showvikdbz/code-reviews · Datasets at Hugging Face

text stringlengths 1 5.2k	labels listlengths 1 1
let me know your thoughts on the most recent patchset it should remove the disconnect we currently have between the pd task selecting it is own cc values and the charge manager selecting them	[ 0 ]
i would delete this comment as it is useless	[ 0 ]
bullshit whitelist no longer needed by making attackbys return 1 on the right places.	[ 1 ]
can you symmetrically verify that we regain the account if we get the refresh token back if yes then optional can you split the test into two or three tests	[ 0 ]
this init is redundant remove	[ 0 ]
please remove this white spaces	[ 0 ]
should be if not sample filter start timestamp or not sample filter end timestamp	[ 0 ]
I'm a moron and checked the wrong branch. I thought this was against master \U0001f926\u200d\u2642	[ 1 ]
THAT'S WHAT I'M SAYING, DUDE!	[ 1 ]
never use echo n in scripts you can use printf s instead	[ 0 ]
All those of us who were affected by this - and the ridiculous amount of time it took to fix - speak up - https://github.com/elementor/elementor/issues/16227	[ 1 ]
Dammit, theres a copy and paste issue	[ 1 ]
There are a lot of IndexOf* methods in BCL that get heavily SIMD optimized. The method name describes the SIMD operation. I agree that a method like `TrimAny` can be better.	[ 1 ]
useless if statement here is fp promoted is guaranteed to be true in the else clause	[ 0 ]
This is still an issue in Oct 2019, kind of ridiculous that nothing has been done about this in over 2 years.	[ 1 ]
i believe most users should be satisfied with default dn choice this way we allow user to just flip the switch and try it on without selecting another dn we have the same setup for dumb member	[ 0 ]
nit best not to indent these bullets and best to have spaces between each bullet	[ 0 ]
Why must this be handled two different ways within windows platforms? Is windows really that screwed up that it doesn't have a single function to do this across its versions?	[ 1 ]
char p and not char p so it should be incrementing the pointer to the char	[ 0 ]
cp r sample policy dir conf dir should be ok	[ 0 ]
maybe `getPrivates` ... `getPrivateProps` sounds a bit weird	[ 1 ]
potentially stupid question here should this be hidden true	[ 0 ]
Oh crap i forgot, thanks.	[ 1 ]
not sure what the best way here is let it leak it will eventually finish terminate it dead that s what i did here send wm close window may prompt the user for confirmation we do not really want that the use case here is things like doing some background network pinging etc user has no idea what is running console apps seem to not do what you expect here the default behaviour will kill the console window without affecting the main process of the app if you wanted to act on a failed attempt with a heavy handed terminateprocess you need to block even longer or delay stuff on the event loop we do not want to do either here generateconsolectrlevent you cannot control the process group this signal is sent to it goes to everybody in the calling process group that s bad very bad i guess the only intended use case is when a dying process wants to warn it is children i could use if i launched a subprocess inside a process group that then launched the command that seems wasteful and wee bit non trivial	[ 0 ]
:+1: @rubbish have you started working on this?	[ 1 ]
self tunnel types tunnel types or	[ 0 ]
Damn, thats exactly what I called it when i started on drlg_l4, but decided to rename it back to `PlaceMiniSets` :D	[ 1 ]
nit you could use your stringbuilder above to do this concatenation	[ 0 ]
enable trusts or use trusts	[ 0 ]
Does not always work. Try the 'Baby croc' ride from Wacky Worlds and it still won't go beyond 5 boats.	[ 1 ]
it's almost like you removed every fucking line with SSticker in it! but that would be absurd and idiotic.	[ 1 ]
goddamnit guys it's so huge and obnoxious and it's STILL huge and obnoxious with this just a bit less so	[ 1 ]
\U0001f914 then that variable name is also kind of shitty	[ 1 ]
> when you put data into the GUI for the first time, the y scale is always rubbish I have this problem on other branches as well not just this one	[ 1 ]
not really, it's more of a failsafe IN CASE something weird happenned, like, the guy strapping C4 on the gun, or some stupid shit like that.	[ 1 ]
this is incompliant with python	[ 0 ]
It's even commented out, the heck is this?	[ 1 ]
blank line between bullets for consistency	[ 0 ]
its only survival pods, not whiteship so I dunno, pods are also tiny as fuck, max 3 tiles	[ 1 ]
it's ridiculous that this IS NOT in `docker ps` ffs	[ 1 ]
This isn't "left over ZAS shit", it means that closed doors will kill fires on their turf, which is perfectly intended.	[ 1 ]
Fuck it we ball is a meme	[ 1 ]
ok we can start here but we should be converging where we require multilib usedep in all the ebuilds that use this	[ 0 ]
should be sdks should be etoews	[ 0 ]
This pr is no longer relevant due to substantial code changes in the meantime. Apologies for the ridiculous delay in getting to this.	[ 1 ]
tmp nullptr	[ 0 ]
can t this be an arch specific library function i hate to see all these config arch ifdefs appearing everywhere	[ 0 ]
Welp I guess I still suck at Github. Just wanted to approve the text changes, lol. Oh well.	[ 1 ]
this line should be removed	[ 0 ]
i think you need to indent the s to make them recognized as bullets	[ 0 ]
the create static client is improperty the ops has no ideas about that term i prefer to use x for glance num retries as	[ 0 ]
noqa should be removed	[ 0 ]
it did not work because it manipulated a wrong linktarget object the way to get the linktarget from a library is call gb library get linktargetname library but it is cleaner this way	[ 0 ]
kill the struct make the server a pure function called schemahandler and then use handlerfunc schemahandler when adding it in run	[ 0 ]
throw invaliduserdataexception checkstate should be used for programming error	[ 0 ]
will this be exposed to the user if yes use an enum if not you can make it an aidl constant to share with your daemon client	[ 0 ]
you should sob	[ 0 ]
Nominally 80 in the standards file, but if they format weird, I will bend it	[ 1 ]
the goals is to abstract all the members here behave similarly with their private sublists uid statistics begin end for pid as is pid statistics begin end for gone pids if i switch to gonepids even more so with regards to making the code look simpler and not have to wrap	[ 0 ]
this is oldcode, git fucky with the diff but ill change it	[ 1 ]
er to clarify with or without this change ssl renegotiate in boringssl always failed boringssl cannot initiate renego only accept it as a client which means this line is not reachable in boringssl ports so it does not actually matter that ssl set state is strictly speaking a no op in boringssl reaching into the ssl and messing with it is state machine is kind of insane openssl s renego api is terrible given that i am not sure what the test is doing when compiled against boringssl since it apparently does not notice but this change should not make things worse	[ 0 ]
>What part of this was a simple matter of precision, asshole?!	[ 1 ]
shoot i was hoping it was the other way is there a way to write a seccomp filter across multiple lines it s too long to include all the or statements on a single line	[ 0 ]
use an accessor and if the caller does not care about the difference between nil and return a string instead of a string	[ 0 ]
one issues is that this results in a bunch of logs unnecessarily this is not a big deal but what about just testing if the parsed args id is a valid uuid in which case we treat it as a uuid otherwise we assume it is a name and try to search for it this way there is only one query either way and no uneeded log errors	[ 0 ]
this class represents a single pair so it should be called markerpair	[ 0 ]
Hm, yeah this is a bit different from all the other manifest renderers. I'm gonna work on fixing this up; I must've fucked something up when replacing some of the code	[ 1 ]
useless braces	[ 0 ]
actually i think it is useless as far as i remember create will execute flush for you	[ 0 ]
Is `ConcurrentDictionary`2.GetOrAdd` the secret here? Is something wacky going on across threads?	[ 1 ]
changing this would break the api we should catch the httpbadrequest in the shell py call and raise the right exception from there also tbh i do not really mind the error to be there this code as much as i hate it is short circuiting a check that should happen server side to begin with	[ 0 ]
i think that bare precise not contains the required php packages modules jenkins manifests slave pp standard packages jenkins params php cli package for community portal build if bare false packages common packages standard packages else packages common packages	[ 0 ]
Bloody visual studio just not helping with the automatic includes :(	[ 1 ]
missing here otherwise i guess it is okay	[ 0 ]
is it really ok to kill edp	[ 0 ]
i will bye remove conf file	[ 0 ]
nit i would call this agent az	[ 0 ]
this can go in the same bug fix as the quickstart sh patch	[ 0 ]
somehow i find function misleading here this is not even a functor object maybe have the visitor as you had earlier just a suggestion	[ 0 ]
i am just being lazy i will change it to support both	[ 0 ]
julien could you please explain this a little it s clear with admin but for project id who is owner and why it is determined with project id sorry if the question is stupid but i need to know	[ 0 ]
clear defalt value in config py	[ 0 ]
also `cannot` over `can not`. English is weird sometimes XD	[ 1 ]
consider rename to current l agents	[ 0 ]
holy shit that would've been bad good catch	[ 1 ]
not super sure i see the justification for this. canisters already contain a shit ton of gas, and this doesnt really make a lot of sense when miners currently can just get you all the gas you need anyway	[ 1 ]
well shit you're right	[ 1 ]
Hmm, damn :/ could you possible update Enigmatic Graves yourself and see if that helps? If not, I hope to get an update out tomorrow evening (~24 hours from now) with an updated version of Enigmatic Graves.	[ 1 ]
Closing because learned this is ridiculous	[ 1 ]
andrea corrected me ignore my stupidity	[ 0 ]
Sucks that we have to wait at least another 3 months before we can use this in our extensions to work around issues in Joomla. Which means I'll have to implement some kind of work-around, which defies the reason for needing this in core for me.	[ 1 ]
adc read max	[ 0 ]
I tried renaming it to nitrogen dioxide in another pr and people said it was shit. Nitryl sounds better and means nitrogen dioxide when inside a carbonic chain. tl;dr Nitryl is better name	[ 1 ]
this really should be collapsed down into just log error le	[ 0 ]
nit why not elif file content length and content length file content length	[ 0 ]
stray blank line	[ 0 ]
fwiw i uploaded a fix for build qemu android sh here c	[ 0 ]
get rid of this delete here	[ 0 ]
missing db mode var influxdb grafana hiera influxdb grafana db mode influxdb grafana mysql mode	[ 0 ]
in practice the wiz just uses repulse or something and steals their shit though	[ 1 ]

let me know your thoughts on the most recent patchset it should remove the disconnect we currently have between the pd task selecting it is own cc values and the charge manager selecting them

[ 0 ]

i would delete this comment as it is useless

[ 0 ]

bullshit whitelist no longer needed by making attackbys return 1 on the right places.

[ 1 ]

can you symmetrically verify that we regain the account if we get the refresh token back if yes then optional can you split the test into two or three tests

[ 0 ]

this init is redundant remove

[ 0 ]

please remove this white spaces

[ 0 ]

should be if not sample filter start timestamp or not sample filter end timestamp

[ 0 ]

I'm a moron and checked the wrong branch. I thought this was against master \U0001f926\u200d\u2642

[ 1 ]

THAT'S WHAT I'M SAYING, DUDE!

[ 1 ]

never use echo n in scripts you can use printf s instead

[ 0 ]

All those of us who were affected by this - and the ridiculous amount of time it took to fix - speak up - https://github.com/elementor/elementor/issues/16227

[ 1 ]

Dammit, theres a copy and paste issue

[ 1 ]

There are a lot of IndexOf* methods in BCL that get heavily SIMD optimized. The method name describes the SIMD operation. I agree that a method like `TrimAny` can be better.

[ 1 ]

useless if statement here is fp promoted is guaranteed to be true in the else clause

[ 0 ]

This is still an issue in Oct 2019, kind of ridiculous that nothing has been done about this in over 2 years.

[ 1 ]

i believe most users should be satisfied with default dn choice this way we allow user to just flip the switch and try it on without selecting another dn we have the same setup for dumb member

[ 0 ]

nit best not to indent these bullets and best to have spaces between each bullet

[ 0 ]

Why must this be handled two different ways within windows platforms? Is windows really that screwed up that it doesn't have a single function to do this across its versions?

[ 1 ]

char p and not char p so it should be incrementing the pointer to the char

[ 0 ]

cp r sample policy dir conf dir should be ok

[ 0 ]

maybe `getPrivates` ... `getPrivateProps` sounds a bit weird

[ 1 ]

potentially stupid question here should this be hidden true

[ 0 ]

Oh crap i forgot, thanks.

[ 1 ]

not sure what the best way here is let it leak it will eventually finish terminate it dead that s what i did here send wm close window may prompt the user for confirmation we do not really want that the use case here is things like doing some background network pinging etc user has no idea what is running console apps seem to not do what you expect here the default behaviour will kill the console window without affecting the main process of the app if you wanted to act on a failed attempt with a heavy handed terminateprocess you need to block even longer or delay stuff on the event loop we do not want to do either here generateconsolectrlevent you cannot control the process group this signal is sent to it goes to everybody in the calling process group that s bad very bad i guess the only intended use case is when a dying process wants to warn it is children i could use if i launched a subprocess inside a process group that then launched the command that seems wasteful and wee bit non trivial

[ 0 ]

:+1: @rubbish have you started working on this?

[ 1 ]

self tunnel types tunnel types or

[ 0 ]

Damn, thats exactly what I called it when i started on drlg_l4, but decided to rename it back to `PlaceMiniSets` :D

[ 1 ]

nit you could use your stringbuilder above to do this concatenation

[ 0 ]

enable trusts or use trusts

[ 0 ]

Does not always work. Try the 'Baby croc' ride from Wacky Worlds and it still won't go beyond 5 boats.

[ 1 ]

it's almost like you removed every fucking line with SSticker in it! but that would be absurd and idiotic.

[ 1 ]

goddamnit guys it's so huge and obnoxious and it's STILL huge and obnoxious with this just a bit less so

[ 1 ]

\U0001f914 then that variable name is also kind of shitty

[ 1 ]

> when you put data into the GUI for the first time, the y scale is always rubbish I have this problem on other branches as well not just this one

[ 1 ]

not really, it's more of a failsafe IN CASE something weird happenned, like, the guy strapping C4 on the gun, or some stupid shit like that.

[ 1 ]

this is incompliant with python

[ 0 ]

It's even commented out, the heck is this?

[ 1 ]

blank line between bullets for consistency

[ 0 ]

its only survival pods, not whiteship so I dunno, pods are also tiny as fuck, max 3 tiles

[ 1 ]

it's ridiculous that this IS NOT in `docker ps` ffs

[ 1 ]

This isn't "left over ZAS shit", it means that closed doors will kill fires on their turf, which is perfectly intended.

[ 1 ]

Fuck it we ball is a meme

[ 1 ]

ok we can start here but we should be converging where we require multilib usedep in all the ebuilds that use this

[ 0 ]

should be sdks should be etoews

[ 0 ]

This pr is no longer relevant due to substantial code changes in the meantime. Apologies for the ridiculous delay in getting to this.

[ 1 ]

tmp nullptr

[ 0 ]

can t this be an arch specific library function i hate to see all these config arch ifdefs appearing everywhere

[ 0 ]

Welp I guess I still suck at Github. Just wanted to approve the text changes, lol. Oh well.

[ 1 ]

this line should be removed

[ 0 ]

i think you need to indent the s to make them recognized as bullets

[ 0 ]

the create static client is improperty the ops has no ideas about that term i prefer to use x for glance num retries as

[ 0 ]

noqa should be removed

[ 0 ]

it did not work because it manipulated a wrong linktarget object the way to get the linktarget from a library is call gb library get linktargetname library but it is cleaner this way

[ 0 ]

kill the struct make the server a pure function called schemahandler and then use handlerfunc schemahandler when adding it in run

[ 0 ]

throw invaliduserdataexception checkstate should be used for programming error

[ 0 ]

will this be exposed to the user if yes use an enum if not you can make it an aidl constant to share with your daemon client

[ 0 ]

you should sob

[ 0 ]

Nominally 80 in the standards file, but if they format weird, I will bend it

[ 1 ]

the goals is to abstract all the members here behave similarly with their private sublists uid statistics begin end for pid as is pid statistics begin end for gone pids if i switch to gonepids even more so with regards to making the code look simpler and not have to wrap

[ 0 ]

this is oldcode, git fucky with the diff but ill change it

[ 1 ]

er to clarify with or without this change ssl renegotiate in boringssl always failed boringssl cannot initiate renego only accept it as a client which means this line is not reachable in boringssl ports so it does not actually matter that ssl set state is strictly speaking a no op in boringssl reaching into the ssl and messing with it is state machine is kind of insane openssl s renego api is terrible given that i am not sure what the test is doing when compiled against boringssl since it apparently does not notice but this change should not make things worse

[ 0 ]

>What part of this was a simple matter of precision, asshole?!

[ 1 ]

shoot i was hoping it was the other way is there a way to write a seccomp filter across multiple lines it s too long to include all the or statements on a single line

[ 0 ]

use an accessor and if the caller does not care about the difference between nil and return a string instead of a string

[ 0 ]

one issues is that this results in a bunch of logs unnecessarily this is not a big deal but what about just testing if the parsed args id is a valid uuid in which case we treat it as a uuid otherwise we assume it is a name and try to search for it this way there is only one query either way and no uneeded log errors

[ 0 ]

this class represents a single pair so it should be called markerpair

[ 0 ]

Hm, yeah this is a bit different from all the other manifest renderers. I'm gonna work on fixing this up; I must've fucked something up when replacing some of the code

[ 1 ]

useless braces

[ 0 ]

actually i think it is useless as far as i remember create will execute flush for you

[ 0 ]

Is `ConcurrentDictionary`2.GetOrAdd` the secret here? Is something wacky going on across threads?

[ 1 ]

changing this would break the api we should catch the httpbadrequest in the shell py call and raise the right exception from there also tbh i do not really mind the error to be there this code as much as i hate it is short circuiting a check that should happen server side to begin with

[ 0 ]

i think that bare precise not contains the required php packages modules jenkins manifests slave pp standard packages jenkins params php cli package for community portal build if bare false packages common packages standard packages else packages common packages

[ 0 ]

Bloody visual studio just not helping with the automatic includes :(

[ 1 ]

missing here otherwise i guess it is okay

[ 0 ]

is it really ok to kill edp

[ 0 ]

i will bye remove conf file

[ 0 ]

nit i would call this agent az

[ 0 ]

this can go in the same bug fix as the quickstart sh patch

[ 0 ]

somehow i find function misleading here this is not even a functor object maybe have the visitor as you had earlier just a suggestion

[ 0 ]

i am just being lazy i will change it to support both

[ 0 ]

julien could you please explain this a little it s clear with admin but for project id who is owner and why it is determined with project id sorry if the question is stupid but i need to know

[ 0 ]

clear defalt value in config py

[ 0 ]

also `cannot` over `can not`. English is weird sometimes XD

[ 1 ]

consider rename to current l agents

[ 0 ]

holy shit that would've been bad good catch

[ 1 ]

not super sure i see the justification for this. canisters already contain a shit ton of gas, and this doesnt really make a lot of sense when miners currently can just get you all the gas you need anyway

[ 1 ]

well shit you're right

[ 1 ]

Hmm, damn :/ could you possible update Enigmatic Graves yourself and see if that helps? If not, I hope to get an update out tomorrow evening (~24 hours from now) with an updated version of Enigmatic Graves.

[ 1 ]

Closing because learned this is ridiculous

[ 1 ]

andrea corrected me ignore my stupidity

[ 0 ]

Sucks that we have to wait at least another 3 months before we can use this in our extensions to work around issues in Joomla. Which means I'll have to implement some kind of work-around, which defies the reason for needing this in core for me.

[ 1 ]

adc read max

[ 0 ]

I tried renaming it to nitrogen dioxide in another pr and people said it was shit. Nitryl sounds better and means nitrogen dioxide when inside a carbonic chain. tl;dr Nitryl is better name

[ 1 ]

this really should be collapsed down into just log error le

[ 0 ]

nit why not elif file content length and content length file content length

[ 0 ]

stray blank line

[ 0 ]

fwiw i uploaded a fix for build qemu android sh here c

[ 0 ]

get rid of this delete here

[ 0 ]

missing db mode var influxdb grafana hiera influxdb grafana db mode influxdb grafana mysql mode

[ 0 ]

in practice the wiz just uses repulse or something and steals their shit though

[ 1 ]