#musicbrainz

ruaok: now that I have mbserver working, what the heck can I do with it? :)

custom queries :)

fix bugs.

:-)

ha

hey ru, would a "holding database" for the cdbaby data be feasible?

do you have any bugs for mb_server newbs? ;)

feasible yes.

massive pain in the ass. yes.

lol

will the data go out of sync. yes.

hmmm

do I want to code that? no.

What if a special DQ were created "Imported"? Releases can't be changed to it, only away from it - would make those from the import immediately avail, easy to have a query for, and once fixed up, we just change the DQ from "Imported

to "Normal"? Would do just about the same thing...

that's a LOT easier.

nigh trivial.

Data becomes avail immediately for those who want that, but it's segregated enough that those of us who would prefer it be cleaned first can still easily query for it

It had occurred to me earlier that those basically are identical in situation to the unvoted ones we had talked about when I proposed the DQ change :)

wah

hmm ok i guess luks iwll not pop up before later this day

MrQwerty` is now known as MrQwerty

ruaok: ayt?

morning warp. what's up?

hello!

just a comment on the cdbaby thing

k

before doing anything with those 180k current cdbaby releases not in the db, shouldn't we create something for new stuff in cdbaby to enter MB as add edits?

otherwise there will be a gap somewhere

warp: They are welcome to use the MB site for adding? :D

(unless we all decide on a freedb import feature sort of thing, in which case none of it matters... :)

given the feedback from the last few days, that's what I am leaning towards.

either a slow trickle that gets voted on or import piece by piece.

ok, makes sense. probably the easiest to code and manage too.

yep.

its all about CD Baby doing the work. :-D

ruaok: well, if you get a slow trickle... i'd say start trickling in the new releases first,... then at some point increase the speed a bit by also trickling in the back catalogue.

yeah, but I would like to see our backlog improve before making it worse

backlog of open edits you mean?

Yeah, I don't know how slow the trickle would have to be to have all the imports checked.

My guess would be far on the negative side. 8)

srotta: that's why i am voting on having the trickle speed be the speed that new releases are added to cdbaby. if we can't managed that,... then we're doomed anyway :)

yep.

ruaok: congrats on the future of music summit thing btw, looks like a good plan

thanks... but no congrats. its something we had to fork cold hard cash over for.

i know :)

but its non-profit to non-profit and we're basically get things at cost.

Any javascript gurus around by any chance?

hello

is there a debian repository for musicbrainz programs?

I think there's one for Ubuntu.

And I remember someone mentioning they would work in Debian unstable as well.

But I haven't actually tested it. luks might have more information.

it won't work on debian

that is, not without rebuilding. the source packages would

Oh, ok.

so, what do I have to do to get oicard on debian? I apologise for my noobnes, I just switched to debian :)

apt-get install picard will get you picard 0.7.2

is it in the official repository?

yes

I don't see it, and I'm running testing

It's in experimental.

umm

Apparently.

I'm quite sure it was in unstable as well

weird

�t least packages.debian.org only lists it in experimental.

yep, I see

pae: apt-get install python-qt4 python-mutagen libavformat-dev libavcodec-dev libofa0-dev

then download http://musicbrainz.org/ftpmirror/pub/musicbrain..., unpack it and run ./setup.py install

ok

"

apt-get install python-dev

Anone here decent with javascript? Been banging my head against this problem for a few hours now, and I'm sure it's prob just something dumb

you can uninstall those -dev packages after installing picard

BrianFreud: I speak JS quite fluently :)

luks: unless you install with aptitude, then it will have marked the libs as orphans and remove them when the dev packages are removed

I don't, which may be the problem... Trying to write a script to open all links on a page at 5 second intervals. I managed to work around the lack of a "wait" function, but right now, I get 0 from document.getElementsByTagName("a") no matter what I do.

BrianFreud: at the time of running document.getElementsByTagName("a");, there are no links in the document

try something like window.onload = function() { allAnchors = document.getElementsByTagName("a"); }

That's kind of what I thought might be the problem, but I couldn't figure out where to put those two vars to make them happen *after* the body did

so window.onload will happen after the body loads?

yes

ok - now, do I need to use function() { var allAnchors = to make the variable global?

no, var allAnchors in the function will make a local variable

ImportError: No module named astrcmp

that's what I get when I run tagger.py

ah, ok, I had it backwards then

thanks , I'll give it a shot, see if it works this time :)

pae: if you installed it with "./setup.py install" then run "picard"

ah

if you want to use ./tagger.py then use ./setup.py build_ext -i

I still get the same error

when running just picard

can you pastebin the whole traceback?

just a sec

ah, you are still running the local one

cd /home/erik/programi_src/musicbrainz/picard-0.9.0alpha14/ && ./setup.py build_ext -i

ok

ok, it starts ok now, but I get warnings saying that libofa can't be found (it's installed) and that no decoders are found

oops

I think `cd /home/erik/programi_src/musicbrainz/picard-0.9.0alpha14/ && ./setup.py config && ./setup.py build_ext -i` should fix that

ok :)

Hi, excuse me, I haven't understood how freeDB is related with musicbrainz...

luks: it still can't find libofa, but I have to go now, bbl

YES! Finally works! :):):)

Finally realized that the DOM references starting with 0, so var thisLink = allAnchors[i].getAttribute("href"); was failing because it was undefined :P

what are you going to with the javascript?

to do

I got fed up with linky and the like - it's either one by one, or all at once with them

I wanted something that could be a lot slower - ie, not suck all system resources at once

luks: i've hacked on another existing mp4 taglib plugin and have now support for all musicbrainz tags as well in addition to all other metadata (support for '----'-type boxes)

which one?

luks: hmm it also came from amarok but it dates back to 2002 and has no dependency on any external lib

i didn't use it as i thought libmp4v2 is superior (heh)

hm, why didn't you use the one i wrote?