text
stringlengths 1
5.2k
| labels
listlengths 1
1
|
|---|---|
fc related code should be removed | [
0
]
|
What actually happened It killed me like a faggot even though I fed it. >Steps to reproduce if possible Kill with his grace. | [
1
]
|
don't use UITileList, it's crap just use MatrixManager.GetAt<RegisterItem>(pos) and you'll get a nice list of RegisterItems, no need to call GetComponent<> yourself | [
1
]
|
i think returning an ass error is a good thing to do will add | [
0
]
|
this should refer to cinder init cinder base has been deprecated for months | [
0
]
|
i guess you want to execute dockerctl shell nailgun supervisorctl restart nailgun here | [
0
]
|
self mock ip is already a mock object you should do this self mock ip get routers return value router | [
0
]
|
That's just wacky. As I said in the issue report, bizarre that it would even miscompile in debug mode in MSVC. Will test it on Android later, I haven't actually checked. | [
1
]
|
it is ok we will remove this script entirely soon | [
0
]
|
use die | [
0
]
|
"Better luck this morning" turned out to be awful foreshadowing, as my day got eaten by a cascade of "why does the same gfortran module work fine on one computer and illegal-instruction on another" rooted nonsense. Sorry. Assuming that's all straight now I'll take a crack tonight. | [
1
]
|
current user must have permission to run kvm vms with libvirt | [
0
]
|
since you are not passing in anything to trace wsgi or doing any conditionals in it i think you can get rid of outer and instead use trace wsgi f | [
0
]
|
what the heck is this :) | [
1
]
|
Actually it's pretty weird, `AudioDriverAndroid` *is* used in `java_glue.cpp`... That's very fishy. | [
1
]
|
Failed test results so far: * Galaxy S8 (Android 9, API 28): Misdetecting scoped storage, broken folder browser and wacky content URIs. Should have been using the old ways. | [
1
]
|
pls daddy using home directory sucks. your submissive breed-able crab femboy | [
1
]
|
blank line must precede bullet list | [
0
]
|
@SimenB the problem here is more about tooling. In phabricator diff view, the individual piece of the line will be highlighted in bright red/green, so you know that only the number changed. It sucks that github is still not doing that :( | [
1
]
|
`people show up wth more puppies` -> `people show up with more puppies` | [
1
]
|
This is starting to get ridiculous. It's already been 10 days since I created this PR to fix the annoying problem that I have caused by splitting the typings into separate files, and it's still not merged. What are we waiting for? | [
1
]
|
please set up lower bounds | [
0
]
|
shouldn t this be ldr w sp to avoid loading the bytes above and corrupting the method parameter | [
0
]
|
stupid oversight on my part i will add that back | [
0
]
|
how about we replace that with new stringbuilder append resource getprefix concatenation with empty string looks ugly to me and triggers idea warning | [
0
]
|
wtf is this really how it works? | [
1
]
|
yes it was the job of passing management false i will revert it as you requested but please note that nobody uses this argument except for this particular function update own password | [
0
]
|
can you put a note that this is a hack to get around a bug in old layoutlib code i would actually rather avoid these hacks but if our users have old versions of layoutlib that we cannot upgrade then i guess we have to | [
0
]
|
me neither stupid mistake thanks for kindly reviewing will rebase to last master | [
0
]
|
I kinda like the forsenics office/room change but giving IH a fucking chem machine? why would they need it?? You should try and interact with medbay, hell IH already get their own mini medbay so giving them their own chemistry makes no fucking senses this PR makes me puke | [
1
]
|
maybe the value for this should be maintained from executing env if it exists | [
0
]
|
magistrate doesn't need to be armored out the ass. This is also inconsistent with regular magistrates, who have literally zero armor (not even armored jumpsuits, like officers). | [
1
]
|
it would make sense to include asses result of the current implmentation | [
0
]
|
It is for sorting by weight. Unfortunately golang is absolutely disgusting when it comes to generic functions. See first example here https://golang.org/pkg/sort/. | [
1
]
|
i was wondering myself i am leaning towards naming this bvt tast and in the vm smoke tast because bvt smoke is an implementation detail of how tast achieves must pass mustpass might be better than must pass just for laziness | [
0
]
|
sorry i missed this comment earlier i have run this several times and have not run into that but that does not mean it is not an issue maybe you could pastebin your failing puppet run | [
0
]
|
and even that mspn might have a chance to be an empty string maybe you should try robert s suggestion | [
0
]
|
not a dumb question at all i copy pasted this from one of my daily driver configs using home is much more consistent | [
0
]
|
i still think you need a comment in the entry that is wrong rather than at the top but hopefully we can just switch to parsing proc pid exe instead | [
0
]
|
this line is a duplicate and should be removed | [
0
]
|
this comment should be lower where the attribute is actually set to false and you can pull the default setting to true out of the case platform block so you do not have to repeat it for ubuntu | [
0
]
|
Holy crap this was hard | [
1
]
|
typo ass ymetric made my morning also slightly confused as to why there is not a asymmetric constant defined in the secrettype class | [
0
]
|
thanks that is an interestin use of eq | [
0
]
|
you probably wanted die here eerror is not fatal | [
0
]
|
You dont actually need locate() in here if bloody_floors is a list, hell you could just do bloody_floors -= T and it wouldn't actually matter whether or not it was in the list | [
1
]
|
I can't believe you fags killed my teammate, apologise to him now. Also I can't use files.riichi.moe so I just stuck it on catbox, you fix it. | [
1
]
|
i c i c tunnel | [
0
]
|
no i am just an idiot i shall fix | [
0
]
|
not opposed but the syntax of this is kinda brutal makes me wonder if the fraction of a second savings are really worth it given the scripts usually run for minutes also could this be made global and reused for the same thing below | [
0
]
|
is this tobefixed comment obsolete the bug referenced is closed | [
0
]
|
consider an enum class for cleanliness | [
0
]
|
(I'm not convinced it's a good idea to warn on image actual layout being wrong at the time of vkUpdateDescriptorSets; however I think VkDescriptorImageInfo::imageLayout != VK_IMAGE_LAYOUT_GENERAL for a storage image descriptor is pretty much guaranteed nonsense) | [
1
]
|
What do you write about nonsense ? Filemanager is in 'develop' branch only ! You cannot test this branch but you are changing code in this branch ? Omg... | [
1
]
|
yes fields class attribute shadows the import in the class body so you cannot do the dictproxyfield assignment as a descriptor this should be changed in all object files really as shadowing like that is dumb and a bug really | [
0
]
|
hidden delete seems very tricky should not we just mark new slave as failed error if it gets corrupted | [
0
]
|
as noted below i think you need to assert what you support here right now if a user does rescue on an instance booted from volume it looks to me like the following will happen root vhd path will return a non existent path self vmutils detach vm disk will silently do nothing new disk with invalid path with be added this definitely will not work as the user intends as it is unsupported at the moment that is fine but it would be much nicer to the user to tell them that with a clear error message rather than leaving them wondering why their rescued instance does not work also i have not looked at the interaction with unrescue but it is critical that the above mess can be unwound with unrescue if it currently work it is presumably just luck tl dr not supporting everything is fine but i think you need to explicitly check that root vhd path exists and there is only disk if not you need to error out early this might be generally bad seems like the sort of thing you would want to know if it did not work separate issue though i do not know if this will explode or not | [
0
]
|
(The CI results are nonsense here, merging anyway.) | [
1
]
|
oh fuck i didn't think it would come up as a review woops | [
1
]
|
need to be removed | [
0
]
|
@david: Thanks! I hoped there will be more elegant methods than casts. Did not know about these ``#pragma's`` but I agree it is even more clumsy than the casts. | [
1
]
|
Space Asshole fits into fewer black prompts, is the thing | [
1
]
|
by default it should return none you do not specifcally need the none here should this not be an empty dict | [
0
]
|
this two lines does nothing mock s object s return value is mock object by default | [
0
]
|
nit better to use flattreetraversal nextsibling host next two is not intuitive | [
0
]
|
what that is what you were saying i m a dummy got them mixed up will fix | [
0
]
|
Oh damn. That's a problem... | [
1
]
|
prefer to mock linux utils execute | [
0
]
|
Damn! Seems like I forgot the binding... I don't know how it could work without it. | [
1
]
|
This way of coding it makes it skip all other checks. Vox can suck skeletons dry. \U0001f60f | [
1
]
|
if we had already decided on the features for a release extra risky code is less chance if there is riskiness as part of features then it is understandable and should be given more time to stabilise rather than not landing that feature the only thing left out is a critical bugs that should be allowed to get in based on it is merit | [
0
]
|
why the fuck does it do this. why is it not one icon with dirs | [
1
]
|
what you really do is abort sys exit you may change this to failed to resize stateful partition because new size dm is out of range dm dm abort | [
0
]
|
who the fuck removed the potassium from bananas | [
1
]
|
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
]
|
use this input mt report pointer emulation input true and kill single touch id | [
0
]
|
source demo openrc sh | [
0
]
|
Sorry about the ridiculous delay in getting a review here. Will take a look. | [
1
]
|
This is getting ridiculous /override Code Style / style / Lint | [
1
]
|
yeah this should not work at all according to my understanding maybe the flush is making this funky | [
0
]
|
should be enable service h eng h api h api cfn h api cw | [
0
]
|
What the _fuck_. I fixed that at least three times. Thanks for the heads-up, I'll fix it. | [
1
]
|
this script is actually just sh | [
0
]
|
elif ne then die don t know what to do with | [
0
]
|
the UI is legitimately awful too and is super buggy, whole thing needs a rework | [
1
]
|
mark sorry but i am not completely understanding why the fact that testing with an instance object does not matter here to just dumb this down for my own sake are you saying that once i change conductor manager compute stop to handle instance as a dict instead of object this test becomes useful | [
0
]
|
Yeah, this is quite inexcusable. It's a no-brainer feature, but I should really call the lack of support here a bug. We're having to use Jupyter notebooks with C# interactive so that we can import nuget dependencies for the project. That's just ridiculous :) | [
1
]
|
garbage | [
0
]
|
``` Configuration path "security.access_control" cannot be overwritten. You have to define all options for this path, and any of its sub-paths in one configuration section. ``` Come on! This is ridiculous. | [
1
]
|
clarify that this is a string containing a fully qualified manila share host name ex openstack backend poola | [
0
]
|
let s capitalise it to be consistent | [
0
]
|
this should be part of is create group below | [
0
]
|
am guessing without checking that in the puppet apache class it is service manage and not manage service like everbody else in the world | [
0
]
|
you are right it should not be changed here thank you | [
0
]
|
adding printout in board customize install sh where we will know if we overrode anything | [
0
]
|
disregard i see your comment to daryl about why this cannot just be a dump of the object s dictionary i would probably still like to see something a bit more elegant here maybe a list of values for the request kwargs that is iterated to build the dictionary but that is totally personal preference | [
0
]
|
for a safer selection of ciphersuites you probably want to crib from the puppet openstackid vhost once the machine is using something to terminate tls that supports ecdhe then this should be edited to remove the rsa aesgcm rsa aes so that pfs only is guarenteed sslciphersuite ecdh aesgcm dh aesgcm ecdh aes dh aes ecdh aes dh aes rsa aesgcm rsa aes aes anull enull md dss psk srp sslhonorcipherorder on | [
0
]
|
per hacking standars the falcon import should be moved down below standard lib imports with a blank line between it and the imports above and below | [
0
]
|
useless this line could be removed | [
0
]
|
needs global static url | [
0
]
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.