#metabrainz

/

      • zerodogg has quit
      • zerodogg joined the channel
      • zerodogg has quit
      • mara42 joined the channel
      • zerodogg joined the channel
      • mara42 has quit
      • mara42 joined the channel
      • mara42 has quit
      • mara42 joined the channel
      • minimal has quit
      • pite_ has quit
      • Zhele has quit
      • Zhele joined the channel
      • BrainzGit
        [listenbrainz-android] 14pranavkonidena opened pull request #431 (03dev…MOBILE-203/UserPages-1.1): Mobile 203/user pages 1.1 https://github.com/metabrainz/listenbrainz-andr...
      • Kladky joined the channel
      • dseomn has quit
      • dseomn joined the channel
      • yellowhatpro[m]
        yvanzo: bitmap , It just hit me that if we are creating triggers, then we are expecting it to log the `internet_archive_urls` table right?
      • But since we have 2 tables, edit_data and edit_note to apply the trigger on, and we will have to tranform the rows, how we are gonna do that using trigger.
      • I suppose the advantage was, that it would still be logging to the `internet_archive_urls`, but we still need transforming the rows from the 2 tables before that right?
      • * yvanzo: bitmap , It just hit me that if we are creating triggers, then we are expecting it to log the `internet_archive_urls` table right?
      • But since we have 2 tables, edit\_data and edit\_note to apply the trigger on, and we will have to tranform the rows, how we are gonna do that using trigger.
      • I suppose the advantage was, that it would still be logging to the `internet_archive_urls` when the container is down, but we still need transforming the rows from the 2 tables before that right?
      • dseomn_ joined the channel
      • dseomn has quit
      • dseomn_ is now known as dseomn
      • mara42 has quit
      • mara42 joined the channel
      • aerozol[m]
        Jade: sorry about the late reply, my week has been absolute chaos. Are you happy to meet and have a chat about the project on any of: Tuesday evening, Wednesday morning, Thursday evening, Friday morning
      • This assumes you are in Europe somewhere! I am in New Zealand so I have shifted my available times forward 12 hours, if you are somewhere else I can work out when I’m available in your times :D
      • FYI I am happy to video chat, but if you are more comfortable we can just have a text meeting
      • pranav: Liking the video of load more in action! Don’t forget to replace the 100% blacks - and let me know if you think the colours need tweaking
      • pranav[m]
        aerozol: yup will implement the colors today, wanted to get all elements first
      • aerozol[m]
        All good, no rush, just making sure you hadn’t forgotten. And curious to see what it looks like!
      • yvanzo[m]
        Hi yellowhatpro yes the transformation is still needed and that is the focus of this week.
      • yellowhatpro[m]
        Right, so that has to be done using postgres only right?
      • And it all has to be defined in the trigger itself
      • yvanzo[m]
        Not necessarily, triggers can just store table name and row id.
      • yellowhatpro[m]
        so does that mean, the transformations, or getting the URLs from the tables, have to be done somewhere between listen/notify
      • given we have the row ids and table names
      • yvanzo[m]
        what with listen/notify?
      • yellowhatpro[m]
        so the idea was to run notify on the internet_archive_urls table, and fetch x URLs, and send it to the listener, which will archive the URLs
      • yvanzo[m]
        it means that an intermediate table would be needed before getting to the *_urls table indeed.
      • yellowhatpro[m]
        yeah right
      • I was just thinking, is there any way we can just rip the URLs from edit* tables, while we are dumping them in internet_archive_urls?
      • But it seems a bit tricky, given the structure of edit_data, so it seems an intermediate table will be needed.
      • yvanzo[m]
        It would be tricky to implement URLs extraction using Postgres functions, and there would be a number of drawbacks: (1) It would run in our main Postgres instance and if something goes wrong it could be more tedious to turn off. (2) It wouldn’t be able to process external sources of data in the future.
      • yellowhatpro[m]
        But if we create a new table where we dump row ids and table names, we still need to somehow read this table and then make transformations, and then send the result to `internet_archive_urls` table.
      • Doesn
      • * But if we create a new table where we dump row ids and table names, we still need to somehow read this table and then make transformations, and then send the result to `internet_archive_urls` table.
      • Doesn't it seem like the polling task with one extra step?
      • yvanzo[m]
        Yes, and with the ability to handle non-linear rows.
      • yellowhatpro[m]
        ohh cool. (what's non-linear rows btw in this context ?? sorry for nub doubt 😅)
      • yvanzo[m]
        When an edit is added, it’s just a new row. When a row is updated, it’s row id stays the same. For example, if we want to archive external links from editor’s bio.
      • Polling would just consider new rows, not updated rows.
      • yellowhatpro[m]
        aaah i see
      • got the point
      • so it's pretty much the same ig, just a step added to the polling task.... (full message at <https://matrix.chatbrainz.org/_matrix/media/v3/...>)
      • Kladky has quit
      • Instead of polling, I would check if there is anything else we can do though
      • Kladky joined the channel
      • Jade[m]
        [@aerozol](https://matrix.to/#/@aerozol:matrix.org) Don't worry about it! 😄
      • Yeah, I'm in the UK - so 11 hours behind you. Tuesday evening sounds good to me?
      • And I'm happy to video chat - just send me the link
      • yvanzo[m]
        yellowhatpro: Don’t get too distracted with moving to triggers (in the long term), transformation(s) can be implemented/tested/documented separately already.
      • discordbrainz
        <12Aerozol> @jade: Sorry, I mixed up the times - I can do your Monday evening, anytime after 9pm, or Tuesday morning, 10-12am (I have a few things on sorry)
      • Jade[m]
        That's OK :)
      • Monday works for me
      • yvanzo[m]
      • Jade[m]
      • aerozol[m]
        Great, see you then! We accidentally picked the MB meeting I think, ok to stay on after?
      • Jade[m]
        Yep!
      • it's a few hours after I think?
      • SigHunter has quit
      • SigHunter joined the channel
      • fletchto99 has quit
      • fletchto99 joined the channel
      • SigHunter has quit
      • SigHunter joined the channel
      • SigHunter has quit
      • SigHunter joined the channel
      • discordbrainz
        <12Aerozol> Oh yeah sorry, a few hours after, it's in my calendar because that's when I take the notes. See you then 👌
      • atj[m]
        <mayhem[m]> "atj: now that that mini-spec..." <- not at this stage no, but i'm sure there will be later!
      • reosarevok[m] joined the channel
      • reosarevok[m]
      • Oh, by now it's just my local cache, it seems
      • Good. aerozol that's finally solved then it seems. Thanks for letting me know
      • yvanzo[m]
        Hi reosarevok , about that, we can update beta after having merged your PR.
      • reosarevok[m]
        Ok, let me quickly update it while I'm here :)
      • yvanzo: should we merge https://github.com/metabrainz/musicbrainz-serve... for now and later on review whether we want to extend it to any text attributes?
      • yvanzo[m]
        No need to rush it if there is an unresolved conversation.
      • rimskii[m]
        Hi aerozol !... (full message at <https://matrix.chatbrainz.org/_matrix/media/v3/...>)
      • also lucifer I almost finished apple music import integration !
      • reosarevok[m]
        yvanzo: ok, will update the chat one and that's it for now
      • mara42 has quit
      • yvanzo: https://github.com/metabrainz/musicbrainz-serve... should be fine now I think?
      • yvanzo[m]
        reosarevok: yup, I can lengthen variable names if you are fine with that too.
      • reosarevok[m]
        Go for it :)
      • mara42 joined the channel
      • lucifer[m]
        <rimskii[m]> "also lucifer I almost finished..." <- awesome 💯
      • mara42 has quit
      • mara42 joined the channel
      • yvanzo[m]
        <reosarevok[m]> "Go for it :)..." <- Pushed. You can’t review your own PR but you can give a feedback still. :)
      • mara42 has quit
      • mara42 joined the channel
      • pite joined the channel
      • pranav[m] uploaded an image: (1217KiB) < https://matrix.chatbrainz.org/_matrix/media/v3/download/matrix.org/AdGVOBoQHkFpuKlYxaJtINca/Screenshot%202024-06-14%20at%204.59.04%E2%80%AFPM.png >
      • pranav[m]
        aerozol[m] (IRC): here following means from user perspective or user perspective? For eg -> If I am viewing ur user page, shud i see follow or following if I have followed them or not or you?
      • mara42 has quit
      • reosarevok[m]
        yvanzo: lgtmbdnt 🚢
      • Sophist-UK joined the channel
      • pranav[m]
        s/[m]/:/, s/(IRC)//
      • BrainzGit
        [musicbrainz-server] 14yvanzo merged pull request #3287 (03master…enter-the-matrix): Update IRC references to (live) chat https://github.com/metabrainz/musicbrainz-serve...
      • yvanzo[m]
        Updating beta…
      • mara42 joined the channel
      • Done updating beta branch and servers.
      • minimal joined the channel
      • bitmap[m]
        <yellowhatpro[m]> "so it's pretty much the same ig,..." <- > <@yellowhatpro:matrix.org> so it's pretty much the same ig, just a step added to the polling task.... (full message at <https://matrix.chatbrainz.org/_matrix/media/v3/...>)
      • yellowhatpro[m]
        yeah right, extracting URL, filtering different URLs would be pain indeed.
      • I'll just create a different branch for it then and make it a side-quest
      • zas has quit
      • zas joined the channel
      • mara42 has quit
      • bitmap[m]
        sure, I'd suggest not introducing yet another intermediate table just so you can use triggers for edit_note though. polling is the simpler route in that case
      • mara42 joined the channel
      • pranav[m] uploaded an image: (1244KiB) < https://matrix.chatbrainz.org/_matrix/media/v3/download/matrix.org/WtXBDwECqwLBJrzAUVhCfOJg/Screenshot%202024-06-14%20at%207.43.08%E2%80%AFPM.png >
      • pranav[m]
        aerozol: Just one suggestion, could we use the existing similar user component which looks as above instead of the new one? Or should I change it everywhere?
      • yvanzo[m]
        bitmap: Can test.mb.o be updated after beta (+ MBS-13579) again or does it require any special attention when updating?
      • BrainzBot
        MBS-13579: Seeding release labels for an existing release is only rarely working https://tickets.metabrainz.org/browse/MBS-13579
      • yvanzo[m]
        (Or is the PR addressing this ticket ready to merge?)
      • bitmap[m]
        it's not ready but I haven't had time to fix it yet, you can feel free to overwrite test.mb
      • yvanzo[m]
        Thanks, will do.
      • pranav[m]
        lucifer: Could u lemme know the web socker URL i need to connect to for getting the listening now? I am server rejected WebSocket connection: HTTP 400.. I must be using the wrong url
      • lucifer[m]
        pranav: how are you trying to connect?
      • yvanzo[m]
        I’m keeping your current patch for that ticket on top of beta for the test branch.
      • pranav[m]
        I am using the asyncio and websocket package in python just to test the connection and events, here is the code... (full message at <https://matrix.chatbrainz.org/_matrix/media/v3/...>)
      • lucifer[m]
        well if there is a http 400, then the response body will have more info.
      • socket.io client would probably work out of the box, we have a socket.io server which supports websockets to be exact.
      • pranav[m]
        Alright, the url is correcr right?
      • Ill try using the socketio client
      • I have prev exp with that, all i can say is copilot sucks :(
      • lucifer[m]
        the url seems right yes.
      • you can see an example in the tests on how to do it, you should https://listenbrainz.org/socket.io/ as your url with socket.io
      • register a callback for listen events like this: https://github.com/metabrainz/listenbrainz-serv...
      • pranav[m]
        alright, thanks much appreciated
      • lucifer[m]
        no worries
      • pranav[m]
        lucifer: how do I get my musicbrainz id? to emit it on connection?
      • lucifer[m]
        pranav: its your username
      • pranav[m]
        But then, I am getting that the socket is connected, however I am not getting any events fired back