Datasets:

showvikdbz

/

code-reviews

like 0

nit get rid of useless return value

how come you do not set closed true always should you also null out contents

typo in commit message which test should be executed when ran within multiusertestcase

Take this You Lesbian

ok come on man how'd you fuck this one up

you do not need this block to be in the else since your if will jump to fail

was going to do that originally but got bummed out because returning multiple values from a function sucks before c++17, will change

I am on oC10 core stuff at the moment. This can wait until the "flaky smoke test" stuff is sorted out. This is only rewording to tidy things up. I need to look and see what rubbish I did in the refactoring commit https://github.com/owncloud/web/pull/6077/commits/4e8674250333e046d66cb27779d5fdaf00a1a95d

Huh, bizarrely, dualshock controller works if you remap it, but all keys come us as like mouse.X and so on. Not ideal. I'll deal with this tomorrow and make a 1.12.2 ... darn it.

Yuck. 20% chance of the emag working to unlock the cover? If you successfully disable the ERT borg you shouldn't be countered by some RNG nonsense \U0001f44e

rtt request count k

i do not know such a case i will run tests with notreached as a separate cl

Any support plans? SQL Notebook without parameterization is clumsy enough to write.

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

in ha mode multirole is redundant

Microsoft has shifted its focus to WinUI 3, give it up, rubbish WPF.

done though in a scripting language like sh it is not clear to me what advantage compressed rootfs might have over rootfs xz

Yep ugly shit. Refactored that as well

Actually, my problem was simply build_network being called every tick. Commenting it out didn't seem to break anything, but, again, we've overhauled shit.

i think lxml is a third party library

it will raise notimplementederror you can pull this patch and try it out btw i am going to get rid of this block of code anyways in my next patch

nitpick parameter is not that long to type is parameter blank would sound better

Doesn't this already fucking exist? Are you porting blindly?

should be after line though seems line useless

max days

arggregate should really be aggregate here and below

kill this carries exactly zero information if there is ever anything useful move to header file format as doxygen documentation

As someone that has worked on tracked vehicles for years, yes, they would tear up wooden flooring, carpets, grass, etc. I have no problem with any of that. Tracks fuck shit up.

why is regexp quote necessary here why not just metering secret metering secret

Deliver messages (posts, replys) to users who alluded in this message. For example: post with text \"faggot --> @username\" will delivered to @username even he doesn't subscribed to poster

This behavior is ridiculous. Can anyone explain why this isn't documented at the very least?

i meant length i typed too fast too silly me

you assign the value just to echo it i would not bother as you can implement this func in one line awk f print nf version file

isn t system api s dbus service constants h violating this then

kmaxvalue bottom right

This is ridiculous..

err assertraises result lib buildscriptfailure commands runbuildscript self root self cmd

none i think that this should be a uuid

there is autorequire l l

remove initcap on example

okay this is really my lack of ability to articulate or plain stupidity i think what i wanted to reflect was one can do a create firewall without supplying a router id or port id list and in that case the firewall will be in a pending create state the logical resource is created but has no binding to any router or ports once we update the resource with a router or router list of ports these columns will be populated so in that sense it is not required but that only implies logical resource creation the pending create state is in line with the current implementation

what style guides are we working with x y is preferred

like omg using @ when talking to someone is for losers

add register a new datastore

> Blob Zombie Concert Well fucking done.

I added it to the version with int not enum and it compiles but gives me a large scale and nonsense output

rather than have a note there should be a bullet which should be first that says to set auth plugin

run tests sh should be run tests sh with an underscore instead of a dash

should have a docimpact if we end up using a new conf option

sorry i was misunderstood i fixed this to use a fk with cascade delete

because i am an idiot

it seems dumb that we even have a cli for a thing that the nova api does not support we should probably have a separate patch to deprecate or just remove that cli since it is busted

Damn looks like we do need it

there s no vm here so vm lastrun should just be var lastrun

arm processors is shit , so it need to increase physics step

I can't even imagine how disgusting the movable buckle cuffs would be with secborgs

realized essex should be removed

goto fail sys

let us make this method private ie get trans wrapped session key and move it down with the other private methods

i think using generics would actually make this more simple fn validate formats t u f ptr const t num usize f f libvdaresult vec u where f fnmut t libvdaresult u do validation convert unsafe slice from raw parts ptr size iter map f collect libvdaresult vec and in the caller let input formats validate formats vda cap input formats vda cap num input formats inputformat new let output formats validate formats vda cap outut formats vda cap num output formats outputformat new

can you kill this extra chars line

is there a need to run this through the il n library will there even be a translation for this since it is a new string template that you are creating which i would assume does not have an equivalent translation that already exists i am fine with it but just seems sort of useless especially if the yum il n will not have a underlying translation anyway

it looks right in the ide :^) I probably used spaces instead of tabs or some shit

i do not think this works go will garbage collect the slice cced you on an email asking about it edit apparently it works

this should be pimp pcjkoptions lock if pimp pimp reset new svtcjkoptions impl pcjkoptions pimp

oversight lol idk how that got on this branch wtf!

this is getting silly how about if none not in self driver name self driver format or if any self driver name self driver format or any of the other far more readable and maintainable ways of doing this

lmao t ```suggestion \tfor(var/area/shuttle/shuttle_area as anything in shuttle_areas) ``` lmao tg what the fuck

It looks like spotless screwed up your formatting here.

That's what I was getting at this whole time, although I guess it's a pretty lousy diagnostic. Still, it doesn't seem like we need to do *better* than what we do for `let`.

globals with destructors guess it does not matter for the dynamic linker

We need a general proc for this shit.

this should still be a refreshonly otherwise you will migrate on every puppet run iirc

And yeah you could screw with them, but you can already do that without my pr, just screwdrive the button and take them out

useless comment delete please

fold this into the if you do not need this bool anymore

Isn't this just a drinking glass? Gonna look weird, What about `bone-break`, do we use that?

Looks like someone before me screwed up the ids. I will fix this

these pmu access structs should be static

should we kill those too

you should pass the entire datastore object not it is properties

this is a django setting it belongs in the next section

`glasses` again. Change `sucking`. Remove the extra space at the end of the string.

i want to split ovs core mask and ovs pmd core mask chunks

same thing this will stick the properties to a java util

is not case insensitive so that is not an equivalent change while we are at it let us get rid of that ugly straircase effect align all the pfont under the one on the if line if pfont pfont pfont

Well yes, merging-vs-rebasing is a hot topic in itself, there're both pros & cons to each... But from the perspective of project maintainer \u2014 I find that *merging master into feature branch* produces absolutely disgusting history graphs. It takes way more effort to make any sense of such history (should anything need go wrong and need reverting, cherry-picking, cross-merging, etc) \u2014 compared to the neat linear rebased-feature-branches.

I'm a moron, nvm

the call to madvise will already have corrupted errno see posix fallocate cpp for how to do this

agree that this line should prabably be removed

damn I'm blind. Cool.

hinductionvaranalysis induction analysis

these mocks should be done in test copy image to volume instead of here since they have no use here

Remove and see what happens, this shit shouldn't be necessary

h git commit title fix sqlalchemy api to make instance destroy also delete system metadata should be under chars

i begin with apologies because i do not know much about cassandra and it is client support but i have a proper stupid question for you why does not this class if not module exist in the client a lot of this seems to be ceilometer agnostic

can you please add a some information here about why may i suggest the copyrights held by openstack llc were transferred to the foundation when it was created

Didn't you just fuck up spider queens massively doing this?

not much gained in trying to make this a function and macros are evil so i am not sure there is a good way here skip back a few patch sets to see some attempts at fixing this but none look good natural since this is just the dump code i would prefer leaving it like this

stub out self disk adapter run vios command as root