#musicbrainz

I'm trying to set up a MB dev server using xubuntu

http://svn.musicbrainz.org/mb_server/trunk/INSTALL contains a nice list of required modules but how do I know which module to actually install based on names like 'Algorithm::Diff'?

luks, you would know, wouldn't you? ;-)

usually libpackage-subpackahe-perl

so libalgorithm-diff-perl

and if the package doesn't exist, than 'sudo cpan'

and 'install Algorithm::Diff'

s/than/then

speaking of which, http://xkcd.com/c149.html

so I could use cpan to track down the package name as well?

yes, cpan can install all of them for you

suitable for xubuntu?

well, using native debian package is usually better

for example i'm not sure if cpan handles uninstalling

Prodoc: do you want a database or dev server?

I was browsing for packages in Synaptic to get them installed easily but I'm having a hard time finding the package names

because if you want to import the DB dumps, don't use SVN trunk

dev server

just translate them with the libpackage-subpackage-perl scheme

libhtml-mason-perl, etc.

tried that but couldn't find half of them that way ;-) I 'll keep trying then and create a list of the names for others to use if they run into the same problem

I have a list on http://wiki.musicbrainz.org/LukasLalinsky/Ubunt..., but it's outdated

I'm following the steps provided in the file above, that's no good?

INSTALL from SVN?

I'll update that one than

aye

that should be the most up-to-date one

ah

there is a lot of DB schema changes in SVN trunk

what should I use instead? /how

if you want to import the database, use the latest release branch

so DB schema of the codebase and the dump matches

this one

ok, cheers

but SVN trunk is more fun! labels and stuff :)

I'll be happy to use that one as well, I just want an environment to add and fix things

will I run into a lot of trouble keeping things sync?

no

but if you want full database, you will need to import it to the old DB schema and upgrade it

to add and fix things you don't need full database, though

How can I prevent a full db to be installed?

in step 5, instead of downloading the data and "./admin/InitDb.pl --createdb --echo --import mbdump*.tar.bz2", use "./admin/InitDb.pl --createdb --clean"

that will create you an empty database

you can probably also ignore memcached for now

and perl modules like RDFStore

also, on ubuntu it's easier to use 'apache-perl', instead of 'apache'

I've installed apache2

that won't work

darn

you need apache 1 and the corresponding mod_perl package

ok

in case of your suggested step 5, will the db be completely empty?

it will have only the default data

like 'ModBot' or 'rob' users :)

ok :-)

shouldn't be much of a problem for dev work

it's easier to work with empty db

I can stuff in some stuff when needed

how much trouble will I run into when installing apache next to apache2?

that shouldn't be problem at all

just make it run on a different port

or I can take the risk of trying to get it to run on apache2, not really a reason why it shouldn't imo

there is a reason

mod_perl for apache2 is different than for apache

and it will be a lot of work to change this?

no idea

luks: in picard, is the "submit tagged information to musicbrainz" a global operation or does it only do it on the currently-selected track/album?

it's global

k cool

lol -> http://xkcd.com/c149.html

I also only 'sudo' get it :P

hi

what was the name of the old wikisoftware? usemod?

usemode?

http://web.archive.org/web/20030803204330/http:... one?

(looks like usemod to me)

takk

damn you luks for getting me addicted to http://xkcd.com/

:P

not me! :)

It's short enough to read in an evening, no problem.

Schlock Mercenary, on the other hand...

darn, who was it then :/

hmm

for some reason the "tagger" link isn't showing up next to releases in mb.org anymore, how do i get it back?

Jugdish: launch picard and it will appear!

picard is already running

hmm, that's odd

indeed

gah!

luks: any idea?

Jugdish: You have to have launched the page from picard.

Use the search box at the top to get it quickly.

oh

You can also add tport=8000 to the end of your urls if you want it on a page now.

aha, good to know

luks, your applying a patch on libtext-unaccent-perl while http://wiki.musicbrainz.org/DebianServerSetup just installs the module

is the patch still required

?

can someone peek at the open edits on Gym Class Heroes? http://musicbrainz.org/mod/search/results.html?...

do we have other albums in MB where we include the track numbers with the track titles?

ah, found the reason

also both your list in the wiki and the DebianServerSetup list don't specifically install Apache::Session::File, Apache::AuthDigest, Bundle::Apache, Devel::Peek, Digest::MD5, HTTP::Negotiate and Time::ParseDate

do one of the following modules have something to do with that? libapache-reload-perl, liburi-perl, libtime-hires-perl, libxml-parser-perl, libapache-request-perl, libtime-modules-perl?

these don't get specifically mentioned in the INSTALL file but both wiki pages list them

Prodoc: ignore the wiki pages

there are all very outdated

remnant of the previous server requirements?

+s

yep

also, Text::Unaccent doesn't need patching anymore

ok, cool

but what about the mentioned modules from the install file which the wiki pages don't list? I can't seem to find a module to install for them

I assumed that libtime-modules-perl would be for Time::ParseDate, could this be correct?

some of them are probably parts of other packages

or am I better off bothering ruaok with this? ;-)

or better using apt-cache :)

"apt-cache search Time::ParseDate" returns libtime-modules-perl