showvikdbz/code-reviews · Datasets at Hugging Face

text stringlengths 1 5.2k	labels listlengths 1 1
ideally we would be able to let the label fill it is parent with maybe a tiny bit of padding between elements i tested in spanish really quickly and on my nexus x the home button and share button text gets ellipsized it d be nice if we gave it a little bit more room it seems like we should have the space screenshot	[ 0 ]
if all the libupnp stuff is currently bundled - it is by accident - no need to have all this libupnp crap in the bundle - sry can't do lone comments because github sucks - you know which lines i mean i hope	[ 1 ]
should we just drop out early if featureutilities isnotouchmodeenabled infobar supportstouchlessmode is there logic we need to execute if we are not actually goign to show the infobar	[ 0 ]
Oh damn, i didn't saw the UpdateOwner call - You're right with this one. Good catch!	[ 1 ]
to be execute to be executed	[ 0 ]
Congratulations you just fixed a fuck ton of lag caused by humans bleeding.	[ 1 ]
please remove trailing spaces p sed ri s s	[ 0 ]
mistral	[ 0 ]
Just make these four goddamn lines into a proc so you don't copypaste the ENTIRE section	[ 1 ]
i am stupid of course we do not have cssvalues here we could just add a create function that takes a string as you can see the cssvalue one immediately converts to string anyway static cssunsupportedstylevalue create const string value return makegarbagecollected cssunsupportedstylevalue value	[ 0 ]
bro just roll it randomly between 5 and 100 instead of using the shittiest die to exist	[ 1 ]
worth noting the security model in jenkins is such that it is not really safe to mix trusted and untrusted slaves and it is also not generally safe to reuse untrusted slaves as such we keep our trusted reusable slaves bound to a separate jenkins master from our nodepool managed slaves if you have a similar split you can simply avoid running this for the jenkins masters handling your reusable slaves and only use it for jenkins masters handling your nodepool managed slaves	[ 0 ]
good damn idea, I'll make it shake the mob. what do you mean by message to the user? some feedback to the mob who did it? shouldn't they see the message? or is visible message pwning me	[ 1 ]
remove blankline	[ 0 ]
i am thinking about aborting here in the case when flags point to existence of the r addend field is handled by read group fields and this part of code should be unreachable	[ 0 ]
Might as well max it out to `0xFF`, it won't cost any memory on the MCU as it just defines the range of values the host should expect. And someone, somewhere, will really want to send an `0xF2` for whatever wacky idea they have.	[ 1 ]
just make sure that no one else reads the current selected client and attempts to an operation on them e g maybe go to the memory view and attempt a heap dump gc on that dead client	[ 0 ]
bad merge you re shadowing oslo config cfg	[ 0 ]
how will the user know which host succeeded	[ 0 ]
i think handler kind is a list so probably handler kind would work	[ 0 ]
oh in local post install cmd we have to use local installed module instead of	[ 0 ]
This system is moronic and I don't like its extension. It should be using material IDs and then getting sheet type from the material datum, not doing this weird path nonsense.	[ 1 ]
was going to ask why you did not do this as a query string parameter then i realized i was being dumb because nobody in their right mind should do a delete off a get method so yes this a handy trick	[ 0 ]
we already have the api group why register it again just do conf register opts your opts group api	[ 0 ]
how do you plan to lock varedits without locking the entire flag entirely? (and dont lock it entirely damnit)	[ 1 ]
check pid	[ 0 ]
Ehh I don't know and I honestly don't care too much. I guess it's a limit to prevent opening 9000 files and trying to do some nasty shit.	[ 1 ]
you re going to want to kill me but redrobot said these should be in the format assertequal expected returned he dinged me on that for mine i do not know if worth ask redrobot	[ 0 ]
What the unholy shit is this? Just make it go by dir and check only once if that dir is blocked. This is so overengineered.	[ 1 ]
Yeah I know that you dislike @Deprecated :smile: Anyways, it's ok with me. Break all the shit. :+1:	[ 1 ]
Ah shit, yeah, you're right, I forgot to put H. before the functions!	[ 1 ]
Oooh damn. Okay then!	[ 1 ]
Bah the ternary operator is crap.	[ 1 ]
we should maybe send sigterm to run oci and let that relay the sigkill to it is child	[ 0 ]
Can we please do this without overloading `--net` anymore? That flag has gotten ridiculous to parse	[ 1 ]
can you kill the leading spaces here	[ 0 ]
bad sym l	[ 0 ]
lol what the fuck happened here (seems alright to me now though)	[ 1 ]
it should break if something happens, cause if merging gases goes wrong we should know (null mixes or other weird stuff)	[ 1 ]
If the diff sucks, see https://github.com/roligugus/suricata/blob/clang-format-proposal-v4/doc/devguide/codebase/code-style.rst for how it's displayed	[ 1 ]
As I understand it no, there was no existing recipe for a "Meat Sandwich" I thought about letting the player craft it cold but using raw meat in an sandwich sounded disgusting. I did however use the BLT as a base for it.	[ 1 ]
so the underscore forms of the option names are all deprecated since this is a new option let us leave out fixed ips	[ 0 ]
Whoops thought i fixed that.... This happend when my VM was dying and copy and paste were getting weird	[ 1 ]
you are not appending any new datasets in this class is this a wip	[ 0 ]
we should not need to do this nonsense now we have migrate data obj livemigratedata detect implementation in the compute manager but it needs a closer look	[ 0 ]
for httperror you may want to parse the error code it is quite common to get those errors should be retried	[ 0 ]
this prop should not be called so it show tight coupling to one of the possible parents you should name it according to what this prop controls like renderlogslink of discardingpossible or nodeoperationspossible	[ 0 ]
Done, changes kinda blew up but see if you like it more.	[ 1 ]
should these params be lined up the only reason i say should is b c it is such a pain	[ 0 ]
also need to invalidate it just like the other validate s	[ 0 ]
that s because pipelinemanager used to get a publisher manager as arg and created transformermanager itself now both have to be created before and passed as arg i have reorganized this code a bit and remove the useless ext name arg	[ 0 ]
this is ugly but all it does is modify the render pass id we do support cor for the root added some braces to make it clearer	[ 0 ]
i think this needs to become an array of structs with each struct containing the struct s c x i c periph id and anything else you need	[ 0 ]
commit message is not a chat this sentense should be removed	[ 0 ]
This is a lot worse than it used to be. weak_mod gets applied every tick but only gets removed once. Also... division ugh... disgusting	[ 1 ]
I hope you plan for these to be constructable in some way because in case of destruction or it being stolen the player is left screwed over.	[ 1 ]
there are white spaces which need to removed	[ 0 ]
damn yeah we'll have to drop them here and store them for this...	[ 1 ]
A screw up on my end, sent a change to the wrong place. Please disregard.	[ 1 ]
Should we close this in favor of #2226? @nicgrayson @rubbish	[ 1 ]
this tlog line should be removed there is one below line	[ 0 ]
that shouldn't be possible in any way, what the fuck byond	[ 1 ]
not working but i have a dirty hack for it	[ 0 ]
dumb question in nova when is the quota taken away from a tenant before this deploying is started or after it is successful	[ 0 ]
Ok tested. Before=rubbish. After=better. Please merge	[ 1 ]
> @ondbyte Ours too. Building the product just prints rubbish on the console. Literally everything that uses colors.js is affected	[ 1 ]
wtf is that ? comparing a string and nullptr ?	[ 1 ]
Wow fuck you too buddy.	[ 1 ]
this command should be removed	[ 0 ]
you call this exact proc three times, when overlay_fullscreen should only be called when shit actually needs to be updated	[ 1 ]
Just use collision for the particle accelerator. If it updates every tick it's gonna take up 1ms of server time again which is ridiculous.	[ 1 ]
1. Why the hell would you put this on the MC. 2. Hardcode for snow map specifically.	[ 1 ]
please call this image api instead of glance api	[ 0 ]
maybe simply trust context reauthentication auth method cfg conf reauthentication auth method	[ 0 ]
make your own damn PR to fix it	[ 1 ]
@Grossley chill ur tits dude, just makin an issue here. You dont got to get all gamered up over here.	[ 1 ]
type audit avc denied read for pid comm mediaserver scontext u r mediaserver s tcontext u r mediaserver s tclass socket	[ 0 ]
can this command hang what happens if tgtd dies	[ 0 ]
There even is an `addMessageWithSound` method, this is so weird.	[ 1 ]
ditto i do not see how storing tar file is better than storing tar gz	[ 0 ]
you need to synchronize between rpres start stop and rpres remove the spinlock will not do it	[ 0 ]
your indents are kinda fucked here	[ 1 ]
screw those new-age data size specifiers	[ 1 ]
`var/turf/mob` is an incredibly cursed variable name	[ 1 ]
Could you note where the actual change is instead of a fuck ton of var renames?	[ 1 ]
ditto server client py	[ 0 ]
what about resources getresourceleaves item for item in node values	[ 0 ]
you should probably include mm h instead	[ 0 ]
subnet request subnet cidr is already an ipnetwork instance therefore the call at line will be useless even if harmless as well	[ 0 ]
our utils execute totally ignores run as root kwarg i guess the correct solution is to fix utils execute rather than to cripple it is callers	[ 0 ]
holy shit thank you for killing this	[ 1 ]
rico i do not see any issue here if find gets some executions we can expect them to be there for delete	[ 0 ]
oops mistake only swift blah is correct as on l path is used in case of swift blah path is blah in case of swift blah it is empty string	[ 0 ]
But it never did add firestacks, that was Hellwater... which I think was just an admin meme.	[ 1 ]
mock guest has all of his methods mocked it is not necessary to override them s mock guest get xml desc return value self dom xml	[ 0 ]
we should not be importing neutron any where here we want to get rid of this drive r	[ 0 ]
how is this actually interpreted by vm concierge as an upper bound on disk size	[ 0 ]
okay i am going to remove address sanitizer config extra shared libraries in my next upload	[ 0 ]
rather than doing this have you considered doing lazy joined on dns nameservers and host routes for reference see c having said that i am gonna make the same comment i made on a few other reviews i personally would like to get us to think about having rally scenarios accompany these types of patches ideally we would have a patch that introduces a rally scenario that sets a baseline and a dependent patch that shows the performance gain	[ 0 ]
Tbh, this edit means less unification than more unification. If the user forgets the arguments \u2013 it's hard to help. Otherwise we have to check and make sure that each and every single argument is in place across all exchanges. Which is nonsense, cause we assume the user can code and can read and follow the documentation.	[ 1 ]

ideally we would be able to let the label fill it is parent with maybe a tiny bit of padding between elements i tested in spanish really quickly and on my nexus x the home button and share button text gets ellipsized it d be nice if we gave it a little bit more room it seems like we should have the space screenshot

[ 0 ]

if all the libupnp stuff is currently bundled - it is by accident - no need to have all this libupnp crap in the bundle - sry can't do lone comments because github sucks - you know which lines i mean i hope

[ 1 ]

should we just drop out early if featureutilities isnotouchmodeenabled infobar supportstouchlessmode is there logic we need to execute if we are not actually goign to show the infobar

[ 0 ]

Oh damn, i didn't saw the UpdateOwner call - You're right with this one. Good catch!

[ 1 ]

to be execute to be executed

[ 0 ]

Congratulations you just fixed a fuck ton of lag caused by humans bleeding.

[ 1 ]

please remove trailing spaces p sed ri s s

[ 0 ]

mistral

[ 0 ]

Just make these four goddamn lines into a proc so you don't copypaste the ENTIRE section

[ 1 ]

i am stupid of course we do not have cssvalues here we could just add a create function that takes a string as you can see the cssvalue one immediately converts to string anyway static cssunsupportedstylevalue create const string value return makegarbagecollected cssunsupportedstylevalue value

[ 0 ]

bro just roll it randomly between 5 and 100 instead of using the shittiest die to exist

[ 1 ]

worth noting the security model in jenkins is such that it is not really safe to mix trusted and untrusted slaves and it is also not generally safe to reuse untrusted slaves as such we keep our trusted reusable slaves bound to a separate jenkins master from our nodepool managed slaves if you have a similar split you can simply avoid running this for the jenkins masters handling your reusable slaves and only use it for jenkins masters handling your nodepool managed slaves

[ 0 ]

good damn idea, I'll make it shake the mob. what do you mean by message to the user? some feedback to the mob who did it? shouldn't they see the message? or is visible message pwning me

[ 1 ]

remove blankline

[ 0 ]

i am thinking about aborting here in the case when flags point to existence of the r addend field is handled by read group fields and this part of code should be unreachable

[ 0 ]

Might as well max it out to `0xFF`, it won't cost any memory on the MCU as it just defines the range of values the host should expect. And someone, somewhere, will *really* want to send an `0xF2` for whatever wacky idea they have.

[ 1 ]

just make sure that no one else reads the current selected client and attempts to an operation on them e g maybe go to the memory view and attempt a heap dump gc on that dead client

[ 0 ]

bad merge you re shadowing oslo config cfg

[ 0 ]

how will the user know which host succeeded

[ 0 ]

i think handler kind is a list so probably handler kind would work

[ 0 ]

oh in local post install cmd we have to use local installed module instead of

[ 0 ]

This system is moronic and I don't like its extension. It should be using material IDs and then getting sheet type from the material datum, not doing this weird path nonsense.

[ 1 ]

was going to ask why you did not do this as a query string parameter then i realized i was being dumb because nobody in their right mind should do a delete off a get method so yes this a handy trick

[ 0 ]

we already have the api group why register it again just do conf register opts your opts group api

[ 0 ]

how do you plan to lock varedits without locking the entire flag entirely? (and dont lock it entirely damnit)

[ 1 ]

check pid

[ 0 ]

Ehh I don't know and I honestly don't care too much. I guess it's a limit to prevent opening 9000 files and trying to do some nasty shit.

[ 1 ]

you re going to want to kill me but redrobot said these should be in the format assertequal expected returned he dinged me on that for mine i do not know if worth ask redrobot

[ 0 ]

What the unholy shit is this? Just make it go by dir and check only once if that dir is blocked. This is so overengineered.

[ 1 ]

Yeah I know that you dislike @Deprecated :smile: Anyways, it's ok with me. Break all the shit. :+1:

[ 1 ]

Ah shit, yeah, you're right, I forgot to put H. before the functions!

[ 1 ]

Oooh damn. Okay then!

[ 1 ]

Bah the ternary operator is crap.

[ 1 ]

we should maybe send sigterm to run oci and let that relay the sigkill to it is child

[ 0 ]

Can we please do this without overloading `--net` anymore? That flag has gotten ridiculous to parse

[ 1 ]

can you kill the leading spaces here

[ 0 ]

bad sym l

[ 0 ]

lol what the fuck happened here (seems alright to me now though)

[ 1 ]

it should break if something happens, cause if merging gases goes wrong we should know (null mixes or other weird stuff)

[ 1 ]

If the diff sucks, see https://github.com/roligugus/suricata/blob/clang-format-proposal-v4/doc/devguide/codebase/code-style.rst for how it's displayed

[ 1 ]

As I understand it no, there was no existing recipe for a "Meat Sandwich" I thought about letting the player craft it cold but using raw meat in an sandwich sounded disgusting. I did however use the BLT as a base for it.

[ 1 ]

so the underscore forms of the option names are all deprecated since this is a new option let us leave out fixed ips

[ 0 ]

Whoops thought i fixed that.... This happend when my VM was dying and copy and paste were getting weird

[ 1 ]

you are not appending any new datasets in this class is this a wip

[ 0 ]

we should not need to do this nonsense now we have migrate data obj livemigratedata detect implementation in the compute manager but it needs a closer look

[ 0 ]

for httperror you may want to parse the error code it is quite common to get those errors should be retried

[ 0 ]

this prop should not be called so it show tight coupling to one of the possible parents you should name it according to what this prop controls like renderlogslink of discardingpossible or nodeoperationspossible

[ 0 ]

Done, changes kinda blew up but see if you like it more.

[ 1 ]

should these params be lined up the only reason i say should is b c it is such a pain

[ 0 ]

also need to invalidate it just like the other validate s

[ 0 ]

that s because pipelinemanager used to get a publisher manager as arg and created transformermanager itself now both have to be created before and passed as arg i have reorganized this code a bit and remove the useless ext name arg

[ 0 ]

this is ugly but all it does is modify the render pass id we do support cor for the root added some braces to make it clearer

[ 0 ]

i think this needs to become an array of structs with each struct containing the struct s c x i c periph id and anything else you need

[ 0 ]

commit message is not a chat this sentense should be removed

[ 0 ]

This is a lot worse than it used to be. weak_mod gets applied every tick but only gets removed once. Also... division ugh... disgusting

[ 1 ]

I hope you plan for these to be constructable in some way because in case of destruction or it being stolen the player is left screwed over.

[ 1 ]

there are white spaces which need to removed

[ 0 ]

damn yeah we'll have to drop them here and store them for this...

[ 1 ]

A screw up on my end, sent a change to the wrong place. Please disregard.

[ 1 ]

Should we close this in favor of #2226? @nicgrayson @rubbish

[ 1 ]

this tlog line should be removed there is one below line

[ 0 ]

that shouldn't be possible in any way, what the fuck byond

[ 1 ]

not working but i have a dirty hack for it

[ 0 ]

dumb question in nova when is the quota taken away from a tenant before this deploying is started or after it is successful

[ 0 ]

Ok tested. Before=rubbish. After=better. Please merge

[ 1 ]

> @ondbyte Ours too. Building the product just prints rubbish on the console. Literally everything that uses colors.js is affected

[ 1 ]

wtf is that ? comparing a string and nullptr ?

[ 1 ]

Wow fuck you too buddy.

[ 1 ]

this command should be removed

[ 0 ]

you call this exact proc three times, when overlay_fullscreen should only be called when shit actually needs to be updated

[ 1 ]

Just use collision for the particle accelerator. If it updates every tick it's gonna take up 1ms of server time again which is ridiculous.

[ 1 ]

1. Why the hell would you put this on the MC. 2. Hardcode for snow map specifically.

[ 1 ]

please call this image api instead of glance api

[ 0 ]

maybe simply trust context reauthentication auth method cfg conf reauthentication auth method

[ 0 ]

make your own damn PR to fix it

[ 1 ]

@Grossley chill ur tits dude, just makin an issue here. You dont got to get all gamered up over here.

[ 1 ]

type audit avc denied read for pid comm mediaserver scontext u r mediaserver s tcontext u r mediaserver s tclass socket

[ 0 ]

can this command hang what happens if tgtd dies

[ 0 ]

There even is an `addMessageWithSound` method, this is so weird.

[ 1 ]

ditto i do not see how storing tar file is better than storing tar gz

[ 0 ]

you need to synchronize between rpres start stop and rpres remove the spinlock will not do it

[ 0 ]

your indents are kinda fucked here

[ 1 ]

screw those new-age data size specifiers

[ 1 ]

`var/turf/mob` is an incredibly cursed variable name

[ 1 ]

Could you note where the actual change is instead of a fuck ton of var renames?

[ 1 ]

ditto server client py

[ 0 ]

what about resources getresourceleaves item for item in node values

[ 0 ]

you should probably include mm h instead

[ 0 ]

subnet request subnet cidr is already an ipnetwork instance therefore the call at line will be useless even if harmless as well

[ 0 ]

our utils execute totally ignores run as root kwarg i guess the correct solution is to fix utils execute rather than to cripple it is callers

[ 0 ]

holy shit thank you for killing this

[ 1 ]

rico i do not see any issue here if find gets some executions we can expect them to be there for delete

[ 0 ]

oops mistake only swift blah is correct as on l path is used in case of swift blah path is blah in case of swift blah it is empty string

[ 0 ]

But it never did add firestacks, that was Hellwater... which I think was just an admin meme.

[ 1 ]

mock guest has all of his methods mocked it is not necessary to override them s mock guest get xml desc return value self dom xml

[ 0 ]

we should not be importing neutron any where here we want to get rid of this drive r

[ 0 ]

how is this actually interpreted by vm concierge as an upper bound on disk size

[ 0 ]

okay i am going to remove address sanitizer config extra shared libraries in my next upload

[ 0 ]

rather than doing this have you considered doing lazy joined on dns nameservers and host routes for reference see c having said that i am gonna make the same comment i made on a few other reviews i personally would like to get us to think about having rally scenarios accompany these types of patches ideally we would have a patch that introduces a rally scenario that sets a baseline and a dependent patch that shows the performance gain

[ 0 ]

Tbh, this edit means less unification than more unification. If the user forgets the arguments \u2013 it's hard to help. Otherwise we have to check and make sure that each and every single argument is in place across all exchanges. Which is nonsense, cause we assume the user can code and can read and follow the documentation.

[ 1 ]