Datasets:

showvikdbz

/

code-reviews

like 0

You say fortnite, dont use the max sensitivity in game, work between the ingame sensitivity and the app Deadzone f11/f12. I'm currently playing with 0,7 X and Y, ADS is about 0.6, and if works fine for me. Its just a matter of you sitting down a couple of hours adjusting to your like. some games are trash to adjust like PUBG, the sensitivity there sucks so much

May want to test this, I've a feeling if it's full it will print out every damn one of these when examining.

self fuel web is fuelwebclient instance self fuel web client is nailgunclient instance

should nuke the whitespace around the operator here

i wonder if we ever want to seriously consider stopping the nonsense of shadowing id just a thought

^^ This - and if you're just smooshing the dupont ends onto an XH2.54 socket, it is a crappy connection at best. If you can get someone to crimp a 5 pin XH2.54 plug onto your BLTouch, then it'll likely never fail again...

High standards good, ugly code bad. Happy to fix this for you. :)

It's not in there in the original MSO version, these chucklefucks added it

i still we need not this check i guess the driver will failed the abort live migration when the migration is not running anymore then the failed will record by instance action so this check did not have too much help for race condition

this shit aint brown and red for like 4 years

silly nit per hacking these need to be double quotes also the summary goes on the first line check the convention

Formatting is messed up here, please come back to fix this.

Why are rocket turrets commented out of the shitty update? They sound brilliantly shite

Atom in general is disgusting, but that's an OK feature

Hmm that's weird, does this vary between compilers?

i have fixed this and added a comment explaining the three possible cases that now exist

this line is probably leftover if so should be killed

drm framebuffer unreference cannot do at irq context so that we create a queue to do vsync handler but i see the armada drm driver it use a queue fb unref work to unref framebuffer so other things we can direct do at irq context i think it would be faster or see my changes c

your silence is maddening going to fucking kill self and change it now

os service token should be ok token

this should be processutils execute in fact please submit this fix as a separate review

Damn. Overlooked that xD

subclass oslo cache backends mongo

prefer using kmaxvalue store closed as the last value and use the argument uma histogram enumeration go histograms howto enum histograms

remove useless chatter

oh that is annoying so to actually use this we would have to define glibc

hm since we have separate processes one per listener we cannot assume that we can only monitor the amphora at this level my assumption is restarting the haproxy process would be handled by the agent and would happen before the agent reports a listener issue to the health manager i am assuming that if the agent reports to the controller that the listener is failed it has exhausted the recovery options it has available and it needs the controller to fail over the listener

they are useless but i think we should clean this up it with another patch

remember these are example events you re allowed to take some shortcuts and hardcode values that the test does not particularly care about for example i would let tid pid and ppid ptid you could maybe even have the parent pid tid be hardcoded to the init process

you might have a race condition here you probably want to disable the users first and foremost before doing anything else a user could potentially be logged in and creating tokens and credentials while you are deleting them and you could end up with one hanging around after you thought yoyu had removed them all this is non transactional after all

this should be removed and used invalidbenchmarkconfig invalidtaskexception

using os path operations on urls is bad form it assumes that the path separator is which is the sort of monkey business that os path exists to prevent put another way a url is not a filesystem path and cannot reliably be manipulated as such

this comment is useless it s my mistake

this seems useless as the above glgeterror has cleared the error state

nit maybe navinovariationsynthetic since users are not disqualified from the normal navi onboarding

10?! Rubbish. :+1:

Something wacky is happening with the merges, I'm pretty sure you didn't mean to add 8,000+ lines to item_groups.json

when you do not set a default value i would like to see a test to fail if no value has been configured

because translate cinder exception is using functools wraps it grabs name off of the function i figured mock would be able to handle that call just fine but for some reason it doesn t i guess it needs to be a string but mock returns a mock object more info here

small performance optimization instead of getting ref you should use count and if count is raise exception

boo, well that sucks :(

Fuck damn... Merged it without seeing your comment

there are no plugins of this type, this interface kind of sucks but its either this or creating a new class of editor plugins

add vpd links which enumerates link varables in dump vpd log also check if they are symlinks

i think the lt and gt are not needed here and should be removed

Use the argument ya filthy dog

haha yeah that would be an accidental untabbing in vim i will undo that hunk

you need to remove it from existing environments

If I had to guess, it's because stack code is from before qdel, so anything calling use() could experience a sudden long ass pause for del to work it's thing

source admin openrc sh

fuck if I knew, this file is literally a copy pasta from the machinery process with a lower interval.

ca cert params get ca cert may be better

UNBLOCK ME ON DISCORD FAGGOT

this should be die or really just take out the whole switch and assert bus like in mtk spi init

the white spaces will be removed in the next patch set

lube just needs a fucking nerf its ridiculous how much every vent clog makes

i agree its a little weird, but I think context will be enough. no one calls these insect, its always bugs

This was caused by #1644. Changing `axes.get_title()` to `axes.get_title().get_text()` around line 49 in `figureoptions.py` fixes it. Seems a bit ridiculous to me.

cls instead of combinationevaluator must be used here

Damn that is beautiful.

i got rid of all of these without even adding them to feedbackprivatedelegate after getting rid of the debugd d bus calls it seemed superfluous

ur mom lol

this part of code is then useless

Ansible's AWS modules are so broken it's getting ridiculous. And the unwillingness to accept patches is beyond belief.

Damn irrational numbers!

why not just call this absoluteresource to avoid the need for the odd dummy class below

yeah the audio service can live as long as the browser process that is how i chose this value based on browser uptime metrics

the fuck, this isn't an excuse to let shit code into the repo @Cheridan

Seriously? Do you guys want to take the entire repository over?

stupid question why was this moved down here

all of them? It sounds like the interface is a pain in the ass

so this is an odd comment but i think it is easier to read if you add a colon after these lines then include the above example after that i was reading this all up side down by default and it was really confusing

I believe I have got this sorted. I had to fall back to loading the active theme right at the start, but this should be a fixed-length(ish) task, getting longer only if a theme has a ridiculous number of files.

done

I was banned for**: Naming a crossbow faggot fucker, I want to clarify that I am gay and the only person I was going to use it on was an IRL friend of mine who is also gay and cool with me calling them faggot. We were planning to have a bow duel and I thought it would be funny.

before template is dom if if androidrunning like storage html

suggest move getting the dumb member id to a separate line so that this is readable

i think you may be able to get rid of apn type key

i would prefer if sub name pid file s s pid self uuid sub name else pid file s pid self uuid

so you are saying that part of this change is to kill the party is that a good thing

domain name domain name

done v lol drive by

The fact that this isn't OOP is bad The fact that this isn't at least an is_type_in_list is worse It's not your fault but this is disgusting

Whoa, double opposite-direction for-loop, that's some hardcore shit!

as far as i remember the does not work in python is it intended btw why format at all why just not to use six text type getattr db object db id name

let s set them to null like you did above

see we want this because its so fucking awful and cringey

useless outer parentheses

This combinatorial explosion of 2^N cases is definitely a bad idea. It was kinda bad with 4 cases, but 8 is ridiculous. I rewrote this to have N cases with OR gates, and will merge that version.

This is wacky. Can we leave a comment explaining what this is doing?

you forgot to add the parameters to handle the mocks the line should read def test destroy instance without vm ref self mock call mock get

oh yes so stupid mistake thank you for observation fixed

the comments in jnihelp h are a bit more vague not just for the mac unlike the art macros h as long as jnihelp h is paranoid enough to define temp failure retry scopedfd will be just as paranoid and include jnihelp h

Weird, I swear I removed these.

Oh shit, good catch, I'll just snip that out.

only common mocks should be here

done also added comments to all the other sh files

actually nevermind that'll fuck the glob

extra fk

Dude what, huge change