workspace
stringclasses
1 value
channel
stringclasses
1 value
sentences
stringlengths
1
3.93k
ts
stringlengths
26
26
user
stringlengths
2
11
sentence_id
stringlengths
44
53
timestamp
float64
1.5B
1.56B
__index_level_0__
int64
0
106k
pythondev
help
I had intended to be “verbose” when explaining my problem :slightly_smiling_face:
2017-10-23T15:06:09.000071
Derek
pythondev_help_Derek_2017-10-23T15:06:09.000071
1,508,771,169.000071
97,903
pythondev
help
:smile:
2017-10-23T15:07:01.000478
Meg
pythondev_help_Meg_2017-10-23T15:07:01.000478
1,508,771,221.000478
97,904
pythondev
help
How do people feel about functional testing in addition to unit testing? Think it is overkill or not needed until certain points in a project?
2017-10-23T15:25:35.000062
Vita
pythondev_help_Vita_2017-10-23T15:25:35.000062
1,508,772,335.000062
97,905
pythondev
help
It's not overkill
2017-10-23T15:45:29.000379
Ronni
pythondev_help_Ronni_2017-10-23T15:45:29.000379
1,508,773,529.000379
97,906
pythondev
help
there was a pycon panel about this with heavyweights and they will all in support, for slightly different reasons
2017-10-23T15:50:41.000706
Seema
pythondev_help_Seema_2017-10-23T15:50:41.000706
1,508,773,841.000706
97,907
pythondev
help
it was more about linting and humans than about functional testing though, very nuanced
2017-10-23T15:52:27.000603
Seema
pythondev_help_Seema_2017-10-23T15:52:27.000603
1,508,773,947.000603
97,908
pythondev
help
I prefer _writing_ unit tests b/c they're faster and easier to write, but just like a child prefers cookies to roasted broccoli (mmm delicious!) it doesn't mean it's sufficient
2017-10-23T16:04:25.000114
Ronni
pythondev_help_Ronni_2017-10-23T16:04:25.000114
1,508,774,665.000114
97,909
pythondev
help
can anyone suggest popular IoC (inversion of control) package to use? PyPI provides too much to choose from <https://pypi.python.org/pypi?%3Aaction=search&amp;term=dependency+injection&amp;submit=search>
2017-10-23T16:14:12.000103
Kellye
pythondev_help_Kellye_2017-10-23T16:14:12.000103
1,508,775,252.000103
97,910
pythondev
help
Is there any real reason to using an ORM (like sqlalchamy) besides not having to write queries yourself? I personally have my own module for storing/executing/returning queries and at times it runs faster than the ORM, obviously because a human is better at constructing complex queries than a computer is (for now). So besides the obvious, are there other reasons you guys find yourselves using one?
2017-10-23T16:23:48.000378
Lory
pythondev_help_Lory_2017-10-23T16:23:48.000378
1,508,775,828.000378
97,911
pythondev
help
how maintainable is your own module :slightly_smiling_face:
2017-10-23T16:24:23.000459
Meg
pythondev_help_Meg_2017-10-23T16:24:23.000459
1,508,775,863.000459
97,912
pythondev
help
scalable? works with schema changes without tinkering?
2017-10-23T16:24:38.000075
Meg
pythondev_help_Meg_2017-10-23T16:24:38.000075
1,508,775,878.000075
97,913
pythondev
help
I mean, sure there are reasons to use raw sql over an ORM
2017-10-23T16:24:58.000350
Meg
pythondev_help_Meg_2017-10-23T16:24:58.000350
1,508,775,898.00035
97,914
pythondev
help
but those should be limited to areas where the optimization is needed
2017-10-23T16:25:13.000568
Meg
pythondev_help_Meg_2017-10-23T16:25:13.000568
1,508,775,913.000568
97,915
pythondev
help
not as a first reaction
2017-10-23T16:25:27.000263
Meg
pythondev_help_Meg_2017-10-23T16:25:27.000263
1,508,775,927.000263
97,916
pythondev
help
<@Lory> I generally try to avoid ORMs since I think they are often the wrong approach to the general problem of database abstraction, so you're not alone, but I'm sure you'll get _plenty_ of arguments for why you should use one
2017-10-23T16:25:42.000244
Issac
pythondev_help_Issac_2017-10-23T16:25:42.000244
1,508,775,942.000244
97,917
pythondev
help
<@Issac> how maintainable is that, though?
2017-10-23T16:26:11.000386
Meg
pythondev_help_Meg_2017-10-23T16:26:11.000386
1,508,775,971.000386
97,918
pythondev
help
for just myself and a few friends that use it, it's fairly easy. Just deifne your boiler plate queries and then call them through a few defined methods that give me all results or one result or whatever and have them preformatted to how I like my data structured
2017-10-23T16:26:51.000429
Lory
pythondev_help_Lory_2017-10-23T16:26:51.000429
1,508,776,011.000429
97,919
pythondev
help
schema changes require manual intervention lol
2017-10-23T16:26:59.000047
Lory
pythondev_help_Lory_2017-10-23T16:26:59.000047
1,508,776,019.000047
97,920
pythondev
help
<@Meg> you mean not using an ORM? I would tend to argue that the types of abstractions that ORMs introduce is increasing the complexity of the code and introduces more maintenance overhead than not using one, although like anything else in development you can do it badly and incur a lot of pain
2017-10-23T16:28:03.000236
Issac
pythondev_help_Issac_2017-10-23T16:28:03.000236
1,508,776,083.000236
97,921
pythondev
help
sorry no, my small module
2017-10-23T16:28:25.000251
Lory
pythondev_help_Lory_2017-10-23T16:28:25.000251
1,508,776,105.000251
97,922
pythondev
help
<@Meg> There's a middle ground between ORM and raw SQL (like <http://knexjs.org/> for node, not sure of a good example in python) that helps with maintainability while being a more simple abstraction
2017-10-23T16:28:45.000342
Antionette
pythondev_help_Antionette_2017-10-23T16:28:45.000342
1,508,776,125.000342
97,923
pythondev
help
django ORM works similar, I think
2017-10-23T16:28:58.000422
Meg
pythondev_help_Meg_2017-10-23T16:28:58.000422
1,508,776,138.000422
97,924
pythondev
help
as well as sqlalchemy, in which you can write raw sql and it’ll handle it for you
2017-10-23T16:29:13.000507
Meg
pythondev_help_Meg_2017-10-23T16:29:13.000507
1,508,776,153.000507
97,925
pythondev
help
My biggest gripe is that it feels awkward to use one in the code, and that's just because of a lack of familiarity with using ORM's, So I figured what does it _really_ matter if I just use my own
2017-10-23T16:29:19.000113
Lory
pythondev_help_Lory_2017-10-23T16:29:19.000113
1,508,776,159.000113
97,926
pythondev
help
and for your own projects, thats fine
2017-10-23T16:29:31.000221
Meg
pythondev_help_Meg_2017-10-23T16:29:31.000221
1,508,776,171.000221
97,927
pythondev
help
if you’re working on non-super critical performance in a team, then I would definitely make you justify the case of not using an ORM if I were interviewing you
2017-10-23T16:30:22.000283
Meg
pythondev_help_Meg_2017-10-23T16:30:22.000283
1,508,776,222.000283
97,928
pythondev
help
and it just that, I'm trying to get into free lance web application design so I'm slowly building my own modules for things I feel are bloated like ORM's, Login modules, etc...
2017-10-23T16:31:12.000284
Lory
pythondev_help_Lory_2017-10-23T16:31:12.000284
1,508,776,272.000284
97,929
pythondev
help
sqlalchemy was started because the author was tired of hibernate’s bloatedness :slightly_smiling_face:
2017-10-23T16:31:42.000545
Meg
pythondev_help_Meg_2017-10-23T16:31:42.000545
1,508,776,302.000545
97,930
pythondev
help
Like most things you just need to be aware of the trade-offs, the once that have been mentioned incl. Imagine working on a large team or needing lots of people to be able to access and modify things in the database (like will happen many many times with a fairly complex dynamic website) and trusting all developers to get the SQL correct. That's a maintainability nightmare.
2017-10-23T16:32:07.000122
Mallie
pythondev_help_Mallie_2017-10-23T16:32:07.000122
1,508,776,327.000122
97,931
pythondev
help
bingo
2017-10-23T16:32:18.000382
Meg
pythondev_help_Meg_2017-10-23T16:32:18.000382
1,508,776,338.000382
97,932
pythondev
help
that there is my primary argument against
2017-10-23T16:32:28.000026
Meg
pythondev_help_Meg_2017-10-23T16:32:28.000026
1,508,776,348.000026
97,933
pythondev
help
especially if the actual performance difference is a few percents overall in load and query time
2017-10-23T16:32:50.000242
Meg
pythondev_help_Meg_2017-10-23T16:32:50.000242
1,508,776,370.000242
97,934
pythondev
help
And you may find that you only need to do one or two db calls, which is great, and you may find that as you maintain more and more projects all that "custom" SQL boilerplate wastes time.
2017-10-23T16:32:51.000232
Mallie
pythondev_help_Mallie_2017-10-23T16:32:51.000232
1,508,776,371.000232
97,935
pythondev
help
furthermore, as a freelance web dev, <@Lory>, you have to consider the ‘when’ case, when another developer takes over your project
2017-10-23T16:33:59.000224
Meg
pythondev_help_Meg_2017-10-23T16:33:59.000224
1,508,776,439.000224
97,936
pythondev
help
how easy is it for the other person/team to get up and running with your setup?
2017-10-23T16:34:18.000111
Meg
pythondev_help_Meg_2017-10-23T16:34:18.000111
1,508,776,458.000111
97,937
pythondev
help
i think you should decide what is most important to you when working with the ORM or the core.
2017-10-23T16:34:36.000361
Johana
pythondev_help_Johana_2017-10-23T16:34:36.000361
1,508,776,476.000361
97,938
pythondev
help
ORM is business objects and Core is more schema-centric.
2017-10-23T16:34:46.000463
Johana
pythondev_help_Johana_2017-10-23T16:34:46.000463
1,508,776,486.000463
97,939
pythondev
help
if you want to be down in the weeds and know your db schema you should use core. if you want to deal with things like business objects use the ORM.
2017-10-23T16:35:10.000396
Johana
pythondev_help_Johana_2017-10-23T16:35:10.000396
1,508,776,510.000396
97,940
pythondev
help
I would agree with <@Meg> there, freelancing does come with the burden of not just doing the most technically right thing, but the most expected thing since a lot of what your client is paying for is something that can serve as a baseline for future devs to continue to modify and support the application
2017-10-23T16:35:22.000377
Issac
pythondev_help_Issac_2017-10-23T16:35:22.000377
1,508,776,522.000377
97,941
pythondev
help
it depends on the level of abstraction you want. I’ve used both quite extensively.
2017-10-23T16:35:27.000138
Johana
pythondev_help_Johana_2017-10-23T16:35:27.000138
1,508,776,527.000138
97,942
pythondev
help
i try to avoid raw sql as much as possible.
2017-10-23T16:36:43.000244
Johana
pythondev_help_Johana_2017-10-23T16:36:43.000244
1,508,776,603.000244
97,943
pythondev
help
speaking for myself, at my current place, I had to figure out what the hell the previous dev was thinking when he put together the query structure for a very important part of the API. He just had python joining strings to construct the query, and execute with raw sql, using cursors, etc. When I moved it over to using django ORM, the actual query time went down about 50 milliseconds per query… without indexes
2017-10-23T16:37:03.000063
Meg
pythondev_help_Meg_2017-10-23T16:37:03.000063
1,508,776,623.000063
97,944
pythondev
help
much better to use a DSL like sqalchemy core.
2017-10-23T16:37:15.000033
Johana
pythondev_help_Johana_2017-10-23T16:37:15.000033
1,508,776,635.000033
97,945
pythondev
help
I think ORMs are problematic because they unnecessarily couple several parts of the system that should be decoupled (datatype definitions, constraint solving, persistence, and transport of network attached data). My personal experience is that in projects of any size they often end up causing more harm than good, and architecturally I dislike the types of applications that tend to have them central to the application
2017-10-23T16:37:37.000374
Issac
pythondev_help_Issac_2017-10-23T16:37:37.000374
1,508,776,657.000374
97,946
pythondev
help
not all ORMs are created equal tho.
2017-10-23T16:38:08.000674
Johana
pythondev_help_Johana_2017-10-23T16:38:08.000674
1,508,776,688.000674
97,947
pythondev
help
sqlalchemy’s orm is significantly better than others i’ve used.
2017-10-23T16:38:46.000103
Johana
pythondev_help_Johana_2017-10-23T16:38:46.000103
1,508,776,726.000103
97,948
pythondev
help
for one it uses the unit of work pattern so you aren’t going to commit anything until it all works.
2017-10-23T16:39:12.000349
Johana
pythondev_help_Johana_2017-10-23T16:39:12.000349
1,508,776,752.000349
97,949
pythondev
help
<@Issac> no argument there with your points about complexity and coupling. But is that really significant, though?
2017-10-23T16:39:18.000597
Meg
pythondev_help_Meg_2017-10-23T16:39:18.000597
1,508,776,758.000597
97,950
pythondev
help
and personally, in my projects, I’ve had to drop to raw sql a handful of times when performance was required
2017-10-23T16:39:54.000377
Meg
pythondev_help_Meg_2017-10-23T16:39:54.000377
1,508,776,794.000377
97,951
pythondev
help
<http://aosabook.org/en/sqlalchemy.html>
2017-10-23T16:40:40.000239
Johana
pythondev_help_Johana_2017-10-23T16:40:40.000239
1,508,776,840.000239
97,952
pythondev
help
I really don’t like writing SQL, and only do it when the profiling shows there will be a distinct difference
2017-10-23T16:40:41.000246
Meg
pythondev_help_Meg_2017-10-23T16:40:41.000246
1,508,776,841.000246
97,953
pythondev
help
and that’s after optimizing the db structure, adding indexes, and doing everything to reduce the number of queries executed
2017-10-23T16:41:20.000217
Meg
pythondev_help_Meg_2017-10-23T16:41:20.000217
1,508,776,880.000217
97,954
pythondev
help
<@Meg> obviously I think it's significant, I don't expect everyone would agree with my personal weight of the tradeoffs though.
2017-10-23T16:41:49.000082
Issac
pythondev_help_Issac_2017-10-23T16:41:49.000082
1,508,776,909.000082
97,955
pythondev
help
oh, certainly. Like I said, there’s valid arguments for and against
2017-10-23T16:42:11.000305
Meg
pythondev_help_Meg_2017-10-23T16:42:11.000305
1,508,776,931.000305
97,956
pythondev
help
though, am wondering
2017-10-23T16:42:23.000487
Meg
pythondev_help_Meg_2017-10-23T16:42:23.000487
1,508,776,943.000487
97,957
pythondev
help
I tend to think that, performance aside, anything that decouples the system and facilitates me reasoning algebraically about my application with a minimum of leaky abstractions and corner cases is a huge win to me
2017-10-23T16:42:25.000601
Issac
pythondev_help_Issac_2017-10-23T16:42:25.000601
1,508,776,945.000601
97,958
pythondev
help
but hey, I'm coming from haskell :laughing:
2017-10-23T16:42:47.000096
Issac
pythondev_help_Issac_2017-10-23T16:42:47.000096
1,508,776,967.000096
97,959
pythondev
help
well, time for me to head out, got a meetup to get to
2017-10-23T16:43:57.000078
Meg
pythondev_help_Meg_2017-10-23T16:43:57.000078
1,508,777,037.000078
97,960
pythondev
help
nice talking with you, <@Issac> :slightly_smiling_face:
2017-10-23T16:44:07.000338
Meg
pythondev_help_Meg_2017-10-23T16:44:07.000338
1,508,777,047.000338
97,961
pythondev
help
have a good meetup
2017-10-23T16:44:25.000087
Issac
pythondev_help_Issac_2017-10-23T16:44:25.000087
1,508,777,065.000087
97,962
pythondev
help
thanks! its put on by ESRI, and I know a few of the map/geo people going to be there
2017-10-23T16:44:57.000612
Meg
pythondev_help_Meg_2017-10-23T16:44:57.000612
1,508,777,097.000612
97,963
pythondev
help
Hey all, in Python 2.7, using `unittest.assertRaisesRegexp`, how come it’s giving me false regex matching failures for my own exception classes, but working fine with something like `ValueError`?
2017-10-23T17:33:12.000331
Micaela
pythondev_help_Micaela_2017-10-23T17:33:12.000331
1,508,779,992.000331
97,964
pythondev
help
in the fail message it says: &gt; AssertionError: “Failed to find the needed file(s) matching 00000000-000000-xxxxxxxx” does not match “Failed to find the needed file(s) matching 00000000-000000-xxxxxxxx”
2017-10-23T17:33:48.000164
Micaela
pythondev_help_Micaela_2017-10-23T17:33:48.000164
1,508,780,028.000164
97,965
pythondev
help
When… those… are… identical… ?
2017-10-23T17:34:09.000126
Micaela
pythondev_help_Micaela_2017-10-23T17:34:09.000126
1,508,780,049.000126
97,966
pythondev
help
It’s not some weird text encoding comparison that’s failing, is it?
2017-10-23T17:34:48.000064
Micaela
pythondev_help_Micaela_2017-10-23T17:34:48.000064
1,508,780,088.000064
97,967
pythondev
help
Actually nevermind, it’s something else going wrong somewhere, if I change the code to raise a ValueError instead I still get the same matching fail
2017-10-23T17:41:22.000363
Micaela
pythondev_help_Micaela_2017-10-23T17:41:22.000363
1,508,780,482.000363
97,968
pythondev
help
Can you post the code that is failing?
2017-10-23T17:41:57.000170
Antionette
pythondev_help_Antionette_2017-10-23T17:41:57.000170
1,508,780,517.00017
97,969
pythondev
help
If you have that exact string as the regex to match, you might need to escape the parens
2017-10-23T17:42:40.000466
Antionette
pythondev_help_Antionette_2017-10-23T17:42:40.000466
1,508,780,560.000466
97,970
pythondev
help
Hi guys, i’m writing a module to read/write/edit files on target fpt server, I know python has supported ftp lib. But i just curious if I look for the folder files (in the server) and try doing nasty stuff with the file like normal file, and the ftp server can detect change, update it back to the ftp user? I dont mind write the module in the same server with fpt, but are there any drawback if I go down this path?
2017-10-24T00:19:40.000003
Hermina
pythondev_help_Hermina_2017-10-24T00:19:40.000003
1,508,804,380.000003
97,971
pythondev
help
What book do you guys recommend for a new python programmer that doesn't have programming experience.
2017-10-24T00:25:30.000102
Bruno
pythondev_help_Bruno_2017-10-24T00:25:30.000102
1,508,804,730.000102
97,972
pythondev
help
Think Python 2e.
2017-10-24T00:30:14.000057
Rosa
pythondev_help_Rosa_2017-10-24T00:30:14.000057
1,508,805,014.000057
97,973
pythondev
help
A friend of mine was also new to programming and he found it really helpful.
2017-10-24T00:30:27.000163
Rosa
pythondev_help_Rosa_2017-10-24T00:30:27.000163
1,508,805,027.000163
97,974
pythondev
help
I'd have to go with <https://automatetheboringstuff.com/> , partly because the book is free to read (the physical book still costs), partly because it's accessible and partly because it doesn't presume. It lacks real apps and takes awhile to get going but it's free and if they put down the book for something better, no harm done.
2017-10-24T00:38:32.000092
Seema
pythondev_help_Seema_2017-10-24T00:38:32.000092
1,508,805,512.000092
97,975
pythondev
help
The automate book still assumes you know basic programming concepts
2017-10-24T00:46:14.000136
Bruno
pythondev_help_Bruno_2017-10-24T00:46:14.000136
1,508,805,974.000136
97,976
pythondev
help
the book does? I thought it was based off the site which has things like ```You run the interactive shell by launching IDLE, which you installed with Python in the introduction. On Windows, open the Start menu, select All Programs ▸ Python 3.3, and then select IDLE (Python GUI). On OS X, select Applications ▸ MacPython 3.3 ▸ IDLE. On Ubuntu, open a new Terminal window and enter idle3. A window with the &gt;&gt;&gt; prompt should appear; that’s the interactive shell. Enter 2 + 2 at the prompt to have Python do some simple math. &gt;&gt;&gt; 2 + 2 4 The IDLE window should now show some text like this: Python 3.3.2 (v3.3.2:d047928ae3f6, May 16 2013, 00:06:53) [MSC v.1600 64 bit (AMD64)] on win32 Type "copyright", "credits" or "license()" for more information. &gt;&gt;&gt; 2 + 2 4```
2017-10-24T00:51:00.000098
Seema
pythondev_help_Seema_2017-10-24T00:51:00.000098
1,508,806,260.000098
97,977
pythondev
help
<@Bruno> it doesn't the first part is all about basic concept :slightly_smiling_face:
2017-10-24T02:44:32.000192
Ciera
pythondev_help_Ciera_2017-10-24T02:44:32.000192
1,508,813,072.000192
97,978
pythondev
help
`file(s)` is a regex that matches the string `files`, but not the string `file(s)`.
2017-10-24T04:12:00.000420
Louis
pythondev_help_Louis_2017-10-24T04:12:00.000420
1,508,818,320.00042
97,979
pythondev
help
`assertRaisesRegexp` kinda sucks if you just want to match literal strings, sadly. To do that reliably, you need something like: ``` assertRaisesRegexp(exc, r'^{}$'.format(re.escape(expected_message)), …) ```
2017-10-24T04:14:48.000063
Louis
pythondev_help_Louis_2017-10-24T04:14:48.000063
1,508,818,488.000063
97,980
pythondev
help
Sometimes, it's easier to use the context manager instead: ``` with self.assertRaises(SomeException) as cm: do_something() e = cm.exception &lt;do assertions on e&gt; ```
2017-10-24T04:15:58.000072
Louis
pythondev_help_Louis_2017-10-24T04:15:58.000072
1,508,818,558.000072
97,981
pythondev
help
Hey all, Is there a way for me to be able to run a cleanup script when I use `deactivate` on my python envs? I set some environment variables in my `.env` that I would like to clear away when I'm done working on something.
2017-10-24T04:29:41.000483
Kayce
pythondev_help_Kayce_2017-10-24T04:29:41.000483
1,508,819,381.000483
97,982
pythondev
help
I’m trying to write an application that will search for indentation made with 4 spaces or with tabs, and replace them with 2 spaces. In have found 2 options for doing this: 1. with fileinput, but this seems to have some limitations and I don’t know if it can do what I want it to do 2. with re regular expressions which one would you recommend?
2017-10-24T04:32:05.000337
Ines
pythondev_help_Ines_2017-10-24T04:32:05.000337
1,508,819,525.000337
97,983
pythondev
help
<@Ines> I don't think that regex will help you since the structure of languages is more based off of context sensitive grammars. Regular expressions (finite automata) aren't powerful enough to represent the evaluate same set of strings as CSG's. (I could be wrong though, it's been years since I've done anything with computer science theory).
2017-10-24T04:37:56.000261
Kayce
pythondev_help_Kayce_2017-10-24T04:37:56.000261
1,508,819,876.000261
97,984
pythondev
help
<@Ines> Oh wait, what am I talking about. I totally missed your question (It's well past my bedtime). - get lines from your input file - count how many spaces there are before the first character in the line
2017-10-24T04:44:49.000333
Kayce
pythondev_help_Kayce_2017-10-24T04:44:49.000333
1,508,820,289.000333
97,985
pythondev
help
would you please re-read my question :slightly_smiling_face: I know what I have to do, I just don’t know the easiest(/best) way to do it in python :slightly_smiling_face:
2017-10-24T04:45:53.000231
Ines
pythondev_help_Ines_2017-10-24T04:45:53.000231
1,508,820,353.000231
97,986
pythondev
help
- divide that count by four, then multiply that by two. - take the substring that's all text, prepend ((space_count / 4) * 2)
2017-10-24T04:46:11.000290
Kayce
pythondev_help_Kayce_2017-10-24T04:46:11.000290
1,508,820,371.00029
97,987
pythondev
help
I’d probably recommend simply using a regex, then you don’t have to count
2017-10-24T04:47:36.000159
Junita
pythondev_help_Junita_2017-10-24T04:47:36.000159
1,508,820,456.000159
97,988
pythondev
help
replacing four spaces or a tab with two spaces will work, regardless of how many multiples of four or tabs (or even a mix) there is
2017-10-24T04:47:58.000249
Junita
pythondev_help_Junita_2017-10-24T04:47:58.000249
1,508,820,478.000249
97,989
pythondev
help
<@Junita> thank you! I was thinking about fileinput as that will save me the trouble of opening the file, reading its content, fix the indentation, write back the content, close the file
2017-10-24T04:52:00.000194
Ines
pythondev_help_Ines_2017-10-24T04:52:00.000194
1,508,820,720.000194
97,990
pythondev
help
also note it might just be easier to do this with a tool like sed, unless you want to do use this as a learning exercise
2017-10-24T04:52:48.000087
Junita
pythondev_help_Junita_2017-10-24T04:52:48.000087
1,508,820,768.000087
97,991
pythondev
help
I think it will confuse the regex. Assume the dashes are spaces, and the underscores are the replacement. Such that `----` gets evaluated to `_`. What happens with `--------` (eight spaces)? Does it get evaluated like `_----` then `__`? Or does something weird happen like `-_---` where the regex wont capture the other 4 spaces.
2017-10-24T04:53:18.000154
Kayce
pythondev_help_Kayce_2017-10-24T04:53:18.000154
1,508,820,798.000154
97,992
pythondev
help
but to be honest, I haven’t found how to search for patterns using line.replace… this is why I switched to re
2017-10-24T04:53:54.000251
Ines
pythondev_help_Ines_2017-10-24T04:53:54.000251
1,508,820,834.000251
97,993
pythondev
help
I say screw it and divide space count by two and the append the text.
2017-10-24T04:54:07.000330
Kayce
pythondev_help_Kayce_2017-10-24T04:54:07.000330
1,508,820,847.00033
97,994
pythondev
help
learning exercise and also I want a script that can be used independently
2017-10-24T04:54:30.000172
Ines
pythondev_help_Ines_2017-10-24T04:54:30.000172
1,508,820,870.000172
97,995
pythondev
help
Forsure!
2017-10-24T04:54:48.000068
Kayce
pythondev_help_Kayce_2017-10-24T04:54:48.000068
1,508,820,888.000068
97,996
pythondev
help
One sec, I'm whipping something up
2017-10-24T04:54:54.000234
Kayce
pythondev_help_Kayce_2017-10-24T04:54:54.000234
1,508,820,894.000234
97,997
pythondev
help
&gt; I think it will confuse the regex.
2017-10-24T04:55:34.000325
Suellen
pythondev_help_Suellen_2017-10-24T04:55:34.000325
1,508,820,934.000325
97,998
pythondev
help
I think regexes are not easily confused %)
2017-10-24T04:55:46.000179
Suellen
pythondev_help_Suellen_2017-10-24T04:55:46.000179
1,508,820,946.000179
97,999
pythondev
help
They are a powerful tool with maths 'n stuff backing them up
2017-10-24T04:56:03.000149
Suellen
pythondev_help_Suellen_2017-10-24T04:56:03.000149
1,508,820,963.000149
98,000
pythondev
help
<@Kayce> it’s pretty easy, you make a match (such as `^(\s{4}|\t)`) and then feed it into `re.sub`, which can handle multiple replaces: <https://docs.python.org/3/library/re.html#re.sub>
2017-10-24T04:56:19.000214
Junita
pythondev_help_Junita_2017-10-24T04:56:19.000214
1,508,820,979.000214
98,001
pythondev
help
exactly this
2017-10-24T04:58:33.000480
Suellen
pythondev_help_Suellen_2017-10-24T04:58:33.000480
1,508,821,113.00048
98,002