#musicbrainz

/

      • uptown joined the channel
      • 2013-01-01 00129, 2013

      • ianmcorvidae
        it attempting to highlight it as prolog is comically bad
      • 2013-01-01 00130, 2013

      • ianmcorvidae
        anyway I don't see anything immediately wrong with it
      • 2013-01-01 00155, 2013

      • Freso
        ianmcorvidae: I set it to Perl three times, no change. :| It's plain text now.
      • 2013-01-01 00102, 2013

      • ianmcorvidae
        it's not perl
      • 2013-01-01 00106, 2013

      • ianmcorvidae
        so that may be related as well
      • 2013-01-01 00131, 2013

      • Freso
        I know, but the Perl parser still at least marked out the strings.
      • 2013-01-01 00144, 2013

      • Freso
        Without wrongly marking anything else.
      • 2013-01-01 00150, 2013

      • Freso likes colours o/
      • 2013-01-01 00151, 2013

      • reosarevok
        Freso: ok, not completely true
      • 2013-01-01 00113, 2013

      • reosarevok
        French participles do sometimes have gender agreement. Should be possible to avoid them in translation though :)
      • 2013-01-01 00107, 2013

      • Freso
        ianmcorvidae: Alright. Committed and pushed. Testing now.
      • 2013-01-01 00130, 2013

      • ianmcorvidae continues investigating how half of musicbrainz appears to be broken
      • 2013-01-01 00109, 2013

      • reosarevok
        How is that different from every other day? ;)
      • 2013-01-01 00115, 2013

      • tyteen4a03 joined the channel
      • 2013-01-01 00124, 2013

      • ianmcorvidae
        a normal day it's a lot less broken than this :P
      • 2013-01-01 00143, 2013

      • reosarevok
        Fine, fine
      • 2013-01-01 00111, 2013

      • nikki
        I think by the time cron stops emailing me every hour, I'm going to think something broke on my end instead :P
      • 2013-01-01 00154, 2013

      • nikki
        'cause my emails about missing thumbnails have been replaced with emails about no new cover art isntead :P
      • 2013-01-01 00103, 2013

      • ianmcorvidae
        okay, one damn thing fixed
      • 2013-01-01 00118, 2013

      • ianmcorvidae
        not the biggest problem right now, but in a couple hours there might be new search indexes again :P
      • 2013-01-01 00152, 2013

      • reosarevok
        well, that's a big one for editors :p
      • 2013-01-01 00101, 2013

      • reosarevok
        (not for customers I guess :/ )
      • 2013-01-01 00133, 2013

      • ianmcorvidae
        well, I mean, most people use our search servers :P not many people set up their own
      • 2013-01-01 00153, 2013

      • ianmcorvidae
        however, yes, replication being broken is still a bit more pressing :P
      • 2013-01-01 00114, 2013

      • Freso
        "Result: FAIL" D:
      • 2013-01-01 00131, 2013

      • Freso
        "Failed 1/343 subtests" - man. :(
      • 2013-01-01 00109, 2013

      • Freso
        Whut? Not code changed in this branch.
      • 2013-01-01 00113, 2013

      • Freso
      • 2013-01-01 00147, 2013

      • reosarevok
        Maybe you failed the attitude test :p
      • 2013-01-01 00100, 2013

      • ianmcorvidae
        can't find the actual error
      • 2013-01-01 00114, 2013

      • Freso
        ianmcorvidae: "Failed test 'content has added timestamp'
      • 2013-01-01 00116, 2013

      • Freso
        "
      • 2013-01-01 00118, 2013

      • Freso
        And
      • 2013-01-01 00125, 2013

      • Freso
        "Failed test 'content has last modified timestamp'"
      • 2013-01-01 00104, 2013

      • ianmcorvidae
        ...
      • 2013-01-01 00109, 2013

      • ianmcorvidae
        the year changed
      • 2013-01-01 00117, 2013

      • Freso
        Ha.
      • 2013-01-01 00121, 2013

      • nikki
        haha
      • 2013-01-01 00140, 2013

      • ianmcorvidae
        I wonder if that's actually it, lol
      • 2013-01-01 00158, 2013

      • Freso
        Looks like a reasonable reason.
      • 2013-01-01 00105, 2013

      • Freso
        They check for "X years ago".
      • 2013-01-01 00114, 2013

      • Freso
        Since we've moved a year, X would now be X+1
      • 2013-01-01 00126, 2013

      • ianmcorvidae
        yeah, but it should probably be mocking the time
      • 2013-01-01 00138, 2013

      • ianmcorvidae
        or, y'know, calculating the correct thing to be checking for :P
      • 2013-01-01 00134, 2013

      • Freso
        ianmcorvidae: Yeah, hold on. Let me see if I can fix the test. :)
      • 2013-01-01 00135, 2013

      • |Frederik joined the channel
      • 2013-01-01 00133, 2013

      • Freso
        $_[-1] = DateTime->new( year => 2012, month => 01, day => 01 );
      • 2013-01-01 00112, 2013

      • Freso
        ocharles: For some reason your timelock wrapper isn't working. :|
      • 2013-01-01 00149, 2013

      • ocharles
        Freso: hm?
      • 2013-01-01 00115, 2013

      • ocharles
        oh, it's a new year and the tests have broke?
      • 2013-01-01 00121, 2013

      • Freso
        ocharles: Yep.
      • 2013-01-01 00135, 2013

      • Freso
        ocharles: https://pastee.org/nsm8b - from line 4528
      • 2013-01-01 00137, 2013

      • ocharles
        which test?
      • 2013-01-01 00138, 2013

      • ocharles
        thanks
      • 2013-01-01 00159, 2013

      • ianmcorvidae
        break all the things (apparently)
      • 2013-01-01 00124, 2013

      • ocharles
        hrm
      • 2013-01-01 00157, 2013

      • |Frederik
        How do I enter a CD track in muscibrainz which actually consists of a mash-up of two different songs in itself?
      • 2013-01-01 00150, 2013

      • |Frederik
        (in practice: in http://musicbrainz.org/release/e6d76625-0d0c-4daf…, track 15 and 16 of disc 2 are in reality squeezed together in track 15)
      • 2013-01-01 00102, 2013

      • Freso
        Ha. I thought I was in -devel. :D
      • 2013-01-01 00114, 2013

      • Freso
        |Frederik: Track A / Track B
      • 2013-01-01 00130, 2013

      • Freso
        |Frederik: Optionally, Track A / [silence] / Track B
      • 2013-01-01 00141, 2013

      • nikki
        the former
      • 2013-01-01 00147, 2013

      • Freso
        If there's a significant amount of silence in between.
      • 2013-01-01 00156, 2013

      • nikki disagrees with the latter
      • 2013-01-01 00109, 2013

      • |Frederik
        thanks. there's no silence between them :-)
      • 2013-01-01 00123, 2013

      • nikki
        also it looks like freso is talking to himself or something >_<
      • 2013-01-01 00133, 2013

      • Freso
        nikki: I know!
      • 2013-01-01 00139, 2013

      • |Frederik
        but what about the artist? artist 1 / Artist B too?
      • 2013-01-01 00149, 2013

      • nikki
        |Frederik: that's what I would do
      • 2013-01-01 00149, 2013

      • Freso
        Tricky.
      • 2013-01-01 00155, 2013

      • Freso
        But yeah.
      • 2013-01-01 00159, 2013

      • |Frederik
        thanks
      • 2013-01-01 00159, 2013

      • jacobbrett
        Sucker DJ's / Soul Providers feat. Michelle Shellers
      • 2013-01-01 00134, 2013

      • tyteen4a03 joined the channel
      • 2013-01-01 00118, 2013

      • Freso
        I really dislike how "Track A / Track B" is both used for separate recordings on the same Recording/Track, but also from separate Works in a set/medley for one Recording/Track. :/
      • 2013-01-01 00129, 2013

      • Freso
        *also for
      • 2013-01-01 00139, 2013

      • nikki
        that's what we have relationships for :P
      • 2013-01-01 00142, 2013

      • Freso
        Unfortunately, I have no good suggestion for how to fix it.
      • 2013-01-01 00145, 2013

      • Freso
        Still.
      • 2013-01-01 00152, 2013

      • Krystof
        I have a stand-alone CD which is almost certainly the exact same CD (but a different release) to CD 3 of http://musicbrainz.org/release/446be6d6-8775-4120…
      • 2013-01-01 00120, 2013

      • nikki
        the best thing to do imo is to just add the appropriate relationships, then the information is there to do whatever you want with it if you don't like the way it's named
      • 2013-01-01 00141, 2013

      • Freso
        Because I wouldn't be doing that as it is. :)
      • 2013-01-01 00156, 2013

      • Freso
        Regardless of how it's named. :)
      • 2013-01-01 00156, 2013

      • Krystof
        I would like to add a new release for this stand-alone CD, but pointing to the existinng recordings, and ideally without having to type in the track list again
      • 2013-01-01 00130, 2013

      • nikki
        you can search for an existing disc if you click the "add disc" button on the tracklist tab
      • 2013-01-01 00134, 2013

      • jacobbrett
        Krystof: Create a new release, use that 80disc release as the tracklist basis, delete the other CDs, set disc 3 to disc 1
      • 2013-01-01 00159, 2013

      • jacobbrett
        Or what nikki said
      • 2013-01-01 00116, 2013

      • Krystof
        separately, should I add my disc-id to disc 3 even though it's not the exact same release?
      • 2013-01-01 00150, 2013

      • jacobbrett
        Krystof: The disc will be considered unique, though they will share the tracklist. Add your disc ID to the new release's CD.
      • 2013-01-01 00130, 2013

      • Freso
        jacobbrett: The DiscID will be attached to the tracklist, which, if he does what you're suggestion, will be the same for the two mediums.
      • 2013-01-01 00140, 2013

      • Freso
        (To my understanding, anyway.)
      • 2013-01-01 00141, 2013

      • |Frederik
        I guess I now need someone to approve this edit before the change is made in the database: http://musicbrainz.org/edit/20349760
      • 2013-01-01 00143, 2013

      • nikki
        no, it won't
      • 2013-01-01 00149, 2013

      • nikki
        disc ids are attached to mediums
      • 2013-01-01 00125, 2013

      • jacobbrett
        Krystof: But i realise that's not what you were asking, so yes, though only if you're sure it's the same master
      • 2013-01-01 00153, 2013

      • sezuan joined the channel
      • 2013-01-01 00152, 2013

      • Krystof
        do my new release and the CD 3 medium from the collection share the tracklist?
      • 2013-01-01 00110, 2013

      • Krystof
        if I add performance credit ARs to one, will they be reflected / present in the other?
      • 2013-01-01 00129, 2013

      • reosarevok
        Krystof: if the recordings are the same, yes
      • 2013-01-01 00142, 2013

      • reosarevok
        (but if you add them to the recordings, not to the release)
      • 2013-01-01 00134, 2013

      • Krystof
        ha, yes, good point
      • 2013-01-01 00145, 2013

      • Krystof
        that is I think what I meant
      • 2013-01-01 00126, 2013

      • Krystof
        (I once drew a diagram of a data structure for a score notation editor that stayed on my office window for years, needing constant reference. I feel that the current mb schema might need a diagram of similar complexity...)
      • 2013-01-01 00157, 2013

      • reosarevok
        haha
      • 2013-01-01 00146, 2013

      • reosarevok
      • 2013-01-01 00156, 2013

      • ianmcorvidae
        we need to update that :P
      • 2013-01-01 00117, 2013

      • Krystof
        like that, only with rounder boxes
      • 2013-01-01 00129, 2013

      • Krystof
        (also, that has too much complexity: everything has an "id", I get it, but for my purposes that information is already present by the fact that there's a box. I understand that for a database user/developer they need to know which are primary and which are foreign keys)
      • 2013-01-01 00135, 2013

      • nikki wonders if we can convince luks to update it :P
      • 2013-01-01 00154, 2013

      • ianmcorvidae
        well, we don't necessarily need *him* to, the source file is available
      • 2013-01-01 00104, 2013

      • ianmcorvidae
        we're just lazy sods who haven't gotten to it :P
      • 2013-01-01 00145, 2013

      • nikki
        sure. but he presumably has the diagram program and source file already :P
      • 2013-01-01 00114, 2013

      • nikki
        Krystof: artist_credit_name doesn't have an id :P
      • 2013-01-01 00124, 2013

      • Krystof
        there's always one
      • 2013-01-01 00134, 2013

      • nikki
        :D
      • 2013-01-01 00114, 2013

      • Krystof
        but ok, yes, that diagram would have told me the answer to my question
      • 2013-01-01 00128, 2013

      • Krystof
        assuming that it wasn't out-of-date in a misleading way for my question, anyway :-)
      • 2013-01-01 00146, 2013

      • ianmcorvidae
        nah, I think it's mostly smallish stuff
      • 2013-01-01 00151, 2013

      • ianmcorvidae
        like release group secondary types
      • 2013-01-01 00151, 2013

      • luks should convert the diagram to visio or something :)
      • 2013-01-01 00102, 2013

      • ianmcorvidae
        hah
      • 2013-01-01 00122, 2013

      • ianmcorvidae
        and make it even less likely any of us linux users will update it? :P
      • 2013-01-01 00145, 2013

      • luks
        are there any sensible linux tools for this kind of diagrams?
      • 2013-01-01 00121, 2013

      • ianmcorvidae
        I honestly don't know, other than dbmodel I've really not needed anything quite like that
      • 2013-01-01 00129, 2013

      • ianmcorvidae
        most things I do graphviz is fine for
      • 2013-01-01 00143, 2013

      • Freso
        \o/ graphviz
      • 2013-01-01 00103, 2013

      • Freso
        But, yea,h not sure if graphviz can do stuff like that.
      • 2013-01-01 00106, 2013

      • Freso
        *yeah,
      • 2013-01-01 00149, 2013

      • JoeLlama joined the channel
      • 2013-01-01 00138, 2013

      • Krystof
        graphviz can do a lot
      • 2013-01-01 00145, 2013

      • Krystof
        another possibility is tikz
      • 2013-01-01 00136, 2013

      • gekl joined the channel
      • 2013-01-01 00157, 2013

      • Krystof
        ditaa also looks possible
      • 2013-01-01 00120, 2013

      • Freso
        LibreOffice's Draw can do flowcharts.
      • 2013-01-01 00141, 2013

      • luks
        yeah, this is exactly why I had to write my own tool :)
      • 2013-01-01 00103, 2013

      • luks
        hm, but anyway, what in the diagram is out of date?
      • 2013-01-01 00141, 2013

      • nikki
        url.description was dropped, as was url.ref_count
      • 2013-01-01 00119, 2013

      • nikki
        work.language was added
      • 2013-01-01 00143, 2013

      • nikki
        the three alias tables now have sort_name (linking to the same entity_name table as the name column), the six date columns and a primary_for_locale field
      • 2013-01-01 00128, 2013

      • nikki
        oh, and a type column linking to artist_alias_type, label_alias_type and work_alias_type
      • 2013-01-01 00106, 2013

      • nikki
        work.iswc has gone and has been replaced by a new iswc table
      • 2013-01-01 00103, 2013

      • nikki
        likewise artist.ipi is now the artist_ipi table and label.ipi is the label_ipi table
      • 2013-01-01 00150, 2013

      • nikki
        er, artist.ipi_code and label.ipi_code I mean
      • 2013-01-01 00149, 2013

      • nikki
        release_group.type now links to release_group_primary_type.id and there's a new release_group_secondary_type table along with release_group_secondary_type_join to link it with release_group
      • 2013-01-01 00135, 2013

      • voiceinsideyou joined the channel
      • 2013-01-01 00154, 2013

      • nikki
        ohh, and the language table. iso_code_2 -> iso_code_1, iso_code_3b -> iso_code_2b, iso_code_3t -> iso_code_2t and a new column iso_code_3
      • 2013-01-01 00104, 2013

      • luks
        and I thought I know at least somehow what is happening in MB... :)
      • 2013-01-01 00121, 2013

      • voiceinsideyou1 joined the channel
      • 2013-01-01 00142, 2013

      • nikki
        hehe