#musicbrainz-devel

:)

yes please

my local checkout is broken because I need to upgrade the database or something.

nikki: there's a temporary hack in place now.

thanks

:)

"Couldn't load class (MusicBrainz::Server) because: Couldn't instantiate component "MusicBrainz::Server::Model::MB", "Couldn't load class (MusicBrainz::Server::Data::EditNote) because: Couldn't load class (MusicBrainz::Server::Email::AutoEditorElection) because: Couldn't load class (MusicBrainz::Server::Email::Role) because:"

perl sounds like a whiney 5 year old today

warp: I'm going to fix jenkins if you're not already doing so

(at least the DBDefs changes)

ocharles: yes, I know nothing of your jenkins yet.

our!@

but the musicbrainz-server workspace is /var/lib/jenkins/jobs/musicbrainz-server (master)/workspace

as long as you're the only one who touched it it's yours :)

luks: in pyacoustid, can I change acoustid.lookup() to only return track ids if meta=None?

Project musicbrainz-server (master) build #46: STILL FAILING in 19 min: http://ci.musicbrainz.org/job/musicbrainz-serve...

ocharles: hrm. is there a good way to see which commits are in 'production' which aren't in 'master' ?

yes, git diff master...production

same as a branch for review :)

ok, jenkins is back to just failing on stuff that's broken :)

ocharles: that's a diff, I want to see commit hashes.

git log then

ocharles: I think what I want is git log master...production --oneline --no-merges --right-only

ocharles: but that still gives me a huge list of commits, which doesn't seem right, we should have that much difference between production and master.

s/should/shouldn't/

hrglgrmpf: sorry, I don't know much about pyacoustid

ok... never mind :-)

do you mean it's trying to default to something else if you pass None?

if so, try meta=''

or something like that

or maybe '0'

that gives a exception :-)

(when parsing is on)

is it wanted that parse_lookup_result() doesn't include acoustids?

or in other words: should it be easily possible with the API to get the track ids, or not?

(the python API)

hrglgrmpf: I guess so

hrglgrmpf: but pyacoustid was written for beets, so it might need some work before it's generally usable

ocharles: http://beta.musicbrainz.org/edit/14911584 someone been using data::dumper?

luks: ok

guys... if an artist has more then one youtube link, there's any manner to get the most important?

FilipeTriworks, what would be "the most important"

?

It is kinda strange for an artist to have two official channels, but I guess you have an example :)

the official one, most seen, which have more videos

alicia keys 3 youtube channels

*have 3*

heh

All 3 are official

Indeed

right but 2 of them are feeded by labels

the first one is personal

we don't have a way to make a distinction :/

I guess the clear options are to either query YouTube for the number of videos, or show them all

The distinction is rare enough for nobody to have wondered about it before

Because the "personal" filter doesn't really cut it either...

(DOGOTVOFFICIAL and DOGOUNCENSORED are two channels from the same Italian band, the latter showing a video that the label refused to have in the main one)

(you'd ideally be able to show both)

(as the video in the latter is not in the other at all)

wi thought that all links could have some kind of weight associated

*I*

thank you all once again

FilipeTriworks, that sounds like something you should do in-house

(I can see the point though :) )

(but it'd need a kind of automation we don't really use anywhere at the moment AFAIK)

yes I'll do :)

nikki: certainly lookls like it

hi..qq..if i use latest dump ftp://ftp.musicbrainz.org/pub/musicbrainz/data/... ..I suppose it expects to have Jan 12 patch installed..

ianmcorvidae, reosarevok, warp, nikki, murdos: https://github.com/ocharles/CAA-spec/blob/maste...

this is my draft of a CAA specification - it might work well to work from a repository for the spec as we can use GitHub issues to highlight problems in the spec, and branch it to make ammendments and do reviews on pull requests

ijabz you too!

ocharles: sounds like a good idea, I'll have a look soon.

https://github.com/ocharles/CAA-spec README added

if people like the sound of this process (not my specification), I will start adding what I see as open issues

stuff that's either not covered, or is too vague

and I'll also post to the devel mailing list

but now, it's meeting time

ocharles: "If the edit is accepted, the Cover Art Archive does not change." should be rejected I guess

(last sentence)

ocharles: so front doesn't have a filetype?

hi <ocharles> when u get a chance ..get u look at this error that i got on my other VM host while install patch..sudo cpanm --installdeps . https://gist.github.com/1627561

ocharles: it doesn't seem to say how to get the smaller versions of files

ocharles: sounds like a perfect process!

ocharles you'll pleased to know I bought 'Pragmatic Guide to Hit' a few weeks ago

Git

Though having read it my view is that whilst indeed Git is powerful, and I can see it being very useful on a project such as the Linux Kernel it seems it can all too easily be abused to create overly complicated scenarios

But on that note, teatime

anyone know how i can resolve this issue...Building and testing Encode-Detect-1.01 ... Building Encode-Detect cc -Isrc -I/usr/lib/perl/5.10/CORE -fPIC -x c++ -Iinclude -c -D_REENTRANT -D_GNU_SOURCE -DDEBIAN -fno-strict-aliasing -pipe -fstack-protector -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -g -o src/nsEscCharsetProber.o src/nsEscCharsetProber.cpp cc: error trying to exec 'cc1plus': execvp: No su

ocharles: aw, no OPTIONS support in CAA? :)

warp: yes, I plan to open an issue about that

but as I said, I wanted comments on the process, not the spec :)

ijabz: oh, cool :)

Swarup: you shouldn't need Encode::Detect

ok..should i let it skip and continue.. with --prompt option

ocharles: where should comments on the spec go? github issues?

hi ruaok!

If you haven't, please see http://tickets.musicbrainz.org/browse/MBS-4192

:)

warp: yea, open an issue for things that are wrong/contentious

memcached kicked.

ruaok: http://chatlogs.musicbrainz.org/musicbrainz-dev... may be of interest to you

coffee reading. :)

nikki: but good catch though btw, I'll open an issue about that

grrr, irc problems

I vote we move to XMPP multi-user chat!

so we can have MUC problems instead!

lol

ocharles / ruaok: one of you probably should answer in http://tickets.musicbrainz.org/browse/MBS-4031

reosarevok: I'm on that today.

yay, just been made maintainer of Web.MusicBrainz, Haskell bindings to /ws/2

lol

Didn't have enough with two jobs and a side project?

spoke up too loudly, did we?

reosarevok: BookBrainz continues to be blocked until I can figure out a way to bracket QuickCheck, or until i give up trying. and I'll be damned if I give up!

(but I'll probably give up if I can't figure it out soon)

Can someone point me in the right direction? I want to merge two hashes that have significant duplicated keys, but I want the values merged together by adding the two values together.

Everything I find online deals with merging and keeping one of the two values

navap: there is no way to do that without being explicit

do you want every key added up?

$a{$_} += $b{$_} for keys %b; # might work for that

I think I want something like $new{$_} = $a{$_} + $b{$_}

But I also want to copy over any keys that are unique to %a or %b

Does that make sense?

This is for my webalizer parser. I've got a hash built for each month's stats and I want to now add everything up and create a hash for each year.

navap: my %new = map { $_ => ($a{$_} || 0) + ($b{$_} || 0) } uniq (keys %a, keys %b); # then, probably

Hi Rob..need help on nginx setup on prod vm host..seeing some errors..

hi

paste me some error messages