#musicbrainz-devel

yup

btw, who can deploy the picard website ? there are few changes in requirements

I'm guessing ruaok and bitmap can, at least.

I could do it

Though ruaok might not know how. :)

ding

streaming_extractor_music is not very "streaming", it seems to be accumulating a lot of data

uses more than 2G RAM on a longer DJ mix

should i keep 1.2 versions on download page for now ? i think it is nice to have stable version, stable version - 1, dev version, and the rest on ftp

luks: I've wondered about that too

zas: I'd remove them

doesn't it need an awful lot of context to compute temporally sensitive things?

(unlike fpcalc, etc?)

well, from the name I'd have expected it works over some windows, not accumulating global data

bitmap, luks : https://github.com/musicbrainz/picard-website/p...

since we don't have 1.3 i386 package for macosx yet, i revamped mac osx in two sections

luks: Well, it averages the entire track for some values (e.g., key, tempo/bpm), so it will need to remember some things.

alastairp might know more.

zas: looks good to me

hm, I guess the picard website stuff should be put on transifex

nikki: yes, but for now there is only an experimental french translation, we just need a picard-website on transifex, i'll make needed changes after, but the main issue is how we'll handle documentation/plugins translations, plugins have no support for translated descriptions, i'm not sure how to handle documentation translations (for me the simpler is sphinx + readthedocs hosting + link from website or better if possible)

ideas welcome ofc ;)

for everything else we do all the translations via transifex, so it's not exactly a surprise that we don't have any translations yet other than the experimental french one :P

I'm not sure what happens with adding stuff to transifex, ian usually does that

* reosarevok: mbs-6201: SQL for event collections

* reosarevok: mbs-6201: unfuck the tab name stuff

woo

should be another one in another 11 minutes or whatever for merging master in (advance warning for a bit of commitspam, though it should be better this time since it hasn't been neglected as long :P)

chirlu`: if you wouldn't mind, could you check if your stuff for the insert methods and such needs any updating on the schema change branch (probably for events)? I think you'd be faster at finding anything that might be relevant since that was your work

Only what I only wrote in comments, I think.

which comments?

https://bitbucket.org/metabrainz/musicbrainz-se... for events

(and the reply)

Can’t find the one for tags now, but “sub _table” can/should be removed from Data::Area.

oh right, that's still just in beta

Ah, https://bitbucket.org/metabrainz/musicbrainz-se... for tags.

yeah, that one's still in review so I figure it's bitmap's problem

bitmap: can you deploy updated picard website ?

:-)

though the fact the insert stuff is still in beta means it needs a bit more time before it's anyone's problem, I guess :P

zas: yes, I was just about to :)

(since schema-changes-2014-q4 tracks master)

Project musicbrainz-server_schema-change-2014-q4 build #18: SUCCESS in 12 min: http://ci.musicbrainz.org/job/musicbrainz-serve...

* Michael Wiencek: Load LinkAttributeType roots and factor out serialization for edits

* Ulrich Klauer: MBS-7750: Mark videos "(video)" instead of "(true)"

* nikki: Select coordinates using a map

* Ian McEwen: MBS-7886: don't allow ratings to be included for places

* Ulrich Klauer: Update edit search tests

* Ulrich Klauer: MBS-7869: Replace DISTINCT clause in edit queries

* Ulrich Klauer: Add limit clause for subscribed-entity edits queries

* Ulrich Klauer: Factor out the limit on edit queries into a constant

* Ulrich Klauer: MBS-7869: Order edit listings by id instead of open_time

* Ian McEwen: Update translations from transifex.

* Ulrich Klauer: Fix index update script

* Ulrich Klauer: Reorganize the string cleanup functions

* nikki: Use existing editor data in place tests

* nikki: Add alt tags to images which are missing them

* Ulrich Klauer: MBS-7915: Show tracklist on "Set track durations" page

* Ulrich Klauer: Get rid of an unncessary list

* Ian McEwen: Fix up a variety of tests.

* Ulrich Klauer: Un-inline code from the Unicode::Encoding plugin

what do I know?

Freso:

Ian McEwen: Update event tabs to use %ENTITIES-based system.

https://github.com/michiwend/gomusicbrainz musicbrainz in go, written by some tomahawk guys

michiwend: ^

neat

zas: the i18n patch broke the uwsgi deployment and I'm not really sure how to fix it

I like their logo

whoo, all schema change patches are now in review or merged :)

bitmap: how did it break uswgi deployment ?

michiwend: oh, you're in here too. ohwell :)

Ohai alastairp, indeed I am ;)

kepstin-laptop: thx :)

luks: it might be that the algorithms are configured to work on frame-frame or window-window, but I think it then stores stuff in memory until all algorhtms have run, and then dumps them to the file

maybe there are more efficient ways of storing output data before moving on to the next algorithm

ruaok: the contribute page is a little unclear that you don't need to download the vm

I had a few questions about that

bitmap: it is due to blueprints, but it is the same structure than critiquebrainz, do you have access to its uswgi setup ?

alastairp: ok, one sec.

zas: yeah, I'll take a look at those

Ian McEwen: Update events in %ENTITIES so they'll properly appear in sitemaps.

bitmap: you need to have a app = create_app() somewhere (see run.py), i guess this is what is missing

yeah, I've just made that change

in the logs now I see https://pastee.org/znwd5

templates were moved, but i guess it is related to uswgi config, do you have any app_name somewhere ?

or does it embed templates somehow as described at http://uwsgi-docs.readthedocs.org/en/latest/Emb...

i know nothing about uswgi

zas, bitmap: https://gist.github.com/mayhem/e8c3172feca887cd... :(

Mac, 10.7.5

hm, odd

I wonder why py2app didn't bundle that

er, it did

wtf

waft:/Applications/MusicBrainz Picard.app/Contents->md5 ./Frameworks/libdiscid.0.dylib

MD5 (./Frameworks/libdiscid.0.dylib) = 0ace1ade9b0989f50c50536f32ee71f1

I rember some patch that was done to fix some py2app ctypes issue or something like that. Let me see if I can find that

ok, bed time.

nn all!

night

ruaok: started with geordi now :p

(thought you'd like to know)

Sleep well!

This is the patch: https://github.com/JonnyJD/python-discid/commit...

but I remember having that patched in py2app or something like that (that's why it isn't in python-discid master)

ah

iirc I have py2app 0.9

make sure this patch is included in your version of py2app

can't say which version that landed right now; I can't find that on bitbucket

I can see those changes in /usr/local/lib/python2.7/site-packages/py2app/ so I guess it's in 0.9

hm, then I don't know what the exact issue is this time, but probably related to https://github.com/JonnyJD/python-discid/pull/42 or maybe it should be packaged in a different place or a regression

you have a minor typo on the acousticbrainz website

at least, i'm going to guess it's a typo

If you have a collection of music files with MBIDs, please one of the submission tools to help us bootstrap this project.

Do I really have to please a submission tool?

bitmap: you might be able to use the patch/commit from python-discid pull #42 for a quick fix for this release. If you can't find the problem directly now.

I suppose some anthopomorphizations of the tools might be pleased by being used, which would have the same effect, but :P

was hoping for a dynamically linked executable, but a vm will do :)

JonnyJD: okay, I don't have a CD drive accessible from my mac, so someone else would have to test it

bitmap: you don't have to test with a CD drive, just that it can build and find discid. The rest should work fine then.

oh, I see

I was confused where the error was occurring

for me the app was picking up the lib in /usr/local/lib, hence not crashing

ah yeah, this is about not having system-wide libdiscid installed I think (only packaged one)

https://github.com/JonnyJD/python-discid/commit... fixed it

cool for now

I am pretty sure this is something py2app should be fixing on their end (again), but good for this release

yeah

I've already had to patch py2app to build things

(for https://bitbucket.org/ronaldoussoren/py2app/iss...)

bitmap: I opened a ticket for python-discid for now to track this: https://github.com/JonnyJD/python-discid/issues/43

okay, thanks :)