#metabrainz

/

      • BrainzGit
        [bookbrainz-site] snyk-bot opened pull request #525 (master…snyk-fix-b6798bfc0e9af55863442307e9ef9ad1): [Snyk] Security upgrade chart.js from 2.9.3 to 2.9.4 https://github.com/bookbrainz/bookbrainz-site/pul…
      • 2020-10-20 29432, 2020

      • travis-ci joined the channel
      • 2020-10-20 29432, 2020

      • travis-ci
        Project bookbrainz-site build #3504: passed in 4 min 30 sec: https://travis-ci.org/bookbrainz/bookbrainz-site/…
      • 2020-10-20 29432, 2020

      • travis-ci has left the channel
      • 2020-10-20 29450, 2020

      • goldenshimmer
        Hi! Is there a way to filter charts by artist/release in Listenbrainz? (see my most listened tracks/releases from an artist, or most listened tracks from a release).
      • 2020-10-20 29438, 2020

      • goldenshimmer
        Started writing up a feature request, but figured I'd double check that it didn't already exist somewhere I'm overlooking it...
      • 2020-10-20 29458, 2020

      • nelgin joined the channel
      • 2020-10-20 29422, 2020

      • nelgin
        yvanzo, just checking in to see if there's any developments on the live indexing?
      • 2020-10-20 29437, 2020

      • nelgin
        (when you're around anyway) :)
      • 2020-10-20 29439, 2020

      • supersandro2000 has quit
      • 2020-10-20 29459, 2020

      • supersandro2000 joined the channel
      • 2020-10-20 29439, 2020

      • heyoni has quit
      • 2020-10-20 29413, 2020

      • heyoni joined the channel
      • 2020-10-20 29421, 2020

      • psychoadept has quit
      • 2020-10-20 29432, 2020

      • supersandro2000 has quit
      • 2020-10-20 29438, 2020

      • yvanzo
        hi nelgin: nothing has been released about live indexing yet, but more changes are being brought to search stuff.
      • 2020-10-20 29413, 2020

      • v6lur joined the channel
      • 2020-10-20 29401, 2020

      • Gore joined the channel
      • 2020-10-20 29402, 2020

      • Gazooo79494 has quit
      • 2020-10-20 29444, 2020

      • Gazooo79494 joined the channel
      • 2020-10-20 29450, 2020

      • sumedh joined the channel
      • 2020-10-20 29454, 2020

      • sumedh has quit
      • 2020-10-20 29450, 2020

      • sumedh joined the channel
      • 2020-10-20 29452, 2020

      • sumedh has quit
      • 2020-10-20 29452, 2020

      • rdswift has quit
      • 2020-10-20 29442, 2020

      • rdswift joined the channel
      • 2020-10-20 29452, 2020

      • kepstin has quit
      • 2020-10-20 29453, 2020

      • maxlath[m] has quit
      • 2020-10-20 29406, 2020

      • reg[m] has quit
      • 2020-10-20 29409, 2020

      • joshuaboniface has quit
      • 2020-10-20 29412, 2020

      • dotmavriq[m] has quit
      • 2020-10-20 29413, 2020

      • MagnusSvensson[m has quit
      • 2020-10-20 29415, 2020

      • maverickfox3[m] has quit
      • 2020-10-20 29415, 2020

      • tripathyprateek[ has quit
      • 2020-10-20 29415, 2020

      • goldenshimmer has quit
      • 2020-10-20 29402, 2020

      • goldenshimmer joined the channel
      • 2020-10-20 29414, 2020

      • CatQuest
        hm. is there a better way to say "receiving air" to?
      • 2020-10-20 29414, 2020

      • CatQuest
        like there is aflute that receives air fro ma bellow. ther is blows but but it doesnt work in this sentece
      • 2020-10-20 29429, 2020

      • CatQuest
        playing, airing. thingy
      • 2020-10-20 29444, 2020

      • CatQuest
        "a row of flute-pipes receiving air from hand operated bellows"
      • 2020-10-20 29451, 2020

      • CatQuest
        ah, "feed air" or "areated"
      • 2020-10-20 29433, 2020

      • reg[m] joined the channel
      • 2020-10-20 29433, 2020

      • tripathyprateek[ joined the channel
      • 2020-10-20 29434, 2020

      • maxlath[m] joined the channel
      • 2020-10-20 29434, 2020

      • kepstin joined the channel
      • 2020-10-20 29434, 2020

      • MagnusSvensson[m joined the channel
      • 2020-10-20 29434, 2020

      • joshuaboniface joined the channel
      • 2020-10-20 29440, 2020

      • maverickfox3[m] joined the channel
      • 2020-10-20 29440, 2020

      • dotmavriq[m] joined the channel
      • 2020-10-20 29425, 2020

      • adhawkins has quit
      • 2020-10-20 29429, 2020

      • adhawkins joined the channel
      • 2020-10-20 29404, 2020

      • d4rkie joined the channel
      • 2020-10-20 29437, 2020

      • D4RK-PH0ENiX has quit
      • 2020-10-20 29405, 2020

      • reosarevok
        If a sperm is wasted / God gets quite areated (?)
      • 2020-10-20 29407, 2020

      • CatQuest
        irate
      • 2020-10-20 29426, 2020

      • MajorLurker
        "activated by"
      • 2020-10-20 29439, 2020

      • CatQuest
        i went with "feed by" becasue it fit better and was shorter :D
      • 2020-10-20 29418, 2020

      • CatQuest
        hm. is it jsut me or is the locale slit shorter now and lacks thingsl iek "old english" and other historical language
      • 2020-10-20 29432, 2020

      • CatQuest
        and isgh. still not sardinian or latin :/
      • 2020-10-20 29446, 2020

      • CatQuest
        i miss sanskrit as an option too :/
      • 2020-10-20 29454, 2020

      • CatQuest
      • 2020-10-20 29421, 2020

      • CatQuest
        !m catquest for \o/ actually doing isntrument tickets :P
      • 2020-10-20 29421, 2020

      • BrainzBot
        You're doing good work, catquest for \o/ actually doing isntrument tickets :P!
      • 2020-10-20 29445, 2020

      • d4rkie has quit
      • 2020-10-20 29433, 2020

      • D4RK-PH0ENiX joined the channel
      • 2020-10-20 29425, 2020

      • ruaok
        pristine___: my daily jam today is much more diverse than it has been as of late. could it be because we finallly got new listens imported last week?
      • 2020-10-20 29431, 2020

      • pristine___
        Possibly. Because of the imports, I noticed that recs were kinda constant, I talked about it to ishaanshah a few weeks back.
      • 2020-10-20 29412, 2020

      • ruaok
        I'll be very interested to see if I like this list -- if I do then we're on a very good track, I think.
      • 2020-10-20 29426, 2020

      • alastairp
        hi ruaok
      • 2020-10-20 29434, 2020

      • ruaok
        hiya
      • 2020-10-20 29437, 2020

      • alastairp
        a few things - can you give me write access to the troi repo?
      • 2020-10-20 29456, 2020

      • ruaok wonders why alastairp doesn't already
      • 2020-10-20 29424, 2020

      • alastairp
        and: have you thought about option flags to patches? I'm finding that i want to add a flag not an argument. not sure the best way to do it
      • 2020-10-20 29400, 2020

      • alastairp
        thinking out loud, I wonder if we could coopt argparse or click for the arg definitions
      • 2020-10-20 29402, 2020

      • ruaok
        I haven't, but I did expand the information that is passed to patches, did you see that?
      • 2020-10-20 29418, 2020

      • alastairp
        I did (and fixed my now-broken patch)
      • 2020-10-20 29420, 2020

      • ruaok
        I tried it with click but....
      • 2020-10-20 29427, 2020

      • alastairp
        heh, right
      • 2020-10-20 29439, 2020

      • alastairp
        let me mull on it
      • 2020-10-20 29446, 2020

      • ruaok
        after spending a couple of hours on trying to get that to work, I decided to just skip it.
      • 2020-10-20 29452, 2020

      • ruaok
        argparser might be more useful.
      • 2020-10-20 29408, 2020

      • ruaok
        but, I agree, options would be good.
      • 2020-10-20 29424, 2020

      • alastairp
        right, you should be able to just pass around an argparse object
      • 2020-10-20 29428, 2020

      • ruaok
        I'm also thinking about adding support for pipeline elements that are user specific.
      • 2020-10-20 29458, 2020

      • ruaok
        e.g always filter the following artists out of my recommendations, if you find .fucktheseartistsrc
      • 2020-10-20 29421, 2020

      • alastairp
        where do you want to track bugs and notes? gh issues?
      • 2020-10-20 29437, 2020

      • ruaok
        lets stick to jira.
      • 2020-10-20 29445, 2020

      • ruaok
        but, not 100% where in jira.
      • 2020-10-20 29449, 2020

      • alastairp
        new project with separate components for each repo?
      • 2020-10-20 29450, 2020

      • ruaok
        +sure
      • 2020-10-20 29443, 2020

      • ruaok
        I'm leaning towards not components inside LB.
      • 2020-10-20 29446, 2020

      • ruaok
        CoreTeam now has admin access to the Troi repo.
      • 2020-10-20 29454, 2020

      • ruaok
        (was only read)
      • 2020-10-20 29411, 2020

      • alastairp
        thanks
      • 2020-10-20 29415, 2020

      • BrainzGit
        [listenbrainz-server] mayhem merged pull request #1143 (rec-component-feedback…rec-feedback-db-api): Recommendation feedback [DB + API] https://github.com/metabrainz/listenbrainz-server…
      • 2020-10-20 29424, 2020

      • alastairp
        sorry, wasn't clear what you meant about LB components?
      • 2020-10-20 29454, 2020

      • ruaok
        I am suggesting we make a "troi" component inside the LB project.
      • 2020-10-20 29410, 2020

      • ruaok
        and a recommendations one for the spark related work.
      • 2020-10-20 29420, 2020

      • yvanzo
        You can even use subcomponents if needed.
      • 2020-10-20 29448, 2020

      • yvanzo
        We have components hierarchy in Tickets.
      • 2020-10-20 29451, 2020

      • alastairp
        OK, I don't mind doing it like that
      • 2020-10-20 29405, 2020

      • alastairp
        ruaok: can you take a look at main/master branches in troi?
      • 2020-10-20 29426, 2020

      • alastairp
        looks like your recent commits you pushed to master
      • 2020-10-20 29442, 2020

      • pristine___
        > I am suggesting we make a "troi" component inside the LB project.
      • 2020-10-20 29446, 2020

      • ruaok
        sigh. habits.
      • 2020-10-20 29452, 2020

      • pristine___
        +1
      • 2020-10-20 29400, 2020

      • alastairp
        hands typing without you knowing what they were writing :)
      • 2020-10-20 29412, 2020

      • ruaok
        yep.
      • 2020-10-20 29429, 2020

      • ruaok
        let me make the components, then I'll fix troi.
      • 2020-10-20 29438, 2020

      • ruaok
        I need to delete the master branch from my own repo.
      • 2020-10-20 29458, 2020

      • ruaok
        components troi and recommendations created.
      • 2020-10-20 29411, 2020

      • ruaok
        alastairp: ok, I merged master into main and pushed. sanity check me please?
      • 2020-10-20 29422, 2020

      • iliekcomputers
        ishaanshah: hey
      • 2020-10-20 29433, 2020

      • iliekcomputers
        Let's chat about the spark imports when you're around
      • 2020-10-20 29432, 2020

      • alastairp
        ruaok: I still see 3 commits in origin/master not in main: https://github.com/metabrainz/troi-recommendation…
      • 2020-10-20 29433, 2020

      • ruaok
        ok, should be fixed now. thanks!
      • 2020-10-20 29452, 2020

      • alastairp
        thank you
      • 2020-10-20 29455, 2020

      • alastairp
        first PR up
      • 2020-10-20 29449, 2020

      • ruaok
        reading
      • 2020-10-20 29411, 2020

      • ruaok
        perhaps I dont understand your comment on line 35 but the block on line 42 should catch required args that are missing, no?
      • 2020-10-20 29441, 2020

      • alastairp
        the issue seemed to be that `args` was `[]`
      • 2020-10-20 29457, 2020

      • alastairp
        so it was iterating over inputs, and then trying to access `args[0]`
      • 2020-10-20 29456, 2020

      • alastairp
        maybe there was something buggy the time that I ran it, and what is supposed to happen is that args gets set to [None, None] if nothing is added to the commandline
      • 2020-10-20 29418, 2020

      • alastairp
        in fact, I may have run into this bug before I updated from origin... let me double-check
      • 2020-10-20 29450, 2020

      • ruaok
        ah, I see. yes.
      • 2020-10-20 29405, 2020

      • ruaok
        I think improving the block on line 42 will give better error messages.
      • 2020-10-20 29415, 2020

      • alastairp
        yeah, right. I'm just opening some tickets, including one to look into argparse. might be better to leave that check until we see if we can integrate argparse?
      • 2020-10-20 29431, 2020

      • ruaok
        ok
      • 2020-10-20 29458, 2020

      • ruaok
        then i see no issues on #10.
      • 2020-10-20 29423, 2020

      • ruaok
        well, there are issues on #10. #10 downing street. but your PR seems fine. :)
      • 2020-10-20 29444, 2020

      • alastairp
        nice
      • 2020-10-20 29402, 2020

      • alastairp
        thanks for all of the heavy work on the pipeline
      • 2020-10-20 29413, 2020

      • alastairp
        pretty nice to just run a command and have a bunch of stuff happen
      • 2020-10-20 29421, 2020

      • ruaok
        np. feels good no?
      • 2020-10-20 29437, 2020

      • ruaok
        just curious, why did you decide to go for mfccs and not (also) gfccs? (what whatever the other one was)
      • 2020-10-20 29426, 2020

      • alastairp
        no reason yet, just getting the pipeline going and then I want to work out a way of trying all metrics and displaying them somehow so that we can get an idea of how well each one works
      • 2020-10-20 29410, 2020

      • ruaok
        ok.
      • 2020-10-20 29431, 2020

      • ruaok
        want to collaborate on that? I just posted an error I got when running your code. should I dive in?
      • 2020-10-20 29448, 2020

      • alastairp
        give me 5 mins and I'll open a ticket pointing to the error if you want
      • 2020-10-20 29409, 2020

      • alastairp
      • 2020-10-20 29426, 2020

      • alastairp
        there's a bug here - if the lookup element can't find an item, it leaves it in the pipeline
      • 2020-10-20 29436, 2020

      • alastairp
        then the playlist element just assumes that it exists
      • 2020-10-20 29457, 2020

      • ruaok
        yes.
      • 2020-10-20 29413, 2020

      • ruaok
        I'm torn on what the right beheaviour should be.
      • 2020-10-20 29419, 2020

      • ruaok
        remove?
      • 2020-10-20 29425, 2020

      • ruaok
        make it an option to remove?
      • 2020-10-20 29408, 2020

      • alastairp
        the even stranger thing, is when I use the lookup endpoint to find data for an mbid that is reported as missing, it gets returned! (and there isn't a redirect either, so that's not the problem)
      • 2020-10-20 29456, 2020

      • ruaok
        recording_lookup element you mean?
      • 2020-10-20 29457, 2020

      • alastairp
        maybe the lookup element should set a flag, and the playlist element (or any other terminal elemtns) can be configured to skip these ones?
      • 2020-10-20 29407, 2020

      • alastairp
        yes, right