#metabrainz

lucifer: did that, in fact I even forked the repo again with another account, cloned it and set up a fresh env

if it doesn't then try sharing docker logs in a gist

*doesn't work

lucifer: Ok, thanks. Those things make sense only when we find these. 😉

haha yes :D

yvanzo: btw there is more weird stuff buried in mb-solr about this. eclipselink moxy is also a jaxb implementation.

pinkman_[m]: I added a comment on LB-1143 with what I envision, but let me know if you want to discuss in more details

but lets ignore it for now and first get the upgrades to prod. can cleanup later.

LB-1143: Import playlist manually on the front-end https://tickets.metabrainz.org/browse/LB-1143

Sure thing

[mmd-schema] 14yvanzo opened pull request #34 (03master…search-681): SEARCH-681 (Java binding): Regenerate after RNG change https://github.com/metabrainz/mmd-schema/pull/34

<lucifer> "if it doesn't then try sharing..." <- ```... (full message at <https://matrix.moviebrainz.org/_matrix/media/v3...>)

[mmd-schema] 14yvanzo opened pull request #35 (03search-681…improve-regen): Improve the regeneration of Java binding https://github.com/metabrainz/mmd-schema/pull/35

lucifer, reosarevok: Please let me know if this works for you too locally ^

s/```/``` lucifer /

s/```/````/

monkey: So is there any modal component we can reuse for the import playlist?

There is one you can base yourself off of, but not sure about reusing directly

Let me fish it out for you

This one here: https://github.com/metabrainz/listenbrainz-serv...

mayhem[m], bitmap, yvanzo, reosarevok: reminder that we have OAuth stuff meeting in ~3hrs from now.

@yvanzo I was able to generate 765 entities using sql-gen. For now, I just created simple enums for cube and point. These two were the only unknown types for me, and the ones that I may have to alter later. Now what should be the next step?

Should we work on postgres functions?

Also `select count(*) from information_schema.tables;` told there are 719 entities.

Hi , I had a doubt.. is the Dialogs function kinda like a snack bar management system?

jasje (IRC): ^

yellowhatpro: Isn’t it already providing some functions for reading/writing the database?

yellowhatpro: What can be the extra structs then?

pranavkonidena[m: kinda but not exactly

jasje (IRC): oh, ok ! Also how r we using the same fn name for the compostable widget and then calling it again inside the same fn? Isn’t that a bad practice?

Something like this ^

<pinkman_[m]> "```..." <- > <@matrixtest7103:matrix.org> ````... (full message at <https://matrix.moviebrainz.org/_matrix/media/v3...>)

Hi! duplicating my question

I'm working on LB-1488, and can't test how YouTube video works on BrainzPlayer. How can I enable the YouTube videos? I tried almost every song, but it gives me the error "Not enough info to search on Youtube

We are missing a track title, artist, or album name to search on Youtube"

LB-1488: Allow YT video to be re-sized https://tickets.metabrainz.org/browse/LB-1488

kubrimskii: i wonder if that error is misleading. do you have a youtube api key available locally?

also i would suggest to join using matrix so that you can stay connected.

Im still figuring out how to connect to matrix, since when i checked it said that it is no longer available for Libera.Chat (?? pls correct me if i am mistaken)

lucifer I am trying to play youtube videos on the website itself, not locally actually

Ok i will check it once again

kubrimskii: just create a matrix account, and then follow this link https://matrix.to/#/#metabrainz:moviebrainz.org

checking the youtube thing

oh thanks !

Also, an idea for improvement: we can change the music bar in BrainzPlayer to move more smoothly,

currently it moves every 2 seconds which looks strange

@yvanzo it did create functions related to schema. I was curious about the functions we get by executing`\df` in psql. https://usercontent.irccloud-cdn.com/file/4pZeO...

@yvanzo extra structs for?

lucifer: hey, do you have a link to your oauth document handy, and was there a particular agenda for today?

pranavkonidena[m: the function of that layer is to separate concerns

when a composable function instantiates a view-model while composing, it cannot render previews

for that reason we need to separate the view-model

Using the same name is not a bad practice as long as both of those functions don't mean something else relative to each other

thats why polymorphism

exists

yellowhatpro: SQL functions don’t seem to be needed in the binding, are they?

Ahh thanks a lot for explaining it !

yellowhatpro: The extra as in the difference between 765 and 719.

<kubrimskii> "Also, an idea for improvement..." <- @irc_lucifer:moviebrainz.org @irc_monkey:moviebrainz.org may i add this one to tickets ?

bitmap: https://docs.google.com/document/d/18ixQpjdjNh8...

rimskii[m]: monkey would know.

rimskii[m]: Yes, I think that would be a good ticket

ok I will add it then

and would like to do it by myself ahaha

rimskii[m]: easy! assign it to yourself after creating it :)

thanks lucifer

lucifer: any sources for mock data for https://github.com/metabrainz/listenbrainz-serv...

and its tests

oh shit

wrong link

jasje: i was reviewing it a few days ago but then forgot, will do and let you know.

@yvanzo, on checking again, I found that 719 are the total entities regardless of the schema, containing `event_art_archive`,`wikidocs`,`pg_catalog`, `cover_art_archive`, `musicbrainz`, `statistics`, `report`, `documentation`, `json_dump`, `information_schema`, `sitemaps`. I generated entities of musicbrainz type, which contains 382 tables.

sql-gen creates 2 files for each entity, one has the struct, the other one has methods

Should I generate schema for types other than musicbrainz as well?

fucking fish hasa godamn anime attac https://twitter.com/mischiefanimals/status/1753... lol

-> fish <--

bitmap, reosarevok: please give a try to https://github.com/metabrainz/mmd-schema/pull/35 whose purpose is to make it work for everyone 🤝

yessiree

yellowhatpro: It’s good to focus on `musicbrainz` schema (or even just a small part of it) for now, but binding some other schemas too can be considered later on.

yellowhatpro: Did you start thinking about how many hours you would like to dedicate to it? (90/175/350)

Ok musicbrainz schema it is then for now.

No not yet, I was currently exploring rust and databases, and this was a great plaything. I'll have to explore the project a bit more to tell the hours tho

is a*

Also it would be more interesting to build a small app on it rather than just having a binding that isn’t used at all.

An example app?

Yeah will try to work on it as well.

<11YoMo12> @lucifer I remember we discussed Listenbrainz matching for Hebrew. Can you refresh my memory what I was supposed to do to make sure they match. https://listenbrainz.org/user/frenchiejnr/ https://musicbrainz.org/recording/81920b2e-f7ca... You can see the last listens on this user that are not matched and they are in MB.

YoMo12: i think there is nothing that you can do at the moment, its more of a limitation of how LB automapper works.

bitmap, yvanzo, reosarevok, mayhem: hi! OAuth meet.

hi hi!

Hi!

hello

we have 3 items on agenda:

1. Do we migrate users before OAuth apps or vice versa or do we do both together?

2. Depending on the route we intend to take, discuss changes needed in the MB server.

3. How does local development adapt to OAuth changes?

starting off with 1.

yellowhatpro: Picking from other ideas, there are a lot of background tasks that can do: creating subscription emails, generating data reports…

eventually we want to have both user registration and oauth app registration in MeB.org (with the data being stored in MeB db as well).

we can do users first, oauth later or the reverse or both together. the doc lists the steps needed for each briefly.

migrating users will be an MB schema change, FWIW, so unless we plan to delay things, then probably migrate them after?

we could probably do it without schema change i think.

schema change yes, but those tables are not replicated, yes?

would need additional handling in the application layer.

editor is not replicated

we could migrate users by copying the data from MB to MeB DB and then making MB read the user data from MeB db if a certain flag is set?

so that you could do it independently of a schema change

yeah, that is an option

editor public data is not replicated but it is dumped still

(and exported)

I *think* and this is not a strong opinion, like option 2 better. The idea of having the lesser used projects pave the way for MB as well as having OAuth reviewed seems pretty smart.

but they will be in separate DB's, and MB makes use of many joins with the editor table which may read things like email

i see

if we do OAuth before users then MeB Oauth provider needs to updated to read from MB in the meanwhile

until the data is migrated from MB to MeB.

Yes, it is probably safer to do it gradually

which makes sense since MeB.org is not used by others like MB.org.

MB could also be a proxy to MeB OAuth for a while

that sounds fine to me but i am not sure if it comes with security implications

yvanzo[m]: that was something else I wanted to discuss

maybe we could ask this to the external security reviewers as well.

that would still be like jumping off a cliff, but with a safety net

mayhem: I think option 2 is better, too