text
stringlengths 1
5.2k
| labels
listlengths 1
1
|
|---|---|
the vblock should be x not xf bytes big
|
[
0
] |
these mojom enums should live in storage common quota too
|
[
0
] |
i agree that looks like a comment i would write myself it feels stupid to remove it from this patch then add it back in the next patch but it is the right thing to do
|
[
0
] |
why not port forward UPNP is a bag of rubbish and if you have 2 PCs doing chia UPNP will not work.
|
[
1
] |
null class loader is still handled by classlinker visitclassroots
|
[
0
] |
to go with the rest i think these should be access ip
|
[
0
] |
bool is main frame false
|
[
0
] |
there should be an invalidhttpresponseheader added to exceptions
|
[
0
] |
This is the color luminosity, which I screwed up on the abbreviation (Y is luma, not luminosity). But looking up more color terms, just calling it `brightness` or `luminosity` is probably best.
|
[
1
] |
remove the white space
|
[
0
] |
this missed the point a bit i was suggesting argv be configurable so you can pass in a dummy tlsdate since you hardcode usr sbin which makes any sort of isolated testing nearly impossible
|
[
0
] |
i would refactor this so none is an early abort rather than if key
|
[
0
] |
Gah, you're right - I'd forgotten about that in the damn shrinker merry-go-round. If we have to twiddle Assetic anyway we may as well just go straight to JSqueeze, I guess? I'll send a different PR in a bit.
|
[
1
] |
@jancborchardt you can already try out this branch to get a (somewhat clumsy) feeling of the interaction.
|
[
1
] |
what other rsa s are you now picking up oh the odd rsa ones i might just avoid them
|
[
0
] |
argument should be trust id trust id
|
[
0
] |
remove the black space
|
[
0
] |
target ld toolchain prefix clang host exeext
|
[
0
] |
nit should be delete a container
|
[
0
] |
This line break is weird.
|
[
1
] |
some aspects will change e g the fk i think it is ok to list here
|
[
0
] |
annoying nits weight properties
|
[
0
] |
@Sebastianv650 you must start paying more attention to the direction Marlin is going. Nevertheless the idea that using objects makes the source code slower is nonsense, it all depends of the "stupidity" of the coder. A real stupid one can even make ASM code slow. ;-) At the end of day great feature and great work, congrats !
|
[
1
] |
this is node s uuid
|
[
0
] |
we also need to check that the source origin is not being spoofed and kill the renderer if it is see
|
[
0
] |
you should use ipwrapper netns delete
|
[
0
] |
oh right it will be the class name not the node name cool
|
[
0
] |
Ugh, that's annoying :(
|
[
1
] |
Oh that sucks @tralezab we should find a new Jerma joke
|
[
1
] |
calculate the scalar first, then do the if thing. this is gonna fuck some people over
|
[
1
] |
sys stderr or sys stdout i do not see stderr used
|
[
0
] |
will fix addresspool silly me done
|
[
0
] |
temp vol ref destroy
|
[
0
] |
Change to `if (!TailCircle.SamplePlaysOnlyOnHit || TailCircle.IsHit)`
|
[
1
] |
this change is useless
|
[
0
] |
this exec should be removed we do not need to minify
|
[
0
] |
we do not need auth version auth protocol i think
|
[
0
] |
Damn... I tried to look smart... \U0001f604
|
[
1
] |
prefer oustring sext odb
|
[
0
] |
please get rid of backslash use brackets
|
[
0
] |
Its some random item a bee contributor added. I'm also gonna be honest, I have no idea what the hell `BESPOKE` means
|
[
1
] |
even though this solution is acceptable it is still not clear to me the difference in behavior between this approach and raising shareresourcenotfound given the code below l if you always throw shareresourcenotfound no if else condition as in your code manager will do exactly what you are doing it will swallow the exception and proceed if it is not deleting path as you are checking for such in share status it will throw exception just as are you doing since manager is only swallowing the exception when deleting the share the only scenario which i think this is valid is if this library is used outside of manila which is what i understood based on our irc conversation but by using share status here i am assuming it will not be used outside of manila manager code path so i am still confused as why always throwing shareresourcenotfound does not achieve the same behavior as you are implementing in your driver code please correct me if i am missing something
|
[
0
] |
this is pretty cool
|
[
0
] |
astr astr replaceall strictly speaking it should be replaceat but that does not matter here right
|
[
0
] |
This lerp is totally fucked. 1. This needs to divide. 2. It doesn't factor in round start time 3. This should just USE LERP
|
[
1
] |
i think you can simplify by defining safedelete that accepts f args and kwargs and calls it while catching notfound then in create resource you call self addcleanup self safedelete data id and in create bandwidth limit rule you call it with the additional policy id the advantage is that you do not have to define the wrapper in line the disadvantage is that any new create method will have to know to use self addcleanup with self safedelete and that it will not be automatic like you did here with overriding addcleanup i am not in love with the idea of overriding addcleanup and i think that any new create method people will just copy paste something else in this file anyway so overall i like my suggestion better
|
[
0
] |
(yes, some combinations are a bit ridiculous, please vote for merging of permissions here https://github.com/owncloud/core/issues/12970
|
[
1
] |
no need to chain filter by this way can put them all into one filter by vlan id vlan id switch ip switch ip port id port id
|
[
0
] |
only the h for the cc should be here the rest should be below and sorted
|
[
0
] |
damn you were faster :)
|
[
1
] |
It's below "good for 1 point". Thick skin is better. At 4 points, it's utter trash.
|
[
1
] |
Gay Piece of Shit handling so fucking op, this 2nd after R34
|
[
1
] |
that sound u make when u shit
|
[
1
] |
dump somewhere that this is a catch stack map
|
[
0
] |
i would rename this method to set api extensions path because this is what it is update may give the impression that a path has been added to this aim it might be better to have an append logic
|
[
0
] |
sorry if i am not being clear i am suggesting that you change activatetabat to take enough context so that tabstripmodel can entirely control calling to the latency recorder what is the user gesture argument should become an enum tabstripmodel activatetabsource kmouse ktouch kotherusergesture knongesture tabstripmodel activatetabsource source kotherusergesture if event type ui et mouse pressed source kmouse model activatetabat model index source if you do this then i do not think you need to expose the latency recorder at all
|
[
0
] |
Once-class-per-file is ridiculous. We have modern editors with all kinds of features for managing large files. I'm much happier navigating within a few files than having to deal with hundreds of tabs. I vote we suppress this rule.
|
[
1
] |
nit we only send down one rect why framerect s changed
|
[
0
] |
consider using std string out that you only overwrite if getstring is successful see below
|
[
0
] |
well this it this page that s a silly reminder now is it now i would suggest you remove it
|
[
0
] |
ahaha, yeah kind of ugly workaround that was using for testing. Forgot about it. Will use your formula instead xD
|
[
1
] |
you can also simulate pausing a vm using kill stop pid kill cont pid although it is not officially supported
|
[
0
] |
Truly weird stuff, there seem to be no end to wacky edge cases :) nice fix.
|
[
1
] |
instead of overwriting the value after fetching why not modify this get test firewall policy attrs method to accept another parameter for audited like below def get test firewall policy attrs self name firewall policy audited audited
|
[
0
] |
this change is just a bug flags help is a string and will never be equal to flags true if we just want help to return failure there are much better ways
|
[
0
] |
this is unused and should be removed
|
[
0
] |
not your fault but i think that this flag should have instead been named acpi ac and made to default to being on then boards like this one could unset it in their overlays
|
[
0
] |
Note that this is not a Kilobyte, but a Kibibyte. But I looked it up aaaaand apparently they use Kio fo that. Huge WTF from me, will change.
|
[
1
] |
Did this get sucked in from another PR?
|
[
1
] |
yeah this is chicken and eng issue android build top is set by envsetup sh i think you should state that envsetup sh even lunch must be run before running this script
|
[
0
] |
this is not rendering properly
|
[
0
] |
this is prettier than my split_at() crap...
|
[
1
] |
i would like to reuse here consts oswl resource types
|
[
0
] |
kiall after a lot of investigation we found the better way to add tracing to all rpc services without this dirty hack we can just add metaclass to base manager class which i am going to do now
|
[
0
] |
Why should alphabetical sorting of the reagents list be an issue? It's bloody eyecandy and doesn't effect playstyles at all.
|
[
1
] |
The assumption here is that you're not insane, and rather meant to write code that made sense. For example, it's likely that if the compiler flags a warning here, you meant to write `&&` instead of `||`, or `===` instead of `!==`, or one of your operands was incorrect. This logic has found *multiple* bugs in existing well-tested codebases. It's a good trade-off vs just allowing nonsense code for the sake of allowing it.
|
[
1
] |
useless
|
[
0
] |
> https://www.reddit.com/r/Android/comments/j3zgmm/managing_files_in_the_androiddata_folder_on/ OMG, WTF. Another shitty design decision from Google under the sauce of "security". It looks like that they want to have only messengers, Facebook and Instagram on the platform. Apps that work with files? Meh.
|
[
1
] |
oh crap that one was for debugging to get an idea about where its currently at
|
[
1
] |
will this trash the sparseness it is an important step but there are other ways for us to go about it this is good because it is simple
|
[
0
] |
you need to add check services false here because after compute node deletion we will have services in down shown by nova
|
[
0
] |
the false is arbitrary and non obvious specify it as a kwarg e g catalog false
|
[
0
] |
this should not be needed get project raises projectnotfound for an invalid project id
|
[
0
] |
outstanding
|
[
0
] |
use tc getcc from the toolchain funcs eclass you need to also pass in cppflags and ldflags you should line wrap this command so it is one arg per line put die at the end of the command
|
[
0
] |
value kuuidbase
|
[
0
] |
do not use tempfile directly in chromite sources instead do a with block from chromite lib import osutils with osutils tempdircontextmanager as tmp path blah
|
[
0
] |
i think l l is one command better use
|
[
0
] |
why do not we use a common method execute cmd to execute all comands
|
[
0
] |
@kotarou3, browser, OS taskbar/dock/menubar, showdown's headers/footers, and your tournaments bar add up to over 300px on average. 200px is approximately half of the remaining 400pxish, which is ridiculous.
|
[
1
] |
the dumps are concurrent in separate processes we just handle the output from them serially can you think of a better name for this
|
[
0
] |
Siltbreaker will breathe Disintegration Blast at you through Respawn Immunity and kill you, cocksucker
|
[
1
] |
What the.. how did that even get in there?! Goddamn git screwing up again. That is part of an entirely different pull request, I don't know why that's here.
|
[
1
] |
Stupid vscode screwed up again :(
|
[
1
] |
shouldn t we reuse the stuff we have in silent call there is a logic which updates action logs also i think we can reuse check which we already have by just inheriting new class checkbeforedeploymenttask as checkbeforeprovisioningtask l l l
|
[
0
] |
you have not just deprecated the devtest overcloud sh build only functionality you have completely removed it keep the deprecation warning and maybe have this call out directly to devtest overcloud images sh directly to preserve the old behavior for a while
|
[
0
] |
but my varialbe is in line i think we should use the same place
|
[
0
] |
I am probably in the top 0.1 % of household income globally and my phone (iphone 6) doesnt support webgl2. Dont be ridiculous.
|
[
1
] |
so i would leave these in for when ironic params ipxe rom dir is undef that way if someone tries to use it on an environment where we do not have the file copying occuring they will not wonder why it does not work
|
[
0
] |
(Something went wrong with my rebasing strategy here, not expecting a review on this nonsense right now!)
|
[
1
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.