#musicbrainz-devel

navap: I liked those help "baloons" on careers.stackoverflow.com

gotta login and "create new CV" to see it

teh canadians are funny http://flowingdata.com/wp-content/uploads/yapb_...

i'm off to snooze

bb!

nn

I swear. I need a "have to ponder this" status in jira.

I have too many things that need to be considered in the grand scheme of things.

warp: http://careers.stackoverflow.com/cv/edit has the same kind of stick/unstick div that ac3 needs and it seems quite fast in my slow machine.

pronik: We don't have kind of room though, and the icons I'm referring to wouldn't be just for forms.

aCiD2: pft. perl?

The footer says "Contact" but the menu says "Contact Us", which should it be?

navap: the principle is the same (they hook $(window).scroll() in cv-edit.js), but I can't easily check if it feels as slow as ac3... as the other computer is now in the basement.

(actually, I can have a look at both on the netbook soonish)

warp, I've been too busy w/work to look... what is it you're actually using the scroll hook to do?

brianfreud: try http://frob.nl/~warp/ac3/edit.php and scroll down.

morning!

fsov morning

warp: I see. Why not just use css?

brianfreud: how could you do that with css?

You have it start as static, then switch to fixed if it would otherwise scroll off the page.

oh right, so it still needs JS

no

that's the current behaviour

I assume it only opens when you click on something to open it.

How can you change from static to fixed without javascript?

So why not just simply 'position: fixed;' position the box instead? Then it'll stay in whatever position, so long as it is open, no matter where you scroll, and no need for any js or capturing the scroll events

Because we don't want it to stay fixed in the same loctation

We want it to original stay as part of the page, but as you scroll down to stay visible

that's the current behaviour, except for at the very top of the page?

aCiD2: I guess my question is, esp given the overhead in pulling that trick off, why?

but that's not the same, is it :)

Anyhow, I don't notice a slowdown on my box.

nd cache that for the later checks.

+sub view : Chained('/user/base') PathPart('ratings')

that's the diff

aCiD2: awesome. that's a pretty subtle typo.

Yea

copy pasta ftw

meh

please pull from acid2/mbs-66

mbs66*

my postgres sorting thing runs out of memory when I try to create an index for track_name.

lol

I have the feeling I'm leaking memory, but I don't see anything wrong in the code.

(and the same thing will happen on test, as it only has 4gb too)

aCiD2: how are you at C? :)\

about as useful as an elephant trying to thread a needle

codereview is down?

aCiD2: haha

hm, I forgot how to create the test database.

aCiD2: 2 minutes 40 seconds for a "prove -l t/" run.

real 2m41.758s

warp: on your new box?

yes

givvvve

so refreshing

very nice :)

What is the difference between the two ratings paths? The difference is too subtle for me it seems.

script/musicbrainz_server -r

watch 'curl --silent http://bb:3000/ws/js/artist?q=Shih&limit=10 ; echo ""'

navap: one said PathPath the other said PathPart

and edit away and stuff happens :)

arghh haha

navap: yea, I actually pasted it here and thought it was a whitespace thing until warp noticed there was a difference :D

warp++ :p

:)

Codereview is still down :(

:(

I say when codereview is down we just commit without review, yay!

j/k

MWAHAHAA

oh...

haha

The diffs could always be just checked manually.

i did want to review navap's branch though, I got it running and looked at the code and didn't see anything crazy

but I did notice one thing - the edit list looks kind abroke

it's missing a lot of padding and looks cramped

Yeah I only fixed the content, not the styling on the mass edit page.

ah, okies

My main focus was on the individual edit page. I didn't want to spend too much time on the mass edit page because I wasn't sure how I wanted it to look.

I just wanted to manku sure it was using the proper templates.

yea, I think this is enough for one review

Here's something that should be quite quick to review. The last commit on that branch is whitespace changes. http://github.com/navap/mb_server/compare/maste...

aCiD2: btw, I have to force install Test::WWWW::Mechanize, Catalyst::Plugin::StackTrace and Text::Unaccent this time.

s/have/had/

warp: Text::Unaccent isn't even needed now

it's still 'use 'd somewhere

nope.. I removed all that... I'm sure

ollie [at debian]:~/musicbrainz-server$ ack-grep 'Text::Unaccent' lib | wc -l

0

it's probably because your branches aren't on master

(they are branched off older commits)

Makefile.PL:requires 'Text::Unaccent' => '1.080';

lib/MusicBrainz/Server/Validation.pm:use Text::Unaccent qw( unac_string );

yea, it's what I said :)

and nice machine name :P

it's small, black and fast.

(all my machines are named after mario characters)

that's a cool naming scheme :)

the server is slow, heavy and grey. hence thwomp.

mine are named after moons

boring.

my top secret gaming website project is bulbasaur. If that ever takes off I don't have to think about names at all, I just pick them one by one from the pokedex.

heh

I don't expect to need 400+ hostnames any time soon.

;)

hahaha

(i could match the last octet of the ip adress to the pokedex number)

warp. you're a nerd.

no shit :)

hehe

haha that's quite geeky

Yay, XLD finished ripping 1 track off a cd after taking 5 hours

the dvd drive in the previous box wasn't working very well anymore.

how come luks, I thought you were setup?

well, mb_server is a moving target

You don't dance only once in your file, you gotta keep dancing to get good ;)

one week later and I need to download the other half of internet :)

Are you sure you need to though? I mean, some version numbers might have changed... but maybe they didn't need to

"sudo cpan Test::Mock" installs lots of packages

yay for my dependencies

plus it fails a test

oo, please tell me which

ah, the fix for that hasn't gone to cpan yet

make sure you have MooseX::ChainedAccessors::Accessor installed

btw