#metabrainz

/

      • Nizar has quit
      • 2021-03-23 08230, 2021

      • c1e0 joined the channel
      • 2021-03-23 08232, 2021

      • c1e0 has quit
      • 2021-03-23 08205, 2021

      • MRiddickW has quit
      • 2021-03-23 08251, 2021

      • xps2 joined the channel
      • 2021-03-23 08200, 2021

      • xps2 has quit
      • 2021-03-23 08218, 2021

      • reosarevok
        yvanzo: https://github.com/metabrainz/musicbrainz-server/… has a lot of different, not-really-related chunks, so I'm not sure how to make it into a ticket
      • 2021-03-23 08243, 2021

      • reosarevok
        I guess I could just make a generic ticket listing all the different bits that are affected, or one ticket per chunk
      • 2021-03-23 08212, 2021

      • c1e0 joined the channel
      • 2021-03-23 08213, 2021

      • yvanzo
        hi reosarevok: as you prefer, these chunks will be merged/released at the same time anyway.
      • 2021-03-23 08215, 2021

      • davic joined the channel
      • 2021-03-23 08254, 2021

      • adhi001 joined the channel
      • 2021-03-23 08256, 2021

      • CatQuest
        hommik 🙋
      • 2021-03-23 08209, 2021

      • sumedh joined the channel
      • 2021-03-23 08250, 2021

      • ruaok
        mooin!
      • 2021-03-23 08216, 2021

      • sumedh has quit
      • 2021-03-23 08226, 2021

      • Madtrick joined the channel
      • 2021-03-23 08236, 2021

      • iliekcomputers
        alastairp: you merged the metrics PR, exciting!
      • 2021-03-23 08229, 2021

      • ruaok
        indeed, I am very much looking forward to using that.
      • 2021-03-23 08223, 2021

      • alastairp
        iliekcomputers: yes, hoping that I could make a release so that in the meeting yesterday I could say "oh, and I deployed metrics and got it running in LB"
      • 2021-03-23 08225, 2021

      • alastairp
        but not quite :-P
      • 2021-03-23 08248, 2021

      • alastairp
        but yeah, I'll get a PR in today to update LB
      • 2021-03-23 08205, 2021

      • sumedh joined the channel
      • 2021-03-23 08231, 2021

      • okno_
        Hello all, I just published a new musicbrainz_rs version, featuring some builder pattern fashioned syntax and a brand new crate documation : https://docs.rs/musicbrainz_rs/0.3.0/musicbrainz_…
      • 2021-03-23 08241, 2021

      • okno_
        If anyone is in the mood to write some doc, many entities still needs to be documented. Let me know if you are interested !
      • 2021-03-23 08233, 2021

      • Madtrick1 joined the channel
      • 2021-03-23 08207, 2021

      • Madtrick has quit
      • 2021-03-23 08222, 2021

      • Madtrick1 has quit
      • 2021-03-23 08250, 2021

      • Madtrick joined the channel
      • 2021-03-23 08209, 2021

      • Madtrick has left the channel
      • 2021-03-23 08239, 2021

      • Madtrick joined the channel
      • 2021-03-23 08253, 2021

      • Madtrick has left the channel
      • 2021-03-23 08212, 2021

      • Madtrick joined the channel
      • 2021-03-23 08238, 2021

      • adhi001 has quit
      • 2021-03-23 08232, 2021

      • nifemi joined the channel
      • 2021-03-23 08202, 2021

      • nifemi has quit
      • 2021-03-23 08245, 2021

      • Nizar joined the channel
      • 2021-03-23 08244, 2021

      • Madtrick has quit
      • 2021-03-23 08210, 2021

      • Madtrick joined the channel
      • 2021-03-23 08233, 2021

      • adhi001 joined the channel
      • 2021-03-23 08234, 2021

      • Etua joined the channel
      • 2021-03-23 08200, 2021

      • Etua has quit
      • 2021-03-23 08250, 2021

      • shivam-kapila
        Good evening Mr_Monkey. Whats left for #$1320?
      • 2021-03-23 08255, 2021

      • shivam-kapila
        #1320*
      • 2021-03-23 08220, 2021

      • Etua joined the channel
      • 2021-03-23 08233, 2021

      • c1e0_ joined the channel
      • 2021-03-23 08221, 2021

      • c1e0 has quit
      • 2021-03-23 08202, 2021

      • Mr_Monkey
        Hai shivam-kapila. I think just a review from alastair , when he finds some time.
      • 2021-03-23 08218, 2021

      • nifemi joined the channel
      • 2021-03-23 08229, 2021

      • milkii has quit
      • 2021-03-23 08203, 2021

      • ruaok
        alastairp: I've been working with the ThreadPoolExecutor... its a bit ... different. had I written my own and suffered more of your ridicule, I would've been done already. 🤣
      • 2021-03-23 08247, 2021

      • goldenshimmer has quit
      • 2021-03-23 08253, 2021

      • piti has quit
      • 2021-03-23 08226, 2021

      • kepstin has quit
      • 2021-03-23 08256, 2021

      • reg[m] has quit
      • 2021-03-23 08210, 2021

      • joshuaboniface has quit
      • 2021-03-23 08213, 2021

      • Cyna[m] has quit
      • 2021-03-23 08227, 2021

      • piti joined the channel
      • 2021-03-23 08241, 2021

      • bharatkalluri[m] has quit
      • 2021-03-23 08255, 2021

      • lorenzuru has quit
      • 2021-03-23 08256, 2021

      • AkshatTiwari[m] has quit
      • 2021-03-23 08256, 2021

      • bitmunch has quit
      • 2021-03-23 08212, 2021

      • SamThursfield[m] has quit
      • 2021-03-23 08214, 2021

      • BrainzGit has quit
      • 2021-03-23 08201, 2021

      • milkii joined the channel
      • 2021-03-23 08259, 2021

      • alastairp
        is that a comment or a complaint? :)
      • 2021-03-23 08232, 2021

      • ruaok
        a complaining comment. :)
      • 2021-03-23 08254, 2021

      • alastairp
        I'll be interested to see how it turns out
      • 2021-03-23 08218, 2021

      • ruaok
        you and me both. been debugging all afternoon.
      • 2021-03-23 08228, 2021

      • alastairp
        I know you can bash out low-level pool management code, I've seen it multiple times. if this is shorter then maybe that's a positive?
      • 2021-03-23 08201, 2021

      • ruaok
        I think once we have the knowledge of how to make this work, this will be easier in the long run.
      • 2021-03-23 08218, 2021

      • ruaok
        the documentation is a bit too pythonic to really make it clear what is expected.
      • 2021-03-23 08237, 2021

      • alastairp
        right, it kind of says "do this and magic will happen" ?
      • 2021-03-23 08204, 2021

      • ruaok
        yes. the futures list... or is a dict? is.. muddy.
      • 2021-03-23 08236, 2021

      • ruaok
        I have yet to see a thread fire off. after several hours of trying.
      • 2021-03-23 08248, 2021

      • ruaok
        the only threads running are the ones I create by hand, lol.
      • 2021-03-23 08209, 2021

      • alastairp
        oh yeah, in the example at https://docs.python.org/3/library/concurrent.futu… the `future_to_url` dict?
      • 2021-03-23 08236, 2021

      • alastairp
        I think it's because a future doesn't have its inputs available on the object
      • 2021-03-23 08243, 2021

      • ruaok
        yeah. crap like that. I can't use any of the standard magic, its far too simplistic. borderline stupid.
      • 2021-03-23 08232, 2021

      • ruaok
        and a more realistic example turns out to be a challenge.
      • 2021-03-23 08237, 2021

      • milkii has quit
      • 2021-03-23 08258, 2021

      • milkii joined the channel
      • 2021-03-23 08201, 2021

      • goldenshimmer joined the channel
      • 2021-03-23 08249, 2021

      • ruaok
        wow. it seems that it practically traps everything running in the thread.
      • 2021-03-23 08212, 2021

      • ruaok
        `{<Future at 0x7f2c0a0a0550 state=finished raised AssertionError>: 1}`
      • 2021-03-23 08233, 2021

      • ruaok
        assertion errors are trapped.
      • 2021-03-23 08221, 2021

      • ruaok
      • 2021-03-23 08227, 2021

      • ruaok
        that finally works.
      • 2021-03-23 08229, 2021

      • alastairp
        cool!
      • 2021-03-23 08232, 2021

      • alastairp
        what sets self.done to True?
      • 2021-03-23 08235, 2021

      • kepstin joined the channel
      • 2021-03-23 08200, 2021

      • ruaok
        the terminate function.
      • 2021-03-23 08211, 2021

      • alastairp
        (what calls terminate?)
      • 2021-03-23 08230, 2021

      • ruaok
      • 2021-03-23 08241, 2021

      • alastairp
        ah, neat. missed that
      • 2021-03-23 08242, 2021

      • ruaok
        its really only called if the user cancels it.
      • 2021-03-23 08224, 2021

      • ruaok
        I guess the container ending will cause... whatever. I don't care. those items will not get ack'ed and thus for the next run will be re-run until acked.
      • 2021-03-23 08228, 2021

      • alastairp
        right, so if there are no more items on the queue it just hangs around a bit more
      • 2021-03-23 08241, 2021

      • ruaok
        so far. yes.
      • 2021-03-23 08253, 2021

      • ruaok
        I've not implemented the historical listen lookup yet.
      • 2021-03-23 08208, 2021

      • ruaok
        the priority queue is also interesting.
      • 2021-03-23 08223, 2021

      • ruaok
        you can't do (priority, dict) and do a put()
      • 2021-03-23 08234, 2021

      • ruaok
        if the two priorities are equal, it will compare the dict and fail.
      • 2021-03-23 08254, 2021

      • ruaok
        so, priorities need to be unique. sigh.
      • 2021-03-23 08221, 2021

      • ruaok
        so i used positive priorities for incoming listens and negative priorities for historical listens.
      • 2021-03-23 08223, 2021

      • joshuaboniface joined the channel
      • 2021-03-23 08200, 2021

      • ruaok
        hacky. I wonder why they implemented it like this.
      • 2021-03-23 08226, 2021

      • Cyna[m] joined the channel
      • 2021-03-23 08221, 2021

      • _lucifer
        you can wrap it inside a class that ignores comparision on dict
      • 2021-03-23 08214, 2021

      • ruaok
        yeah, but I find that decision for the default a bit questionable.
      • 2021-03-23 08218, 2021

      • _lucifer
        that happens because they break ties on priority values by comparing the values. since `dict`s are unordered the error occurs.
      • 2021-03-23 08204, 2021

      • bharatkalluri[m] joined the channel
      • 2021-03-23 08259, 2021

      • reg[m] joined the channel
      • 2021-03-23 08207, 2021

      • SamThursfield[m] joined the channel
      • 2021-03-23 08202, 2021

      • bitmunch joined the channel
      • 2021-03-23 08222, 2021

      • AkshatTiwari[m] joined the channel
      • 2021-03-23 08213, 2021

      • elomatreb[m] joined the channel
      • 2021-03-23 08255, 2021

      • Nizar has quit
      • 2021-03-23 08232, 2021

      • lorenzuru joined the channel
      • 2021-03-23 08207, 2021

      • reosarevok
        yvanzo: ok, found a few chunks it made sense for me to split it into. https://github.com/metabrainz/musicbrainz-server/… :)
      • 2021-03-23 08257, 2021

      • Nizar joined the channel
      • 2021-03-23 08242, 2021

      • nifemi has quit
      • 2021-03-23 08257, 2021

      • c1e0_ has quit
      • 2021-03-23 08212, 2021

      • Etua has quit
      • 2021-03-23 08219, 2021

      • yvanzo
        reosarevok: it looks great, thank you!
      • 2021-03-23 08258, 2021

      • reosarevok
        np!
      • 2021-03-23 08206, 2021

      • reosarevok
        chaban: do you feel https://tickets.metabrainz.org/browse/MBS-10667 is a problem then? As in, should all rels with the same entity be grouped first, in your opinion, and only if all the relationships match between two entities then the two entities should be grouped?
      • 2021-03-23 08207, 2021

      • BrainzBot
        MBS-10667: Release-URLs are sometimes split across multiple relationship groups
      • 2021-03-23 08222, 2021

      • Madtrick1 joined the channel
      • 2021-03-23 08222, 2021

      • reosarevok
        I'm not sure whether either option is better than the other, personally :)
      • 2021-03-23 08209, 2021

      • Madtrick has quit
      • 2021-03-23 08233, 2021

      • Cyna[m]
        reosarevok: any ticket open which I can pick up ?
      • 2021-03-23 08248, 2021

      • Cyna[m]
        for MB
      • 2021-03-23 08207, 2021

      • CatQuest
        reosarevok: I'm not chaban, but I would actually except that "thekirkz" link was grouped together with all the things, and then the "tnsrecords" one alone as download for free because that would be clearer. , everything is on the same link after all so no matter if I want to stream or download or buy there is one link. that there is a *second* link (which may or may not do other things should only be grouped whne it does exactly the same
      • 2021-03-23 08207, 2021

      • CatQuest
        as the other one)
      • 2021-03-23 08239, 2021

      • sumedh has quit
      • 2021-03-23 08228, 2021

      • Cyna[m]
        reosarevok, bitmap you on ?
      • 2021-03-23 08254, 2021

      • Cyna[m]
        Im not very good at UI styling. here is a propotype of a simple notification clickable. Clicking on the notification area opens up a preview of past notifications with its preview and clicking on clear reverts back to main page. https://www.figma.com/proto/HVBTB4PAk7MBkfEqhkk9K…
      • 2021-03-23 08214, 2021

      • Cyna[m]
        Do let me know if this does not work or any of your suggestions :)
      • 2021-03-23 08240, 2021

      • Cyna[m]
      • 2021-03-23 08255, 2021

      • rdswift has quit
      • 2021-03-23 08202, 2021

      • chaban
        I don't think MBS-10667 is a problem per se, it's just a random weirdness I've noticed I guess.
      • 2021-03-23 08203, 2021

      • BrainzBot
        MBS-10667: Release-URLs are sometimes split across multiple relationship groups https://tickets.metabrainz.org/browse/MBS-10667
      • 2021-03-23 08234, 2021

      • elomatreb[m]
        Picard question regarding this tag: https://picard-docs.musicbrainz.org/en/appendices… - the docs show it using the ID3 frame `TXXX:Artists` (lowercase), but Picard seems to write `TXXX:ARTISTS` (all caps). Which is correct?
      • 2021-03-23 08248, 2021

      • rdswift joined the channel
      • 2021-03-23 08256, 2021

      • chaban
        The relationship grouping seems to be weird in general. On this recording vocals and guitar targeting the same artist are put in a single row: https://musicbrainz.org/recording/9efc8631-c3ea-4…
      • 2021-03-23 08246, 2021

      • chaban
        While on this one multiple roles of an artist are spread across: https://musicbrainz.org/recording/7a9cb96e-84c8-4…
      • 2021-03-23 08208, 2021

      • chaban
        Oh wait, is it because of the dates?
      • 2021-03-23 08216, 2021

      • chaban
        Seems it works as intended then
      • 2021-03-23 08214, 2021

      • chaban
        Depending on which info you are interested in there is too much data squeezed in one-size-fits-all layout. Say, you want to know which roles a person had, in that case grouping by instrument would be harder to read than grouping by person then showing the instruments.