showvikdbz/code-reviews · Datasets at Hugging Face

text stringlengths 1 5.2k	labels listlengths 1 1
@poettering You're being kind of absurd and stupid here. Supplying a broken timeserver as a default is inane and silly. Please accept the pull request. https://github.com/systemd/systemd/issues/439	[ 1 ]
Don't do this brian wtf	[ 1 ]
damn i want to make an orm someday	[ 1 ]
Der. Absoluten. Fukken.	[ 1 ]
nit remove this is win block and expand the one above to cover this action	[ 0 ]
is not this weird should not they be using include build native test and getting this for free	[ 0 ]
the following cls remove all userptr and mmap buffers only dmabuf will be used	[ 0 ]
ef adn the caller cannot know he is dealing with a usim	[ 0 ]
remove get allowed nic networkgroups here	[ 0 ]
uidevtoolsdomagent const dom agent	[ 0 ]
Why does `SINGLE_LINE` mode use `UIKeyboardTypeEmailAddress`. It's weird.	[ 1 ]
seems slave will be enough here	[ 0 ]
I mean, this is a bit of a wacky solution, but would it work to home X to max instead of min? Doing that should leave the BL-touch in the print area, and then it would work as a z-endstop.	[ 1 ]
it is not part of base extract them to separate files config py or parameters py and capacity log py	[ 0 ]
who the fuck set this on initialize	[ 1 ]
Dang it, those darn herobrines.	[ 1 ]
Well I believe you are using the wrong language then. There is way more rubbish in PHP to make your eyes water than that...	[ 1 ]
this can also make some other thread spuriously start registering as the main thread it also only fools one check but many other things still look like it is the main thread like tls and what the fuck maynotbemainthread so this is a very limited illusion if it is just for your particular tests maybe your code could provide it is own bypass named ismainthreadortest or something that can be overridden ideally though we would actually run it off the main thread since if the code cares about threading we should probably test how it deals with threading this also seems like it would give tsan a better opportunity to detect bugs	[ 0 ]
Thanks @heartsucker, it seems like you were beaten by a few minutes and this was just opened for the same reasons https://github.com/ansible/ansible/issues/32092. I am going to close this issue so that we don't have duplicates opened for the same thing.	[ 1 ]
delete dead code since you are here	[ 0 ]
o sfinae shakes fist	[ 0 ]
what the fuck does this mean	[ 1 ]
not required python will garbage collect it for you	[ 0 ]
same docstring issue here	[ 0 ]
how about this one polling interval	[ 0 ]
argument should be trust id trust id	[ 0 ]
i guess all of this and maybe more is unused code and should be removed	[ 0 ]
done same thing the whole function is useless now	[ 0 ]
can you test without these special run bindmounts the above run run should pull in the entire run filesystem including run lvm and run lvmmetad pid	[ 0 ]
holy fuck genius 10/10 changing asap	[ 1 ]
yeah but it affects the outcome of the method because it leaves instance device metadata unset but looking at instance save it looks like it will be handled properly so that is fine	[ 0 ]
stupid question will this data stream always always match the data written on line and on or is it possible that we will in a later version change the data we write read and then this can go crazy if that is a possibility i would recommend making the first field you write out an integer currently to state that this is serialization format in the read code read the int back and make sure it is if it is not do not attempt to read the object stream since it is written by a later tool	[ 0 ]
host state fakes fakehoststate host compute should probably be moved to setup as it is the same for all tests	[ 0 ]
Damn, forgot abut this, will commit a fix.	[ 1 ]
This is getting ridiculous.	[ 1 ]
yep just like the health monitor controller you re really into killing bogons are not you i will get it added as i have done for the rest	[ 0 ]
small nit put bl status watchdog above bl status csum valid to maintain the bit order above	[ 0 ]
How could I say that ? Please fix your translation... Most of it is a complete nonsense.	[ 1 ]
Any update on this? It's getting ridiculous at this point it's been 2 weeks and no fix?	[ 1 ]
add a dummy javadoc	[ 0 ]
oops this is a useless comment i forgot to remove	[ 0 ]
this is a really complicated data structure re think your test design here from your comment this seems like it should be as follows a pid should have a mapping to it is averages across timesteps so this could be type foo set pid vector double then all you need to have is a set foo s	[ 0 ]
i am not sure about this it is probably not a good idea to make germd willing to kill any pid that gets sent to it even if it is not something that the current process started	[ 0 ]
This would seem to confirm my own findings: some ESP8266 boards are rubbish. I'm sure you'll agree that unless a problem can be replicated on the reference board it can't be considered a bug. I wonder if your "wemos?" board is actually from WeMos. Their D1 mini is good (in my experience).	[ 1 ]
ahh, damn, I'm breaking one of my conventions. fixed	[ 1 ]
> Can this get merged?? I'd be happy to know myself, I stopped contributing to PowerShell because of ridiculous review times. It's not the only MR I have stuck in review for over 6 months.	[ 1 ]
there is no need for a mutex any function you call after opening the hw module will be executed after the initializers are finished i have double checked this	[ 0 ]
typo in commit message useles useless	[ 0 ]
`http`? `digicert`? No clue what this is, but it reads weird :)	[ 1 ]
the character here is not necessary and cold potentialy because problems	[ 0 ]
stephen do we need an allow rule here for execution of system bin run as runas exec	[ 0 ]
this can also make some other thread spuriously start registering as the main thread it also only fools one check but many other things still look like it is the main thread like tls and what the fuck maynotbemainthread so this is a very limited illusion if it is just for your particular tests maybe your code could provide it is own bypass named ismainthreadortest or something that can be overridden ideally though we would actually run it off the main thread since if the code cares about threading we should probably test how it deals with threading this also seems like it would give tsan a better opportunity to detect bugs	[ 0 ]
jb sigmask should be jblen	[ 0 ]
true though any app doing work while installing a new split is insane and if you are insane there are already enough ways to crash yourself the copyof sounds like a good fix for this though	[ 0 ]
is there maybe any way to check or cleanup lxd and btrfs database manually? because when i re-install lxd, it is back to ~10 second migration but after 20 migrations its back to ridiculous time again	[ 1 ]
right : D damn I wanted to use const for the first time	[ 1 ]
nullptr	[ 0 ]
Say what you will, though you're getting pretty darn good with TB :)	[ 1 ]
you could make this final and remove the null assigment	[ 0 ]
holy moly cipsoft, what the hell XD	[ 1 ]
nitpick trailing spaces here and below and in util sh	[ 0 ]
apologies if this is another silly observation the pipeline has the proxy logging filter twice what s the reason for that	[ 0 ]
convert all subprocess to cros build lib runcommand	[ 0 ]
this should be delete oper delete	[ 0 ]
ditto but particularly in this case a user id may not be a uuid if it is coming from ldap or some other backend	[ 0 ]
i'm laughing a fucking lot holy shit i will use this and get immediately slapped	[ 1 ]
also looks like you can just get rid of mapsize and use page size instead	[ 0 ]
and fuck those vars too.	[ 1 ]
i guess google c naming would be hkdfsha not that we are remotely consistent about this personally i think hkdf and sha look silly but ah well	[ 0 ]
can we have unit tests equivalent to unifiedmessagecenterviewtest stackingcounterlayout unifiedmessagecenterviewtest stackingcounternotaffectingmessageviewbounds unifiedmessagecenterviewtest stackingcounterremovedwithnotifications	[ 0 ]
is puppet openstack still relevant should we continue mentioning it here	[ 0 ]
done ouch awfull mistake	[ 0 ]
i think invalid nic argument s s can also be part of nic error and thus wud help u to get rid of err msg	[ 0 ]
this test failed on my saio because i already had a container in the test account is it possible to perform the check and be tolerant of existing containers	[ 0 ]
user id conf castellan user id	[ 0 ]
this will never get called because the proto calling functions will not get here they dump the protos cached then return just put btif debug bond event dump stats body in here	[ 0 ]
this should be a warning or info msg	[ 0 ]
line needs a modularization def has device dut vid pid also why not use lsusb d that will make the code very simple	[ 0 ]
Technically you could set initialData to your data and then set the staleTime to something ridiculous. Then your data will never refresh and always pull from the cache.	[ 1 ]
i would suggest we call this self module registry	[ 0 ]
it was used by the unit test but i have just realized i was being a giant idiot and refactored it away	[ 0 ]
What the fuck happened here?	[ 1 ]
iqn self test volume	[ 0 ]
you can stop inheriting from test l agent scheduler l schedulerbasemixin then in line use helpers set agent admin state simpler less usages of that stupid mixin win win	[ 0 ]
this is a little lame returning a string i will change it to a mock	[ 0 ]
i think the adb bug is fixed now did you try the old adb shell rm instead of adb shell sh c	[ 0 ]
this is dumb we have to wait a second or before calling the confirm resize of the instance because we will get an error from the call saying the resize failed even though it had not problem this is likely because the task description or id is set after the verify resize state shows up this needs a little more research on why	[ 0 ]
There is more info on HeadOutlet issues at #37293. Overall it appears to be a very clumsy solution.	[ 1 ]
Oh. Also this fucking runtimes if the sharer ISN'T datum/gas_mixture/immutable	[ 1 ]
For what it's worth, I have a simply ridiculous habit of purchasing cards whose max long since passed. I have no idea why I do it. But some kind of visual indicator that the requirement is unmet might save my bacon the next time I try to do this.	[ 1 ]
we do not have procfs h on android if we add it on bionic and platforms this hack and couple of others are not required it should be added to old platforms as well to remove these hacks completely note that glibc comment for procfs h says anyway the whole purpose of this file is for gdb and gdb only don t read too much into it don t use it for anything other than gdb unless you know what you are doing	[ 0 ]
sp tx vo state vo wait tx video stable so you can remove this and put the break in the else block above	[ 0 ]
you should use versionutils to report that this is deprecated you can see an example in cinder scheduler manager py	[ 0 ]
nit policy m get network policy is already a mock so you can do self policy m get network policy return value old qos policy	[ 0 ]
Wait shit wrong pr this isnt the laser beam	[ 1 ]
this has to be the or else gerrit will not know to track the submodule also it is a stupid broken git ini file so i think it is supposed to have leading tabs	[ 0 ]
nit you do not need any of get calls here as there will both of these hashes in any case	[ 0 ]
i have set the mod wsgi home to horizon dir so this should not be a problem	[ 0 ]
good point i will whip something up	[ 0 ]
This commit is causing abnormally long boot times in games. Could we get a fix, or if it will take a while, a revert until it's fixed. It's a bit ridiculous to be waiting nearly 20 seconds for game to start.	[ 1 ]

@poettering You're being kind of absurd and stupid here. Supplying a broken timeserver as a default is inane and silly. Please accept the pull request. https://github.com/systemd/systemd/issues/439

[ 1 ]

Don't do this brian wtf

[ 1 ]

damn i want to make an orm someday

[ 1 ]

Der. Absoluten. Fukken.

[ 1 ]

nit remove this is win block and expand the one above to cover this action

[ 0 ]

is not this weird should not they be using include build native test and getting this for free

[ 0 ]

the following cls remove all userptr and mmap buffers only dmabuf will be used

[ 0 ]

ef adn the caller cannot know he is dealing with a usim

[ 0 ]

remove get allowed nic networkgroups here

[ 0 ]

uidevtoolsdomagent const dom agent

[ 0 ]

Why does `SINGLE_LINE` mode use `UIKeyboardTypeEmailAddress`. It's weird.

[ 1 ]

seems slave will be enough here

[ 0 ]

I mean, this is a bit of a wacky solution, but would it work to home X to max instead of min? Doing that should leave the BL-touch in the print area, and then it would work as a z-endstop.

[ 1 ]

it is not part of base extract them to separate files config py or parameters py and capacity log py

[ 0 ]

who the fuck set this on initialize

[ 1 ]

Dang it, those darn herobrines.

[ 1 ]

Well I believe you are using the wrong language then. There is way more rubbish in PHP to make your eyes water than that...

[ 1 ]

this can also make some other thread spuriously start registering as the main thread it also only fools one check but many other things still look like it is the main thread like tls and what the fuck maynotbemainthread so this is a very limited illusion if it is just for your particular tests maybe your code could provide it is own bypass named ismainthreadortest or something that can be overridden ideally though we would actually run it off the main thread since if the code cares about threading we should probably test how it deals with threading this also seems like it would give tsan a better opportunity to detect bugs

[ 0 ]

Thanks @heartsucker, it seems like you were beaten by a few minutes and this was just opened for the same reasons https://github.com/ansible/ansible/issues/32092. I am going to close this issue so that we don't have duplicates opened for the same thing.

[ 1 ]

delete dead code since you are here

[ 0 ]

o sfinae shakes fist

[ 0 ]

what the fuck does this mean

[ 1 ]

not required python will garbage collect it for you

[ 0 ]

same docstring issue here

[ 0 ]

how about this one polling interval

[ 0 ]

argument should be trust id trust id

[ 0 ]

i guess all of this and maybe more is unused code and should be removed

[ 0 ]

done same thing the whole function is useless now

[ 0 ]

can you test without these special run bindmounts the above run run should pull in the entire run filesystem including run lvm and run lvmmetad pid

[ 0 ]

holy fuck genius 10/10 changing asap

[ 1 ]

yeah but it affects the outcome of the method because it leaves instance device metadata unset but looking at instance save it looks like it will be handled properly so that is fine

[ 0 ]

stupid question will this data stream always always match the data written on line and on or is it possible that we will in a later version change the data we write read and then this can go crazy if that is a possibility i would recommend making the first field you write out an integer currently to state that this is serialization format in the read code read the int back and make sure it is if it is not do not attempt to read the object stream since it is written by a later tool

[ 0 ]

host state fakes fakehoststate host compute should probably be moved to setup as it is the same for all tests

[ 0 ]

Damn, forgot abut this, will commit a fix.

[ 1 ]

This is getting ridiculous.

[ 1 ]

yep just like the health monitor controller you re really into killing bogons are not you i will get it added as i have done for the rest

[ 0 ]

small nit put bl status watchdog above bl status csum valid to maintain the bit order above

[ 0 ]

How could I say that ? Please fix your translation... Most of it is a complete nonsense.

[ 1 ]

Any update on this? It's getting ridiculous at this point it's been 2 weeks and no fix?

[ 1 ]

add a dummy javadoc

[ 0 ]

oops this is a useless comment i forgot to remove

[ 0 ]

this is a really complicated data structure re think your test design here from your comment this seems like it should be as follows a pid should have a mapping to it is averages across timesteps so this could be type foo set pid vector double then all you need to have is a set foo s

[ 0 ]

i am not sure about this it is probably not a good idea to make germd willing to kill any pid that gets sent to it even if it is not something that the current process started

[ 0 ]

This would seem to confirm my own findings: some ESP8266 boards are rubbish. I'm sure you'll agree that unless a problem can be replicated on the reference board it can't be considered a bug. I wonder if your "wemos?" board is actually from WeMos. Their D1 mini is good (in my experience).

[ 1 ]

ahh, damn, I'm breaking one of my conventions. fixed

[ 1 ]

> Can this get merged?? I'd be happy to know myself, I stopped contributing to PowerShell because of ridiculous review times. It's not the only MR I have stuck in review for over 6 months.

[ 1 ]

there is no need for a mutex any function you call after opening the hw module will be executed after the initializers are finished i have double checked this

[ 0 ]

typo in commit message useles useless

[ 0 ]

`http`? `digicert`? No clue what this is, but it reads weird :)

[ 1 ]

the character here is not necessary and cold potentialy because problems

[ 0 ]

stephen do we need an allow rule here for execution of system bin run as runas exec

[ 0 ]

this can also make some other thread spuriously start registering as the main thread it also only fools one check but many other things still look like it is the main thread like tls and what the fuck maynotbemainthread so this is a very limited illusion if it is just for your particular tests maybe your code could provide it is own bypass named ismainthreadortest or something that can be overridden ideally though we would actually run it off the main thread since if the code cares about threading we should probably test how it deals with threading this also seems like it would give tsan a better opportunity to detect bugs

[ 0 ]

jb sigmask should be jblen

[ 0 ]

true though any app doing work while installing a new split is insane and if you are insane there are already enough ways to crash yourself the copyof sounds like a good fix for this though

[ 0 ]

is there maybe any way to check or cleanup lxd and btrfs database manually? because when i re-install lxd, it is back to ~10 second migration but after 20 migrations its back to ridiculous time again

[ 1 ]

right : D damn I wanted to use const for the first time

[ 1 ]

nullptr

[ 0 ]

Say what you will, though you're getting pretty darn good with TB :)

[ 1 ]

you could make this final and remove the null assigment

[ 0 ]

holy moly cipsoft, what the hell XD

[ 1 ]

nitpick trailing spaces here and below and in util sh

[ 0 ]

apologies if this is another silly observation the pipeline has the proxy logging filter twice what s the reason for that

[ 0 ]

convert all subprocess to cros build lib runcommand

[ 0 ]

this should be delete oper delete

[ 0 ]

ditto but particularly in this case a user id may not be a uuid if it is coming from ldap or some other backend

[ 0 ]

i'm laughing a fucking lot holy shit i will use this and get immediately slapped

[ 1 ]

also looks like you can just get rid of mapsize and use page size instead

[ 0 ]

and fuck those vars too.

[ 1 ]

i guess google c naming would be hkdfsha not that we are remotely consistent about this personally i think hkdf and sha look silly but ah well

[ 0 ]

can we have unit tests equivalent to unifiedmessagecenterviewtest stackingcounterlayout unifiedmessagecenterviewtest stackingcounternotaffectingmessageviewbounds unifiedmessagecenterviewtest stackingcounterremovedwithnotifications

[ 0 ]

is puppet openstack still relevant should we continue mentioning it here

[ 0 ]

done ouch awfull mistake

[ 0 ]

i think invalid nic argument s s can also be part of nic error and thus wud help u to get rid of err msg

[ 0 ]

this test failed on my saio because i already had a container in the test account is it possible to perform the check and be tolerant of existing containers

[ 0 ]

user id conf castellan user id

[ 0 ]

this will never get called because the proto calling functions will not get here they dump the protos cached then return just put btif debug bond event dump stats body in here

[ 0 ]

this should be a warning or info msg

[ 0 ]

line needs a modularization def has device dut vid pid also why not use lsusb d that will make the code very simple

[ 0 ]

Technically you could set initialData to your data and then set the staleTime to something ridiculous. Then your data will never refresh and always pull from the cache.

[ 1 ]

i would suggest we call this self module registry

[ 0 ]

it was used by the unit test but i have just realized i was being a giant idiot and refactored it away

[ 0 ]

What the fuck happened here?

[ 1 ]

iqn self test volume

[ 0 ]

you can stop inheriting from test l agent scheduler l schedulerbasemixin then in line use helpers set agent admin state simpler less usages of that stupid mixin win win

[ 0 ]

this is a little lame returning a string i will change it to a mock

[ 0 ]

i think the adb bug is fixed now did you try the old adb shell rm instead of adb shell sh c

[ 0 ]

this is dumb we have to wait a second or before calling the confirm resize of the instance because we will get an error from the call saying the resize failed even though it had not problem this is likely because the task description or id is set after the verify resize state shows up this needs a little more research on why

[ 0 ]

There is more info on HeadOutlet issues at #37293. Overall it appears to be a very clumsy solution.

[ 1 ]

Oh. Also this fucking runtimes if the sharer ISN'T datum/gas_mixture/immutable

[ 1 ]

For what it's worth, I have a simply *ridiculous* habit of purchasing cards whose max long since passed. I have no idea why I do it. But some kind of visual indicator that the requirement is unmet might save my bacon the next time I try to do this.

[ 1 ]

we do not have procfs h on android if we add it on bionic and platforms this hack and couple of others are not required it should be added to old platforms as well to remove these hacks completely note that glibc comment for procfs h says anyway the whole purpose of this file is for gdb and gdb only don t read too much into it don t use it for anything other than gdb unless you know what you are doing

[ 0 ]

sp tx vo state vo wait tx video stable so you can remove this and put the break in the else block above

[ 0 ]

you should use versionutils to report that this is deprecated you can see an example in cinder scheduler manager py

[ 0 ]

nit policy m get network policy is already a mock so you can do self policy m get network policy return value old qos policy

[ 0 ]

Wait shit wrong pr this isnt the laser beam

[ 1 ]

this has to be the or else gerrit will not know to track the submodule also it is a stupid broken git ini file so i think it is supposed to have leading tabs

[ 0 ]

nit you do not need any of get calls here as there will both of these hashes in any case

[ 0 ]

i have set the mod wsgi home to horizon dir so this should not be a problem

[ 0 ]

good point i will whip something up

[ 0 ]

This commit is causing abnormally long boot times in games. Could we get a fix, or if it will take a while, a revert until it's fixed. It's a bit ridiculous to be waiting nearly 20 seconds for game to start.

[ 1 ]