text
stringlengths 1
5.2k
| labels
listlengths 1
1
|
|---|---|
done i think they came from a messed up rebase
|
[
0
] |
it just so happens that this dependency is only mostly dead the chromeos input method kcapslockkey constant is used which ought to be removed in a separate cl
|
[
0
] |
maybe dumb question will ever print as something like nope
|
[
0
] |
you dropped the fips configuration check out the functional tests failures
|
[
0
] |
@jonpday The author want a solution, then I gave him a solution to log back in. He obviously can change the password lifetime to whatever he want after that. Your comment is ridiculous, I don't want to fight with you. So have fun.
|
[
1
] |
Adding a new parameter without increasing the formspec version will throw the error `Invalid TabHeader element ...` and become unusable on older clients. It sucks.
|
[
1
] |
useless break
|
[
0
] |
i want to ensure that none of the fields get dropped for e g if resource state got dropped accidentally in the code there should be a failure
|
[
0
] |
useless
|
[
0
] |
yes you should check that you are actually not returning anything in the parent id field
|
[
0
] |
here and below mark as private this lastgrantedchunk
|
[
0
] |
Disable collision on Earth at exit door to airport that trolls (god gaymers)
|
[
1
] |
odd missed in commit let me re upload i must have done something stupid
|
[
0
] |
theres still a fuck ton of duplication across all these, i get why its done but at the very least put all of these in their own folder
|
[
1
] |
we can now use get first node by host for old compat which is slightly less horrible than this
|
[
0
] |
self region name region name
|
[
0
] |
suggest just returning the session here and create the client instance later tuples are nasty when undocumented i thought this was an error at first
|
[
0
] |
you can just use nextcommitpositionoutput a a b b
|
[
0
] |
use unsignedbytes toint
|
[
0
] |
stupid me my editor shows them the same color and tricks me everytime p
|
[
0
] |
Changing the suggestion, then, and updating the top post. Making non-GUI the default, and add a `--allow-gui` (less ridiculous than `--yes-gui`). Changing that behaviour to be the reverse is a suggestion for another time (if ever).
|
[
1
] |
That's creepy. Hope I didn't mess something weird up. The loading stuff is all new code, so I expect there might be new bugs.
|
[
1
] |
a x also i think you will need to set the user bit setuid to be able to send signals to pid setuid sbin shutdown unless you will be calling this as root
|
[
0
] |
fuck no, use the new smoothing system
|
[
1
] |
Why would you make value, then... the fuck. just write the numbers instead of values like `AddComponent(/datum/component/shielded, 1, 2, 3, etc)`
|
[
1
] |
this start end pid filename are something that can go one heirarchy up that way you will not need to repeat these again and again for each snapshot these are constant for a document
|
[
0
] |
weird flex but this is just code churn to do exactly the same thing
|
[
1
] |
updated comment unloading loopy libs deserves verefication
|
[
0
] |
can we extract this to a function all the exit points are die statements so should be fine
|
[
0
] |
nit this kinda feels like an sshhost thing hot a host thing for example if there were an httphost we almost certainly would not expect it to automatically run within a bash shell environment if the user wanted that they would do something like sh c inline shell script i am not sure what the escaping behavior should be then feel free to leave it here though
|
[
0
] |
these are incorrect for sklrvp and should probably be removed
|
[
0
] |
No reason, brain fart
|
[
1
] |
this one is good we spit out a warning a die in a fire if this is below
|
[
0
] |
Hah. I hate this behavior of `Split` so much.
|
[
1
] |
Reason number 1 why YAML is shit: it enforces soft tabs.
|
[
1
] |
NACK any concept related to edits for racial/color concerns. "blackness" and "whiteness" are natural concepts and none of us should be wasting our time behaving like racists and attempting to cure it with passive-aggressive naming nonsense. Go do some real help if you're concerned.
|
[
1
] |
Can we change "Shit" to "Crap"? I don't believe we have curse words on the Gatsby blog.
|
[
1
] |
actually this seems to be the problem l as using the triggers a refresh of apache which puppet tries to do via systemd
|
[
0
] |
wrapping with exception here is not right this should be side effect n rpc remoteerror
|
[
0
] |
why did you remove instance id setting to context
|
[
0
] |
should be an elif as device owner can only be a single value
|
[
0
] |
because these vars are not used i will remove them
|
[
0
] |
Turf code is stupid weird and all floors have a floor_tile subtype.
|
[
1
] |
lest kill these two lines too
|
[
0
] |
the engine and connection is managed by get session and uses lazy loading it is better to use them rather than write twice
|
[
0
] |
i am fine with removing that revocations do refresh the cache of the revocation list done in the next patchset
|
[
0
] |
why run tests sh should invoke it at all i think it is a test s responsibility
|
[
0
] |
i do not like iteratively building things that i can declare in place
|
[
0
] |
diese uebersetzung verliert den interessanten teil des kommentars
|
[
0
] |
Yeah, it's rubbish. I'd be tempted to leverage headless chrome via the Puppeteer library to modify the links. I guess the alternative would be to modify NextJS to give it the support I described earlier in this thread. I don't think the core devs are interested in making those changes themselves.
|
[
1
] |
s examplei example s iscsi ilo iscsi ilo driver
|
[
0
] |
tavrc msg p avrc
|
[
0
] |
i think here elevating the context is not necessary are you sure this is necessary you re querying the db bypassing any sort of utility function that uses the context
|
[
0
] |
just to make sure are pig configs correct for mapreduce
|
[
0
] |
cosmetic nit failed pid avoid using
|
[
0
] |
it is a little gross that pthreads had this state for dlerror
|
[
0
] |
4mb is ridiculous huge and it took quite a while to load on my internet connection(probably Github throttles).
|
[
1
] |
this may be a dumb question but should not this be metadata key self metadata key value self metadata value
|
[
0
] |
this block needs to be wrapped in if debugblur
|
[
0
] |
That's a mix of `camelCase` and `SCREAMING_SNAKE_CASE`, that is not nice.
|
[
1
] |
is the bdm an object here if so then why not bdm destination type
|
[
0
] |
can you explain why this was chosen why not a traditional setup using f this is quick abi it does not have to follow native conventions if they are ugly
|
[
0
] |
gosh damn it, @RemieRichards
|
[
1
] |
Let me tell you I tried my damned hardest to avoid it, but in the end it didn't work out. At least the only lag you'll get if any is just for a moment (when the shot hits).
|
[
1
] |
this block belongs in stop opendaylight the bits in unstack sh go here
|
[
0
] |
Total fail - this PR is rubbish, we need this code still (like @thomaseizinger said).
|
[
1
] |
typo burdon should be burden
|
[
0
] |
the names are not mandated plankton and twinkie are not drp but both of them have src and snk pdos so i created a separate vif for src and snk
|
[
0
] |
Fuck u don't send me email again bastards
|
[
1
] |
There's no need for this else check right? all it's going to do is increase the amount of nested ifs, which can get ugly fast.
|
[
1
] |
oh i see what you mean yes i think that is correct onchildprocessdied will eventually be called when the other client unbind and i guess we get killed by the oom since cplh keeps it bound so when it does a restart it does not reuse it updated the comments
|
[
0
] |
This formatting is really weird.
|
[
1
] |
I don't think such a method will be implemented because I've seen it proposed before, but I will say that this use-case sounds like a terrible bastardization of "Config". Why aren't you just caching the data returned in the API calls? That's what the cache is for.
|
[
1
] |
sorry did not see this earlier rather use the pimp member to do this here and elsewhere further down calling lock repeatedly seems inefficient to me when we already have a shared ptr member
|
[
0
] |
you should probably throw a fatal error if the path contains a double quote something that will abort the build it should not happen in normal cases
|
[
0
] |
All of the code here is some of the most complicated in this codebase. I don't really understand it. Pinning / dealing with futures and getting it all to work alongside the event-based UI is crazy.
|
[
1
] |
oh the default return value is default a sentinel unittest mock default so let us be explicit about mock soft shutdown return value true
|
[
0
] |
i am not sure how useful keeping track of changes is for objects that get saved as json so i propose to either add a flag that will allow novaobject subclasses to not track this at all if they do not care or additionally make obj reset changes recurse into subobjects like obj what changed does but having this code here for the sake of being pedantic seems wrong to me
|
[
0
] |
imo dirname readlink e is more robust
|
[
0
] |
subnet id myfixed ips get subnet id
|
[
0
] |
hmm interesting does the inspect mode work with touchscreen you can try that with ctrl shit c in devtools and then touching on the page should reveal the element in devtools if that works i guess we can remove the handling also note that we do not really care about the primary pointer any would work for us i guess
|
[
0
] |
Did you accidentally re-indent a ton of stuff for some reason?
|
[
1
] |
according to self wait until ports state implementation self agent check ovs status reset mock seems useless
|
[
0
] |
This is awful. It might be better (but more invasive) to pass url_root directly to cockpit_web_response_new.
|
[
1
] |
Can't byond handle it somehow? Adding that logic here is adding unreadable shitcode
|
[
1
] |
Nevermind I'm full of shit
|
[
1
] |
Damn, again reviewed 3 days ago and forgot to submit.
|
[
1
] |
How can I make Travis fail? So that I can see the **Details** button of another fail. Well...it's ridiculous. But how can I see detailed message of the build when _All checks have failed_?
|
[
1
] |
yes it is probably useless as i got rid of the shared mem acquire for this case in an earlier version and forgot to remove this
|
[
0
] |
the problem is that if i assign the value to a variable which is not being used anywhere else the compiler throws a warning about a useless assignment so i just use a variable which is used earlier in the code to fool the compiler i will add a comment to this respect
|
[
0
] |
Make sure all this shit gets updated
|
[
1
] |
should domain id be token domain id i do not see domain id used in here
|
[
0
] |
this should be for obj content in matching ds objects
|
[
0
] |
this print statement is useless where is the output expected to go stdout will never be looked at
|
[
0
] |
it mean message storage default storage
|
[
0
] |
fyi an hardcoded should be a hardcoded
|
[
0
] |
dummy
|
[
0
] |
whoops fat fingers fixed
|
[
0
] |
nit put the full url or put these each on their own bug line gerrit seems to only convert the first one into a link and i am too lazy to copy paste
|
[
0
] |
I can't believe how clumsy it is to change your mind about what the root node should be. Just let me drag a child above the top node and reparent everything under it.
|
[
1
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.