#musicbrainz-devel

/

      • andreypopp joined the channel
      • 2013-02-01 03254, 2013

      • voiceinsideyou joined the channel
      • 2013-02-01 03229, 2013

      • andreypopp joined the channel
      • 2013-02-01 03201, 2013

      • voiceinsideyou joined the channel
      • 2013-02-01 03205, 2013

      • andreypopp joined the channel
      • 2013-02-01 03224, 2013

      • ijabz joined the channel
      • 2013-02-01 03212, 2013

      • warp
        hello!
      • 2013-02-01 03255, 2013

      • ijabz_ joined the channel
      • 2013-02-01 03235, 2013

      • ocharles
        ianmcorvidae: you are being a bit too hard I think, seeing as 7digital run half there business on mono :p
      • 2013-02-01 03238, 2013

      • ocharles
        but I agree, it's still pretty shitty
      • 2013-02-01 03231, 2013

      • warp
        warp has changed the topic to: toasty | http://musicbrainz.org/#devel | agenda: reviews, MBS-5800 (reotab/nikki), MBS-5677 (ian), move to bitbucket instead (Freso), triage (warp) | to ask rob: ws meeting
      • 2013-02-01 03247, 2013

      • andreypopp joined the channel
      • 2013-02-01 03205, 2013

      • Freso
        Oh, there's a meeting tonight, yes?
      • 2013-02-01 03212, 2013

      • Freso makes sure to have soup ready
      • 2013-02-01 03219, 2013

      • warp
        Freso: schema change call for participation meeting
      • 2013-02-01 03223, 2013

      • Freso
        warp: Yes yes. I was just checking that I didn't remember the date wrong. :)
      • 2013-02-01 03218, 2013

      • ocharles
        what, a meeting tonight?
      • 2013-02-01 03219, 2013

      • ocharles
        :/
      • 2013-02-01 03226, 2013

      • ocharles
        when was that decided?
      • 2013-02-01 03200, 2013

      • warp
        ocharles: around the time of the previous schema change release.
      • 2013-02-01 03210, 2013

      • ocharles
        we decided all the way back then to have a meeting on friday?
      • 2013-02-01 03215, 2013

      • ocharles
        bleh
      • 2013-02-01 03217, 2013

      • warp
        no
      • 2013-02-01 03232, 2013

      • warp
        we decided all the way back then to have a meeting on february 1st. (without looking at which week day it was :)
      • 2013-02-01 03244, 2013

      • ocharles
        oh
      • 2013-02-01 03258, 2013

      • ocharles
        well, i won't be there for it
      • 2013-02-01 03203, 2013

      • ocharles
        at least, i doubt i will :)
      • 2013-02-01 03208, 2013

      • warp
        ocharles: it's only the call for participation. so it's only to announce that everyone should make sure their favourite schema change tickets are actionable
      • 2013-02-01 03231, 2013

      • warp
        ocharles: I probably won't be there for it either. I will have just checked-in to the hotel for FOSDEM probably.
      • 2013-02-01 03238, 2013

      • ocharles
        ooo, fun!
      • 2013-02-01 03243, 2013

      • andreypopp joined the channel
      • 2013-02-01 03214, 2013

      • Leftmost joined the channel
      • 2013-02-01 03232, 2013

      • Leftmost
        Hmm. Can I get debug output out of ProcessSubscriptions in any way? I'm running it and nothing is happening, but I'm not sure if there's just no error and I'm missing groundwork, or if it's complaining for some reason and I'm not seeing it.
      • 2013-02-01 03224, 2013

      • luks
        isn't there a --verbose option?
      • 2013-02-01 03200, 2013

      • Leftmost
        Doesn't appear to be doing anything. In fact, throwing printf statements into ProcessSubscriptions is getting me nothing.
      • 2013-02-01 03211, 2013

      • warp
        Leftmost: how about warn?
      • 2013-02-01 03213, 2013

      • warp
        I don't know if those scripts are affected by the LOGGER_ARGUMENTS settings in DBDefs
      • 2013-02-01 03240, 2013

      • Leftmost
        No output from warn either.
      • 2013-02-01 03244, 2013

      • Leftmost
        Even sticking a die in the first line gets me nothing.
      • 2013-02-01 03251, 2013

      • warp
        then something is going wrong :)
      • 2013-02-01 03219, 2013

      • Leftmost
        Clearly. I'm running it with `carton exec -- ./admin/ProcessSubscriptions' just as it's done in daily.sh.
      • 2013-02-01 03258, 2013

      • warp
      • 2013-02-01 03247, 2013

      • Leftmost
        Did exactly that and nothing. I'm not familiar enough with carton to know what could be going wrong. Any ideas?
      • 2013-02-01 03256, 2013

      • warp
        hm, no, not really.
      • 2013-02-01 03217, 2013

      • warp
        other musicbrainz stuff does run?
      • 2013-02-01 03234, 2013

      • Leftmost
        Yep. Server is up and running.
      • 2013-02-01 03240, 2013

      • warp
        weird!
      • 2013-02-01 03254, 2013

      • possible joined the channel
      • 2013-02-01 03256, 2013

      • warp
        Leftmost: you can run carton exec -- /bin/bash to get a shell with carton's variables. that way you can inspect the environment to see if anything looks wrong there.
      • 2013-02-01 03213, 2013

      • warp
        Leftmost: but if the musicbrainz server is running fine, that seems to be correct.
      • 2013-02-01 03219, 2013

      • possible has left the channel
      • 2013-02-01 03236, 2013

      • ijabz joined the channel
      • 2013-02-01 03207, 2013

      • Leftmost
        Odd. If I run it with an option it doesn't actually have it gives me usage output, but I have no clue where that comes from.
      • 2013-02-01 03220, 2013

      • warp
        Leftmost: that is done automatically by MooseX::Getopt (used in lib/MusicBrainz/Script/SubscriptionEmails.pm)
      • 2013-02-01 03215, 2013

      • Leftmost
        How the hell can it get there but not die if I tell it to die?
      • 2013-02-01 03248, 2013

      • warp
        are you editing the right file? :)
      • 2013-02-01 03223, 2013

      • warp
        (I have two or three musicbrainz-server checkouts and sometimes get it wrong ;)
      • 2013-02-01 03250, 2013

      • Leftmost
        D'oh.
      • 2013-02-01 03200, 2013

      • Leftmost
        Ahem.
      • 2013-02-01 03248, 2013

      • Leftmost
        I'm not at my best at this hour.
      • 2013-02-01 03257, 2013

      • Leftmost
        Thanks for, uhh... debugging my problem. :-P
      • 2013-02-01 03212, 2013

      • warp
        :D
      • 2013-02-01 03223, 2013

      • djce joined the channel
      • 2013-02-01 03213, 2013

      • Leftmost
        There, output as expected, and even output I wanted to see.
      • 2013-02-01 03217, 2013

      • Leftmost
        Now hopefully the email will arrive or I'll have to poke ianmcorvidae into telling me why it didn't. :-P
      • 2013-02-01 03239, 2013

      • andreypopp joined the channel
      • 2013-02-01 03229, 2013

      • ocharles
        Good morning
      • 2013-02-01 03211, 2013

      • SexyKen-FN joined the channel
      • 2013-02-01 03213, 2013

      • warp
        ocharles: "We need to know how many operations this script will incur, and whether or not that will affect replication packets." what should I do to determine that?
      • 2013-02-01 03220, 2013

      • warp
      • 2013-02-01 03254, 2013

      • ocharles
        warp: setting up replication locally, running the script, and then poking the replication tables and maybe producing a replication packet and checking the resulting size
      • 2013-02-01 03239, 2013

      • warp
        ooph, ok.
      • 2013-02-01 03250, 2013

      • ocharles
        it's not much work, don't worry :)
      • 2013-02-01 03208, 2013

      • ocharles
        you just need to install dbmirror (from our github account) and then some SQL in admin, and that's about it
      • 2013-02-01 03222, 2013

      • ocharles
        you will need a full, representative, dump to run the script against though too
      • 2013-02-01 03233, 2013

      • ocharles
        so if you're ran it locally once and nuked your data, you'll need to reimport
      • 2013-02-01 03253, 2013

      • Leftmost
        Morning, ocharles.
      • 2013-02-01 03239, 2013

      • warp
        ocharles: oh with setting up replication locally you mean generating the replication packets. I was thinking of setting up replication as in a client of the replication packets.
      • 2013-02-01 03249, 2013

      • ocharles
        oh, no - you don't need to go that far
      • 2013-02-01 03203, 2013

      • ocharles
        if the packet you produce is around the same order of magnitude, I'd say :ship: it
      • 2013-02-01 03227, 2013

      • warp
        but I am currently nuking my data, so I still have to do a fresh import.
      • 2013-02-01 03230, 2013

      • warp
        (which won't be done before I have to get ready for fosdem)
      • 2013-02-01 03237, 2013

      • Leftmost
        I'm not really sure what the purpose of load_subscription() is in lib/MusicBrainz/Script/SubscriptionEmails.pm. Can anyone lend any insight?
      • 2013-02-01 03225, 2013

      • warp
        Leftmost: if you look in Entity::ArtistSubscription, you'll see it has an artist_id and an artist property.
      • 2013-02-01 03254, 2013

      • Leftmost
        So it takes the ID and fills in the artist object?
      • 2013-02-01 03255, 2013

      • warp
        Leftmost: $self->c->model('Artist')->load($subscription) in this case will take that artist_id, load the data from the database and assign it to artist.
      • 2013-02-01 03207, 2013

      • warp
        yes, something like that.
      • 2013-02-01 03245, 2013

      • warp
        c->model('Artist')->load refers to the load() method of MusicBrainz::Server::Data::Artist
      • 2013-02-01 03237, 2013

      • Leftmost
        Yeah, I was looking at that too. I just wasn't sure I understood what load_subobjects() was doing and whether that would work for collections, since collections have no similar load() function.
      • 2013-02-01 03219, 2013

      • warp
        I don't think there is currently any entity with a collection_id or something like that, so having a ->load would be a bit pointless.
      • 2013-02-01 03255, 2013

      • Leftmost
        warp, I'm implementing collection subscriptions.
      • 2013-02-01 03229, 2013

      • warp
        ok, so you've created CollectionSubscription.pm ?
      • 2013-02-01 03237, 2013

      • Leftmost
        Yeah.
      • 2013-02-01 03245, 2013

      • warp
        then you probably get to be the one to implement a ->load on Data::Collection as well :)
      • 2013-02-01 03209, 2013

      • Leftmost
        Yep. :)
      • 2013-02-01 03229, 2013

      • Leftmost
        Would load_subobjects() work on Collection since it's a CoreEntity?
      • 2013-02-01 03216, 2013

      • warp
        I don't know, I haven't really touched those parts of the code much.
      • 2013-02-01 03245, 2013

      • Leftmost
        ocharles, any idea?
      • 2013-02-01 03232, 2013

      • reosarevok joined the channel
      • 2013-02-01 03259, 2013

      • reosarevok
        heh
      • 2013-02-01 03206, 2013

      • reosarevok
      • 2013-02-01 03203, 2013

      • nikki joined the channel
      • 2013-02-01 03216, 2013

      • Leftmost
        What the heck do you use an Excel file with every MB song for? What can open that without exploding?
      • 2013-02-01 03224, 2013

      • warp
        let's implement a webservice for it so someone can click save as on a link to http://musicbrainz.org/ws/all_the_things.xls
      • 2013-02-01 03240, 2013

      • ocharles
        Leftmost: I can look in a bit
      • 2013-02-01 03245, 2013

      • luks
        heh, it has 17 bids
      • 2013-02-01 03254, 2013

      • Leftmost
        Alright, thanks.
      • 2013-02-01 03249, 2013

      • nikki joined the channel
      • 2013-02-01 03202, 2013

      • warp
        Excel 2007 has maximum of 1,048,576 rows, afaics this hasn't changed in Excel 2010 or Excel 2013.
      • 2013-02-01 03239, 2013

      • reosarevok
        warp: apparently nobody told 17 people! :p
      • 2013-02-01 03243, 2013

      • Leftmost
        We only have 13,200,110 rows in track. :-P
      • 2013-02-01 03208, 2013

      • warp
        I was looking at recording, which is still about 12,000,000
      • 2013-02-01 03257, 2013

      • reosarevok
        I like how one tag is "data entry"
      • 2013-02-01 03207, 2013

      • reosarevok
        :p
      • 2013-02-01 03222, 2013

      • warp
        for about $50
      • 2013-02-01 03216, 2013

      • Leftmost
        I would guess that the rate of track creation would exceed the ability of anyone to enter tracks.
      • 2013-02-01 03258, 2013

      • reosarevok
        Leftmost, you killjoy
      • 2013-02-01 03200, 2013

      • reosarevok
        :p
      • 2013-02-01 03245, 2013

      • Leftmost
        That's what I'm here for.
      • 2013-02-01 03227, 2013

      • djce joined the channel
      • 2013-02-01 03226, 2013

      • alastairp
        reosarevok: man, I'm so tempted :-P
      • 2013-02-01 03209, 2013

      • ocharles
        aw yeahhhh
      • 2013-02-01 03214, 2013

      • ocharles
        musicbrainz-data-service can create relationships
      • 2013-02-01 03221, 2013

      • Freso
        ocharles: :)
      • 2013-02-01 03225, 2013

      • ocharles gets a cup of tea and gets back to Leftmost
      • 2013-02-01 03238, 2013

      • Freso
        In Soviet Russia, relationships can create musicbrainz-data-service.
      • 2013-02-01 03203, 2013

      • Freso
        Beat that, ocharles.
      • 2013-02-01 03216, 2013

      • ocharles
        In Soviet Lambda, Haskell creates everything.
      • 2013-02-01 03201, 2013

      • Leftmost sips mead.
      • 2013-02-01 03208, 2013

      • reosarevok
        at the beginning was the monad?
      • 2013-02-01 03210, 2013

      • derr84 joined the channel
      • 2013-02-01 03222, 2013

      • Freso
        warp: Is there a reason you haven't assigned you to http://tickets.musicbrainz.org/browse/MBS-5801 ?
      • 2013-02-01 03238, 2013

      • Freso
        warp: I do I just need to poke you more about it?
      • 2013-02-01 03207, 2013

      • ocharles
        ok Leftmost, what are you trying to do?
      • 2013-02-01 03240, 2013

      • Freso
        That's a grand question, dude.
      • 2013-02-01 03209, 2013

      • voiceinsideyou joined the channel
      • 2013-02-01 03227, 2013

      • Leftmost
        I've created a CollectionSubscription class and I'm trying to load the collection corresponding to a CollectionSubscription's ID attribute.
      • 2013-02-01 03231, 2013

      • Leftmost
        I tried copypastaing the MusicBrainz::Server::Data::Artist load() method over to Collection, but right now it's telling me that validation failed for 'Collection'.
      • 2013-02-01 03258, 2013

      • reosarevok
        Go go go developing-through-copypasting!
      • 2013-02-01 03205, 2013

      • MBJenkins
        Oliver Charles: Add support for submitting relationship trees
      • 2013-02-01 03220, 2013

      • reosarevok
        (not sarcastic or anything :) )
      • 2013-02-01 03255, 2013

      • Leftmost
        Admittedly, I don't really understand how load_subobjects() works.
      • 2013-02-01 03253, 2013

      • ocharles
        Why do you think you need to load the collection, out of interest?
      • 2013-02-01 03256, 2013

      • ocharles
        Also, paste code
      • 2013-02-01 03224, 2013

      • Leftmost
        To get the collection name for subscription emails.
      • 2013-02-01 03250, 2013

      • ocharles
        Ok, on wards to the pasting :)
      • 2013-02-01 03254, 2013

      • ocharles
        onwards...
      • 2013-02-01 03254, 2013

      • Leftmost
      • 2013-02-01 03204, 2013

      • Leftmost
        If there's anything that's unclear or you need more context, let me know.
      • 2013-02-01 03246, 2013

      • ocharles
        Hm