showvikdbz/code-reviews · Datasets at Hugging Face

text stringlengths 1 5.2k	labels listlengths 1 1
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 ]
you can complete these shelving tasks	[ 0 ]
i think it would be more logical not to store the results after runner abort because the user expects that the execution will stop after the first sla fail but okay i have removed this	[ 0 ]
mock get domain return value dom mock	[ 0 ]
private with a getter and setter that sets the dirty bit	[ 0 ]
I don't think it's MetalLB anymore. It could be something wacky with Calico networking on RKE. I tried creating a cluster that didn't use MetalLB, and got the same results.	[ 1 ]
Go up or down, try to move left or right while still pressing W or S or the directional key equivalents if you're an unrobust faggot.	[ 1 ]
is this a mistake or am i too stupid to understand some clever trick here seems to me you meant if retry is none or attempt retry if so sounds like tests need improvement	[ 0 ]
Oh man, that is heroic work. Trailing commas suck for real.	[ 1 ]
If you are at 12.5x chance you failure you are playing an experimental before #49046 which fixes the absurd failure rates. And that one was merged two week ago.	[ 1 ]
we should remove osl seterrorreporting to sal osl unx compat cxx in a follow up commit	[ 0 ]
I know this wasnt in the PR but god damn why isnt this a switch statement	[ 1 ]
this test case is for tenant user so the project id should be set instead of none at least i would recommend to set dummy values for both user id and project id using tests unit fake constants py	[ 0 ]
it seems weird to store this in moles and not just shot counts	[ 1 ]
put a space between microversion and	[ 0 ]
Probably need this to ultimately be a strncpy. Should check what the max length is, don't want users to cause problems for themselves, stack corruption sucks... -[Unknown]	[ 1 ]
git blame nkinder	[ 0 ]
you do not need the s here as that is the cwd for this func	[ 0 ]
https://github.com/phalcon/cphalcon/pull/3213 fixed. and so i shall close this. but damn, zephir is much easier to code den w/e the hell it is before. i was half guess when i fixed this previously, that concat_VSVSVSVS got me all messed up.	[ 1 ]
You're right, damn copy paste :-/	[ 1 ]
Look, this is getting ridiculous. You say: > for the new UMat class we need a recursive mutex That is not a recursive mutex. So do you need it or not?	[ 1 ]
because i am stupid and failed to resolve the merge conflict correctly	[ 0 ]
this backslash is unnecessary and should be eliminated	[ 0 ]
damn, forgot that one! Fixed in the latest rebase.	[ 1 ]
The indent might be screwed?	[ 1 ]
use virtual function (and same for all ugly cast)	[ 1 ]
clashes with wide used gettext translation function i know another popular idiom dummy for example dummy my var some tuple	[ 0 ]
it should be slave	[ 0 ]
i think the print should be removed	[ 0 ]
Damn you, `git blame`! But seriously, good catch.	[ 1 ]
please remove redundant white spaces	[ 0 ]
I can just drop the refactor. Really losing steam on this nonsense...	[ 1 ]
sorry to be a pita i wonder if this would not be better by getting rid of the large if then else and having arch variants cflags mcpu target combo local arch arm has lpae filter cortex a krait denver cortex a cortex a cortex a target combo ifneq local arch arm has lpae arch variant cflags d arm feature lpae endif local arch arm has lpae local arch arm is non buggy cortex a filter cortex a krait denver cortex a cortex a cortex a target combo ifneq local arch arm is non buggy cortex a arch variant ldflags wl no fix cortex a else arch variant ldflags wl fix cortex a endif local arch arm is non buggy cortex a the code is something of a spaghetti mess currently not aided by the proposed change and the end result is that there is bit rot in areas like mcpu	[ 0 ]
i see suggest revising the comment to reportbadmessage will kill the renderer process but mojo complains if the callback is not run just run it with nonsense arguments	[ 0 ]
the existence of the live migration flag parameter is one of the stupider decisions in nova s past because it assumes all migrations want to use the same set of flags which is wrong we already see example of the wrongness by the need to add a separate block migration flag parameter as such we should not encourage further use of we should have a dedicated live migration compression on off parameter if we want admins to request compression	[ 0 ]
this check useless here	[ 0 ]
you do not need this flag if you have a launchpad bug	[ 0 ]
This is ridiculous, is not there any exact list of permissions / roles to enable to try this stuff out?	[ 1 ]
if not self dbinstance and not self resource id	[ 0 ]
i think this is supposed to be gone also you ran validate json files sh right	[ 0 ]
self scheduled task scheduled task	[ 0 ]
I screwed up! Thanks again!	[ 1 ]
should be acedgevpnassociation	[ 0 ]
if you set block migration	[ 0 ]
set request invalid although that has no behavioral relevance because preemption is the last phase of scheduling so no future pass is going to look atthis request it is still more correct	[ 0 ]
WHAT? Oh hell no, sorry I got stuck in 7.4 apparently. Sorry man	[ 1 ]
missing param image source	[ 0 ]
setdaemon true is for that purpose for daemon thread parent thread can die without waiting for it to finish	[ 0 ]
make sure this module is invoked as the main program if name main parse edid	[ 0 ]
holy shit athath if you post any more people might think you care or something LMAO	[ 1 ]
Well bloody my PR is going to change nothing. I'll leave this review open till I find something else to change	[ 1 ]
Or at least only apply this to walls, is_blocked_turf() will fail on dense mobs, tables, random objects, and honestly a disgusting amount of stuff	[ 1 ]
let us use openrc because we are going to have non root users on slaves	[ 0 ]
i will continue your adventure tomorrow morning if the bizarre failures are still around by then.	[ 1 ]
i would have just made this a separate class for maximal call site simplicity although it honestly does not bother me too much the way it is because shadowframe is already a mess class lockcounttracker std unique ptr lockcountdata data public void begintracking void endtracking void trackmonitorenter void trackmonitorexit bool checkmonitorsunheldorthrow struct shadowframe lockcounttracker lock count tracker	[ 0 ]
i know nit but you are missing a period here	[ 0 ]
the mips gdbserver still builds and i could not work out from the git history where this came from and why and it is not in bionic obviously since this patch is the result of copying this two files from tot bionic	[ 0 ]
if we add this board will we have to start killing off the rest i mean there can only be one	[ 0 ]
should having a python library be listed here although they are not part of the normal release cyle without the library most of the services are pretty much useless	[ 0 ]
Mass replace fucked something up here.	[ 1 ]
Ah shit, I thought I got all of those. Thanks for noticing- will fix in the next batch of these I do.	[ 1 ]
the joke should be the crossed out text alone, not crossed out text and shit grammar because if you don't fix it now someone WILL in the future.	[ 1 ]
i think you need a install r tools pip requires	[ 0 ]
This is a terrible hack. Also NOTE_ON with velocity 0 do exactly the same as NOTE_OFF.	[ 1 ]
should be self d env admin net	[ 0 ]
it s not necessary it s added partly as a why not and partly for consistency with the file import	[ 0 ]
This is literal copypaste shit from the thing for making a spear from a wired rod, so should probably change that as well.	[ 1 ]
This is absolutely fucking awful as someone who experienced a test merge. four shoots to a down officer, a entire charge to not even down a guy with a shield. Stop fucking changing things for little reason.	[ 1 ]
watcher api public admin internal ip	[ 0 ]
rename to requests kwargs because you clobber kwargs above	[ 0 ]
good observation let me take a stab and upload a new patchset	[ 0 ]
error nodes nodes to remove	[ 0 ]
WTF\uff1f Why not using Math.min()?	[ 1 ]
kwargs self parent id name parent id	[ 0 ]
I can only see it breaking in 2 circumstances: 1. the new steps are smaller and you hit the maximum number of iterations (at this point I think its probably safe to say that the start guess is rubbish) 2. the new steps are bigger so it misses a narrow minima. However, I suspect that would use the sqrt(epsilon) which would be a smaller step -> so would still work as long as point 1 does not happen.	[ 1 ]
Our issue was mainly with one module, so we build a node backend just for that one function. Ridiculous. Starting to think eonjs is the way forward now.	[ 1 ]
@DeityLink if you use any sort of non shit code editor they automatically trim whitespace like that.	[ 1 ]
Damn it, I always forget these.	[ 1 ]
>starting lists at one Byond Y Of course it made more sense when it was array offsets but conventions damn it	[ 1 ]
uuuuuh fuck ill fix it	[ 1 ]
I suck at naming......	[ 1 ]
please wrap the ceilometer polling like this ceilometer polling	[ 0 ]
Damn, 1200 pull requests in only half a year!	[ 1 ]
do a call pretty as with the boot img	[ 0 ]
Wait what how the hell did the RPD file end up in this	[ 1 ]
i may sound stupid asking this why we do not have metavar direct macvtap normal baremetal here as well	[ 0 ]
can you get rid of the white space above	[ 0 ]
i think so without the signer these artifacts are completely useless	[ 0 ]
ahh okay that s where i was being stupid	[ 0 ]
seems like it can always be a token acquired by a trust id associated with a user who created the trigger	[ 0 ]
dumb linters	[ 0 ]
Seriously?!?! I have never heard that before.	[ 1 ]
please ass closes bug	[ 0 ]
Shit, Will get to fixing it tomorrow	[ 1 ]
please set up lower bounds	[ 0 ]
this import should be removed	[ 0 ]
bool g crostini ui allowed for testing false	[ 0 ]
Not sure how you catch this shit	[ 1 ]
is regex name false btw i think the same should be done with the files to extract argument see the tarballbuildartifact item above	[ 0 ]
disposing all pending requests instead	[ 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 ]

you can complete these shelving tasks

[ 0 ]

i think it would be more logical not to store the results after runner abort because the user expects that the execution will stop after the first sla fail but okay i have removed this

[ 0 ]

mock get domain return value dom mock

[ 0 ]

private with a getter and setter that sets the dirty bit

[ 0 ]

I don't think it's MetalLB anymore. It could be something wacky with Calico networking on RKE. I tried creating a cluster that didn't use MetalLB, and got the same results.

[ 1 ]

Go up or down, try to move left or right while still pressing W or S or the directional key equivalents if you're an unrobust faggot.

[ 1 ]

is this a mistake or am i too stupid to understand some clever trick here seems to me you meant if retry is none or attempt retry if so sounds like tests need improvement

[ 0 ]

Oh man, that is heroic work. Trailing commas suck for real.

[ 1 ]

If you are at 12.5x chance you failure you are playing an experimental before #49046 which fixes the absurd failure rates. And that one was merged two week ago.

[ 1 ]

we should remove osl seterrorreporting to sal osl unx compat cxx in a follow up commit

[ 0 ]

I know this wasnt in the PR but god damn why isnt this a switch statement

[ 1 ]

this test case is for tenant user so the project id should be set instead of none at least i would recommend to set dummy values for both user id and project id using tests unit fake constants py

[ 0 ]

it seems weird to store this in moles and not just shot counts

[ 1 ]

put a space between microversion and

[ 0 ]

Probably need this to ultimately be a strncpy. Should check what the max length is, don't want users to cause problems for themselves, stack corruption sucks... -[Unknown]

[ 1 ]

git blame nkinder

[ 0 ]

you do not need the s here as that is the cwd for this func

[ 0 ]

https://github.com/phalcon/cphalcon/pull/3213 fixed. and so i shall close this. but damn, zephir is much easier to code den w/e the hell it is before. i was half guess when i fixed this previously, that concat_VSVSVSVS got me all messed up.

[ 1 ]

You're right, damn copy paste :-/

[ 1 ]

Look, this is getting ridiculous. You say: > for the new UMat class we need a recursive mutex That is not a recursive mutex. So do you need it or not?

[ 1 ]

because i am stupid and failed to resolve the merge conflict correctly

[ 0 ]

this backslash is unnecessary and should be eliminated

[ 0 ]

damn, forgot that one! Fixed in the latest rebase.

[ 1 ]

The indent might be screwed?

[ 1 ]

use virtual function (and same for all ugly cast)

[ 1 ]

clashes with wide used gettext translation function i know another popular idiom dummy for example dummy my var some tuple

[ 0 ]

it should be slave

[ 0 ]

i think the print should be removed

[ 0 ]

Damn you, `git blame`! But seriously, good catch.

[ 1 ]

please remove redundant white spaces

[ 0 ]

I can just drop the refactor. Really losing steam on this nonsense...

[ 1 ]

sorry to be a pita i wonder if this would not be better by getting rid of the large if then else and having arch variants cflags mcpu target combo local arch arm has lpae filter cortex a krait denver cortex a cortex a cortex a target combo ifneq local arch arm has lpae arch variant cflags d arm feature lpae endif local arch arm has lpae local arch arm is non buggy cortex a filter cortex a krait denver cortex a cortex a cortex a target combo ifneq local arch arm is non buggy cortex a arch variant ldflags wl no fix cortex a else arch variant ldflags wl fix cortex a endif local arch arm is non buggy cortex a the code is something of a spaghetti mess currently not aided by the proposed change and the end result is that there is bit rot in areas like mcpu

[ 0 ]

i see suggest revising the comment to reportbadmessage will kill the renderer process but mojo complains if the callback is not run just run it with nonsense arguments

[ 0 ]

the existence of the live migration flag parameter is one of the stupider decisions in nova s past because it assumes all migrations want to use the same set of flags which is wrong we already see example of the wrongness by the need to add a separate block migration flag parameter as such we should not encourage further use of we should have a dedicated live migration compression on off parameter if we want admins to request compression

[ 0 ]

this check useless here

[ 0 ]

you do not need this flag if you have a launchpad bug

[ 0 ]

This is ridiculous, is not there any exact list of permissions / roles to enable to try this stuff out?

[ 1 ]

if not self dbinstance and not self resource id

[ 0 ]

i think this is supposed to be gone also you ran validate json files sh right

[ 0 ]

self scheduled task scheduled task

[ 0 ]

I screwed up! Thanks again!

[ 1 ]

should be acedgevpnassociation

[ 0 ]

if you set block migration

[ 0 ]

set request invalid although that has no behavioral relevance because preemption is the last phase of scheduling so no future pass is going to look atthis request it is still more correct

[ 0 ]

WHAT? Oh hell no, sorry I got stuck in 7.4 apparently. Sorry man

[ 1 ]

missing param image source

[ 0 ]

setdaemon true is for that purpose for daemon thread parent thread can die without waiting for it to finish

[ 0 ]

make sure this module is invoked as the main program if name main parse edid

[ 0 ]

holy shit athath if you post any more people might think you care or something LMAO

[ 1 ]

Well bloody my PR is going to change nothing. I'll leave this review open till I find something else to change

[ 1 ]

Or at least only apply this to walls, is_blocked_turf() will fail on dense mobs, tables, random objects, and honestly a disgusting amount of stuff

[ 1 ]

let us use openrc because we are going to have non root users on slaves

[ 0 ]

i will continue your adventure tomorrow morning if the bizarre failures are still around by then.

[ 1 ]

i would have just made this a separate class for maximal call site simplicity although it honestly does not bother me too much the way it is because shadowframe is already a mess class lockcounttracker std unique ptr lockcountdata data public void begintracking void endtracking void trackmonitorenter void trackmonitorexit bool checkmonitorsunheldorthrow struct shadowframe lockcounttracker lock count tracker

[ 0 ]

i know nit but you are missing a period here

[ 0 ]

the mips gdbserver still builds and i could not work out from the git history where this came from and why and it is not in bionic obviously since this patch is the result of copying this two files from tot bionic

[ 0 ]

if we add this board will we have to start killing off the rest i mean there can only be one

[ 0 ]

should having a python library be listed here although they are not part of the normal release cyle without the library most of the services are pretty much useless

[ 0 ]

Mass replace fucked something up here.

[ 1 ]

Ah shit, I thought I got all of those. Thanks for noticing- will fix in the next batch of these I do.

[ 1 ]

the joke should be the crossed out text alone, not crossed out text and shit grammar because if you don't fix it now someone WILL in the future.

[ 1 ]

i think you need a install r tools pip requires

[ 0 ]

This is a terrible hack. Also NOTE_ON with velocity 0 do exactly the same as NOTE_OFF.

[ 1 ]

should be self d env admin net

[ 0 ]

it s not necessary it s added partly as a why not and partly for consistency with the file import

[ 0 ]

This is literal copypaste shit from the thing for making a spear from a wired rod, so should probably change that as well.

[ 1 ]

This is absolutely fucking awful as someone who experienced a test merge. four shoots to a down officer, a entire charge to not even down a guy with a shield. Stop fucking changing things for little reason.

[ 1 ]

watcher api public admin internal ip

[ 0 ]

rename to requests kwargs because you clobber kwargs above

[ 0 ]

good observation let me take a stab and upload a new patchset

[ 0 ]

error nodes nodes to remove

[ 0 ]

WTF\uff1f Why not using Math.min()?

[ 1 ]

kwargs self parent id name parent id

[ 0 ]

I can only see it breaking in 2 circumstances: 1. the new steps are smaller and you hit the maximum number of iterations (at this point I think its probably safe to say that the start guess is rubbish) 2. the new steps are bigger so it misses a narrow minima. However, I suspect that would use the sqrt(epsilon) which would be a smaller step -> so would still work as long as point 1 does not happen.

[ 1 ]

Our issue was mainly with one module, so we build a node backend just for that one function. Ridiculous. Starting to think eonjs is the way forward now.

[ 1 ]

@DeityLink if you use any sort of non shit code editor they automatically trim whitespace like that.

[ 1 ]

Damn it, I always forget these.

[ 1 ]

>starting lists at one Byond Y Of course it made more sense when it was array offsets but conventions damn it

[ 1 ]

uuuuuh fuck ill fix it

[ 1 ]

I suck at naming......

[ 1 ]

please wrap the ceilometer polling like this ceilometer polling

[ 0 ]

Damn, 1200 pull requests in only half a year!

[ 1 ]

do a call pretty as with the boot img

[ 0 ]

Wait what how the hell did the RPD file end up in this

[ 1 ]

i may sound stupid asking this why we do not have metavar direct macvtap normal baremetal here as well

[ 0 ]

can you get rid of the white space above

[ 0 ]

i think so without the signer these artifacts are completely useless

[ 0 ]

ahh okay that s where i was being stupid

[ 0 ]

seems like it can always be a token acquired by a trust id associated with a user who created the trigger

[ 0 ]

dumb linters

[ 0 ]

Seriously?!?! I have never heard that before.

[ 1 ]

please ass closes bug

[ 0 ]

Shit, Will get to fixing it tomorrow

[ 1 ]

please set up lower bounds

[ 0 ]

this import should be removed

[ 0 ]

bool g crostini ui allowed for testing false

[ 0 ]

Not sure how you catch this shit

[ 1 ]

is regex name false btw i think the same should be done with the files to extract argument see the tarballbuildartifact item above

[ 0 ]

disposing all pending requests instead

[ 0 ]