showvikdbz/code-reviews · Datasets at Hugging Face

text stringlengths 1 5.2k	labels listlengths 1 1
is it really useful since we have isinvalid i suggest only keep isvalid and remove isinvalid	[ 0 ]
i do not think so as far as i understand if we kill data on multipath device it will be automatically vanished from all underlying devices	[ 0 ]
nit place public methods before private methods instead of interspersing them	[ 0 ]
won t this line crash with attributeerror since you do not have those methods start stop on your fake manager	[ 0 ]
Yuck. It's 2022. I can't believe we're still dealing with proxy issues like it's 1999. Shouldn't the O/S be responsible for proxy config at this point? Why does every network-capable software need to deal with proxies?	[ 1 ]
don t store this into self it is useless	[ 0 ]
I kinda wanna leave it in for further testing just because of how much of a bitch it is rn, although it can be removed if that's a demand	[ 1 ]
oh look someone who was calling hash cleanup listdir with positional args	[ 0 ]
why the fuck is it a hatchet?	[ 1 ]
should this be sh	[ 0 ]
So you are very damn right all about flags but `do_after` still needs proper calls	[ 1 ]
god dammit, blame d3athrow	[ 1 ]
An integer limit of 64 digits is ridiculous. Personally I prefer removing the integer restriction. Free text field to allow creativity/expression.	[ 1 ]
damn I was sure I did that though, must have messed up a commit	[ 1 ]
rules for init go in init te you do not need the domain trans for dhcp	[ 0 ]
this seems useless i think you can do something like return iter set itertools chain	[ 0 ]
i was not even sure why i include obj obj puffin done apparently i am a lousy makefile writer p	[ 0 ]
i hope we can get rid of the trailing p	[ 0 ]
path should be dummy networking sfc services sfc drivers dummy dummydriver	[ 0 ]
refer to you seem talking about restrictions but there are a lot more restrictions in the spec we do not need to explain all of this kind of restrictions	[ 0 ]
> ridiculous thing, NOTHING got after two weeks. > each day has 2T Plot adding. > ridiculous thing, NOTHING got after two weeks. > each day has 2T Plot adding. I add 200+ T plots everyday\U0001f623	[ 1 ]
always redirect into tmp then mv f tmp otherwise the redirection creates the target even on failure	[ 0 ]
you are totally right the dumb guy who did that was me i am repenting here already c modules openstack project files zuul layout yaml	[ 0 ]
endif android	[ 0 ]
small performance optimization instead of getting ref you should use count and if count is raise exception	[ 0 ]
useless trailing rand name already adds a	[ 0 ]
i am fine with removing that revocations do refresh the cache of the revocation list done in the next patchset	[ 0 ]
nit should be without and	[ 0 ]
After whacking my brain it seems that the commented out code for Sec Berets in the job/security file somehow was screwing with this as well, hopefully I've fixed this.	[ 1 ]
what exception this is useless	[ 0 ]
oh, herp de fucking derp.	[ 1 ]
but why? Could be that it clicks, you go 'oh fuck', and throw it and it does nothing, it's only a minor failure not a critical failure	[ 1 ]
okay u just burned my brain please call it in normal way for flavor args in self config or for flavor config in self config or something like that	[ 0 ]
thanks actually i should remove them because locked is not returned through api	[ 0 ]
nit you should put this new arg as the last one as in test linux ip lib py	[ 0 ]
I have no artistic skills whatsoever, so I won't disgrace this thread with something I cobble together, but I think something like a combination of the SE stack and the current fire from SD's avatar would be cool	[ 1 ]
If you fucking get a mecha drill and use it for this, you deserve to have the ability.	[ 1 ]
good ol paprika, adding a var to all jumpsuit to resolve one fucking issue that he created. Good job... All that was needed was two lines in update_inv_w_uniform().	[ 1 ]
this might make some additional things dirty since if only the virtuals were dirty and the directs were not now all methods would be marked dirty	[ 0 ]
Since this was written, we have gained the federation tester (`https://matrix.org/federationtester/api/report?server_name=<...>`), but it still lacks a UI (https://github.com/matrix-org/matrix-federation-tester/issues/6). The point stands that the errors we get are crap.	[ 1 ]
top level url	[ 0 ]
similar to create recordset could you do a recordset self get priority recordset here instead of calling self get recordset	[ 0 ]
is omnibox output an input section this css class seems to be applying some bottom margin at but the name of it seems inaccuarate no how about simply changing this to omnibox inputs omnibox outputs margin bottom and remove the css class	[ 0 ]
rejected - nonsense	[ 1 ]
nit end with a colon	[ 0 ]
That's weird, because these variables aren't used.	[ 1 ]
theres just a shit ton of boilerplate for this tbh. not opposed to these being in since its already coded but for the future this probably needs an easier way to do it	[ 1 ]
@mhegazy Argh, good point! I have VS on D: but in a different folder, but I also have rubbish `D:\\Program Files (x86)\\Microsoft Visual Studio 14.0` left from other installation of windows %). Now everything works, thanks!	[ 1 ]
i do not think that is needed base timeticks and what the fuck timeticks are just aliases and can be assigned directly	[ 0 ]
not using grubby as per above comment	[ 0 ]
freaking hate that rule	[ 1 ]
dude, don't be salty, it's a one line fix	[ 1 ]
A couple of nits with respect to the original (C&C) mission: - [x] The AI doesn't repair damaged structures. - [x] The airstrikes are absolutely ridiculous. The original at its fastest speed gave one a10 every 2 minutes. We give 3 a10's every 15 seconds.	[ 1 ]
again naive use of strncpy can lead to problems in the future	[ 0 ]
str dict looks a bit like an overkill but i can see how it is more bullet proof this way	[ 0 ]
after the while getopts loop you need this piece of boilerplate shift optind	[ 0 ]
Sucks that we can't optimize away this retain.. I guess that's C++ for you :(	[ 1 ]
once we fix all of them something like the marvell code above should become fatal	[ 0 ]
Yea that isn't right, this looks fucked. Good catch.	[ 1 ]
you do not need this or make it more obvious if container id in self container cache return self container cache container id	[ 0 ]
Oh geez, how did I even do that?	[ 1 ]
Most of this is a copy of Unitology, should be Diablo stuff right?	[ 1 ]
this was the code before, am gonna revamp, but just added 'this->' I think. edit: oh whoops, I see what you mean now. Yeah my bad.	[ 1 ]
This is how it's been prior to the PR, you've always been able to suck monkeys with ckeys	[ 1 ]
on python if ctx request id is a str request id startswith b req fails with a typeerror python req x startswith b req typeerror startswith first arg must be str or a tuple of str not bytes you can try by modifying generate request id of oslo context context py of tox py virtual environment to return a string	[ 0 ]
That change is nonsense, you removed the `==`.	[ 1 ]
this is probably pedantic but is not while true the pythonic idiom or is it really while or is there any preference at all sorry to ask silly questions	[ 0 ]
Ah crap, I knew I forgot something.	[ 1 ]
put all these into ch dashboard rst	[ 0 ]
nit boot device cimc to ironic boot device get first device rn if first device else else none	[ 0 ]
this is a dumb question but this is only called from pxedeploy deploy i was wondering why it was put into a method here maybe it will be used by other stuff later	[ 0 ]
Damn, I see now, yeah, then `\|\|` _is_ needed.	[ 1 ]
please remove unnecessary feel free to do in separate commit better to use count	[ 0 ]
looks like this may not be the right driver to use as a dummy driver for driver loading tests used for security groups is there a better one to use or should a dummy class be created here	[ 0 ]
not actually sure this is useful under what scenario would a developer not want to stop mock patchers on teardown	[ 0 ]
this is shitty code. just do spawn_reagent = pick("polonium", etc) and then reagents.add_reagent(spawn_reagent), you buttface.	[ 1 ]
people have been more anal about "code standards" lately. I always liked single line if() changes all being on the same line, but whatever.	[ 1 ]
yeah this is silly isscreenon does not necessarily mean what you think it means what if there were multiple screens you should be stopping progress from onpause instead the activitymanager will take care of pausing activities when the device goes to sleep	[ 0 ]
label these e g nullptr salt salt len	[ 0 ]
the docstring now clarifies the metadata error bit but the reason backtrack somehow makes portage ignore metadata failures or eapi masking is not quite clear i assume it is portage idiocy either way i was more aiming for a comment right around this line clarifying why backtrack was suppressed if it is to suppress fallback behaviour for metadata errors that is a bit opaque	[ 0 ]
that seems lame in java too but i guess that ship has sailed	[ 0 ]
oh it looks better in my pc chrome i am interested in why i can see a useless indent in my ios edition	[ 0 ]
This needs to be put back somewhere, or nations is more fucked than it already is.	[ 1 ]
Ah, nice catch, I introduced that when trying to refactor the ass code but the code seemed unique enough that bringing it to the same standard didn't seem like an option	[ 1 ]
shebang is sh in the script and x mode it will run from shell but not within the tox ini without sh before it not sure why	[ 0 ]
Damn, I tried to make it as a suggestion you could just accept :/	[ 1 ]
optional use this signedin instead since that is the actual property triggering this change handler	[ 0 ]
i am somewhat of a fan of still having the useless just to make it obvious that we are setting a field but i guess that is a matter of opinion	[ 0 ]
yeah maybe we can kill the empty buffers where they get created	[ 0 ]
why not file dat ass trip n for s in f readlines	[ 0 ]
consider indenting the wrapped lines of the bullets to fix restview syntax	[ 0 ]
wrap testr sh does not exist	[ 0 ]
here should be admin snapshots client reset snapshot status	[ 0 ]
this could be just return verify result line is useless	[ 0 ]
why spawn an external kill process rather than using os kill	[ 0 ]
Oops, screw up on my part!	[ 1 ]
I fucked up and forgot this. Thanks for fixing that.	[ 1 ]
god this is even worse. doing COLOR_BRIGHT_LIME_GREEN	[ 1 ]
Of course, what nonsense, I invite @Tolriq to change a line and we don't waste any more time.	[ 1 ]
I did the hard reset and even restarted everything but this bug is there (along with a shit ton more). Pretty much this extension is just not working for me with all these bugs in 2.x. I fell back to 1.5 to have the same bugs.	[ 1 ]

is it really useful since we have isinvalid i suggest only keep isvalid and remove isinvalid

[ 0 ]

i do not think so as far as i understand if we kill data on multipath device it will be automatically vanished from all underlying devices

[ 0 ]

nit place public methods before private methods instead of interspersing them

[ 0 ]

won t this line crash with attributeerror since you do not have those methods start stop on your fake manager

[ 0 ]

Yuck. It's 2022. I can't believe we're still dealing with proxy issues like it's 1999. Shouldn't the O/S be responsible for proxy config at this point? Why does every network-capable software need to deal with proxies?

[ 1 ]

don t store this into self it is useless

[ 0 ]

I *kinda* wanna leave it in for further testing just because of how much of a bitch it is rn, although it can be removed if that's a demand

[ 1 ]

oh look someone who was calling hash cleanup listdir with positional args

[ 0 ]

why the fuck is it a hatchet?

[ 1 ]

should this be sh

[ 0 ]

So **you are very damn right all about flags** but `do_after` still needs proper calls

[ 1 ]

god dammit, blame d3athrow

[ 1 ]

An integer limit of 64 digits is ridiculous. Personally I prefer removing the integer restriction. Free text field to allow creativity/expression.

[ 1 ]

damn I was sure I did that though, must have messed up a commit

[ 1 ]

rules for init go in init te you do not need the domain trans for dhcp

[ 0 ]

this seems useless i think you can do something like return iter set itertools chain

[ 0 ]

i was not even sure why i include obj obj puffin done apparently i am a lousy makefile writer p

[ 0 ]

i hope we can get rid of the trailing p

[ 0 ]

path should be dummy networking sfc services sfc drivers dummy dummydriver

[ 0 ]

refer to you seem talking about restrictions but there are a lot more restrictions in the spec we do not need to explain all of this kind of restrictions

[ 0 ]

> ridiculous thing, NOTHING got after two weeks. > each day has 2T Plot adding. > ridiculous thing, NOTHING got after two weeks. > each day has 2T Plot adding. I add 200+ T plots everyday\U0001f623

[ 1 ]

always redirect into tmp then mv f tmp otherwise the redirection creates the target even on failure

[ 0 ]

you are totally right the dumb guy who did that was me i am repenting here already c modules openstack project files zuul layout yaml

[ 0 ]

endif android

[ 0 ]

small performance optimization instead of getting ref you should use count and if count is raise exception

[ 0 ]

useless trailing rand name already adds a

[ 0 ]

i am fine with removing that revocations do refresh the cache of the revocation list done in the next patchset

[ 0 ]

nit should be without and

[ 0 ]

After whacking my brain it seems that the commented out code for Sec Berets in the job/security file somehow was screwing with this as well, hopefully I've fixed this.

[ 1 ]

what exception this is useless

[ 0 ]

oh, herp de fucking derp.

[ 1 ]

but why? Could be that it clicks, you go 'oh fuck', and throw it and it does nothing, it's only a minor failure not a critical failure

[ 1 ]

okay u just burned my brain please call it in normal way for flavor args in self config or for flavor config in self config or something like that

[ 0 ]

thanks actually i should remove them because locked is not returned through api

[ 0 ]

nit you should put this new arg as the last one as in test linux ip lib py

[ 0 ]

I have no artistic skills whatsoever, so I won't disgrace this thread with something I cobble together, but I think something like a combination of the SE stack and the current fire from SD's avatar would be cool

[ 1 ]

If you fucking get a mecha drill and use it for this, you deserve to have the ability.

[ 1 ]

good ol paprika, adding a var to all jumpsuit to resolve one fucking issue that he created. Good job... All that was needed was two lines in update_inv_w_uniform().

[ 1 ]

this might make some additional things dirty since if only the virtuals were dirty and the directs were not now all methods would be marked dirty

[ 0 ]

Since this was written, we have gained the federation tester (`https://matrix.org/federationtester/api/report?server_name=<...>`), but it still lacks a UI (https://github.com/matrix-org/matrix-federation-tester/issues/6). The point stands that the errors we get are crap.

[ 1 ]

top level url

[ 0 ]

similar to create recordset could you do a recordset self get priority recordset here instead of calling self get recordset

[ 0 ]

is omnibox output an input section this css class seems to be applying some bottom margin at but the name of it seems inaccuarate no how about simply changing this to omnibox inputs omnibox outputs margin bottom and remove the css class

[ 0 ]

rejected - nonsense

[ 1 ]

nit end with a colon

[ 0 ]

That's weird, because these variables aren't used.

[ 1 ]

theres just a shit ton of boilerplate for this tbh. not opposed to these being in since its already coded but for the future this probably needs an easier way to do it

[ 1 ]

@mhegazy Argh, good point! I have VS on D: but in a different folder, but I also have rubbish `D:\\Program Files (x86)\\Microsoft Visual Studio 14.0` left from other installation of windows %). Now everything works, thanks!

[ 1 ]

i do not think that is needed base timeticks and what the fuck timeticks are just aliases and can be assigned directly

[ 0 ]

not using grubby as per above comment

[ 0 ]

freaking hate that rule

[ 1 ]

dude, don't be salty, it's a one line fix

[ 1 ]

A couple of nits with respect to the original (C&C) mission: - [x] The AI doesn't repair damaged structures. - [x] The airstrikes are absolutely ridiculous. The original at its fastest speed gave one a10 every 2 minutes. We give 3 a10's every 15 seconds.

[ 1 ]

again naive use of strncpy can lead to problems in the future

[ 0 ]

str dict looks a bit like an overkill but i can see how it is more bullet proof this way

[ 0 ]

after the while getopts loop you need this piece of boilerplate shift optind

[ 0 ]

Sucks that we can't optimize away this retain.. I guess that's C++ for you :(

[ 1 ]

once we fix all of them something like the marvell code above should become fatal

[ 0 ]

Yea that isn't right, this looks fucked. Good catch.

[ 1 ]

you do not need this or make it more obvious if container id in self container cache return self container cache container id

[ 0 ]

Oh geez, how did I even do that?

[ 1 ]

Most of this is a copy of Unitology, should be Diablo stuff right?

[ 1 ]

this was the code before, am gonna revamp, but just added 'this->' I think. edit: oh whoops, I see what you mean now. Yeah my bad.

[ 1 ]

This is how it's been prior to the PR, you've always been able to suck monkeys with ckeys

[ 1 ]

on python if ctx request id is a str request id startswith b req fails with a typeerror python req x startswith b req typeerror startswith first arg must be str or a tuple of str not bytes you can try by modifying generate request id of oslo context context py of tox py virtual environment to return a string

[ 0 ]

That change is nonsense, you removed the `==`.

[ 1 ]

this is probably pedantic but is not while true the pythonic idiom or is it really while or is there any preference at all sorry to ask silly questions

[ 0 ]

Ah crap, I knew I forgot something.

[ 1 ]

put all these into ch dashboard rst

[ 0 ]

nit boot device cimc to ironic boot device get first device rn if first device else else none

[ 0 ]

this is a dumb question but this is only called from pxedeploy deploy i was wondering why it was put into a method here maybe it will be used by other stuff later

[ 0 ]

Damn, I see now, yeah, then `||` _is_ needed.

[ 1 ]

please remove unnecessary feel free to do in separate commit better to use count

[ 0 ]

looks like this may not be the right driver to use as a dummy driver for driver loading tests used for security groups is there a better one to use or should a dummy class be created here

[ 0 ]

not actually sure this is useful under what scenario would a developer not want to stop mock patchers on teardown

[ 0 ]

this is shitty code. just do spawn_reagent = pick("polonium", etc) and then reagents.add_reagent(spawn_reagent), you buttface.

[ 1 ]

people have been more anal about "code standards" lately. I always liked single line if() changes all being on the same line, but whatever.

[ 1 ]

yeah this is silly isscreenon does not necessarily mean what you think it means what if there were multiple screens you should be stopping progress from onpause instead the activitymanager will take care of pausing activities when the device goes to sleep

[ 0 ]

label these e g nullptr salt salt len

[ 0 ]

the docstring now clarifies the metadata error bit but the reason backtrack somehow makes portage ignore metadata failures or eapi masking is not quite clear i assume it is portage idiocy either way i was more aiming for a comment right around this line clarifying why backtrack was suppressed if it is to suppress fallback behaviour for metadata errors that is a bit opaque

[ 0 ]

that seems lame in java too but i guess that ship has sailed

[ 0 ]

oh it looks better in my pc chrome i am interested in why i can see a useless indent in my ios edition

[ 0 ]

This needs to be put back somewhere, or nations is more fucked than it already is.

[ 1 ]

Ah, nice catch, I introduced that when trying to refactor the ass code but the code seemed unique enough that bringing it to the same standard didn't seem like an option

[ 1 ]

shebang is sh in the script and x mode it will run from shell but not within the tox ini without sh before it not sure why

[ 0 ]

Damn, I tried to make it as a suggestion you could just accept :/

[ 1 ]

optional use this signedin instead since that is the actual property triggering this change handler

[ 0 ]

i am somewhat of a fan of still having the useless just to make it obvious that we are setting a field but i guess that is a matter of opinion

[ 0 ]

yeah maybe we can kill the empty buffers where they get created

[ 0 ]

why not file dat ass trip n for s in f readlines

[ 0 ]

consider indenting the wrapped lines of the bullets to fix restview syntax

[ 0 ]

wrap testr sh does not exist

[ 0 ]

here should be admin snapshots client reset snapshot status

[ 0 ]

this could be just return verify result line is useless

[ 0 ]

why spawn an external kill process rather than using os kill

[ 0 ]

Oops, screw up on my part!

[ 1 ]

I fucked up and forgot this. Thanks for fixing that.

[ 1 ]

god this is even worse. doing COLOR_BRIGHT_LIME_GREEN

[ 1 ]

Of course, what nonsense, I invite @Tolriq to change a line and we don't waste any more time.

[ 1 ]

I did the hard reset and even restarted everything but this bug is there (along with a shit ton more). Pretty much this extension is just not working for me with all these bugs in 2.x. I fell back to 1.5 to have the same bugs.

[ 1 ]