#metabrainz

/

      • iliekcomputers
        ++
      • 2020-03-30 09024, 2020

      • sweta05 joined the channel
      • 2020-03-30 09037, 2020

      • gr0uch0mars_ joined the channel
      • 2020-03-30 09036, 2020

      • bitmap_ joined the channel
      • 2020-03-30 09047, 2020

      • gr0uch0mars has quit
      • 2020-03-30 09047, 2020

      • adi5026[m] has quit
      • 2020-03-30 09048, 2020

      • bitmap has quit
      • 2020-03-30 09049, 2020

      • leonh has quit
      • 2020-03-30 09049, 2020

      • DjSlash has quit
      • 2020-03-30 09049, 2020

      • _flow_ has quit
      • 2020-03-30 09049, 2020

      • cloverfield has quit
      • 2020-03-30 09050, 2020

      • gr0uch0mars_ is now known as gr0uch0mars
      • 2020-03-30 09052, 2020

      • bitmap_ is now known as bitmap
      • 2020-03-30 09018, 2020

      • leonh joined the channel
      • 2020-03-30 09038, 2020

      • cloverfield joined the channel
      • 2020-03-30 09054, 2020

      • DjSlash joined the channel
      • 2020-03-30 09047, 2020

      • _flow_ joined the channel
      • 2020-03-30 09019, 2020

      • adi5026[m] joined the channel
      • 2020-03-30 09058, 2020

      • ruaok
        alastairp: can you sanity check me on this db connection business? I think you remember this much better than I do:
      • 2020-03-30 09001, 2020

      • ruaok
      • 2020-03-30 09012, 2020

      • ruaok
        from there until
      • 2020-03-30 09013, 2020

      • ruaok
      • 2020-03-30 09055, 2020

      • ruaok
        in this context simply assume that timescale == postgres, because it is.
      • 2020-03-30 09021, 2020

      • ruaok
        self.ts will then cache the connection to the timescale server. is that how we want to run the listenstore?
      • 2020-03-30 09006, 2020

      • alastairp
        the listenstore stuff has changed a lot over the years, so I'm not sure how much I remember
      • 2020-03-30 09021, 2020

      • alastairp
        I'm going to take a look at this page: https://docs.sqlalchemy.org/en/13/core/connection… follow along at home if you want
      • 2020-03-30 09036, 2020

      • ruaok
        where else could I follow along from??
      • 2020-03-30 09043, 2020

      • alastairp
        :(
      • 2020-03-30 09003, 2020

      • alastairp
        in what context is a listenstore used? inside flask when going API submit -> database, and also from database -> api response/website
      • 2020-03-30 09015, 2020

      • alastairp
        also in CLI scripts when reading/writing listens?
      • 2020-03-30 09022, 2020

      • ruaok
        yes, yes, yes.
      • 2020-03-30 09025, 2020

      • ruaok
        thanks for that link. yeah, I need to cache the engine object and make connections since those are pooled.
      • 2020-03-30 09040, 2020

      • alastairp
        so, we call listenbrainz.db.init_db_connection once on app startup
      • 2020-03-30 09006, 2020

      • ruaok
        yes.
      • 2020-03-30 09007, 2020

      • alastairp
        this makes db.engine an engine instance. from there, we use .connect() in each view when we need a handle to a db session
      • 2020-03-30 09019, 2020

      • ruaok
        (to be exact, listenbrainz.ts.init_db_connection)
      • 2020-03-30 09032, 2020

      • ruaok
        alastairp: agreed.
      • 2020-03-30 09051, 2020

      • ruaok
        aight, thanks for the sanity check. let me update the code, then send links again.
      • 2020-03-30 09009, 2020

      • alastairp
        so, if listenbrainz.ts.engine is an engine instance for connecting to the database, _perhaps_ we want to pass in that engine as an arg to TSListenstore.__init__ ?
      • 2020-03-30 09024, 2020

      • alastairp
        instead of calling create_engine again
      • 2020-03-30 09010, 2020

      • alastairp
        because you can call `with engine.connect()` as many times as you want within a single web request
      • 2020-03-30 09016, 2020

      • alastairp
        using the same engine
      • 2020-03-30 09003, 2020

      • ruaok
      • 2020-03-30 09027, 2020

      • Mr_Monkey
        Freso: There's a suspected spammer on Jira: https://tickets.metabrainz.org/secure/ViewProfile…
      • 2020-03-30 09027, 2020

      • Mr_Monkey
        Credit where credit due: first reported by chaban, I missed it then https://chatlogs.metabrainz.org/brainzbot/metabra…
      • 2020-03-30 09027, 2020

      • Mr_Monkey
        Is there a way we can revert all their changes in one fell swoop? (Maybe yvanzo the Jira expert would know ;) ?)
      • 2020-03-30 09028, 2020

      • alastairp
        oh right, it's just using the global,. that's probably OK
      • 2020-03-30 09048, 2020

      • ruaok
        yeah, this is what I had before but it just didn't feel right.
      • 2020-03-30 09012, 2020

      • ruaok
        this is how the DB connections are managed for the other PG tables, I'd guess so.
      • 2020-03-30 09019, 2020

      • ruaok
        k, thanks.
      • 2020-03-30 09028, 2020

      • outsidecontext
        zas: not really. Qt 5.14 had some changes to the networking, see https://doc.qt.io/qt-5/whatsnew514.html . so it might be related to this. but I don't understand how it should make a difference if the underlying network uses a VPN or not, AFAIK it just should just sit above the platform's networking stack. but I have no idea about Windows and VPN really
      • 2020-03-30 09048, 2020

      • alastairp
        yeah, perhaps from a design perspective it might be nice to have self.ts inside that class instance, but if we're just going to instantiate it by passing in listenbrainz.ts.engine, that's probably not changing much
      • 2020-03-30 09017, 2020

      • alastairp
        maybe this engine lifecycle is something to look at in the future, but for now I wouldn't sweat it
      • 2020-03-30 09041, 2020

      • zas
        outsidecontext: afaik, it seems related to connections/disconnections, and Qt *thinks* it isn't connected anymore or the like
      • 2020-03-30 09031, 2020

      • outsidecontext
        zas: that could make sense, it could be related to the new "network connectivity monitoring" then
      • 2020-03-30 09038, 2020

      • zas
        likely, I saw there are new env vars related to this, though it seems complicated to test, especially since it doesn't impact all platforms the same way
      • 2020-03-30 09052, 2020

      • ruaok
        alastairp: +1
      • 2020-03-30 09003, 2020

      • outsidecontext
        maybe it tries to connect to something to check the connectivity and that something is not available over the VPN.
      • 2020-03-30 09021, 2020

      • outsidecontext
        but I don't even know how to reproduce or test this :(
      • 2020-03-30 09007, 2020

      • rohitdandamudi
        yvanzo: Hello, I have shared the proposal. Can you please have a look at it. Thank you
      • 2020-03-30 09057, 2020

      • outsidecontext
        zas: I'm not sure how to deal with this for now. I can't find a fitting bug report on Qt, but I also don't know how to best report this without being able myself to reproduce it. Should we downgrade to Qt 5.13 on Windows? But I think this will introduce regressions regarding high DPI abilities
      • 2020-03-30 09030, 2020

      • byyz joined the channel
      • 2020-03-30 09016, 2020

      • byyz
        hey guys, quick question. If I link my spotify account for scrobbling, will my spotify username be publicly available anywhere?
      • 2020-03-30 09056, 2020

      • iliekcomputers
        byyz: no
      • 2020-03-30 09003, 2020

      • byyz
        oh cool
      • 2020-03-30 09006, 2020

      • byyz
        thanks
      • 2020-03-30 09007, 2020

      • iliekcomputers
        we use oauth and only store the client id and client secret
      • 2020-03-30 09026, 2020

      • byyz
        also if i linked my last.fm to import, its not constantly importing for there right? its just a one time thing?
      • 2020-03-30 09037, 2020

      • byyz
        or do i have to unlink it somehow to avoid double listens with last.fm/spotify?
      • 2020-03-30 09007, 2020

      • iliekcomputers
        yeah, last.fm is a one time thing, but spotify is ongoing. spotify only gives us a max of 50 of your latest listens, so you won't get many historical listens
      • 2020-03-30 09022, 2020

      • byyz
        Will the last 50 overlap?
      • 2020-03-30 09032, 2020

      • byyz
        cause I just imported from Last.fm a second ago
      • 2020-03-30 09018, 2020

      • iliekcomputers
        yeah, that could lead to duplicates in the last 50. there's no good way around this right now.
      • 2020-03-30 09024, 2020

      • byyz
        gotcha
      • 2020-03-30 09027, 2020

      • byyz
        thanks!
      • 2020-03-30 09016, 2020

      • iliekcomputers
        no worries! if you have any feedback / feature requests, feel free to ping us here or open a ticket at tickets.metabrainz.org
      • 2020-03-30 09035, 2020

      • iliekcomputers
        we're generally more responsive here
      • 2020-03-30 09017, 2020

      • yvanzo
        hi rohitdandamudi: shared where?
      • 2020-03-30 09059, 2020

      • prabal has quit
      • 2020-03-30 09032, 2020

      • rohitdandamudi
        yvanzo: I have directly shared the google doc.
      • 2020-03-30 09055, 2020

      • yvanzo
        can you please share the link?
      • 2020-03-30 09002, 2020

      • ruaok is deliciously deleting influx nonsense code
      • 2020-03-30 09034, 2020

      • yvanzo
        Mr_Monkey (cc Freso): no, but I blocked this account just in case.
      • 2020-03-30 09054, 2020

      • ruaok
        shivam-kapila: the function update_listen_counts(), which is being called from tests will no longer be needed. counts will be updated automaticaly. so you can remove calls to this function.
      • 2020-03-30 09035, 2020

      • Mr_Monkey
        Thanks yvanzo !
      • 2020-03-30 09049, 2020

      • rohitdandamudi
      • 2020-03-30 09027, 2020

      • yvanzo
        Mr_Monkey: actually there is a way to batch assign tickets but not to revert a change, so you have to check history for each ticket at first anyway.
      • 2020-03-30 09040, 2020

      • yvanzo
        rohitdandamudi: thanks!
      • 2020-03-30 09002, 2020

      • byyz has left the channel
      • 2020-03-30 09014, 2020

      • Cyna
        Hey Mr_Monkey, you up ?
      • 2020-03-30 09059, 2020

      • Cyna
        I need help setting up bookbrainz-site. The docker-compose doesn't work directly. Tried the ./develop.sh too. Even though its same. It doesn't work either
      • 2020-03-30 09009, 2020

      • Cyna
      • 2020-03-30 09036, 2020

      • Cyna
        I've first run `docker system prune -a` then `./develop.sh`
      • 2020-03-30 09029, 2020

      • zas
        outsidecontext: actually it'd be interesting to know if the issue appeared with 5.14 only, I don't think downgrading to 5.13 is the proper solution, but having 5.13 based packages could help diagnose the exact issue (atm, we aren't even sure it doesn't happen with previous versions of Qt)
      • 2020-03-30 09007, 2020

      • zas
        is it possible to provide such 5.13 based packages for debugging purposes?
      • 2020-03-30 09042, 2020

      • outsidecontext
        sure, we could do some build on the CI on a separate branch, or I build one locally
      • 2020-03-30 09010, 2020

      • Mr_Monkey
        Cyna: Hi! It looks like a permissions issue on the directory. Does docker need to be run with privileges?
      • 2020-03-30 09011, 2020

      • zas
        we could ask users who made those reports to test if it happens with previous Qt version
      • 2020-03-30 09020, 2020

      • Mr_Monkey
        Or maybe adjust the folder's permissions?
      • 2020-03-30 09030, 2020

      • Cyna
        Nah I dont think so
      • 2020-03-30 09033, 2020

      • zas
        if it doesn't we, at least, confirm it appeared with 5.14, and decide what to do
      • 2020-03-30 09046, 2020

      • Cyna
        I use docker without sudo everytime..
      • 2020-03-30 09029, 2020

      • Cyna
        I've tried runnig the script using sudo, it doesn't work either
      • 2020-03-30 09038, 2020

      • alastairp
        Cyna: "EISDIR error occurs when you try to open a file, but the path given is a directory."
      • 2020-03-30 09057, 2020

      • alastairp
        I'd check git status and see if you have any unexpected files in the directory
      • 2020-03-30 09012, 2020

      • Cyna
      • 2020-03-30 09018, 2020

      • Cyna
        Nope nothing...
      • 2020-03-30 09033, 2020

      • Cyna
        Origin -> bookbrainz/bookbrainz-site
      • 2020-03-30 09036, 2020

      • alastairp
        so, it seems that there's _something_ that is being read. It's unfortunate that the stacktrace doesn't include the file that it's trying to open
      • 2020-03-30 09042, 2020

      • Cyna
        Any idea what should I do to make it work ?
      • 2020-03-30 09055, 2020

      • travis-ci joined the channel
      • 2020-03-30 09055, 2020

      • travis-ci
        Project bookbrainz-data-js build #1414: passed in 1 min 50 sec: https://travis-ci.org/bookbrainz/bookbrainz-data-…
      • 2020-03-30 09055, 2020

      • travis-ci has left the channel
      • 2020-03-30 09006, 2020

      • shivam-kapila
        ruaok: Oh by coincidence I thought the same. I am almost done with basic setup. Just filtering some errors and the first test will be good to go. Please wait for some more time.
      • 2020-03-30 09056, 2020

      • mzfr[m] has left the channel
      • 2020-03-30 09010, 2020

      • ruaok
        I'll keep deleting influx code in the meantime.
      • 2020-03-30 09036, 2020

      • sweta05 has quit
      • 2020-03-30 09022, 2020

      • outsidecontext
        zas: I can prepare a build this evening
      • 2020-03-30 09006, 2020

      • iliekcomputers
        are you gonna delete the username escape code i spent days on
      • 2020-03-30 09023, 2020

      • iliekcomputers
        sad! :P
      • 2020-03-30 09026, 2020

      • iliekcomputers
        well, not really
      • 2020-03-30 09045, 2020

      • jwf has quit
      • 2020-03-30 09003, 2020

      • BestSteve has quit
      • 2020-03-30 09006, 2020

      • ShraddhaAg has quit
      • 2020-03-30 09006, 2020

      • leonardo has quit
      • 2020-03-30 09007, 2020

      • jwf joined the channel
      • 2020-03-30 09037, 2020

      • ShraddhaAg joined the channel
      • 2020-03-30 09039, 2020

      • jwf has quit
      • 2020-03-30 09040, 2020

      • ishaanshah[m]1 joined the channel
      • 2020-03-30 09056, 2020

      • jwf joined the channel
      • 2020-03-30 09058, 2020

      • ishaanshah[m] has quit
      • 2020-03-30 09011, 2020

      • leonardo joined the channel
      • 2020-03-30 09016, 2020

      • BestSteve joined the channel
      • 2020-03-30 09024, 2020

      • Freso
        zas: Thanks for helping ruaok ❤️
      • 2020-03-30 09010, 2020

      • Freso
        Mr_Monkey: I don’t know a way to revert them in one go. :( Thank you yvanzo for blocking them.
      • 2020-03-30 09020, 2020

      • leonardo has quit
      • 2020-03-30 09041, 2020

      • Cyna
        Mr_Monkey: What do you think is the best way to let users know if the properties are required or optional. In musicbrainz we use do it by making the property title bold and in BB its done by dividing into sections
      • 2020-03-30 09005, 2020

      • ishaanshah[m]1
        iliekcomputers: ping
      • 2020-03-30 09024, 2020

      • iliekcomputers
        pong
      • 2020-03-30 09002, 2020

      • ishaanshah[m]1
        I was working on rate limiting the MB api
      • 2020-03-30 09003, 2020

      • ishaanshah[m]1
        It seems that no custom headers are passed into Access-Control-Expose-Headers
      • 2020-03-30 09040, 2020

      • iliekcomputers
        i don't know what that means
      • 2020-03-30 09054, 2020

      • ishaanshah[m]1
        Hence I can't access X-RateLimit variables
      • 2020-03-30 09009, 2020

      • ishaanshah[m]1
        I am running into issues with CORS
      • 2020-03-30 09013, 2020

      • iliekcomputers
        ah
      • 2020-03-30 09029, 2020

      • iliekcomputers
        yeah, that sounds like something i should have envisioned :D
      • 2020-03-30 09043, 2020

      • iliekcomputers
        let's do this
      • 2020-03-30 09052, 2020

      • ishaanshah[m]1
        I was reading about it, if cors is enabled no custom header can be accesed
      • 2020-03-30 09058, 2020

      • iliekcomputers
        1. fix the links on the user page using if-else
      • 2020-03-30 09013, 2020

      • iliekcomputers
        2. look into fixing the track mbids on the backend