#musicbrainz

yeah how's it going with that?

i'm interested in using it on a pretty big archive.

I'm fixing some threading issues in the mb wrappers...

then they user will not have to wait at all

very nice

it's text only and it works

because of a server problem it doesn't handle MACs the way I want it too yet

i don't suppose the tagger supports renaming files, or gathering id3 information from filenames where the id3 info is wrong or the trmid isn't in the database, does it?

does renaming

the filename parsing code is not working quite yet

waiting until I hit a batch of file that need it

phidauex (~phidauex@dsl093-192-027.stl1.dsl.speakeasy.net) has joined #musicbrainz

so I have a good test base

hey hey

hehe

i've got a big batch of files that need it!

you want to test on those?

what I'm thinking is... do a FileInfo query, if it doesn't have anything, then parse the filename,

Mutiny: are they on a sshable box? do you mind giving me a an account? can you get all the depends working first?

what depends do you need?

i have a 'catalog' of the files that might be more useful.

filename + path, trmid, size, id3 artist+album+title, encoding parameters, etc.

get tagger.py working on your machine and we'll talk

shower time

ok

and let me know if you have any questions

Mutiny: it would take a lot of work for me to tweak my code to deal with your catalog

okay then, nevermind

i can do it

Mutiny: you code python?

mmhmm

sure do

I'm glad somebody already wrote this tagger code for me. I've got alot of mp3s that need massaging.

massaging indeed :)

i've already 'massaged' a few thousand mp3s on my PC.. its slow work, especially since many of them aren't in the musicbrainz database yet (or rather, they are now, but weren't before)

4.3 GB of sorted stuff....

16.9 to go

then there is the semi-sorted

i've got 32Gb total, i'm not sure exactly how far through it I am. A lot of what I have was already well organized into albums and well tagged, and those are going fast, but I just got several gigs of great hiphop and stuff from a friend of mine, and while the music is awesome, his complete inability to organize or properly tag mp3s has made it really slow to go through

I've got 637gb

you serious?

thats lunacy

:)

I've tried to keep my collection trimmed to stuff I really like. i don't like having a lot of fluff. though i can understand the desire to collect for the sake of it. :)

Mutiny: that's fucked up man... you work and a music store and rip every cd?

no

that would be over 1000 cds uncompressed! :)

only constrained by disk, within a few weeks of upgrading to 1.4tb that'll be filled too

if you submit all those you'll singlehandedly double the number of entries in the mb database ;)

where do you get them? just download? or rip lots?

people uploaded them

ahh

yeah i offered to submit them, but i don't know how interested they are *shrug*

Mutiny: ok... you are my main user then

does tunetagger have support for submitting the TRM ids like the windows client?

heh, with over half a terabyte of mp3s, already cataloged, it might be worth writing a script to parse the data in the catalog :)

phidauex: yes... it does it in batchs of 15 so you don't loose too much if it crashes

excellent

i already 'extracted' all the meta data from the archive

it took ~4 days to generate all the trmids

hot damn

tunetagger also downloads cover art

i noticed that when i was perusing the source :)

very cool

Mutiny: you shouldn't have too much trouble using my code..

Eventually it would be nice for musicbrainz to support all the ID3v2 frames, including the goodies like cover art

you want to make use of the the musicfile.MusicFile object

phidauex: TT is doing just that...

:)

client support, then server support

icepick: i don't expect i will. i'm probably going to give it a run through later today.

after i set up traffic shaping for codecon

I just want the date of the release in MB

after that I don't care

that would be nice to have

composer and interpreted/remixed by would also be nice, and could help with some of the difficulties with classical music

unsure how that would fix in my file naming scheme tho

filenaming is a hassle.. i'm not sure of the best way to do it really :) so far i use /artist/album/artist - tracknumber - songname.mp3|ogg

but its easy to get ridiculously long filenames

how do you pick songs to play?

I usually make an empty playlist, and then start drag'n'dropping files or folders that i want to listen too. organizing in folders the way I do makes it pretty easy for me to just drag in whole albums, which is what i usually do. then i put it on shuffle and go nuts

i also have things organized by genre, but i don't pay attention to the genre tag, i organize them myself

so why does it matter if you filenames are long

burning to cd ;)

over a certain number of characters, i don't remember how many, you stop conforming to the cd filename standards

65

64

special characters ending up in filenames can also be an issue

i'm not sure if some software already does this, but it would be nice to have support for removing special chars and truncating the length of the filename when doing file renaming

so the cdburner doesn't cough with 200 errors about overly-long filenames :)

you could very easyly add it and send me a patch

:)

i need to see about getting tunetagger running first

where are you hung up at the moment?

oh, i haven't even given it a solid try yet :)

what do you mean by 'musicbrainz + python wrapper' dependency?

what os are you on (i think I should know this already )

os x

so it may not work :)

ahh... yes... that is a problem

send robert $$$

:)

so its still reliant on the mb_client libraries?

i chatted robert up yesterday

he said he'd let me know as soon as things are in a usable state

yalaforge (~matthias@krl9-d9bb4fae.pool.mediaWays.net) has joined #musicbrainz

hi

ahoy

Does anyone know if a pkg-config file has chances to be included in libmusicbrainz?

sure

ekkis (~e@adsl-63-194-16-45.dsl.lsan03.pacbell.net) has joined #musicbrainz

Where should I submit it?

is the patch tracker shown on the sf.net page?

icepick: please wait a minute

otherwise just mail it to icepick@icepick.info and I'll add it

icepick: Seems it isn't there. I'll mail the file, thanks.

BTW: NetBSD now has a current musicbrainz package :-)

bye

adios

what kind of a stupid moron messed up the sort name for dave matthews band

lol

there are some wierd things in there..

i've been wrestling with getting dan the automator made into his own artist instead of being trapped as an alias of anandji & kalandji

there is a lot of work to be done, but at least the dream of a clean database is possible, unlike the static freedb

apparently this is correct "matthews, dave band"

but possibly the most fucked up thing I've seen

are you sure thats how it should look? :P

i think the actual title of the band is "The Dave Matthews Band" which would make the sortname "Dave Matthews Band, The"

the name doesn't refer to a person, it refers to a band, which happens to 'belong' to a person

I agree... let's change it

right on

xiidozen_ (trilluser@129.82.85.73) has joined #musicbrainz

I've just d/ld trm-0.21 and tried to build it... the configure complains that I should have the MusicBrainz client >= 2.0.0 but the client on the web site is only 0.2.1... what to do?

what OS?

linux

rh 7.2

ekkis: that's very odd... what url was the 0.2.1 written?

it probably means you need the client library, which should be at 2.0

i think you need the MAD libraries and the libmusicbrainz library to build trm

ekkis: that's exactly what you need

I just installed mad

but the client library... I guess I'll just get the 0.2.1 version then?

get libmusicbrainz 2.0.0 for linux

found it

yeah, that should be what you need

it's on the ftp server, the web page is just old

the documentation is a bit vague on that