showvikdbz/code-reviews · Datasets at Hugging Face

text stringlengths 1 5.2k	labels listlengths 1 1
this entire file's formatting is crappy, someone should do something about it at some point	[ 1 ]
this comment need to be removed	[ 0 ]
devtest overcloud sh	[ 0 ]
does this trim for you or should you trim splitstr before calling	[ 0 ]
You shall be introduced as the ranting lunatic I met on the web :P	[ 1 ]
notice crappy solid edges (not present on PSX)	[ 1 ]
this reference is waaaaay too fucking overt. change that up a good fucking bit	[ 1 ]
oops my silly git management with several clones of this meant that i completely overwrote the changes i made thanks to xiaoqian it should be all good again now	[ 0 ]
should this be in the harouter since you have named it to snat ns name should it not be in dvrhaedgerouter	[ 0 ]
what i failed to mention previously mixing in such pointless cosmetic changes is a poor waste of reviewers time	[ 0 ]
Ahhhhh that makes sense! Again, damn shadowing!!!!!!	[ 1 ]
can you remove these placeholders please	[ 0 ]
may not be able may fail	[ 0 ]
That code makes my brain hurt \u2014 the idea of strdupping "" on error is just bizzare\u2026 Nevertheless, this fix is "obviously correct", so no reason not to merge it.	[ 1 ]
root helper utils get root helper	[ 0 ]
auth plugin auth plugin	[ 0 ]
in that case i suggest actually adding it as a param callers can simply specify a dummy value themselves if they do not care about it	[ 0 ]
please next time do not write trivialfix in commit msg this is useless information here should be why this commit needed	[ 0 ]
OK, if it's got the gears on it should have a point or two of cutting as well. To-hit should be -2 or -3 as the damn thing is gonna move as you swing it, BUT it'd hurt if it hit.	[ 1 ]
> I see no point adding flags (i.e: `--allow-hrtime`) for "security reason" while this is not sorted. Again, this is a ridiculous point. You're contributing nothing.	[ 1 ]
you re passing up probably your only opportunity to use the commit message kill kill	[ 0 ]
i agree that a should be removed	[ 0 ]
this should end in a colon similar comment applies throughout	[ 0 ]
callbacks should be instantiated in init	[ 0 ]
> Which makes me ask why your main auth cookie is set to None? I've seen things man... scary things where people do iframe mashups and other nonsense. But specifically for IdentityServer it's needed for JS clients doing authorization in iframes for silent token renewal.	[ 1 ]
I made my mind. This is trash.	[ 1 ]
The UI for this is pretty ridiculous; if you have more than ~15-20 days in the range then you have to scroll off to the side. Wouldn't be the worst idea to just restrict the date range on the client side.	[ 1 ]
bgp speaker network list unnecessary should be removed	[ 0 ]
doh i am dumb done	[ 0 ]
you moved the do migration list function to shell py looks like you forgot to remove it here	[ 0 ]
Not sure why this was reimplemented manually instead of using `HitResult.IsHit()`. Oh well.	[ 1 ]
Ye, lombok removal would take ridiculous amounts of boilerplate code to be added, and if it's possible to avoid that - let's avoid that.	[ 1 ]
are useless just use if ironic enabled	[ 0 ]
this method name does not look suggestive to me what about check user has access to subtree	[ 0 ]
profileid profileid extra config extra config	[ 0 ]
Gah, the whole mandatory inline thing for non-members keeps tripping me up, will fix.	[ 1 ]
Ooh shit how sloppy of me, good catch.	[ 1 ]
override zuul branch should be used here due to the c landed	[ 0 ]
this is nonsense. There is no tarpitting.	[ 1 ]
i think it should be a namedtuple and not a class	[ 0 ]
ubuntu arch please	[ 0 ]
i blame mst for reintroducing them with adeada cc e ceb a ffe d b a week ago	[ 0 ]
you can probably remove this last sentence	[ 0 ]
it should be "Sweet Bro and Hella Jeff: Expanded Edition" j e f f not jef he is not a cleaning product	[ 1 ]
this is copy pasted verbatim from the neutron patch but it is silly just do for event in self pending events	[ 0 ]
Oh they are in a different topic? Well damn, alright I'll see what I can do	[ 1 ]
these verify token calls need to be avoided when we have a cached token making those calls defeats the purpose of caching	[ 0 ]
silly question why does scanout mean tiling for i is that consistent with the other drivers	[ 0 ]
It did not help... this is ridiculous. https://clickhouse-test-reports.s3.yandex.net/17630/8239df16039b13d83fe5ed2c8773e7c37a213995/functional_stateless_tests_(thread)/stderr.log	[ 1 ]
nit destination file name	[ 0 ]
Lol. C'mon guys. There's plenty of memes in the code like the last of Arse and other Easter eggs. Let's be merciful when killing a process.	[ 1 ]
you rat bastard you compiled three minutes ago	[ 1 ]
that damn bot can you fix it for me	[ 1 ]
initialize binder fd and max threads here so they do not start out with garbage	[ 0 ]
My bad, intent is at a mob level You messed up the define, it's I_HURT not I_HARM	[ 1 ]
you can use wait true istead such hacks	[ 0 ]
Christ what the fuck is this	[ 1 ]
yes this could use text2path and be shorter but how about fuck text2path	[ 1 ]
i am dumb and this actually was being done in copy remote dir local except with a bug basename on rpath instead of lpath i did not notice what was going on because i was pulling directly into the adb source directory which already has a test subdirectory removed this and replaced it with a proper fix	[ 0 ]
aha stupid me for some reason i thought this is printed for constructor s help and that it describes constructor parameters	[ 0 ]
this is a perfectly valid util apparently it is simply not undo friendly	[ 0 ]
you proper english people and all your extra u s	[ 0 ]
nitpick work a ro und	[ 0 ]
polling is kind of ugly but also the simplest way i could think of to wait for something that happens after the timeout would have been incorrectly restarted i could also introduce a filter to wait for a beforeunload ack let me know if you think that is better or if you have other suggestions	[ 0 ]
That benchmark is doing a million loops. Find me an app that ever does a million loops and I'll show you a shitty implementation. This change is a terrible one and the optimization improvements would be negligible in the real world.	[ 1 ]
how does the proof look like that m hproviderdetailscache can never be empty here if the assert is sound the following if is redundant and should be removed	[ 0 ]
I blame BYOND's shitty syntax	[ 1 ]
if you respin then please blow away the last empty line	[ 0 ]
damn. how did this ever work? :open_mouth:	[ 1 ]
i would still argue that setsid should be called as part of init therefore only when pid namespacing is used	[ 0 ]
> (Although it'd be safer to pin the kernel version anyway just in case; one less source of non-determinism.) IMO it starts getting ridiculous. We should not force any kernel/glibc/OS version.	[ 1 ]
device exists should be mocked to return true then there is no need for execute and ipdevice	[ 0 ]
is it ok to have the agent continue it is execution when it is not able to send it is heart beats controller will not be aware of this l agent unless it receives a heartbeat just wondering if there is nothing dependent on agent state information	[ 0 ]
typo in mechanism let s not give people an excuse to file silly typo fixes patches	[ 0 ]
this comment belongs to the removed code and should be removed too	[ 0 ]
optional you dn t write this but i would write const entry this ui directorytree selecteditem entry null	[ 0 ]
instead it could be data template f for f in fields if f in template	[ 0 ]
As I wrote, this is nothing new and has been working like this since tags have been introduced in Joomla 3.1 or 3.2. Yes, it is ugly and sucks big time, but it is not a new behavior. To keep backwards compatibility, I can also already guarantee you, that we wont be changing this in 4.x.	[ 1 ]
this class and selectprojectuser below are unnecessary since they are only used by admin instance launchin	[ 0 ]
this version needs to be incremented see the number after rebase	[ 0 ]
Yeah, I don't think giving sec this shit is good, it just brings back stuncombat without the weaknesses a stunbaton had. NSV is a different server with different needs	[ 1 ]
why would you replace a var name with a worse one wtf	[ 1 ]
the neutron server is monkey patched for all the modules	[ 0 ]
you grep openssl and you mentionned openssl so use openssl library no	[ 0 ]
nr and sys are the two common prefixes for the userspace constants for system calls i forget which is sysv and which is the other one so you need to grep your linux kernel tree for clone or read man clone or whatever what s happening here is that other than flags and fn being the wrong way round bionic clone s arguments are the same as the kernel s syscall that is handled in the next block up here we are stashing the two arguments that bionic clone entry will need later i d like to reorder to remove the need for the later block but some retard made this public abi so i worry i would break something i d also like to kill pthread clone and just use this more general form but since we probably have to keep this for abi compatibility anyway it s worth thinking about simplifying all this for the bit architectures though	[ 0 ]
Thank you very much for your help and will correct these ridiculous mistakes.	[ 1 ]
may be worth describing what authn request is you are mutating it a lot in this function	[ 0 ]
Oh shit I'm so dumb sorry	[ 1 ]
This sort of shit should be in the map files themselves imo. like map_files/Cerestation/fluff.dm	[ 1 ]
it's probably fine to leave it as subtypes but in the future, updatepaths should be written for expected changes (unless you know of a place that actually subtyped this fuckin' mob)	[ 1 ]
ak that was dumb of me not to see that that s why you did it that way	[ 0 ]
reflow without backslash	[ 0 ]
Damn you might have just found another bug that caused documents and the like not to show up on the pinpointer	[ 1 ]
More or less ass than duplicating a bunch of code that doesn't need to be duplicated.	[ 1 ]
i thought you are is samueldmq	[ 0 ]
imo getextraargumentsstartindex is worse extraargument does not tell me anything and suggest that anything could be there which is not true we know precisely what is there	[ 0 ]
ill caveat this with i am a c idiot but why are we not using static cast here	[ 0 ]
Damn! Cheers for picking that up!	[ 1 ]
i wonder if this value should be computed by puppet facter	[ 0 ]
It's ridiculous that you close this ticket when so many people are having the same issue. Obviously there is an issue. Says a lot about the support from Prestashop.	[ 1 ]

this entire file's formatting is crappy, someone should do something about it at some point

[ 1 ]

this comment need to be removed

[ 0 ]

devtest overcloud sh

[ 0 ]

does this trim for you or should you trim splitstr before calling

[ 0 ]

You shall be introduced as the ranting lunatic I met on the web :P

[ 1 ]

notice crappy solid edges (not present on PSX)

[ 1 ]

this reference is waaaaay too fucking overt. change that up a good fucking bit

[ 1 ]

oops my silly git management with several clones of this meant that i completely overwrote the changes i made thanks to xiaoqian it should be all good again now

[ 0 ]

should this be in the harouter since you have named it to snat ns name should it not be in dvrhaedgerouter

[ 0 ]

what i failed to mention previously mixing in such pointless cosmetic changes is a poor waste of reviewers time

[ 0 ]

Ahhhhh that makes sense! Again, damn shadowing!!!!!!

[ 1 ]

can you remove these placeholders please

[ 0 ]

may not be able may fail

[ 0 ]

That code makes my brain hurt \u2014 the idea of strdupping "" on error is just bizzare\u2026 Nevertheless, this fix is "obviously correct", so no reason not to merge it.

[ 1 ]

root helper utils get root helper

[ 0 ]

auth plugin auth plugin

[ 0 ]

in that case i suggest actually adding it as a param callers can simply specify a dummy value themselves if they do not care about it

[ 0 ]

please next time do not write trivialfix in commit msg this is useless information here should be why this commit needed

[ 0 ]

OK, if it's got the gears on it should have a point or two of cutting as well. To-hit should be -2 or -3 as the damn thing is gonna move as you swing it, BUT it'd hurt if it hit.

[ 1 ]

> I see no point adding flags (i.e: `--allow-hrtime`) for "security reason" while this is not sorted. Again, this is a ridiculous point. You're contributing nothing.

[ 1 ]

you re passing up probably your only opportunity to use the commit message kill kill

[ 0 ]

i agree that a should be removed

[ 0 ]

this should end in a colon similar comment applies throughout

[ 0 ]

callbacks should be instantiated in init

[ 0 ]

> Which makes me ask why your main auth cookie is set to None? I've seen things man... scary things where people do iframe mashups and other nonsense. But specifically for IdentityServer it's needed for JS clients doing authorization in iframes for silent token renewal.

[ 1 ]

I made my mind. This is trash.

[ 1 ]

The UI for this is pretty ridiculous; if you have more than ~15-20 days in the range then you have to scroll off to the side. Wouldn't be the worst idea to just restrict the date range on the client side.

[ 1 ]

bgp speaker network list unnecessary should be removed

[ 0 ]

doh i am dumb done

[ 0 ]

you moved the do migration list function to shell py looks like you forgot to remove it here

[ 0 ]

Not sure why this was reimplemented manually instead of using `HitResult.IsHit()`. Oh well.

[ 1 ]

Ye, lombok removal would take ridiculous amounts of boilerplate code to be added, and if it's possible to avoid that - let's avoid that.

[ 1 ]

are useless just use if ironic enabled

[ 0 ]

this method name does not look suggestive to me what about check user has access to subtree

[ 0 ]

profileid profileid extra config extra config

[ 0 ]

Gah, the whole mandatory inline thing for non-members keeps tripping me up, will fix.

[ 1 ]

Ooh shit how sloppy of me, good catch.

[ 1 ]

override zuul branch should be used here due to the c landed

[ 0 ]

this is nonsense. There is no tarpitting.

[ 1 ]

i think it should be a namedtuple and not a class

[ 0 ]

ubuntu arch please

[ 0 ]

i blame mst for reintroducing them with adeada cc e ceb a ffe d b a week ago

[ 0 ]

you can probably remove this last sentence

[ 0 ]

it should be "Sweet Bro and Hella Jeff: Expanded Edition" j e f f not jef he is not a cleaning product

[ 1 ]

this is copy pasted verbatim from the neutron patch but it is silly just do for event in self pending events

[ 0 ]

Oh they are in a different topic? Well damn, alright I'll see what I can do

[ 1 ]

these verify token calls need to be avoided when we have a cached token making those calls defeats the purpose of caching

[ 0 ]

silly question why does scanout mean tiling for i is that consistent with the other drivers

[ 0 ]

It did not help... this is ridiculous. https://clickhouse-test-reports.s3.yandex.net/17630/8239df16039b13d83fe5ed2c8773e7c37a213995/functional_stateless_tests_(thread)/stderr.log

[ 1 ]

nit destination file name

[ 0 ]

Lol. C'mon guys. There's plenty of memes in the code like the last of Arse and other Easter eggs. Let's be merciful when killing a process.

[ 1 ]

you rat bastard you compiled three minutes ago

[ 1 ]

that damn bot can you fix it for me

[ 1 ]

initialize binder fd and max threads here so they do not start out with garbage

[ 0 ]

My bad, intent is at a mob level You messed up the define, it's I_HURT not I_HARM

[ 1 ]

you can use wait true istead such hacks

[ 0 ]

Christ what the fuck is this

[ 1 ]

yes this could use text2path and be shorter but how about fuck text2path

[ 1 ]

i am dumb and this actually was being done in copy remote dir local except with a bug basename on rpath instead of lpath i did not notice what was going on because i was pulling directly into the adb source directory which already has a test subdirectory removed this and replaced it with a proper fix

[ 0 ]

aha stupid me for some reason i thought this is printed for constructor s help and that it describes constructor parameters

[ 0 ]

this is a perfectly valid util apparently it is simply not undo friendly

[ 0 ]

you proper english people and all your extra u s

[ 0 ]

nitpick work a ro und

[ 0 ]

polling is kind of ugly but also the simplest way i could think of to wait for something that happens after the timeout would have been incorrectly restarted i could also introduce a filter to wait for a beforeunload ack let me know if you think that is better or if you have other suggestions

[ 0 ]

That benchmark is doing a million loops. Find me an app that ever does a million loops and I'll show you a shitty implementation. This change is a terrible one and the optimization improvements would be negligible in the real world.

[ 1 ]

how does the proof look like that m hproviderdetailscache can never be empty here if the assert is sound the following if is redundant and should be removed

[ 0 ]

I blame BYOND's shitty syntax

[ 1 ]

if you respin then please blow away the last empty line

[ 0 ]

damn. how did this ever work? :open_mouth:

[ 1 ]

i would still argue that setsid should be called as part of init therefore only when pid namespacing is used

[ 0 ]

> (Although it'd be safer to pin the kernel version anyway just in case; one less source of non-determinism.) IMO it starts getting ridiculous. We should not force any kernel/glibc/OS version.

[ 1 ]

device exists should be mocked to return true then there is no need for execute and ipdevice

[ 0 ]

is it ok to have the agent continue it is execution when it is not able to send it is heart beats controller will not be aware of this l agent unless it receives a heartbeat just wondering if there is nothing dependent on agent state information

[ 0 ]

typo in mechanism let s not give people an excuse to file silly typo fixes patches

[ 0 ]

this comment belongs to the removed code and should be removed too

[ 0 ]

optional you dn t write this but i would write const entry this ui directorytree selecteditem entry null

[ 0 ]

instead it could be data template f for f in fields if f in template

[ 0 ]

As I wrote, this is nothing new and has been working like this since tags have been introduced in Joomla 3.1 or 3.2. Yes, it is ugly and sucks big time, but it is not a new behavior. To keep backwards compatibility, I can also already guarantee you, that we wont be changing this in 4.x.

[ 1 ]

this class and selectprojectuser below are unnecessary since they are only used by admin instance launchin

[ 0 ]

this version needs to be incremented see the number after rebase

[ 0 ]

Yeah, I don't think giving sec this shit is good, it just brings back stuncombat without the weaknesses a stunbaton had. NSV is a different server with different needs

[ 1 ]

why would you replace a var name with a worse one wtf

[ 1 ]

the neutron server is monkey patched for all the modules

[ 0 ]

you grep openssl and you mentionned openssl so use openssl library no

[ 0 ]

nr and sys are the two common prefixes for the userspace constants for system calls i forget which is sysv and which is the other one so you need to grep your linux kernel tree for clone or read man clone or whatever what s happening here is that other than flags and fn being the wrong way round bionic clone s arguments are the same as the kernel s syscall that is handled in the next block up here we are stashing the two arguments that bionic clone entry will need later i d like to reorder to remove the need for the later block but some retard made this public abi so i worry i would break something i d also like to kill pthread clone and just use this more general form but since we probably have to keep this for abi compatibility anyway it s worth thinking about simplifying all this for the bit architectures though

[ 0 ]

Thank you very much for your help and will correct these ridiculous mistakes.

[ 1 ]

may be worth describing what authn request is you are mutating it a lot in this function

[ 0 ]

Oh shit I'm so dumb sorry

[ 1 ]

This sort of shit should be in the map files themselves imo. like map_files/Cerestation/fluff.dm

[ 1 ]

it's probably fine to leave it as subtypes but in the future, updatepaths should be written for expected changes (unless you know of a place that actually subtyped this fuckin' mob)

[ 1 ]

ak that was dumb of me not to see that that s why you did it that way

[ 0 ]

reflow without backslash

[ 0 ]

Damn you might have just found another bug that caused documents and the like not to show up on the pinpointer

[ 1 ]

More or less ass than duplicating a bunch of code that doesn't need to be duplicated.

[ 1 ]

i thought you are is samueldmq

[ 0 ]

imo getextraargumentsstartindex is worse extraargument does not tell me anything and suggest that anything could be there which is not true we know precisely what is there

[ 0 ]

ill caveat this with i am a c idiot but why are we not using static cast here

[ 0 ]

Damn! Cheers for picking that up!

[ 1 ]

i wonder if this value should be computed by puppet facter

[ 0 ]

It's ridiculous that you close this ticket when so many people are having the same issue. Obviously there is an issue. Says a lot about the support from Prestashop.

[ 1 ]