Datasets:

showvikdbz

/

code-reviews

like 0

ah fuck it. Oh also yeah we should keep the var for the living emote, can't hurt anything

why do not you use params instead of user id

This especially and the other hard-coded crap above is the ugly stuff I kept here in the C&C DLL. Not really sure how/if to clean up it up properly to make it work for anything else.

tell jhole he's a faggot

nit add a space between the colon and the bug number

you should specify here your company

nitpick but it is actually not hard to make a jump table in python full cmc self full cmc request request type certificate meta not important but if you wanted to get rid of a big if block

not worth it i think this is just a dcheck the caller also dchecks the url before adding the tab helper so this is really useless

wtf that should not be there ! grrr ZZZ....

Yeah that looks rubbish - let me have a crack at this again

if you make that change maybe kill this since adb socket accept already does the equivalent to this

If anyone would like to change it so that smaller items get sucked _first_, change line to `if(prob(current_size * 5) && hand.w_class <= ((current_size+1)/2) && H.unEquip(hand))`

no need of else here if not added self add initiator to array initiator name if self is initiator associated to host initiator name is false self associate initiator to host initiator name hostid

>AI-Ian >Not A-Ian ONE FUCKING JOB

this seems pretty silly just error out if there is no pdata

obj mock clustercollection all assert called once with

nit useless blank

memcache should be memcached

@Dogrtt Exactly the same here. I followed the answer posted [here](https://developercommunity.visualstudio.com/content/problem/25310/unload-projects-hangs-on-close-solution.html) and it worked wonders. It does suck if I ever need the service though.

commented code should be removed

i think seamicro vendor pxe vendor should not be attributes of this class

Travis bitches because they removed encrypted environment variables so I cannot pull the third-party libraries. I'll look into that.

see my remarks on configure etcd sh

again this should all be stuffed into setup baremetal

build slaves should not be needed for new installs

kill this empty line i think

**GASP** Gimme some brackets dawg

i hope this is not a dumb question but how many cpus do the dsvm tests run with in the check queue i would like to know for sure how many workers we are testing with

done

put please at this block err msg

This API seems like nonsense - if threads can race to set/unset the flag for another thread, it can't be depended on.

i am sorry i did not noticed before pass a file to write fp as an argument to jsonutills dump

it works fine on windows the generation drop down is always shown on focus thus the event here is good enough and announces the pop up on both mac and windows i changed the announcement of the menu item now it declares the whole text there

use orderedlist here this is a high level task flow and not a procedure complete these high level tasks to set up the vnx direct driver para orderedlist listitem para install x you must install the x tool on the x see x para listitem listitem para register para listitem orderedlist

i prefer to remove these all constants at all

are these varchars if so then may as well put like we do for most every other text field

oh damn I forgot it! Thanx

That sucks :-( Thanks for trying...

thanks i knew i was doing something stupid here

is too far from real life put at least

isn t this mocking more than was originally there that is instead of mocking get guest xml as a unit you are now mocking every little piece inside of get guest xml and letting it call down instead is that what you intended edit i am clearly late to this party so feel free to ignore me i would have thought get guest xml would be tested by itself as a unit and then mocked everywhere else i am guessing the old testing was using a faulty dummy xml which was hiding the bug

leading spaces should be removed

thanks, i was looking for it all over the damned place.

i will set magnum guest image url in gate hook sh

this line should be params params

do we still plan to experiment with this let s drop dead experiments

you have lots of boilerplate additions for serialization and deserialization why not write a webtest wrapper that handles that for you

please do not remove the double it s a doxygen comment

you do not have to redefine this just use updatesharemetadatarequestparameter in the wadl

mypy types do not show up in the recipes generated markdown this project supports conversion to py annotations

> This is also default item throwforce also it's a damn cheese wheel give it some HEFT Cheese heft increased to 6, throw the cheese

If I do that how am I supposed to change the icon state of carbon dioxide to a sprite saying "WTF?"

again raising self xenapi failure smells

s to dissapear to be removed

I mean like it seems work, tried to buy it on private server roundstart and it didn\u2019t work, but after triggering meteor event it worked sooo? Is it just shit code that finds a way to work?

wouldn't this say "It has the electronic firing pin installed" , that looks kind of weird why not put \\an and capitalize the firing pin names?

Which version are you using? Are you overwriting any points during any of your writes? I think there's a known issue with 0.13 regarding overwritten points where it can get screwed up and take an absurdly long time to return. I think that's fixed on master and might be fixed with beta2. #6611 for reference.

> tfw goofball has a point time to check the temperature in hell

gotcha then i suggest firewalling against dma or fuse bb phys

should be executed after for loop ends

notfound should be thrown instead of

x role s

Having a lousy day stuck in a system issue. I will push new commits in these two days as soon as my DUT is fixed.

ceph conf and manila keyring

issue with a pid file is that i am getting rid of the jabra vold conf hence there is no rule pre shutdown to clean up the pid file in the next boot we may encounter a stale pid file and thus kill the wrong process

i did not quite get this comment project domain id has a fk to domain id this way this comment is valid for generate root domain below not to this

Oh shit whoops, almost everything is copied in some way.

oh whoops max addr should be a uint t

equal to oustring has to be removed

This is ridiculous. The PR is too small? Let's merge PRs and make progress.

does kill hup reload dnsmasq eith the same pid otherwise we might want to re fetch and update it is value

to be pedantic kick start is a hyphenated word

remove repo it is redundant

this allocates a useless empty slice var cons console

do we do this in other elements the problem with this logic is that only root will have write access to usr lib python so if you run the script as not root you end up having to do the heavy lifting that is cached for you in a pyc pyo

remove vmax

I seem to have figured it out. It won't work with a leading capital. This is sort of ridiculous.

i have taken a stab at populating config sign types in chromeos config py

it should be execution of wf and wf from wb

please also remove these mixins i do not see how they are useful

i will update this and submit a new patch i will convert the other dumb member test to match while i am in there

remove all ds store in your next commit

the word ass not really good here

don't dupe these timeleft checks, do an early return. this is kinda ugly

one global per worker with list of tasks to kill sorted so it could quite efficiently check if it should kill or not

i never claimed to be competent mr spear

spacing is off but we are going to kill this ebuild so i will not sweat it

That's ridiculous. You need it to be unsigned char for the isdigit call, and a signed char for the calculation?!!?

Oh man, that's gonna be a nasty delete. So many Guids

user data should be executed with privileged rights should not need sudo here

john this is not a cut paste the one from libvirt driver is already in utils as convert version to int this is the reverse of it will move it to utils

`std::stack` is so shitty lol

you should use class fakeimage in fake py

kill blank line

Structures required to execute `usbGenerateDisconnectPulse()` are not yet initialized, hence the call to `IOGetByTag(...)` return some rubbish most likely. @mikeller suggests returning the old boilerplate code for now, without using `IOGetByTag()` if I'm correct.

only router id should be conflicted with ids to ignore i mean if we have only router id the ids to ignore should always be none

dumb question cannot we get the value defined in volumes yaml from hiera

stupid question but why the mix of bools and strings representing bools

why was this needed is there code setting values using bdm something whatever if so perhaps fixing it so the writes go into self values would be the right change or perhaps this whole class should be killed off and just use a dict

using subnetpool id none is fine because for every sp sp id none