines a probability distribution over the classes.
Weights
We talked briefly about what weights were earlier but now we’ll look at them in depth.
Weights are the strength of the connection between the various neurons.
We have parallels for this in our own minds. In your brain, you have a series of biological neurons. They’re connected to other neurons with electrical/chemical signals passing between them.
But the connections are not static. Over time some of those connections get stronger and some weaker.
The more electro-chemical signals flowing between two biological neurons, the stronger those connections get. In essence, your brain rewires itself constantly as you have new experiences. It encodes your memories and feelings and ideas about those experiences by strengthening the connections between some neurons.
Source U.S. National Institute of Health — Wikimedia Commons.
Computer based neural networks are inspired by biological ones. We call them Artificial Neural Networks or ANNs for short. Usually when we say “neural network” what we really mean is ANN. ANN’s don’t function exactly the same as biological ones, so don’t make the mistake of thinking an ANN is some kind of simulated brain. It’s not. For example in a biological neural network (BNN), every neuron does not connect to every other neuron whereas in an ANN every neuron in one layer generally connects to every neuron in the next layer.
Below is an image of a BNN showing connections between various neurons. Notice they’re not all linked.
Source: Wikimedia Commons: Soon-Beom HongAndrew ZaleskyLuca CocchiAlex FornitoEun-Jung ChoiHo -Hyun KimJeong -Eun SuhChang-Dai KimJae -Won KimSoon -Hyung Yi
Though there are many differences, there are also very strong parallels between BNNs and ANNs.
Just like the neurons in your head form stronger or weaker connections, the weights in our artificial neural network define the strength of the connections between neurons. Each neuron knows a little bit about the world. Wiring them together allows them to have a more comprehensive view of the world when taken together. The ones that have stronger connections are considered more important for the problem we’re trying to solve.
Let’s look at several screenshots of the Neural Network Playground, a visualizer for TensorFlow to help understand this better.
The first network shows a simple six layer system. What the network is trying to do is cleanly separate the blue dots from the orange dots in the picture on the far right. It’s looking for the best pattern that separates them with a high degree of accuracy.
I have not yet started training the system here. Because of that we can see weights between neurons are mostly equal. The thin dotted lines are weak connections and the thicker lines are strong connections. The network is initialized with random weights as a starting point.
Now let’s take a look at the network after we’ve trained it.
First notice the picture on the far right. It now has a nice blue dot in the middle around the blue dots and orange around the rest of the picture. As you can see it’s done pretty well, with a high degree of accuracy. This happened over 80 “epochs” or training rounds.
Also notice that many of the weights have strong blue dotted lines between various neurons. The weights have increased and now the system is trained and ready to take on the world!
Training Our Neural Net and Optimizing It
Now let’s have the model crunch some numbers. To do that we compile it and set its optimizer function.
optimizer='adam',
metrics=['accuracy'])
It took me a long time to understand the optimizer function because I find most explanations miss the “why” behind the “what.”
In other words, why the heck do I need an optimizer?
Remember that a network has target predictions y and as it’s trained over many epochs it makes new predictions y’. The system tests these predictions against a random sample from the test dataset and that determines the system’s validation accuracy. A system can end up 99% accurate on the training data and only hit 50% or 70% on test images, so the real name of the game is validation accuracy, not accuracy.
The optimizer calculates the gradient (also known as partial derivatives in math speak) of the error function with respect to the model weights.
What does that mean? Think of the weights distributed across a 3D hilly landscape (like you see below), which is called the “error landscape.” The “coordinates” of the landscape represent specific weight configurations (like coordinates on a map), while the “altitude” of the landscape corresponds to the total error/cost for the different weight configurations.
Error landscape
The optimizer serves one important function. It figures out how to adjust the weights to try to minimize the errors. It does this by taking a page from the book of calculus.
What is calculus? Well if you turn to any math text book you’ll find some super unhelpful explanations such as it’s all about calculating derivatives or differentials. But what the heck does that mean?
I didn’t understand it until I read Calculus Better Explained, by Kalid Azad.
Here’s what nobody bothers to explain.
Calculus does two things:
- Breaks things down into smaller chunks, aka a circle into rings.
- Figures out rates of change.
In other words if I slice up a circle into rings:
Courtesy of the awesome Calculus Explained website.
I can unroll the rings to do some simple math on it:
Bam!
In our case we run a bunch of tests, adjust the weights of the network but did we actually get any closer to an better solution to the problem? The optimizer tells us that!
You can read about gradient descent with an incredible amount of detail here or in the Stanford course but you’ll probably find like I did that they’re long on detail and light on the crucial question of why.
In essence, what you’re trying to do is minimize the errors. It’s a bit like driving around in the fog. In an earlier version of this post, I characterized gradient descent as a way to to find an optimal solution. But actually, there is really no way to know if we have an “optimal” solution at all. If we knew what that was, we would just go right to it. Instead we are trying to find a “better” solution that works. This is a bit like evolution. We find something that is fit enough to survive but that doesn’t mean we created Einstein!
Think of gradient descent like when you played Marco Polo as a kid.
You closed your eyes and all your friends spread out in the pool. You shouted out “Marco” and all the kids had to answer “Polo.” You used your ears to figure if you were getting closer or farther away. If you were farther away you adjusted and tried a different path. If you were closer you kept going in that direction. Here we’re figuring out how best to adjust the weights of the network to help them get closer to understanding the world.
We chose the “adam” optimizer described in this paper. I’ve found through brute force changing my program that it seems to produce the best results. This is the art of data science. There is no one algorithm to rule them all. If I changed the architecture of the network, I might find a different optimizer worked better.
Here is a list of all the various optimizers in Keras.
Next we set up TensorBoard so we can visualize how the network performs.
tb = TensorBoard(log_dir='./logs')
All we did was create a log directory. Now we will train the model and point TensorBoard at the logs.
print("Accuracy: %.2f%%" % (score[1]*100))
All right, let’s fire this bad boy up and see how it does!
Epoch 89/100
50000/50000 [==============================] - 3s - loss: 0.4834 - acc: 0.8269 - val_loss: 0.6286 - val_acc: 0.7911
Epoch 90/100
50000/50000 [==============================] - 3s - loss: 0.4908 - acc: 0.8224 - val_loss: 0.6169 - val_acc: 0.7951
Epoch 91/100
50000/50000 [==============================] - 4s - loss: 0.4817 - acc: 0.8238 - val_loss: 0.6052 - val_acc: 0.7952
Epoch 92/100
50000/50000 [==============================] - 4s - loss: 0.4863 - acc: 0.8228 - val_loss: 0.6151 - val_acc: 0.7930
Epoch 93/100
50000/50000 [==============================] - 3s - loss: 0.4837 - acc: 0.8255 - val_loss: 0.6209 - val_acc: 0.7964
Epoch 94/100
50000/50000 [==============================] - 4s - loss: 0.4874 - acc: 0.8260 - val_loss: 0.6086 - val_acc: 0.7967
Epoch 95/100
50000/50000 [==============================] - 3s - loss: 0.4849 - acc: 0.8248 - val_loss: 0.6206 - val_acc: 0.7919
Epoch 96/100
50000/50000 [==============================] - 4s - loss: 0.4812 - acc: 0.8256 - val_loss: 0.6088 - val_acc: 0.7994
Epoch 97/100
50000/50000 [==============================] - 3s - loss: 0.4885 - acc: 0.8246 - val_loss: 0.6119 - val_acc: 0.7929
Epoch 98/100
50000/50000 [==============================] - 3s - loss: 0.4773 - acc: 0.8282 - val_loss: 0.6243 - val_acc: 0.7918
Epoch 99/100
50000/50000 [==============================] - 3s - loss: 0.4811 - acc: 0.8271 - val_loss: 0.6201 - val_acc: 0.7975
Epoch 100/100
50000/50000 [==============================] - 3s - loss: 0.4752 - acc: 0.8299 - val_loss: 0.6140 - val_acc: 0.7935
Test score: 0.613968349266
Accuracy: 79.35%
We hit 79% accuracy after 100 epochs. Not bad for a few lines of code. Now you might think 79% is not that great, but remember that in 2011, that was better than state of the art on Imagenet and it took a decade to get there! And we did that with just some example code from the Keras Github and a few tweaks.
You’ll notice that in 2012 is when new ideas started to make an appearance.
AlexNet, by AI researchers Alex Krizhevsky, Ilya Sutskever and Geoffrey Hinton, is the first orange dot. It marked the beginning of the current renaissance in deep learning. By the next year everyone was using deep learning. By 2014 the winning architecture was better than human level image recognition.
Even so, these architectures are often very tied to certain types of problems. Several of the most popular architectures today, like ResNet and Google’s Inception V3 do only 88% on the tiny CIFAR10 images. They do even worse on the larger CIFAR100 set.
The current state of the art is DenseNet, which won the ImageNet contest last year in 2016. It chews through CIFAR10, hitting a killer 94.81% accuracywith an insanely deep 250 layers and 15.3 million connections! It is an absolute monster to run. On a single Nvidia 1080GTX, if you run it with the 40 x 12 model which hits the 93% accuracy mark you see in the chart below, it will take a month to run. Ouch!
That said, I encourage you to explore these models in depth to see what you can learn from them.
I did some experimenting and managed to hack together a weird architecture through brute force experimentation that achieve 81.40% accuracy using nothing but the build in Keras layers and no custom layers. You can find that on Github here.
50000/50000 [==============================] - 10s - loss: 0.3503 - acc: 0.8761 - val_loss: 0.6229 - val_acc: 0.8070
Epoch 71/75
50000/50000 [==============================] - 10s - loss: 0.3602 - acc: 0.8740 - val_loss: 0.6039 - val_acc: 0.8085
Epoch 72/75
50000/50000 [==============================] - 10s - loss: 0.3543 - acc: 0.8753 - val_loss: 0.5986 - val_acc: 0.8094
Epoch 73/75
50000/50000 [==============================] - 10s - loss: 0.3461 - acc: 0.8780 - val_loss: 0.6052 - val_acc: 0.8147
Epoch 74/75
50000/50000 [==============================] - 10s - loss: 0.3418 - acc: 0.8775 - val_loss: 0.6457 - val_acc: 0.8019
Epoch 75/75
50000/50000 [==============================] - 10s - loss: 0.3440 - acc: 0.8776 - val_loss: 0.5992 - val_acc: 0.8140
Test score: 0.599217191744
Accuracy: 81.40%
We can load up TensorBoard to visualize how we did as well.
tensorboard --logdir=./logs
Now open a browser and go to the following URL:
127.0.1.1:6006
Here is a screenshot of the training over time.
You can see we quickly start to pass the point of diminishing returns at around 35 epochs and 79%. The rest of the time is spent getting it to 81.40% and likely overfitting at anything beyond 75 epochs.
So how would you improve this model?
Here are a few strategies:
- Implement your own custom layers
- Do image augmentation, like flipping images, enhancing them, warping them, cloning them, etc
- Go deeper
- Change the settings on the layers
- Read through the winning architecture papers and stack up your own model that has similar characteristics
And thus you have reached the real art of data science, which is using your brain to understand the data and hand craft a model to understand it better. Perhaps you dig deep into CIFAR10 and notice that upping the contrast on those images would really make images stand out. Do it!
Don’t be afraid to load things up in Photoshop and start messing with filters to see if images get sharper and clearer. Figure out if you can do the same thing with Keras image manipulation functions.
Deep learning is far from a magic bullet. It requires patience and dedication to get right.
It can do incredible things but you may find yourself glued to your workstation watching numbers tick by for hours until 2 in the morning, getting absolutely nowhere.
But then you hit a breakthrough!
It’s a bit like the trial and error a neural net goes through. Try some stuff, get closer to an answer. Try something else and get farther away.
I am now exploring how to use genetic algorithms to auto-evolve neural nets. There’s been a bunch of work done on this front but not enough!
Eventually we’ll hit a point where many of the architectures are baked and easy to implement by pulling in some libraries and some pre-trained weights files but that is a few years down the road for enterprise IT.
This field is still fast developing and new ideas are coming out every day. The good news is you are on the early part of the wave. So get comfortable and start playing around with your own models.
Study. Experiment. Learn.
Do that and you can’t go wrong. | https://www.experfy.com/blog/learning-ai-if-you-suck-at-math-part5-deep-learning-and-convolutional-neural-nets-in-plain-english | CC-MAIN-2019-35 | refinedweb | 5,938 | 66.23 |
In this project, you’ll learn how to build an asynchronous ESP32 web server with the DHT11 or DHT22 that displays temperature and humidity using Arduino IDE.
The web server we’ll build updates the readings automatically without the need to refresh the web page.
With this project you’ll learn:
- How to read temperature and humidity from DHT sensors;
- Build an asynchronous web server using the ESPAsyncWebServer library;
- Update the sensor readings automatically without the need to refresh the web page.
For a more in-depth explanation on how to use the DHT22 and DHT11 temperature and humidity sensors with the ESP32, read our complete guide: ESP32 with DHT11/DHT22 Temperature and Humidity Sensor using Arduino IDE
Watch the Video Tutorial
You can watch the video tutorial or keep reading this page for the written instructions.
Asynchronous Web Server
To build the web server we’ll use the ESPAsyncWebServer library that provides an easy way to build an asynchronous web server. Building an asynchronous web server has several advantages as mentioned in the library GitHub page, such as:
- “Handle more than one connection at the same time”;
- “When you send the response, you are immediately ready to handle other connections while the server is taking care of sending the response in the background”;
- “Simple template processing engine to handle templates”;
- And much more.
Take a look at the library documentation on its GitHub page.
Parts Required
To complete this tutorial you need the following parts:
- ESP32 development board (read ESP32 development boards comparison)
- DHT22 or DHT11 Temperature and Humidity Sensor
- 4.7k Ohm Resistor
- Breadboard
- Jumper wires
You can use the preceding links or go directly to MakerAdvisor.com/tools to find all the parts for your projects at the best price!
Schematic
Before proceeding to the web server, you need to wire the DHT11 or DHT22 sensor to the ESP32 as shown in the following schematic diagram.
In this case, we’re connecting the data pin to GPIO 27, but you can connect it to any other digital pin. You can use this schematic diagram for both DHT11 and DHT22 sensors.
(This schematic uses the ESP32 DEVKIT V1 module version with 36 GPIOs – if you’re using another model, please check the pinout for the board you’re using.)
Note: if you’re using a module with a DHT sensor, it normally comes with only three pins. The pins should be labeled so that you know how to wire them. Additionally, many of these modules already come with an internal pull up resistor, so you don’t need to add one to the circuit.
Installing Libraries
You need to install a couple of libraries for this project:
- The DHT and the Adafruit Unified Sensor Driver libraries to read from the DHT sensor.
- ESPAsyncWebServer and Async TCP libraries to build the asynchronous web server.
Follow the next instructions to install those libraries:
Installing the DHT Sensor Library
To read from the DHT sensor using Arduino IDE, you need to install the DHT sensor library. Follow the next steps to install the library.
- Click here to download the DHT Sensor library. You should have a .zip folder in your Downloads folder
- Unzip the .zip folder and you should get DHT-sensor-library-master folder
- Rename your folder from
DHT-sensor-library-masterto DHT_sensor
- Move the DHT_sensor folder to your Arduino IDE installation libraries folder
- Finally, re-open your Arduino IDE
Installing the Adafruit Unified Sensor Driver
You also need to install the Adafruit Unified Sensor Driver library to work with the DHT sensor. Follow the next steps to install the library.
- Click here to download the Adafruit Unified
Code
We’ll program the ESP32 using Arduino IDE, so make sure you have the ESP32 add-on installed before proceeding:
Open your Arduino IDE and copy the following code.
/********* Rui Santos Complete project details at *********/ // Import required libraries #include "WiFi.h" #include "ESPAsyncWebServer.h" #include <Adafruit_Sensor.h> #include <DHT.h> // Replace with your network credentials const char* > </head> <body> <h2>ESP32 DHT Server</h2> <p> <i class="fas fa-thermometer-half" style="color:#059e8a;"></i> <span class="dht-labels">Temperature</span> <span id="temperature">%TEMPERATURE%</span> <sup class="units">°C</sup> </p> <p> <i class="fas fa-tint" style="color:#00add6;"></i> <span class="dht-labels">Humidity</span> <span id="humidity">%HUMIDITY%</span> <sup class="units">%</sup> </p> </body> > </html>)rawliteral"; // Replaces placeholder with DHT values String processor(const String& var){ //Serial.println(var); if(var == "TEMPERATURE"){ return readDHTTemperature(); } else if(var == "HUMIDITY"){ return readDHTHumidity(); } return String(); } void setup(){ // Serial port for debugging purposes Serial.begin(115200); dht.begin(); //()); }); // Start server server.begin(); } void loop(){ }
Insert your network credentials in the following variables and the code will work straight away.
const char* ssid = "REPLACE_WITH_YOUR_SSID"; const char* password = "REPLACE_WITH_YOUR_PASSWORD";
How the Code Works
In the following paragraphs we’ll explain how the code works. Keep reading if you want to learn more or jump to the Demonstration section to see the final result.
Importing libraries
First, import the required libraries. The WiFi, ESPAsyncWebServer and the ESPAsyncTCP are needed to build the web server. The Adafruit_Sensor and the DHTlibraries are needed to read from the DHT11 or DHT22 sensors.
#include "WiFi.h" #include "ESPAsyncWebServer.h" #include <ESPAsyncTCP.h> #include <Adafruit_Sensor.h> #include <DHT.h>
Setting your network credentials
Insert your network credentials in the following variables, so that the ESP32 can connect to your local network.
const char* ssid = "REPLACE_WITH_YOUR_SSID"; const char* password = "REPLACE_WITH_YOUR_PASSWORD";
Variables definition
Define the GPIO that the DHT data pin is connected to. In this case, it’s connected to GPIO 27.
#define DHTPIN 27 // Digital pin connected to the DHT sensor
Then, select the DHT sensor type you’re using. In our example, we’re using the DHT22. If you’re using another type, you just need to uncomment your sensor and comment all the others.
#define DHTTYPE DHT22 // DHT 22 (AM2302)
Instantiate a DHTobject with the type and pin we’ve defined earlier.
DHT dht(DHTPIN, DHTTYPE);
Create an AsyncWebServerobject on port 80.
AsyncWebServer server(80);
Read Temperature and Humidity Functions
We’ve created two functions: one to read the temperature (readDHTTemperature()) and the other to read humidity (readDHTHumidity()).); } }
Getting sensor readings is as simple as using the readTemperature() and readHumidity() methods on the dht object.
float t = dht.readTemperature();
float h = dht.readHumidity();
We also have a condition that returns two dashes (–) in case the sensor fails to get the readings.
if (isnan(t)) { Serial.println("Failed to read from DHT sensor!"); return "--"; }
The readings are returned as string type. To convert a float to a string, use the String() function.
return String(t);
By default, we’re reading the temperature in Celsius degrees. To get the temperature in Fahrenheit degrees, comment the temperature in Celsius and uncomment the temperature in Fahrenheit, so that you have the following:
//float t = dht.readTemperature(); // Read temperature as Fahrenheit (isFahrenheit = true) float t = dht.readTemperature(true);
Building the Web Page
Proceeding to the web server page.
As you can see in the above figure, the web page shows one heading and two paragraphs. There is a paragraph to display the temperature and another to display the humidity. There are also two icons to style our page.
Let’s see how this web page is created.
All the HTML text with styles included is stored in the index_html variable. Now we’ll go through the HTML text and see what each part does.
The following <meta> tag makes your web page responsive in any browser.
<meta name="viewport" content="width=device-width, initial-scale=1">
The <link> tag is needed to load the icons from the fontawesome website.
">
Styles
Between the <style></style> tags, we add some CSS to style the web page.
>
Basically, we’re setting the HTML page to display the text with Arial font in block without margin, and aligned at the center.
html { font-family: Arial; display: inline-block; margin: 0px auto; text-align: center; }
We set the font size for the heading (h2), paragraph (p) and the units(.units) of the readings.
h2 { font-size: 3.0rem; } p { font-size: 3.0rem; } .units { font-size: 1.2rem; }
The labels for the readings are styled as shown below:
dht-labels{ font-size: 1.5rem; vertical-align:middle; padding-bottom: 15px; }
All of the previous tags should go between the <head> and </head> tags. These tags are used to include content that is not directly visible to the user, like the <meta> , the <link> tags, and the styles.
HTML Body
Inside the <body></body> tags is where we add the web page content.
The <h2></h2> tags add a heading to the web page. In this case, the “ESP32 DHT server” text, but you can add any other text.
<h2>ESP32 DHT Server</h2>
Then, there are two paragraphs. One to display the temperature and the other to display the humidity. The paragraphs are delimited by the <p> and </p> tags. The paragraph for the temperature is the following:
<p> <i class="fas fa-thermometer-half" style="color:#059e8a;"</i> <span class="dht-labels">Temperature</span> <span id="temperature">%TEMPERATURE%</span> <sup class="units">°C</sup> </p>
And the paragraph for the humidity is on the following snipet:
<p> <i class="fas fa-tint" style="color:#00add6;"></i> <span class="dht-labels">Humidity</span> <span id="humidity">%HUMIDITY%</span> <sup class="units">%</sup> </p>
The <i> tags display the fontawesome icons.
How to display icons
To chose the icons, go to the Font Awesome Icons website.
Search the icon you’re looking for. For example, “thermometer”.
Click the desired icon. Then, you just need to copy the HTML text provided.
<i class="fas fa-thermometer-half">
To chose the color, you just need to pass the style parameter with the color in hexadecimal, as follows:
<i class="fas fa-tint" style="color:#00add6;"></i>
Proceeding with the HTML text…
The next line writes the word “Temperature” into the web page.
<span class="dht-labels">Temperature</span>
The TEMPERATURE text between % signs is a placeholder for the temperature value.
<span id="temperature">%TEMPERATURE%</span>
This means that this %TEMPERATURE% text is like a variable that will be replaced by the actual temperature value from the DHT sensor. The placeholders on the HTML text should go between % signs.
Finally, we add the degree symbol.
<sup class="units">°C</sup>
The <sup></sup> tags make the text superscript.
We use the same approach for the humidity paragraph, but it uses a different icon and the %HUMIDITY% placeholder.
<p> <i class="fas fa-tint" style="color:#00add6;"></i> <span class="dht-labels">Humidity</span> <span id="humidity">%HUMIDITY%</span> <sup class="units">%</sup> </p>
Automatic Updates
Finally, there’s some JavaScript code in our web page that updates the temperature and humidity automatically, every 10 seconds.
Scripts in HTML text should go between the <script></script> tags.
>
To update the temperature on the background, we have a setInterval() function that runs every 10 seconds.
Basically, it makes a request in the /temperature URL to get the latest temperature reading.
xhttp.open("GET", "/temperature", true); xhttp.send(); }, 10000 ) ;
When it receives that value, it updates the HTML element whose id is temperature.
if (this.readyState == 4 && this.status == 200) { document.getElementById("temperature").innerHTML = this.responseText; }
In summary, this previous section is responsible for updating the temperature asynchronously. The same process is repeated for the humidity readings.
Important: since the DHT sensor is quite slow getting the readings, if you plan to have multiple clients connected to an ESP32 at the same time, we recommend increasing the request interval or remove the automatic updates.
Processor
Now, we need to create the processor() function, that will replace the placeholders in our HTML text with the actual temperature and humidity values.
String processor(const String& var){ //Serial.println(var); if(var == "TEMPERATURE"){ return readDHTTemperature(); } else if(var == "HUMIDITY"){ return readDHTHumidity(); } return String(); }
When the web page is requested, we check if the HTML has any placeholders. If it finds the %TEMPERATURE% placeholder, we return the temperature by calling the readDHTTemperature() function created previously.
if(var == "TEMPERATURE"){ return readDHTTemperature(); }
If the placeholder is %HUMIDITY%, we return the humidity value.
else if(var == "HUMIDITY"){ return readDHTHumidity(); }
setup()
In the setup(), initialize the Serial Monitor for debugging purposes.
Serial.begin(115200);
Initialize the DHT sensor.
dht.begin();
Connect to your local network and print the ESP32 IP address.
WiFi.begin(ssid, password); while (WiFi.status() != WL_CONNECTED) { delay(1000); Serial.println("Connecting to WiFi.."); }
Finally, add the next lines of code to handle the web server.()); });
When we make a request on the root URL, we send the HTML text that is stored on the index_html variable. We also need to pass the processorfunction, that will replace all the placeholders with the right values.
server.on("/", HTTP_GET, [](AsyncWebServerRequest *request){ request->send_P(200, "text/html", index_html, processor); });
We need to add two additional handlers to update the temperature and humidity readings. When we receive a request on the /temperature URL, we simply need to send the updated temperature value. It is plain text, and it should be sent as a char, so, we use the c_str() method.
server.on("/temperature", HTTP_GET, [](AsyncWebServerRequest *request){ request->send_P(200, "text/plain", readDHTTemperature().c_str()); });
The same process is repeated for the humidity.
server.on("/humidity", HTTP_GET, [](AsyncWebServerRequest *request){ request->send_P(200, "text/plain", readDHTHumidity().c_str()); });
Lastly, we can start the server.
server.begin();
Because this is an asynchronous web server, we don’t need to write anything in the loop().
void loop(){ }
That’s pretty much how the code works.
Upload the Code
Now, upload the code to your ESP32. Make sure you have the right board and COM port selected.
After uploading, open the Serial Monitor at a baud rate of 115200. Press the ESP32 reset button. The ESP32 IP address should be printed in the serial monitor.
Web Server Demonstration
Open a browser and type the ESP32 IP address. Your web server should display the latest sensor readings.
Notice that the temperature and humidity readings are updated automatically without the need to refresh the web page.
Troubleshooting
If your DHT sensor fails to get the readings, read our DHT Troubleshooting Guide to help you fix the issue.
Wrapping Up
In this tutorial we’ve shown you how to build an asynchronous web server with the ESP32 to display sensor readings from a DHT11 or DHT22 sensor and how to update the readings automatically.
If you liked this project, you may also like:
- Learn ESP32 with Arduino IDE (course)
- Build an ESP32 Web Server using Files from Filesystem (SPIFFS)
- ESP32 Web Server – control outputs
- ESP32 Deep Sleep with Arduino IDE and Wake Up Sources
This tutorial is a preview of the “Learn ESP32 with Arduino IDE” course. If you like this project, make sure you take a look at the ESP32 course page where we cover this and a lot more topics with the ESP32.
82 thoughts on “ESP32 DHT11/DHT22 Web Server – Temperature and Humidity using Arduino IDE”
Hello. I keep getting this error when I compile.
Board nodemcuv2 (platform esp8266, package esp8266) is unknown
Error compiling for board NodeMCU 1.0 (ESP-12E Module).
It compiles OK when I select Arduino Nano, but not the ESP8266. Can you give me any suggestions on how to solve this problem?
Thanks,
Steve Tripoli
I changed these lines and it now compiles and uploads.
This is the orignial
// Replace with your network credentials
const char* ssid = “YOUR_SSID”;
const char* password = “YOUR_PASSWORD”;
All I did was to remove the const in both lines.
I’m glad it works! 🙂
Will this work using an ESP8266? I am trying, but this all I get from the serial monitor.
ets Jan 8 2013,rst cause:4, boot mode:(3,6)
wdt reset
load 0x4010f000, len 1384, room 16
tail 8
chksum 0x2d
csum 0x2d
v4ceabea9
~ld
Connecting to MyHome36
Hi Steve.
We have a similar project dedicated to the ESP8266 that you can read here: ESP8266 DHT11/DHT22 Temperature and Humidity Web Server with Arduino IDE
Try this project instead.
I hope this helps.
Regards,
Sara 🙂
Thank you. I did that project, works great thanks. I wanted to display temp of my garage on my phone without reloading the website to get the updated temp.
To display sensor readings without updating the web page, you can take a look at this project with the ESP32, where we build a web server that updates sensor readings without the need to update the web page. You just need to make a few modifications to the code to read your sensor.
Here’s the project:
I hope this helps 🙂
Hi,
This work perfectly, thank you.
I just wants to connect more sensors with the esp32 , for example the DHT11 for temperature and humidity and the PIR motion sensor. It is possible to do it ? and witch pins use to collect data ?
Hi.
Yes, you can do that.
You can use the PIR motion sensor with any GPIO of the ESP32 (except GPIOs 6, 7, 8, 9, 10, and 11).
I hope this helps.
Regards
sara 🙂
It is asking for inside the DHT.h
I cannot find this EXACT library. Please advise.
Thank you
Hi Ray.
You can find the DHT library here:
Found it, got it to compile. However, on runtime, upon going to browser, it displays the page, but it always goes to “FAILED” on ALL readings. I’m trying to find out why it can’t read the DHT11 🙁
It may be a power issue. If you’re supplying 3.3V to the DTH11, try to supply 5V instead, and see if that fixes your issue.
Also, make sure you have everything wired correctly with the 4.7K Ohm pull-up resistor.
Also test powering up the sensor using an external power suppky.
I hope this helps debugging your problem.
Regards,
Sara 🙂
Hi, Suppose I have my sensors outside the house, how to supply power to the ESP8266 module without the USB port? Can batteries do this job?
Hi.
Yes, you can supply power to the ESP8266 using batteries.
You may also want to add deep sleep to your code to save power.
Regards,
Sara 🙂
hello
I loaded this on a esp32, and it works… for a while, then the browser can`t access the web server until I restart the ESP32 again. Any thoughts ?
Sorry, but it does not work here. I always receive “Failed to read from DHT sensor!”. The same circuit works with another library without any failure ()
Any Idea?
Hi Martin.
Sometimes the DHT sensor fails to read the temperature, but I have no idea why that happens?
But do you get good results without failures with that new library?
We need to try that library as see what we get.
Regards,
Sara 🙂
It looks like you now also need the Adafruit_Sensor-master repo to get the DHT repo to work right. It can be found here:.
Without it I was getting a “multiple libraries were found for “wifi.h”” error that wouldn’t allow for a compile.
Or did I miss a step when following the instructions?
Hi Michael.
You are right. You also need to install the adafruit sensor library.
I’ve updated the post to include that.
Thanks for telling us.
Regards,
Sara 🙂
Hi, just wondering how can I save the reading data from the website to a text file for every 30 seconds? I added a line of client.println(“Content-Disposition: attachment; filename=\”test.txt\””); under the line of client.println(“Content-Type: text/html”); , but the webpage stop refreshing after the first text file has been downloaded. Please advise.
Hi Michael.
I’ve never used that method to save data.
We have other examples to save data to a google sheet, to a file on a microSD card, or thingspeak for example:
Take a look at the following tutorials that may help:
– Publish Sensor Readings to Google Sheets (ESP8266 Compatible):
– ESP8266 Daily Task – Publish Temperature Readings to ThingSpeak:
– ESP32 Data Logging Temperature to MicroSD Card:
Regards,
Sara 🙂
Will this work on an ESP8266? I am trying to find what to change to make this work on an 8266 and I cant find anything in the code. I tried just setting the board (under tools->board) to my model ESP8266 (adafruit feather Huzzah esp8266) but it wont take the code.
Hi Josh.
We have a similar tutorial with the ESP8266:
The ESP8266 uses the ESP8266WiFi.h library instead of the WiFi.h library.
I hope this helps.
Regards,
Sara 🙂
Hi,
I adapted this code to send a key status to the browser, but the 30 seconds of auto refresh is to long… Is there a way to send (and show on browser) the data right after the key is pressed or not?
Hi Marcio.
You can change the line 136 on the code:
Regards,
Sara
Sara, it is always the dht11 or dht22 sensors which are limited to low temperatures. Why not do a project that uses the MAX31865 RTD and MAX31855 Thermocouple amplifier boards with the ESP8266. These devices offer outstanding accuracy over an extremely long range of temperatures.
Hi Carl.
Thank you for the suggestion.
We have to try those in the future and maybe come up with a tutorial.
Regards,
Sara
Hi all, good post, but i’ve suggestion.
this app contains 2 async separated calls, for temperature and humidity: why not a single xhr call to read these values then returning an object (ie {“T”:”25,32″, “H”:”76.54″})?
regards
Finally a new project that isn’t MicroPython, thanks, very cool.
Question: is it possible to insert the icons local, that is on my local network that doesn’t have internet access? (Where I live way out in the country internet is VERY expensive so have to go into town and use library internet about once a week)
Can they be downloaded elsewhere and stored on the esp32?
Thanks for all the great tutorials.
dp
Hi David.
Thank you for your interest in our projects.
You can store the icon files on the ESP32 SPIFFS, but they need to occupy very little space. Or you can use a microSD card to store the icons.
However, I don’t have any tutorials that show how to load images from a microSD card or from SPIFFs.
Nonetheless, we have a few tutorials about these subjects that may help you throughout your project:
Alternatively, you can remove the icons from the web server page.
I hope this helps.
Regards,
Sara
David,
you need a css framework (fontawesome) containing images used as background: see the link tag; i you have a pc, you can arrange it as another web server on your lan, download the the framework and copy it on your pc, and the serve it from the pc.
say you pc/server is at 192,168.1.255 ip, change
<link rel="stylesheet" href="…
to
<link rel="stylesheet" href="192.168.1.255/styels/use.fontawesome.com/
enjoy
href=”192.168.1.255/styles/use.fontawesome.com/…
sotty for typo
Bonjour. impossible de compiler le programme. il m’indique:’tcp’p_api_call_data’ incomplete .exit status 1.à vous lire salutations; toutes les librairies sont bien instalés le “SSID” et le “Passeword ” sont OK ,alors?
Hi. when compiling I get this error:
int begin(char* ssid, const char *passphrase);
^
exit status 1
invalid conversion from ‘const char*’ to ‘char*’ [-fpermissive]
Do you know what may be happening?
Thank you
Hi all,
I can replace ESP 32 to ESP 8266 in this project?
Thank all!
Hi.
This project only works with the ESP32.
But you can change the code to include the right libraries to make it compatible with the ESP8266. However, we don’t have that documented at the moment.
You need to use the ESP8266WiFi library instead of the WiFi library, and the ESPAsyncTCP instead of the Async TCP.
Regards,
Sara
Hii
I keep getting this error when I compile.
“invalid conversion from ‘const char*’ to ‘char*’ [-fpermissive]”
Can you give me any suggestions on how to solve this problem?
Thanks,
Udayana
Hi.
You probably don’t have all the necessary libraries installed.
Please check that you have all these libraries installed:
Regards,
Sara
Hi, Sara and Rui,
I have realized this project with the ESP32 and everything was perfect.
I’wanted to try with an ESP8266-01 (1Mb) and I changed the library WiFi.h with the ESP8266WiFi.h
I have also in my library folder the ESPAsyncTCP.h, but is not used directly in the sketch.
But can’t read the sensor (DFH11 or DHT22) with the error “Failed to read from DHT sensor!”.
I have insert the display of the Temp and Humidity at the beginning of the Setup, and run perfectly.
I have also changed the interval time in the Javascript from 10.000 to 20.000.
The sensor is powered with 5v with a pullup resistor of 4.7k.
Also a display after the “dht.readTemperature()” shows “nan” (Not a Number) and a “delay(200)” or more added after the “dht.readTemperature()d” can’t run better.
It’s not so important, but I’d like to understand why.
One time out of twenty I have a correct display of the Temp and Humidity at the beginnig after a “reset”.
I have put a counter instead the Temp value and run correctly also on the Smartphone.
Thanks for any suggestion. I start also to try yours intersting projects with MicroPython.
This project is the best project with ESP and C++ I ever found. Really!!
Roberto
Graz
Austria
Hi Roberto.
Thank you so much for your nice words.
When you get “Failed to read from DHT sensor!” it usually means that the sensor is not wired properly, or it is not being powered properly. If you’re powering your ESP32 using an USB cable directly through your computer USB port, that can be the problem, and that’s why you sometimes get readings and others don’t.
Try power it from a USB hub powered by an external source. Also helps to replace the USB cable with a stronger and/or shorter one.
If you get a “nan” error, it usually means that the sensor didn’t have enough time to get the readings. If you’re using a DHT22, you can only request readings every 2 seconds.
I hope this helps.
Regards,
Sara
I have installed the AsyncTCP-master library but I get the error:
C:\Users\marti\Documents\Arduino\sketchbook\libraries\ESPAsyncWebServer-master\src/ESPAsyncWebServer.h:36:25: fatal error: ESPAsyncTCP.h: No such file or directory
#include
^
I see the file AsyncTCP.h in the library but not ESPAsyncTCP.h Is there another library I need that wasn’t in the article?
Hi Martin.
For the ESP32, you just need these two libraries for the asynchronous web server:
Regards,
Sara
Hi Sara,
following the detailed (thank you!) instructions in “installing ESP32 board …” second part “deleting the espressif folder ..” to get the latest version of it I successful could test the WiFiscan example.
Great!
However:
I failed to compile the “ESP32 DHT11/DHT21 webserver temp & hum …” example from your Web site. Got a whole bunch of error messages, first read as:
… AsyncTCP-952b7eb6ba62f7071f7da2a274d36e06b97de572\src\AsyncTCP.cpp:259:27: error: field ‘call’ has incomplete type ‘tcpip_api_call’
etc. etc.
Searching the WEB I found a lot of comments concerning this issue totaly unintelligble to me :-((
Please give advice how to fix the problen or public a new – compatible – version of the Web-server sketch.
Thank you
Hi.
I’ve never faced that error.
But here: they say to remove the old ESP32-arduino core and install the new one.
If that doesn’t work, I don’t know what you can do to solve the problem.
Sorry that I can’t help much.
Regards,
Sara
Sara, thank you for the very fast response!
As I’m using the latest arduino IDE and the latest espressive file may I ask you for a very short cross check before I have to dig deeper into installation details?
I’m using IDE 1.8.9 and have – according to the instruction – deleted the espressive file in my sketchbook folder to compile the sourcecode from your Web-side.
Thank you so much for your support
Unfortunately, I don’t know what to do from here as I’ve never experienced that problem.
You can also check if you have more than one Arduino IDE installed in your computer that may be conflicting…
Thank you Sara for your hint. I gone thru all the steps in as adviced. Afterwards I could sucessfull compile and run the ESP32 examples WiFiScan, getChipID, LEDCSoftwarefade etc. but still not the async. Webserver sketch.
Did you suceed in compiling the code?
Maybe there is someone out here who can try to compile – no hardware neede for the test just compile.
Thank you for support!
I’ve tested the code and it compiled fine. So, the code is ok.
But I’m using Arduino IDE 1.8.7.
Sara
it looks like an incompabilty of older an newer async tcpip libs:
field ‘call’ has incomplete type ‘tcpip_api_call’ struct tcpip_api_call call;
If you ever update arduino IDE et all pls try to compil again and report the result here. It’s allways a mess with incompatible updates I know.
Thank you for your support!
Thanks so much for this tutorial, it’s fantastic and just what I was looking for.
Do you know if there a way to display a graph of the data on the website so you can view the changes in temp & humidity over the course of a day, instead of just at a point in time??
Thanks for reading and I’m glad you found it helpful! Unfortunately at the moment I don’t have any tutorials on that exact subject…
There is a compilation error that seems to be linked to one of the libraries:
C:\Users\ik\Documents\Arduino\libraries\ESPAsyncWebServer\src\WebAuthentication.cpp:23:17: fatal error: md5.h: No such file or directory
I have the ESPAsyncWebserver installed and there is a WebAuthentication.cpp file.
What could be wrong here
ah I see, md5.h missing. Installed that.. led to hoist of other problems with the adafruit sensor library
Hey thanks for the awesome guide!
I know this is a bit old but there were a few issues.
The link to the AsyncTCP might be incorrect or outdated, I googled and found the files from a different source.
There was a problem with multiple wifi.h locations, but this seems like a general esp32 issue.
Also the credentials for wifi showed errors, I removed the ‘const’ and it worked.
I have one question, the website randomly gets assigned to an local ip address.
Is there any way to keep the ip static so it will be the same?
Thank you for the guide
Hi.
Thank you for pointing out those issues.
I’ve fixed the URLs for the libraries.
Multiple wifi.h libraries is a problem with your libraries directory and how your installation files are organized.
The code worked fine for me as it is. I have to see what is going on.
To assign a static IP address, you can follow this tutorial:
Regards,
Sara
hi there I’m getting this error how do I fix it
Serial port /dev/cu.Bluetooth-Incoming-Port
Connecting…….._____….._____….._____….._____….._____….._____…..____An error occurred while uploading the sketch
_
A fatal error occurred: Failed to connect to ESP32: Timed out waiting for packet header
Hi Matthew.
Just follow this tutorial to solve the issue:
Regards,
Sara
Hi there,
thanks for the fine tutorial.
It’s working, but I’ve little issues:
I set the interval to 12000, both t and h.
Now the h updates every 8 seconds, the t only when I reload the page.
Not so comfortable and I don’t understand – both parts have similar code.
So I thought, it would be nice to display the date and time of the readings too. Unfortunately I don’t know how to integrate this function.
Another question for ESP32: If I interrupt the voltage and connect again, starts it working by itself or must I press the boot- and/or the en-button?
Thanks for your applied time
Siegfried
Update:
After changing interval time and uploading, it doesn’t connect to wifi anymore.
Everything else is the same.
Maybe I should have added this:
18:47:16.866 -> rst:0x1 (POWERON_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT)
18:47:16.866 -> configsip: 0, SPIWP:0xee
18:47:16.866 -> clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
18:47:16.912 -> mode:DIO, clock div:1
18:47:16.912 -> load:0x3fff0018,len:4
18:47:16.912 -> load:0x3fff001c,len:928
18:47:16.912 -> ho 0 tail 12 room 4
18:47:16.912 -> load:0x40078000,len:8424
18:47:16.912 -> ho 0 tail 12 room 4
18:47:16.912 -> load:0x40080400,len:5868
18:47:16.912 -> entry 0x4008069c
18:47:18.316 -> Connecting to WiFi..
18:47:19.299 -> Connecting to WiFi..
18:47:20.282 -> Connecting to WiFi..
18:47:21.313 -> Connecting to WiFi..
18:47:22.295 -> Connecting to WiFi.. and so on
I tried again and again.
Hi.
Did you introduced your network credentials?
It is common to get that error when we upload the code without the credentials. It continuously tries to connect to wifi.
Regards,
Sara
Hi Sara,
I did! It worked before. The code is the same, I only changed interval time. The problem is still after I undid the change.
I tried the ‘const char* ssid = “W…’ lines without ‘const’, wich hint I found somewhere – no other result.
Regards,
Siegfried
I’m sorry to be flashy but I can’t solve the problem.
I made a new sketch, copying your code, only changing the network credentials, did the upload. It started the same way as before: continuously tries to connect to wifi.
I can’t understand because it worked about ten days ago for about one weak. It’s the same computer and other hardware (Win7, DOIT ESP32 Devkit V1, Router WLAN adjustment).
Would you have any suggestions?
Regards,
Siegfried
Hi.
I’m sorry, but I don’t know what can be wrong.
If you are inserting your network credentials properly, it is very difficult to figure out what is wrong.
Can you try any other wi-fi sketch example and see if the ESP32 connects to wi-fi?
In the Arduino IDE, you can select a file from the examples folder.
Regards,
Sara
Hi,
as the ESP32 to google sheets project works fine since may, 19. on the same wifi, I finally
changed the DHT22 to another one – no better result
put it to 5V as I read somewhere – no better result
put it back to 3.3V – miraculously it w o r k s !!!
I don’t understand but obviously sometimes you have to try some crazy things with the computer stuff.
How could I display date and time on the web page too?
Is there any workshop for learning how to get access to my WiFi.localIP from www? After reading the whole google (nearly), all I tried till now failed.
Thanks so much for helping,
Siegfried
Hi.
There are many different ways to display date and time. For example:
– use an RTC to keep track of time and save it in a variable that you publish in your web server.
– get time from NTP server:
– you can also display the last time the page/readings were updated using a javascript function in your web server code as we do in this example:
(I’m not sure if this example is suitable for what you are looking for, it depends on your project).
– you can also search in this topic: esp32.com/viewtopic.php?t=6043
To make your local web server accessible from anywhere in the world you need to port forward your router (this may expose your network to security problems).
You can also use ngrok services.
I hope this helps.
Regards,
Sara
Hi Sara,
I’m very pleased to get your fast and detailed answer.
I’ll trie all this, when it won’t be so hot as these days!
Regards,
Siegfried
I’m going to connect two ultrasonic sensors with ESP module is there anything to change or can I just replacer the coding part which is related to the ultrasonic sensor part?
help me
Hi.
It depends on your application. But if you just add the ultrasonic sensor part, it should work.
Regards,
Sara
Hi.
Dont show nothing in my monitor serial, please help me.
Hi Igor.
After uploading the code, open the Serial Monitor and make sure you have the right baud rate selected. In this case 115200.
Then, press the ESP32 on-board RESET button. It should display the IP address.
I hope this helps.
Regards,
Sara
hi..i am facing an error after uploading the code.i open the serial monitor and there was an error message…BROWNOUT DETECTOR WAS TRIGGERED.pls help me to solve this
Hi.
Please read our troubleshooting guide, bullet 8:
I hope this helps,
Regards,
Sara
hi
this is the first time i play arround with something other than a rasperry pi,
i have a problem with the ESPAsyncWebServer library.
when i try to upload the code i get a error at compiliation. it can not find the ESPAsyncTCP.h file.
Screenshot
i think i copy and rename the librarys right..?
also i just uploaded a onboard LED blink sketch that works…?
hope you can help me noob
Hi Christian.
How did you install the libraries?
You need to install two libraries:
–
–
One of the easiest ways to install the library is downloading the zip file.
Then, in your Arduino IDE, go to Sketch, Include Library, Add .zip library and select the zip files you’ve just downloaded.
I hope this helps,
Regards,
Sara
hi,
I’ve a Question. How to can I combine the DHT values on webpage with an Input-Output module?
Is there any possibility?
Kind regards
Jan
Hi Jan.
You can try to combine the code in this tutorial with this one:
It uses the asyncwebserver library too. Take a look at the “Arduino Sketch” section.
Regards,
Sara
Hi Sara.
Great! Thank you! This is exactly what I was looking for.
Regards
Jan
Can I make it work as an access point and send the sensor data through it? so as not to have the need to be connected to a local network.
Hi Jose.
Yes, you can set it as an access point.
You can follow this tutorial:
Regards,
Sara
Hello everybody,
as I never knew if my server displayed actual data,
I worked out how to display the date and time of sensor readings in european format,
– in case of someone is interested, I did it that way:
First the libraries:
#include
#include
then:
// Define NTP Client to get time
WiFiUDP ntpUDP;
NTPClient timeClient(ntpUDP);
// Variables to save date and time
String formattedDate;
String dayStamp;
String timeStamp;
String newdate;
String zeit;
int splitT;
then added the readZeit function (Zeit=time):
String readZeit() {
while(!timeClient.update()) {
timeClient.forceUpdate();
}
formattedDate = timeClient.getFormattedDate();
splitT = formattedDate.indexOf(“T”);
timeStamp = formattedDate.substring(splitT+1, formattedDate.length()-1);
dayStamp = formattedDate.substring(0, splitT);
String year = dayStamp.substring(0,4);
String month = dayStamp.substring(5,7);
String day = dayStamp.substring(8);
// format for european standard
zeit = day + “.” + month + “.” +year+” “+timeStamp;
Serial.println(“Variable zeit”);
Serial.println(zeit);
return String(zeit);
}
changed the styles a little:
html {
font-family: Arial;
display: inline-block;
margin: 0px auto;
text-align: center;
}
h2 { font-size: 3.0rem; }
p { font-size: 2.5rem;
.units { font-size: 1.5rem; }
.dht-labels{
font-size: 1.5rem;
vertical-align:middle;
}
A new icon for the time:
Zeit
%zeit%
Setinterval for time (I choosed 5 sec for all intervals):
setInterval(function ( ) {
var xhttp = new XMLHttpRequest();
xhttp.onreadystatechange = function() {
if (this.readyState == 4 && this.status == 200) {
document.getElementById(“zeit”).innerHTML = this.responseText;
}
};
xhttp.open(“GET”, “/zeit”, true);
xhttp.send();
}, 5000 ) ;
behind WiFiConnected I added:
// Initialize a NTPClient to get time, offset 3600 for +1 hour
timeClient.begin();
timeClient.setTimeOffset(3600);
while(!timeClient.update()) {
timeClient.forceUpdate();
}
and at last:
server.on(“/zeit”, HTTP_GET, [](AsyncWebServerRequest *request){
request->send_P(200, “text/plain”, readZeit().c_str());
});
Concerning my posts from May, everything works much better since I have a windows10 laptop!
Kind regards,
Siegfried | https://randomnerdtutorials.com/esp32-dht11-dht22-temperature-humidity-web-server-arduino-ide/?replytocom=367459 | CC-MAIN-2019-51 | refinedweb | 6,774 | 66.54 |
I'm trying to purge my WordPress content from "false" carriage return (CR). These are caused after a migration of my content, that now presents from time to time a code that makes the web rendering engine to "paint" a CR where I would like to be nothing. The paragraphs seem to have a double CR because of this, and look too far apart.
I'd like to be able to make a MySQL query in order to get rid of that strings, but at the moment I haven't found the key. What I've tried is
UPDATE wp_posts set post_content = replace (post_content,' ',' ');
But i get
<p> </p>
where before were the strings. This seems not the answer at all. Could it have to be with the ampersand, and in that case, should I use something like or something similar?
I am not sure why would be interpreted as a carriage return, because it should look like a space (Non-Breaking SPace). In any case, I was able to get your SQL to work on a test database I created by changing the string to have double-quotes.
UPDATE wp_posts set post_content = replace (post_content," ","");
Also note, there is no space between the last pair. You want it to replace with nothing, not a space.
mysql> UPDATE wp_posts set post_content = replace (post_content," ",""); Query OK, 7 rows affected (1.34 sec) Rows matched: 15232 Changed: 7 Warnings: 0
The code <p> </p> is frequently used by some GUI HTML editors to represent a carriage return. I would be that what you really need to be searching for and removing is <p> </p>, and not just which is just a non-breaking space character.
<p> </p>
mysql> UPDATE wp_posts set post_content = replace (post_content,"<p> </p>",""); Query OK, 0 rows affected (1.26 sec) Rows matched: 15232 Changed: 0 Warnings: 0
By posting your answer, you agree to the privacy policy and terms of service.
asked
4 years ago
viewed
3595 times
active | http://serverfault.com/questions/246761/mysql-replacing-nbsp-with-nothing-delete-those-please | CC-MAIN-2015-22 | refinedweb | 332 | 77.67 |
Guangrong, To be more precise, the current approach is that there are two levels of “instances”. The first is, using Kubernetes terms, “service” level. The second is “pod” level, where a pod is a container or group of closely coupled containers.
Advertising
Using VES collector as example, the R1 VES container becomes a pod. Then another layer of abstraction is added for addressing platform maturity requirements, to wrap individual pod into service (load balancer type). External world interacts with VES at the service level. Scaling and resilience within a VES service (individual VES pod gets restarted, scaled to more replica instances, etc) are handled by Kubernetes, transparent to outside world, and no DCAE control involvement. If additional VES “service” needs to be deployed, that will involve DCAE control. Instances of service can be identified individually. With this said, there is no current plan to explicitly notify individual service instances about other instances of the same service. Each service instance can probably look into Consul or Kubernetes mechanisms to see if there are other service instances of the same type as itself. Is this something you need for R2 Holmes? Lusheng From: "fu.guangr...@zte.com.cn" <fu.guangr...@zte.com.cn> Date: Thursday, February 8, 2018 at 9:33 PM To: "fu.guangr...@zte.com.cn" <fu.guangr...@zte.com.cn> Cc: "JI, LUSHENG (LUSHENG)" <l...@research.att.com>, "roger.maitl...@amdocs.com" <roger.maitl...@amdocs.com>, "onap-discuss@lists.onap.org" <onap-discuss@lists.onap.org>, "tang.pe...@zte.com.cn" <tang.pe...@zte.com.cn> Subject: 答复: Re: [onap-discuss] [dcae][dcaegen2][holmes] A Question on AutoScaling of DCAE Microservices Lusheng, One more question on this: will the existing microservice instances be notified by DCAE once a new instance is spun up or a redundant instance is destroyed? Say, there's already an instance of Holmes (e.g. Holmes A) and then a new instance (Holmes B) of Holmes is instantiated. Will DCAE tell Holmes A that there's another instance of Holmes named "Holmes B" spun up just now? Guangrong 原始邮件 发件人：付光荣10144542 收件人： <l...@research.att.com>; 抄送人： <roger.maitl...@amdocs.com>; <onap-discuss@lists.onap.org>;唐鹏10114589; 日 期 ：2018年02月05日 09:56 主 题 ：答复: Re: [onap-discuss] [dcae][dcaegen2][holmes] A Question on AutoScaling of DCAE Microservices Thanks Roger and Lusheng for your kind feedbacks. I'll look through the links that Roger pointed. Since Holmes does not have to do anything on metric collecting, I think the top priority for our team is to handle the state of our containers properly after auto scaling. Lusheng, As you know, we have a virtual F2F event this week. So I'm not sure whether we have a chance to discuss this. Please do let me know when you are ready to share. Thank you very much. Regards, Guangrong 发件人： <l...@research.att.com>; 收件人： <roger.maitl...@amdocs.com>;付光荣10144542; 抄送人： <onap-discuss@lists.onap.org>;唐鹏10114589; 日 期 ：2018年02月03日 05:13 主 题 ：Re: [onap-discuss] [dcae][dcaegen2][holmes] A Question on AutoScailing of DCAE Microservices Roger, Thanks for the pointers. Guangrong, The details of the Kubernetes plan for DCAE is still work-in-progress. Here are some of highlights for service components. 1. Will support Kubernetes based scaling and resilience mechanism for dockerized service components. * This implies that the container/containers of a service component will be packaged as pod. The resilience is expected to be provided by Kubernetes cluster. * The Kubernetes based scaling support may need additional support from service component developer. For example, if a service component is stateless, for which each instance behaves exactly the same as the next, are “scaling-ready”. Load can be distributed to any instance and the result would be the same. However, if the service component keeps states, multiple replicas of this service component may have different local states if not handled carefully. The actual mechanism to ensure state synchronization is application dependent. But one typical approach is to push “states” to an external service such as a DB, or persistent volume, or a distributed kv store, etc, and states are loaded into individual replica when needed (e.g. startup) so different replicas get the state view from the same copy. * In terms of multiple replicas subscribing to the same message router topic, there is a way to distribute the load. That is, each replica uses the same “groupid” but different “userid”. Message router will consider a message received by a group when it is received by any user of the group. This way we can avoid the same message being delivered to multiple replicas. 1. Our goal is to maintain the interfaces that how a service component interacts with the rest of DCAE the same, e.g. how your component gets deployed and how your component receives configuration updates, etc. 2. How the scaling trigger arrives and the actual scaling (i.e. more replicas) is handled by external mechanisms. Service components themselves do not need to worry about that. We hope to have more details to share the next week, and set up focus meeting discussing more. Thanks, Lusheng From: Roger Maitland <roger.maitl...@amdocs.com> Date: Friday, February 2, 2018 at 1:48 PM To: "fu.guangr...@zte.com.cn" <fu.guangr...@zte.com.cn>, "JI, LUSHENG (LUSHENG)" <l...@research.att.com> Cc: "onap-discuss@lists.onap.org" <onap-discuss@lists.onap.org>, "tang.pe...@zte.com.cn" <tang.pe...@zte.com.cn> Subject: RE: [onap-discuss] [dcae][dcaegen2][holmes] A Question on Auto Scailing of DCAE Microservices Guangrong, I don’t have an answer in the context of the DCAE controller but OOM/Kubernetes has facilities to help build a Holmes cluster in the containerized version of DCAE (which is being worked on). The cluster can be static (which is believe is what most projects intend for Beijing) or dynamic (the OOM team would love to work with you on this). Here are some links I hope you find useful: * OOM Scaling:<> * K8s auto-scaling:<> Here is a sample of how an auto-scaler is configured: apiVersion: autoscaling/v2beta1 kind: HorizontalPodAutoscaler metadata: name: php-apache namespace: default spec: scaleTargetRef: apiVersion: apps/v1beta1 kind: Deployment name: php-apache minReplicas: 1 maxReplicas: 10 metrics: - type: Resource resource: name: cpu targetAverageUtilization: 50 status: observedGeneration: 1 lastScaleTime: <some-time> currentReplicas: 1 desiredReplicas: 1 currentMetrics: - type: Resource resource: name: cpu currentAverageUtilization: 0 currentAverageValue: 0 The OOM team would be happy to work with you on implementing this. Cheers, Roger From: onap-discuss-boun...@lists.onap.org [mailto:onap-discuss-boun...@lists.onap.org] On Behalf Of fu.guangr...@zte.com.cn Sent: Friday, February 2, 2018 2:58 AM To: l...@research.att.com Cc: onap-discuss@lists.onap.org; tang.pe...@zte.com.cn Subject: [onap-discuss] [dcae][dcaegen2][holmes] A Question on Auto Scailing of DCAE Microservices Lusheng, The Holmes team is currently working on the auto scaling plans for Holmes. We need to confirm something with you. To my understanding, the microservice should only focus on how to maintain and balance its data flow rather than how the docker containers/vms are scaled by their controller. As a DCAE application, I think it's DCAE controller's responsiblity to determine when and how to scale in or scale out Holmes instances. Is that correct? If what my understanding is correct, does DCAE have any specific requirements regarding collecting the status and metrics of its microservices? Regards, Guangrong This message and the information contained herein is proprietary and confidential and subject to the Amdocs policy statement, you may review at<>
_______________________________________________ onap-discuss mailing list onap-discuss@lists.onap.org | https://www.mail-archive.com/onap-discuss@lists.onap.org/msg07465.html | CC-MAIN-2018-09 | refinedweb | 1,274 | 55.95 |
After watching Andrei Alexandrescu’s talk on Going Native 2013, I wanted to take a crack at it myself. The presentation covers how to expand a tuple into individual arguments in a function call. Being a Python programmer I’m a little spoiled by
func(*args) so the ability to do this in C++11 is something I’m eager to use. What I came up with wound up being quite similar, but more flexible. I wanted to make it more generic, to work with
std::pair and
std::array. The version presented in that video is incredibly powerful, but it can go a bit further.
The limitations start at the top level, the
explode free function.
template <class F, class... Ts> auto explode(F&& f, const tuple<Ts...>& t) -> typename result_of<F(Ts...)>::type { return Expander<sizeof...(Ts), typename result_of<F(Ts...)>::type, F, const tuple<Ts...>&>::expand(f, t); }
The
tuple& argument allows a means to use
result_of to figure out the return type, and
sizeof... to determine the size of the
tuple itself. This can be accomplished via other means.
decltype can be used to figure out the return type. It needs more typing, but removes the need for
result_of. As for
sizeof..., there is a
std::tuple_size available which can reach the same end. Using this makes
explode non-variadic. Taking a universal reference, rather than capturing the parameter pack, means different versions for lvalue and rvalue refs aren’t needed.
My initial function (called
expand instead) is:)); }
Some things to note:
std::tuple_sizeworks on
std::pair(yielding 2) and on
std::array(yielding the size of the array).
std::getalso supports
std::pairand
std::array, meaning that now
tuple,
pair, and
arraycan all work in this context.
std::remove_referenceis needed for calling
std::tuple_sizebecause
tupis a universal reference, and
Tupmay deduce to an lvalue reference type
The
decltype goes through each level of the expansion, until much like the original, it hits a base case and does the call.
#include <cstddef> #include <tuple> #include <utility> #include <type_traits> #include <array> template <std::size_t Index, typename Functor, typename Tup> struct Expander { template <typename... Ts> static auto call(Functor&& f, Tup&& tup, Ts&&... args) -> decltype(Expander<Index-1, Functor, Tup>::call( std::forward<Functor>(f), std::forward<Tup>(tup), std::get<Index-1>(tup), std::forward<Ts>(args)...)) { return Expander<Index-1, Functor, Tup>::call( std::forward<Functor>(f), std::forward<Tup>(tup), std::get<Index-1>(tup), std::forward<Ts>(args)...); } }; template <typename Functor, typename Tup> struct Expander<0, Functor, Tup> { template <typename... Ts> static auto call(Functor&& f, Tup&&, Ts&&... args) -> decltype(f(std::forward<Ts>(args)...)) { static_assert( std::tuple_size< typename std::remove_reference<Tup>::type>::value == sizeof...(Ts), "tuple has not been fully expanded"); // actually call the function return f(std::forward<Ts>(args)...); } };)); }
A few examples showing the flexibility.
int f(int, double, char); int g(const char *, int); int h(int, int, int); int main() { expand(f, std::make_tuple(2, 2.0, '2')); // works with pairs auto p = std::make_pair("hey", 1); expand(g, p); // works with std::arrays std::array<int, 3> arr = {{1,2,3}}; expand(h, arr); }
Each level of the call takes one argument at a time off the back of the
tuple using
std::get and the template Index parameter, decrements the index, and recurses. This is a bit hard to imagine, so I’ll illustrate. This sequence is not meant to be taken too literally.
Let’s say I have a
tuple of
string,
int,
char, and
double. I’ll denote this example tuple as
tuple("hello", 3, 'c', 2.0). The expansion would happen something like the following
expand(f, tuple("hello", 3, 'c', 2.0)) -> call<4>(f, tuple("hello", 3, 'c', 2.0)) -> call<3>(f, tuple("hello", 3, 'c', 2.0), 2.0) -> call<2>(f, tuple("hello", 3, 'c', 2.0), 'c', 2.0) -> call<1>(f, tuple("hello", 3, 'c', 2.0), 3, 'c', 2.0) -> call<0>(f, tuple("hello", 3, 'c', 2.0), "hello", 3, 'c', 2.0) -> f("hello", 3, 'c', 2.0)
Of course
std::integer_sequence in C++14 turns all of this on its head. Maybe I should’ve implemented that instead… | https://makecleanandmake.com/2014/06/ | CC-MAIN-2022-05 | refinedweb | 705 | 51.95 |
[
]
Mark Hindess resolved HARMONY-4735.
-----------------------------------
Resolution: Fixed
This works for me now. Please re-open if it is still failing for you.
> [classlib][luni] Needs to enhance encoding setting to print Chinese in Linux
> ----------------------------------------------------------------------------
>
> Key: HARMONY-4735
> URL:
> Project: Harmony
> Issue Type: Bug
> Components: Classlib
> Environment: Linux32 [Fedora]
> Reporter: Chunrong Lai
> Priority: Minor
> Attachments: H4735.println.workaround.patch
>
>
> Below simple example shows the bug.
> public class HChinese {
> public static void main(String argv[]) {
> char []str = {0x4f60, 0x597d, 0x002c, 0x4e16, 0x754c}; // "Hello, World" in Chinese.GB2312
> System.out.println(new String(str));
> System.out.println("你好世界"); // "Hello, World" in Chinese
> }
> }
> To reproduce the bug, start Xwindow with "LC_ALL" and "LANG" set to "zh_CN.GB2312" (You
can quit Xwindow firstly with /sbin/init 3, export LC_ALL=zh_CN.GB2312, export LANG=zh_CN.GB2312,
then run startx) and run HChinese.
> Without the attached patch. Harmony just can not print Chinese. It will work with the
patch.
> However the patch more likes a workaround because it needs hardly code the "GB2312".
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online. | http://mail-archives.apache.org/mod_mbox/harmony-commits/201003.mbox/%3C585051499.434481269352887453.JavaMail.jira@brutus.apache.org%3E | CC-MAIN-2014-23 | refinedweb | 187 | 61.63 |
The rearranger plugin rearranges (reorders) class definitions within a Java file according to rules specified by the user.
Version 0.6 adds several new capabilities:
- Ability to select items by matching name to a regular expression pattern.
- Ability to insert comments between groups of items.
- Added support for native, synchronized, transient and volatile attributes.
- Added ability to detect if a field is initialized to an instance of an anonymous class.
- Added ability to detect class static initializers.
Comments can be emitted conditionally, based on whether or not any items matched the preceding and/or subsequent rules. This can prevent "spurious" comments from appearing.
Before the file is rearranged, any comments matching those you specify in the rules are removed, on the theory that they were generated by a previous Rearranger execution. Be careful what you specify for comments! You are responsible for proper formation of comments; e.g., use // or matching /* */ entries. Comments may be multiline.
Several bugs were fixed:
- Configuration dialog now correctly determines if settings are unchanged
- methods with "getter/setter" and "other" selected were not being selected (the conditions are now OR'd, not AND'ed)
In the process of cleaning things up, some configuration items were renamed (mostly in method attributes); please check your configurations to be sure they are still correct.
The plugin requires EAP build 944 or later. It is available from the IDEA Plugin Manager.
See.
Please let me know if you have any problems. (I changed a lot of code! :)
-Dave
Attachment(s):
RearrangerConfig4.JPG
Not a trivial request, but here is what I'd like to see in the rearranger :
Reformat : reorder extracted methods to match the logical flow
Example :
Before : code to reformat
-
public void outer (){
foo1 ();
foo2 ();
foo3 ();
}
private void foo3 (){ ...}
private void foo1 (){ ...}
private void foo2 (){ ...}
After : reformating action (Requested)
-
public void outer (){
foo1 ();
foo2 ();
foo3 ();
}
private void foo1 (){ ...}
private void foo2 (){ ...}
private void foo3 (){ ...}
note : the same idea could be implemented as
- intention ("Move method to match logical flow")
- code inspection
Alain
Alain,
I'm willing to consider it if you're willing to help me figure out the rules and the exceptions! :)
Is the rule thus: when a private method is referenced by one other method (the 'parent method'), it can be considered an "extracted" method. Extracted methods can be reordered (1) based on order they appear in the parent method, with respect to other extracted methods; and/or (2) moved to a position immediately after the parent method.
So there could be two checkboxes on the 'Method' panel.
1) Move extracted methods (private methods called only by the parent method) immediately below the parent method, wherever it may finally be placed. If this is not checked, private methods are left where they are.
2) Sort extracted methods by the order in which they are called (or in the order in which they appear to be called) by the parent method. If this is not checked but #1 is, then the extracted methods are simply moved in their existing order underneath the parent method.
This would (naturally!) have to happen recursively, wouldn't it? E.g.
Would we ignore private methods called by more than one parent method, or called twice by the same method?
-Dave
Dave,
>Alain,
>I'm willing to consider it if you're willing to help me figure out the rules and the exceptions! :)
>
Fair enough.
Your 2 options sounds good, but I think it's not enough .
option 3 : Depth-first vs breadth-first
-
Looking at your example, where an extracted method call 2 further
extracted methods,
, I can see 2 ways to reorder the methods :
Depth-first
, or Breadth-first
option 4 : move after the last usage.
-
> Would we ignore private methods called by more than one parent
> method, or called twice by the same method?
If people checked the "sort by call-order" option, considering only the
1st call make the most sense to me today, but I may change my mind, with
usage. Otherwise, you could add this - 4th - option :
Automatic mode:
-
You could also add an option to automatically sort classes when you open
them, depending on some rules :
- only the 1st time you open it
(you'd have to store the list of processed classes in the project file)
- age of last modif
- size (nof lines, or nof methods)
- location (regexp on the package and class names)
Alain
Hi,
This is a great plugin, extremely useful for cleaning up other peoples
code:-)
On 2003/10/29 04:32, Dave Kriewall wrote:
This setting does not seem to be saved however.
I have added two items in "Class Member Order":
- all methods whose name does not match 'main' (alphabetized)
- all methods whose name matches 'main'
When I restart IDEA the settings have changed to:
- all methods (alphabetized)
- all methods
As a workaround I don't restart IDEA;-)
Other than that, the name matching works great! All my main methods are
now at the bottom of the class, where they belong:-)
-- Bas
Thanks for the feedback, Bas. I'll fix it right away.
-Dave
OK Bas,
give version 0.7 a try.
-Dave
On 2003/11/04 01:41, Dave Kriewall wrote:
Thanks, that was really fast! A quick test shows that it works perfectly
now. Excellent.
-- Bas
Alain, (and all rearranger users who might find this useful)
I'm about ready to get to work on your request for special handling of private methods. Here's my proposal.
1) I think that the configuration for this is separate from the existing "Class Member Order" panel. What we are really saying is that private methods will be treated by a completely different set of rules. So I'm proposing adding another tab, say "Global Settings" or "Private Methods", in addition to the existing two (class member order, outer class order).
2) On this new tab, user has following top-level choices (say radio buttons):
If the 2nd is chosen, then the user can also specify ordering options:
Examples:
If there is only one private method called by a parent method, and the private method calls no other private method, then none of these rules (2 or 3) make any difference.
If two private methods are called by a parent, and the private methods do not call other private methods, then the depth-first/breadth-first rule makes no difference.
The two private methods would be ordered according to rule 3A, 3B or 3C.
For the following 6 examples (combinations of rules 2 and 3), assume the following source code:
Then the resulting order of method placement is:
Algorithm when "move private methods" option is chosen:
Note: Any private methods that are not referenced by any other method in the class would be placed according to "Class Member Order" rules. These are useless methods (unused code) but may not be in the future.
Does this make sense?
Is there any need to emit special comments? If so, would the ability to emit a comment before and after the children be sufficient?
For example,
Thanks for your thoughts on this.
-Dave
Dave :
Thanks for the energy you are putting into this. I can't wait to the
result.
You don't need much sleep, do you?
Here are a few remarks :
1 : your plan looks ok.
-
but...
2 : it's about "extracted methods", rather than "private methods"
-
The idea behind my initial request it to have the physical order, in the
class, follow the logical flow, in the calling tree. In this context,
being private for a method is just a detail. It's not a requirement.
Another way to say this is :
making a private method public shouldn't change the rearranger result.
Or at least, there should be a way to configure the plugin to behave
this way.
Motivation : sometimes, people would make a private method public, only
to be able to test it directly, rather than through their public parent.
3 : son with 2+ fathers
-
What happens to a (private) method that would be called by 2 other
methods in the same class.
This was in my previous mail.
By default, I would move it after it's first call/father.
For completeness, you could offer an option :
° move after callee after first caller
or
° move after callee after last caller
but I'm not sure it would be useful. I wouldn't use it.
4 : handling of useless private methods : comment and location
-
1.a : special comment before their section
example
// USELESS PRIVATE METHODS
1.b : location
at the end of the "methods" section,
or
at the end of the class.
5 : the case of the overloaded methods/constructors
-
(if you have time...)
Example
Personally, I prefer
This is also valid if you replace the constructors by methods "foo(...)".
5 : OT
This morning, I posted a someway related request, that would fit your
plugin nicely :
Add separate method call and declaration colour for private: coz they
were extracted.
Alain
Oops, typo in point 5. Here is the correct version :
]]>
5 : the case of the overloaded methods/constructors
-
(if you have time...)
Example
Personally, I prefer
This is also valid if you replace the constructors by methods "foo(...)".
]]>
Alain
Alain,
Thanks for the feedback!
2) The reason I thought it would be best to restrict it to private methods is that:
- it's more likely that it is an extracted method
- getters/setters would not be affected.
I'm pretty sure that folks don't view getters/setters as extracted methods, so moving them after the first or last caller wouldn't be appropriate.
On the other hand, I see your point about making private methods public temporarily.
Only compromise I can think of is something like adding a checkbox on the Method dialog that means "exempt from special reordering treatment". So you could make a rule that moves all getter/setters toward the front of the file and exempts them from consideration for the kind of rearrangement we're discussing. Or conversely, have a set of rules that selects methods that we should "consider for special reordering treatment". This is probably better since we may think of other types of exemptions or special reorderings in the future.
So I'll propose that the user can create a list, similar to current selection criteria, that will select methods for special reordering. (The list order will be immaterial.) User could exclude methods by method name, e.g. or could specifically exclude methods of type getter/setter.
By default, the list would be empty so that current behavior is unchanged.
User could add private methods, non-getter/setter methods, or whatever describes your notion of "extracted" methods.
You could even add constructors (peeking ahead to your request below.)
3) son with 2+ fathers -- I snuck in that option in my previous message -- sorry, wasn't very obvious:
4) emitting a comment for useless private methods would be handled by the current rules-based feature.
Any method not selected for special treatment (because, although available for special treatment, it was not called by any other method in the class) remains in the list of items to be rearranged the current way.
So you could have a comment
which is only emitted if an item matches the next rule, and the next rule is
5) Yeah, let's talk about constructors. If we open this up to any methods, not just private methods, then I think constructors would be handled just fine. Only additional way I can think to sort them would be based on number of parameters.
The way you listed it in your amended message is reversed; the "son" constructor (called by 3 parents) comes first, then the parents in some order. To make this constructor example work, it looks like we need a "reverse order" checkbox so that the son comes first, followed by parent.
If the code was:
so that the constructors were cascaded, then the "reverse order" technique would produce:
because a constructor with N parameters calls the constructor with N+1 parameters, so there's a greatGrandfather-Grandfather-Father-Son hierarchy.
OK, it's a bit more work, and a bit less sleep.. ;) but I'll do whatever it takes to make my clientele happy!
-Dave
Dave,
> 2) The reason I thought it would be best to restrict it to private
methods is that:
> - it's more likely that it is an extracted method
> - getters/setters would not be affected.
True, getters/setters are a special case, and they should not take part
in this game.
(Don't forget to use the user's code style to recognize them !)
About your suggestion for point 2 :
> ..the user can create a list,
I'm a little confused here, because I find it quite complex, although
the "problem" is pretty simple, and common enough to deserve a smart and
effortless default handling.
Just to give more weight to this request, here is another real-life
example where a private method becomes public for a good reason :
step 1 : initial code.
-
step 2 : method extraction
-
step 3 : the private method becomes public
-
Later in the client code, I'll need to use directly the second method,
so I'll make it public.
The default handling of the code above must do as if the extracted
method were still private, and place it just after its
caller/father.Otherwise it will become hard to read. And making the code
easier to read is the whole purpose of the plugin.
Some more about overloaded methods :
-
On the the code above, if the extraction had introduced an intermediary
helper method, I think the natural order should not be followed
completely. I would introduce an exception, to keep overloaded methods
close together :
Example : before
-
For the code above, I'd tend to say that overloaded methods should be
kept together :
=> rearrangement should produce :
Example : after
-
public boolean isAccessor (Editor e, File f) ..
public boolean isAccessor (PsiMethod m) ..
private PsiMethod methodAtCaret (Editor e, File f) ..
=> rather than the real callling order :
public boolean isAccessor (Editor e, File f) ..
private PsiMethod methodAtCaret (Editor e, File f) ..
public boolean isAccessor (PsiMethod m) ..
Alain
Alain,
in step 3 you have a method
which to me looks much like a getter method. But you want this to follow its parent, the other "isAccessor" method, because you regard it as an extracted method.
How do I reconcile this with your assertion that getters/setters should not take part in this game?
Do you feel that it should be treated as an extracted method because it is called by another getter (its parent)? (It might be called by other methods also.)
The real problem is knowing when a non-private method (e.g. getters/setters) is to be regarded as an extracted method. These extracted methods are moved along with their parent(s), and aren't subject to the normal rules of rearrangement. I'm not yet convinced that we can come up with definitive criteria for determining which methods are "extracted." That's why I was proposing leaving it up to the user. (When in doubt, punt. :) Let the user decide if certain methods are, in his/her opinion, extracted. I could add additional criteria to make that possible, such as
or others you may come up with.
If you can really nail down what constitutes an extracted method, then I wouldn't have to make the user define it. That would be great, but I have a sinking feeling that it won't suit everybody. The problem doesn't seem simple to me (at the moment).
-Dave
P.S. I am using
to determine if a method is getter or setter.
Dave,
-
I feel a big part of your problems would disappear, and your code would be simpler, if you had an accuccate accessor tester :
> in step 3 you have a method ..boolean isAccessor(PsiMethod m)..
> which to me looks much like a getter method.
NO, look at what the method does/the method's code :it is NOT an getter. It doesn't follow the getter pattern.
>P.S. I am using PropertyUtil.isSimplePropertyGetter(method);
> psi.util.PropertyUtil.isSimplePropertySetter(method);
public int getX() {return notX ;}
public int getX() { x += 1 ; return x ;} //
-
protected static boolean isAccessor ( PsiMethod i_method, CodeStyleManager i_codeStyleManager )
{
return MyPsiMethodUtil.isSetter ( i_method, i_codeStyleManager )
|| MyPsiMethodUtil.isGetter ( i_method, i_codeStyleManager );
}
-
public static boolean isHashCode(PsiMethod i_method)
{
return checkMethodSignature(i_method, "public", "int", "hashCode", 0);
}
public static boolean isToString(PsiMethod i_method)
{
return checkMethodSignature(i_method, "public", "String", "toString", 0);
}
public static boolean isEquals(PsiMethod i_method)
{
final boolean basicsAreOk = checkMethodSignature(i_method, "public", "boolean", "equals", 1);
if (!basicsAreOk)
return false;
final PsiParameter[] parameters = i_method.getParameterList().getParameters();
final boolean paramIsOk = "Object".equals(parameters[0].getTypeElement().getText());
return paramIsOk;
}
-
private static boolean checkMethodSignature(PsiMethod i_method, final String i_expectedModifier, final String i_expectedReturnType, final String i_expectedName, final int i_expectedNofParameters)
{
if (i_method == null)
return false;
final boolean nameIsOk = i_expectedName.equals(i_method.getName());
if (! nameIsOk)
return false;
if (! (nofParameters(i_method) == i_expectedNofParameters))
return false;
final PsiModifierList modifierList = i_method.getModifierList();
if (! i_expectedModifier.equals(modifierList.getText()))
return false;
if (!( i_expectedReturnType.equals(i_method.getReturnTypeElement().getText())))
return false;
return true ;
}
-
public static boolean isGetter(PsiMethod i_method, CodeStyleManager i_codeStyleManager)
{
if (i_method == null)
return false ;
final boolean hasParameters = 1 <= nofParameters(i_method);
if (hasParameters )
return false;
final String name = i_method.getName();
if (name.startsWith("get")) return getterNameMatchedReturnedValue(i_method, name, "get", i_codeStyleManager);
if (name.startsWith("is" )) return getterNameMatchedReturnedValue(i_method, name, "is", i_codeStyleManager);
if (name.startsWith("has")) return getterNameMatchedReturnedValue(i_method, name, "has", i_codeStyleManager);
return false;
}
private static boolean getterNameMatchedReturnedValue(PsiMethod i_method, String i_name, String i_prefix, CodeStyleManager i_codeStyleManager)
{
String i_methodNameTrail = i_name.substring(i_prefix.length());
if (!nameIsWellFormed(i_methodNameTrail))
return false;
if (isAbstract(i_method))
return false;
final String nameFromBody = nameOfVariableReturnedInBody(i_method);
final String propertyName = propertyNameFromMethodTrail(i_methodNameTrail);
final boolean namesMatch = nameFromBody.equals(fieldName(propertyName, i_codeStyleManager));
return namesMatch;
}
private static String nameOfVariableReturnedInBody(PsiMethod i_method)
{
final PsiStatement[] statements = i_method.getBody().getStatements();
final boolean methodBodyIsEmpty = nofStatements(i_method) == 0;
if (methodBodyIsEmpty)
return "";
if ( ! U.isReturnStatement(statements[0]))
return "";
final PsiReturnStatement returnStatement = (PsiReturnStatement) statements[0];
final PsiExpression returnValue = returnStatement.getReturnValue();
//todo : improve
if (null ==returnValue)
return "";
String returnValueText = returnValue.getText();
returnValueText = returnValueText.replaceFirst("this
.", "");
return returnValueText;
}
private static String propertyNameFromMethodTrail(String i_source)
{
final String left = i_source.substring(0,1).toLowerCase();
final String right = i_source.substring(1);
return left + right;
}
-
public static boolean isSetter(PsiMethod i_method, CodeStyleManager i_codeStyleManager)
{
if ( couldBeAsetter(i_method))
return false;
String i_methodNameTrail = i_method.getName().substring("set".length());
if (!nameIsWellFormed(i_methodNameTrail))
return false;
final PsiElement psiElement = i_method.getBody().getStatements()[0].getChildren()[0];
if ( !U.isAssignment(psiElement) )
return false;
PsiAssignmentExpression assignement = (PsiAssignmentExpression) psiElement;
final boolean notAPlainAssignment = PsiJavaToken.EQ != assignement.getOperationSign().getTokenType();
if (notAPlainAssignment)
return false;
final PsiExpression rExpressionRaw = assignement.getRExpression();
if (null == rExpressionRaw)
return false;
String
lExpression = assignement.getLExpression().getText().replaceFirst("this.", ""),
parameterName = i_method.getParameterList().getParameters()[0].getName();
if (!(parameterName.equals(rExpressionRaw.getText())))
return false;
final boolean targetIsProperty = lExpression.equals(fieldName(propertyNameFromMethodTrail(i_methodNameTrail), i_codeStyleManager));
return targetIsProperty;
}
private static boolean couldBeAsetter(PsiMethod i_method)
{
if ( i_method == null ) return false;
if ( i_method.getBody() == null ) return false;
if ( 1 != nofParameters(i_method) ) return false;
if ( 1 != nofStatements(i_method) ) return false;
if ( ! i_method.getName().startsWith("set") ) return false;
return true;
}
private static boolean nameIsWellFormed(String i_methodNameTrail)
{
final boolean methodNameIsTooShort = i_methodNameTrail.length() == 0;
if ( methodNameIsTooShort )
return false;
final char firstCandidateMemberChar = i_methodNameTrail.charAt(0);
if ( Character.isLowerCase(firstCandidateMemberChar))
return false;
return true;
}
public static boolean isAssignment (final PsiElement i_element)
{
return (i_element instanceof PsiAssignmentExpression);
}
private static String propertyNameFromMethodTrail(String i_source)
{
final String left = i_source.substring(0,1).toLowerCase();
final String right = i_source.substring(1);
return left + right;
}
private static String fieldName(String i_propertyName, CodeStyleManager i_codeStyleManager)
{
return i_codeStyleManager.propertyNameToVariableName(i_propertyName,VariableKind.FIELD);
}
private static int nofParameters (PsiMethod i_method)
{
return i_method.getParameterList().getParameters().length;
}
private static int nofStatements (PsiMethod i_method)
{
return i_method.getBody().getStatements().length;
}
private static boolean isAbstract (PsiMethod i_method)
{
return null == i_method.getBody();
}
public static boolean isReturnStatement (final PsiStatement i_statement)
{
return i_statement instanceof PsiReturnStatement;
}
Hi Alain,
Thanks very much for your code sample. I remember you posted something about determining accessors/mutators properly a few weeks back (on OpenAPI forum I think) and someone replied with the solution I used. I had it in the back of my mind to sneak a look at your plugin to see what you were doing. This makes it much easier!
(Don't you wish we had source code for PSI so we could figure out if they already had a utility that would work?)
OK, so we automatically exclude G/Setters, and top level methods (those not called by another method in the program), and constructors (special handling). We should also exclude methods involved in a cycle (a() calls b(), b() calls a() -- some form of recursion happening).
User may also exclude the remaining non-private methods by the number of callers they have (as you suggested, "never, 1, 2+".)
What special handling do you want for equals(), hashcode() and toString()? Just exclusion from special treatment? You can already place them specially anywhere you want by doing a name match in the method criteria. And, for that matter, you can handle any overriding method in the same way.
Today is going to be a tabifier sort of day, but I hope to get to work on this next.
Thanks for your help & suggestions,
-Dave
Dave Kriewall wrote:
>What special handling do you want for equals(), hashcode() and toString()? Just exclusion from special treatment? You can already place them specially anywhere you want by doing a name match in the method criteria.
>
As for the accessors, I give them a special treatment in the
CamouflagePlugin, because they are a special kind of methods that belong
to the same family : canonicals methods.
I'd like them to
- stick together,
- in a specific order (ex: 1: equals, 2: hashCode, 3: toString)
- be identified/movable as a block (to unclutter the main ui : 3 lines
=> 1 line)
=>
- in the main interface, you would simply locate where to place the
group of canonical methods (1 line),
- in another location (another tab), you would choose the order inside
the group.
You could reuse that UI principle ( 1 line for the group in the main ui,
and more details in another tab) for other group/kinds of methods :
- constructors
- accessors
..?
Alain | https://intellij-support.jetbrains.com/hc/en-us/community/posts/206790595--ANN-Rearranger-plugin-new-version-0-6-released | CC-MAIN-2020-05 | refinedweb | 3,675 | 55.13 |
here's one stupid :)
Inside my project mvc4 I have namespace
prj.MVC4.Controllers
where I'm using Server.MapPath(..) without problem
and on prj.MVC4.Models Server.MapPath(...) does not exist on current context.
I'm aware that Server.MapPath resides in System.Web and both namespace and assembly are added into class with using System.Web and System.Web.dll is added to the prj.
on ctrl + . I'm getting Microsoft.SqlServer as suggested namespace to add.
How to fix this?
Post:25Points:175
Re: Server.MapPath does not exist in currect context
The server is a property of the controller, to access it elsewhere while running a web application you can use | https://www.mindstick.com/forum/1589/server-mappath-does-not-exist-in-currect-context | CC-MAIN-2018-34 | refinedweb | 115 | 72.93 |
Home › Forums › WinForms controls › Xceed Grid for WinForms › Xceed data grid for .net and exporting to xlsx format › Reply To: Xceed data grid for .net and exporting to xlsx format
User (Old forums)
Member
August 13, 2009 at 4:46 pm
Post count: 23064
Hi,
Ctrl-W, S to Open Solution Explorer… (in VB.NET, you also need to click on Show All Files to be shown the “References”.) Right-Click “Add references…”, select the .NET tab, scroll down to Xceed.FileSystem.Dll, Click this and Accept / OK.
Also,
Imports Xceed.FileSystem (VB.NET)
or
using Xceed.FileSystem; (C#)
at the top of your file.
Imported from legacy forums. Posted by Ghislain (had 336 views) | http://forums.xceed.com/forums/reply/17233/ | CC-MAIN-2022-05 | refinedweb | 115 | 79.46 |
Introduction: Intel Galileo Interactive LED Matrix
Create an interactive 8x8 LED Matrix and send pictures to be displayed from anywhere around the world!
In this project we are using an Intel Galileo board which Intel has generously provided us with (they're giving a lot of those to unis worldwide). The Intel Galileo seems to be a great device for the job, as it combines a Linux system with built-in Ethernet with an Arduino
We created this interactive project as part of our course, Multimodal Media Madness - Personal Fabrication at RWTH Aachen University, Germany. It consists of two main elements, the Intel Galileo with the LED matrix and a server side application. On the server runs a simple drawing application so anyone who has access to the site can draw pictures, and your LED matrix will display them!
A running version of the drawing app can probably still be found at ruedigerhoffmann.dlinkddns.com/galileo
Step 1: Gather Materials
For this Instructable you'll need:
- 8x8 LED Matrix (RGB FULL COLOR MATRIX, 6*6CM, 8*8 DOTS, C+)
- Intel Galileo Board (we used a Gen 2 board, but a Gen 1 board should be fine)
- High Voltage Source Driver / Transistor (TOSHIBA Bipolar Digital Integrated Circuit Silicon Monolithic TD62783APG)
- 8 bit shift egister (M74HC595B1R)
- Breadboard
- 8*10Ω Resistors
- lots of Breadboard Jumper Wires (Try to get at least 5 different colors, this will make the wiring much easier.)
- a standard network cable (to connect the Galileo to the internet; not shown)
- a webserver with php installed - even a Raspberry PI should work fine
Step 2: Install the LED Matrix and Bit Shift Register
Install the 8x8 LED Matrix on the Breadboard. Have the side with text face the left side of your Breadboard. The LED matrix has two sets of 16 Pins. The Pins for the Vcc of the LEDs are on the column and the Pins for the green LEDs are on the row to the right, the Pins for the blue and the red ones are on the rows to the left.
Place the High Voltage Source Driver close to the right side of the 8x8 LED Matrix.
Step 3: Connect the High Voltage Driver to the Matrix
Connect the High Voltage Source Driver (HVSD) to the Pins of the Vcc columns of the 8x8 LED Matrix. Output 1 goes to LED Pin 32, Output 2 to Pin 31 and so on.
Note that the column pins asadre the two outermost sets of 4 Pins(Pins 17 to 20 and 29 to 32) on one side of the Matrix. To be sure which side this is please refer to the datasheet. Pin1 is labeled as that on the bottom of the Matrix.
Make a connection from the High Voltage Source Drivers VCC and GND to the corresponding bars on your Breadboard.
Step 4: Connect the HVSD to the Galileo
To connect the Input Pins of the High Voltage Source Driver to the output Pins of the Galileo.
- Galileo Output 6 connects to HSVD Input 1
- Galileo Output 5 connects to HSVD Input 2
- Galileo Output 8 connects to HSVD Input 3
- Galileo Output 9 connects to HSVD Input 4
- Galileo Output 10 connects to HSVD Input 5
- Galileo Output 11 connects to HSVD Input 6
- Galileo Output 12 connects to HSDV Input 7
- Galileo Output 13 connects to HSVD Input 8
Make the connections as per the Description. In the Picture the cable from pin 7 has to go to the pin 5.
Note: Due to some yet unsolved issue with pin 7 that caused flickering we had to remap pin7 to pin 5. The Sketch that you upload to the Galileo already takes care of that remapping.
Connect the GND and 5V VCC Pins of the Galileo to the corresponding bars of your Breadboard
Step 5: Protect the LEDs
To protect the LEDs, connect 10 Ohm resistors to the LED pins(If you want to use the RED LEDs you have to connect Pins 1 to 8). On the end of every resistor, put a cable that we'll connect to the 8 bit shift register.
Note: If you use a different LED matrix, you'll propably need other resistors so it still lights up brightly.
Step 6: Connect the LEDs to the Shift Register
Connect every outputs of the 8 bit shift register(Q0 thru Q7) to the other ends of the resistors. If you want to connect the red LEDs as we did, you have to connect Q0 to Pin8, Q1 to Pin7 and so on.
Note, that one output of the Shift Register is not on the same side as the others (as detailed in the Data Sheet)
Step 7: Connect the Shift Register to the Galileo
To make the necessary connections to push the data into the Shift Register connect:
- Data Pin: DS (or SI) from the ShiftRegister to Digital 2 on the Galileo (Picture Yellow)
- Clock Pin: SHCP (or SCK) from the ShiftRegister to Digital 3 on the Galileo (Picture Blue)
- Latch Pin (to trigger output on the Shift Register): STCP (or RCK) rom the ShiftRegister to Digital 4 on the Galileo (Picture Black)
Step 8: Setting Up Your Galileo
The Instructable Galileo - Getting Started should help you get set using the Galileo. You might also want to download Linux to the microSD as it enables the Sketch to start automatically whenever the Galileo gets powered on. Intel's website has a ton of resources too, but it's kinda hidden.
Step 9: Load the Code Onto the Galileo
To breathe life into your LED Matrix, copy and paste our Code into the Arduino IDE and run it. You can also download the .ino file below. The Code uses Interrupts to fill the Shift Register and light up the specific LEDs. Also the "popen" command is used to communicate between the underlying Linux shell and our sketch. The binary file it gets from the server is downloaded by using curl.
NOTE: Of course you'll have to change the URL to your own server.
#include <timerone.h> int on = 0; int last_on = 0; int datapin = 2; int clockpin = 3; int latchpin = 4; byte data[8]; FILE *fp; char input[10]; void setup() { // Initialize the digital pin as an output. // Pin 13 has an LED connected on most Arduino boards); Timer1.initialize(1000); // the timer period is 100000 useconds, that is 0.1 sec Timer1.attachInterrupt(timerIsr,1000); // the callback will be called on each 5th timer interrupt, i.e. every 0.5 sec //initLED(); data[0] = 0b00000111; data[1] = 0b00000001; data[2] = 0b11100111; data[3] = 0b10110001; data[4] = 0b10101011; data[5] = 0b10101000; data[6] = 0b10101000; data[7] = 0b10101000; Serial.begin(115200); } void loop() { //SWAP THE URL PLEASE fp = popen("curl PLEASE INSERT YOUR URL HERE/galileo/picture.txt","r"); if(fp == NULL) { Serial.println("Couldnt run the curl command"); } else { fgets(input,10,fp); } if(pclose(fp) != 0) { Serial.println("fail"); } data[0] = (byte)input[1]; data[1] = (byte)input[2]; data[2] = (byte)input[3]; data[3] = (byte)input[4]; data[4] = (byte)input[5]; data[5] = (byte)input[6]; data[6] = (byte)input[7]; data[7] = (byte)input[8]; delay(3000); } /// -------------------------- /// Custom ISR Timer Routine /// -------------------------- void timerIsr() { if(on==6) on = 8; digitalWrite((13-last_on), LOW); digitalWrite(latchpin,LOW); if(on == 8) shiftOut(datapin, clockpin, MSBFIRST, ~data[6]); else shiftOut(datapin, clockpin, MSBFIRST, ~data[on]); digitalWrite(latchpin,HIGH); digitalWrite( (13-on), HIGH); digitalWrite((13-last_on),LOW); last_on = on; if(on == 8) on =6; on = (on+1)%8; }
Step 10: Set Up the Server Side & Connect the Galileo
On the server runs a simple php script which provides a nice drawing interface. The picture is converted to binary and stored in the file picture.txt, which the Galileo can download.
You'll need some kind of webserver setup with php installed. Just use one of these handy instructables if you haven't got one already!
Just put the file index.php on your server. (Remove the .txt extension first.) You might want to rename it to something like galileo.php if the folder already has an index.* file.
Don't forget to create an empty text file named "picture.txt" which the user "www-data" (or whatever user runs your webserver - Attention: This is normally not the same user that set up the server!) can read and write into. The easiest way (on a linux shell) is the following:
$ cd /path/to/index.php $ touch picture.txt $ sudo chown www-data picture.txt
Start up the Galileo, run the Sketch and connect it to the internet.
Now just browse to the php file in your web browser of choice and start drawing. It even works on mobile devices!
Recommendations
We have a be nice policy.
Please be positive and constructive.
7 Comments
What is your opinion about the 3 objective of this project ?
Hey. I am interested in doing this project. Could you upload the drawing app maybe. Cause the link given is not working
The php can be found at the very last step
I am using a 32 x 64 LED scrolling display.
Great project. I would like to adapt the project to display text on the LED intead. Any ideas?
Cool project! I've always wanted to get started with the Intel boards.
You might want to go for the Edison if you need time-critical stuff. The Galileo is kinda shaky with timer interrupts as the Arduino stuff is just running as a linux processs. | http://www.instructables.com/id/Intel-Galileo-Interactive-LED-Matrix/ | CC-MAIN-2018-17 | refinedweb | 1,588 | 68.7 |
degrees() and radians() in Python
In this tutorial, you will learn about degrees() and radians() methods in Python. By using the math module the data will be change to degrees() and radians() in Python. For mathematical solving, ship direction, etc will be taken.
Degrees():
The degrees are unit of angle measure, all the angels are using degrees basically, you learn that circle contains 360 degrees like that it will be used.
The degree symbol will obtain by using Chr(176) method
Radian():
It is a unit of measure of angels in trigonometry it will use instead of degrees, whereas the circle contains only 360 degrees, a full circle is just over 6 radians. a full circle has 2 pi radians.
Diagram for Degree and Radian:
The above diagram will show the glance description of the Degrees and Radians.
Importing math module:
from the below code are used to import math module
import math
Program on Degrees and Radians in Python:
from the below code about Degrees and Radians:
import math l=[1,2,3] for i in range(len(l)): print(math.degrees(l[i]),"degrees") print(math.radians(l[i]),"radians")
Output:
57.29577951308232 degrees 0.017453292519943295 radians 114.59155902616465 degrees 0.03490658503988659 radians 171.88733853924697 degrees 0.05235987755982989 radians
Explanation
- From the above code importing math module.
- Consider a list having 3 elements by using the math module showing the degrees and radians values.
- By using for loop the data will be shown in degrees and radians format.
References:
For further references about degrees and radians ->click here | https://www.codespeedy.com/degrees-and-radians-in-python/ | CC-MAIN-2021-17 | refinedweb | 260 | 58.08 |
Is space a valid component of a password or not? I am trying to save recent 10 passwords in one string in database and need to find a good delimiter for them. I think space maybe a good candidate. What do you think?
This question came from our site for system and network administrators. depends on your password policy. I know quite some sites/systems where the space is a valid character for a password.
To be on the safe side you could check for spaces within the password and escape those.
Oh, and as a short update: Try to repair the database design. As you have a 1:n relationship, you should save each password separately and connect every entry to the according user.
Spaces are normally valid. I'd be wary of any delimiter, as it's a form of security through obscurity that someone won't crack it or accidentally stumble on it in the future and you'll have to figure out the bug.
I'd use a separate entry for each one.
You don't mention what application this is...if you're making the application, you could try doing something to enforce your own policy that would scrub out and sanitize the entry, or you would more sensibly hash the password (you generally don't want actual passwords saved) and the hash wouldn't have the password in it. Then I suppose you could use whatever delimiter you want as long as it isn't part of the hash namespace of characters.
You can't rely on space, as it is a valid password character on most systems, especially now that pass-phrases are the new passwords.
Depending upon what/how you are doing this, you might be able to use a char with ASCII 0x00, another character not normally found on a keyboard or what about unicode?
Personally, I wouldn't attempt to concatenate them into a single string, I'd probably store a entry for each password.
asked
5 years ago
viewed
1787 times
active
3 years ago | http://superuser.com/questions/262536/are-spaces-valid-components-of-a-password?answertab=active | CC-MAIN-2015-18 | refinedweb | 346 | 71.75 |
[Share]Simple ListView Class
The below is very simple. It's also been shared before by other members here. But just shows how simple it can be to make a list view or dialog list view. Not many lines of real doing something code. With the new beta, 301011 can be less lines. Anyway, has been a few questions about ui.TableView's lately.
I know I have not commented the code. I think comments can actually make it more difficult to understand. I don't mean in a professional environment, but with small snippets like this. Not really sure. But ok, in this example I am showing font family names. Wanted to do something that had some meaning other than a list of numbers. But really if you can get this far with a ui.TableView, you are 90% there. Does not take much from here to create your own data_source or create your own ui.TableViewCell's
# Pythonista Forum - @Phuket2 import ui, itertools def get_font_list(): # from someone on the forum, i think @omz UIFont = ObjCClass('UIFont') return list(itertools.chain(*[UIFont.fontNamesForFamilyName_(str(x)) for x in UIFont.familyNames()])).action = self.my_action self.tbl = tbl self.add_subview(tbl) def my_action(self, sender): self.value = sender.items[sender.selected_row] self.close() if __name__ == '__main__': w, h = 600, 800 f = (0, 0, w, h) style = '' my_list = get_font_list() v = SimpleListView(my_list, frame=f, name='Font List') v.present(style=style) v.wait_modal() print(v.value)
I am having a problem to do a inline sort for the get_font_list func. I did the below, but guessing it's bad/inefficient. Not sure why I can't figure it out. But the below does return the fonts sorted, just in an inefficient way.
def get_font_list(): # from someone on the forum, i think @omz UIFont = ObjCClass('UIFont') ''' return list(itertools.chain(*[UIFont.fontNamesForFamilyName_(str(x)) for x in UIFont.familyNames()])) ''' lst = list(itertools.chain(*[UIFont.fontNamesForFamilyName_(str(x)) for x in UIFont.familyNames()])) return sorted([str(item) for item in lst])
@Phuket2 If you want to sort a
listin-place, you can use
list.sort:
lst = list(...) lst.sort() return lst
Also, I think you can shorten the list comprehension a little, to something like this:
lst = [str(font) for family in UIFont.familyNames() for font in UIFont.fontNamesForFamilyName_(family)] lst.sort() return lst
Then the
strconversion is done in the list comprehension directly and you don't have to do it later.
@dgelessus , thanks. Perfect.
The new func for get_font_list as you say above.
def get_font_list(): # from someone on the forum, i think @omz UIFont = ObjCClass('UIFont') lst = [str(font) for family in UIFont.familyNames() for font in UIFont.fontNamesForFamilyName_(family)] lst.sort() return lst
Hmmm, one day I will get something right 100%. It's not today. I am sure some guys spotted a issue with what I say about using ui.ListDataSource and benefits with accessory items etc. that's all well and good. But if you create the cell yourself, you lose that functionality. As the ui.ListDataSource is creating that magic when it creates the ui.TableViewCell. So when you create the cell yourself that functionality disappears.
But not all is lost. It's not documented, but ui.TableViewCell takes a param. None = default, 'subtitle', 'value1', 'value2'. but ui.TableViewCell creates a slightly different cell layout depending on what Str its passed.
But if @omz creates a new ui.TableViewCell type something like 'listdatasource' , we could possibly have our cake and eat it.
@omz, not sure if this is difficult or not (a new ui.TableViewCell type). But to me it makes sense. It would make ui.ListDataSource a lot more flexible, unless I am missing something, which is very possible
@Phuket2
ui.ListDataSourceis implemented in the Python part of the
uimodule (
site-packages/ui.pyin the standard library section), so you can see for yourself how it does all the "magic" with accessory items
:)
To give a custom cell an accessory item, you need to set its
accessory_typeattribute (this is documented in the
ui.TableViewCelldocs), and to handle tapping on the accessory button, you need to implement the
tableview_accessory_button_tapped(self, tv, section, row)method on your delegate (this part is not documented as far as I know - it's one of those omz secret features).
@ramvee , thanks. You are right about the import. I think because I use that import in the pythonista_startup.py it does not fail for me. Well at least I think that's why.
@dgelessus , lol. Thanks. You are so right. At least I predicted I was going to be wrong. I was just wrong recursively 😱
What threw me off was, if you assign a list of dicts to LDS.items as in the help file (title, image, accessory_type), they appear to be ignored if you create your own cell. I can't get it clear in my head if it should be like that or not. But what you mention works perfectly well.
Ok,) | https://forum.omz-software.com/topic/3597/share-simple-listview-class/12 | CC-MAIN-2021-21 | refinedweb | 831 | 69.07 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.