#musicbrainz

What I'm saying is that I can't enter it with the � in the 'Relate to URL' page

When you paste the URL in, it changes to the escaped form?

Yes

hm, that's odd.

Can you try it? It maybe a setting on my computer

Well, thats why I was saying it's odd :p It works fine for me.

I dunno. Windows character encodings are a mystery to me. I want my Mac back :-(

I see that the wikipedia edit is the same thing, when you entered it, it was automatically escaped.

it should be escaped

I don't even know how to type an �. I had to copy and paste that from when you typed it :-)

When the url is displayed on an artists page though, it will be in the proper characters http://musicbrainz.org/show/url/?urlid=546771

luks: Why should it be escaped?

luks: When I entered the URL, it wasn't escaped http://musicbrainz.org/show/edit/?editid=10294425

navap: because otherwise it's not valid URL

So whats to become of all the URLs that have non-valid characters?

Munger: The wikipedia one is still open, the discogs one is cancelled.

So if I enter the discogs one with an escaped url, will it display right? If so, are you going to vote against it again? :-)

I didn't vote against it because it was escaped, I voted against it because it wasn't pointing to the right place.

And yes, it will be displayed correctly http://musicbrainz.org/show/url/?urlid=546771

That URL is actually escaped, but when it's displayed it is formatted "correctly".

Bloody accents!

He should be forced to spell his name 'Arndray'

LMAO. I have a music player on the iPhone that displays lyrics in sync with the music. My daughter thought they were subtitles for the deaf :-D

heh

Mmm, I've imported all the files from mbdump except puid and puid_join, no indexes, no keys, vacuum full analyze, and it's using 1.8gb.

That sounds far too low.

Seems to be all there. :/

400mb of primary keys, 1.8gb of indexes (wtf).

Well, that was fun. I've completely forgotten what I was going to do.

ARTILLERY.

luks: Still haven't figured out the issue with some oggs, but I've traced it to the fingerprinting routines, and only in some oggs.

http://brianfreud.info/ogg/ <- sample crasher

Crash when doing what, brianfreud?

if you try to analyze it, or if you have auto-analysis on and try to load it

Man, not sure what's worse, a hangover from alcohol or from a root canal.

In picard?

yes

That's funny, it works on my machine.

using .11 with python 2.6

0.11 with Windows.

Apparenttly bundles 2.5.

yeah, I think all the current distributions are using 2.5, not 2.6

using .12dev, using the "python tagger.py" method, picard just locks up

I think I fail at using left joins...

Cba to set-up replication, though.

left joins are never inuitive for me

I normally just do where a=b and c=d etc, but I actually want everything from the first column this time so there goes that approach

nikki: what're you trying to do?

heh

select all of an artist's tracks and any tracks those tracks are linked to (managed that bit) but using a mbid not row id

I could just cheat and do a separate query to get the row id ¬_¬

Tracks those tracks are linked to? que?

relationships

Give me a minute.

sexy

can I please have super-3D interaction of the mb dataset? :)

nikki: I'm beginning to think this is easier done as two queries. :/

Or that I shouldn't be drinking. Or perhaps both!

Muz_: reached the Balmer peak yet?

I'm way beyondit.

does anyone know any artists with a lot of tracks linked together?

I can show you what I'm doing but it's all japanse :P

japanese too

Muz_: :D