#metabrainz

bitmap: the sample reindex completed fine. i am going to run a full reindex on trille now.

[musicbrainz-server] 14mwiencek opened pull request #3499 (03master…fast-ref-to-type): Speed up `Data::Utils::ref_to_type` https://github.com/metabrainz/musicbrainz-serve...

Tickets will go offline for security update in half an hour from now.

Hey everyone, I’m planning to apply for GSoC and working on my proposal for **new statistics graphs** in ListenBrainz. I’d love any feedback or guidance to improve it. Looking forward to contributing and learning! [Here](https://community.metabrainz.org/t/gsoc-2025-development-of-advanced-user-statistics-visualizations/750040) is my draft proposal.

* Hey everyone, I’m planning to apply for GSoC and working on my proposal for **new statistics graphs** in ListenBrainz. I’d love any feedback or guidance to improve it. Looking forward to contributing and learning! [Here](https://community.metabrainz.org/t/gsoc-2025-development-of-advanced-user-statistics-visualizations/750040) is my draft proposal.

ansh lucifer

* Hey everyone, I’m planning to apply for GSoC and working on my proposal for **new statistics graphs** in **ListenBrainz**. I’d love any feedback or guidance to improve it. Looking forward to contributing and learning! [Here](https://community.metabrainz.org/t/gsoc-2025-development-of-advanced-user-statistics-visualizations/750040) is my draft proposal.

ansh lucifer

Welcome holycow23 and thank you for your proposal. Don’t worry about proposed mentors not being online right now, they will catch it later on.

* Hey everyone! I’m planning to apply for GSoC and working on my proposal for **new statistical graphs** in **ListenBrainz**. I’ve previously contributed by developing a couple of new features and would love any feedback or guidance to refine my proposal. Looking forward to contributing and learning!

You can check out my draft proposal here.

ansh lucifer

* Hey everyone! I’m planning to apply for GSoC and working on my proposal for **new statistical graphs** in **ListenBrainz**. I’ve previously contributed by developing a couple of new features and would love any feedback or guidance to refine my proposal. Looking forward to contributing and learning!

You can check out my draft proposal [here](https://community.metabrainz.org/t/gsoc-2025-development-of-advanced-user-statistics-visualizations/750040).

ansh lucifer

* Hey everyone! I’m planning to apply for GSoC and working on my proposal for **new statistical graphs** in **ListenBrainz**. I’ve previously contributed by developing a couple of new features in ListenBrainz and would love any feedback or guidance to improve my proposal. Looking forward to contributing and learning!

Check out my draft proposal [here](https://community.metabrainz.org/t/gsoc-2025-development-of-advanced-user-statistics-visualizations/750040).

ansh lucifer

holycow23[m]: stop that

holycow23[m]: sorry for sounding rude. Your edits are spamming the IRC room. And I think your message is clear.

Sorry for causing the incovenience

No worry, you couldn’t have know, that was rude indeed. Thanks for not over-editing though.

yeah, I should've worded that better

I always forget edits send new messages on IRC 😅 good reminder to try to not use them often

reosarevok: Should the topic be updated to clear the agenda?

Yes, thanks!

Tickets are back.

Aww, I was liking the bike

But thanks :)

Remade my pending PR on Melba.

Although is it normal that there's both a Library and binary target? I find it difficult to justify the presence of a library in there, as it's tailor made for MusicBrainz.

We could extract the IA archiving and internal database into it's own library, but the rest doesn't really fit in

[metabrainz.org] 14mayhem merged pull request #503 (03master…master): Update team.html https://github.com/metabrainz/metabrainz.org/pu...

Hi rustynova, IIRC there was one to target to be used as a daemon and another as a CLI tool.

Also the longer plan is to make it work for any other project (BB…) not just MB.

(But we’re not there yet.)

Oh. For daemons, you still need a binary to act as an actual runnable code. Or else it's just raw rust

As for expanding, it's useless. The code is already in the binary crate, and not calling the library part.

I see the intent, just that it wasn't properly executed. For now i'd suggest to act as if the crate is a library, and when we want to expend, we can extract the core part (database and IA interface) and make multiple children binaries with the custom code for that project

That question was mostly to see if I could update some dependencies/rust version to remove some now useless code.

Libraries should try to not use the latest stuff, but binaries absolutely should

I’m not sure it is really need to have two separate binaries either, but at least you got the idea.

Absolutely, that is very welcome!

We'll see about it when we need it then!

Also migrate to the 2024 edition. With junior code there might have some things getting broken

bitmap: I just noticed that we sort RGs by secondary type id in `_find_by_artist_fast` but by secondary type name in `_find_by_artist_slow`.

I assume we'd want to change both to use child_order with the schema change - although child order is 0 for all secondary types right now AFAICT

Which is also the only reason the list is in alphabetical order when editing / adding RGs

(since that does sort on child_order before name)

Two questions: 1) I assume not, but is the different sort intended in any way? 2) Should we be setting child orders for the types, and if so should we start with alphabetical and keep the same sort we have the form, or id and keep the sort we have in _find_by_artist_fast?

[troi-recommendation-playground] 14mayhem merged pull request #167 (03main…fix-unresolved-releases): Minor cleanup https://github.com/metabrainz/troi-recommendati...

<rustynova[m]> "We'll see about it when we..." <- > <@_discord_492790468321607681:chatbrainz.org> We'll see about it when we need it then!

> Also migrate to the 2024 edition. With junior code there might have some things getting broken

maybe shape it as a library alone but with a (hopefully MeB-project-agnostic) API?

> <@_discord_492790468321607681:chatbrainz.org> We'll see about it when we need it then!

> Also migrate to the 2024 edition. With junior code there might have some things getting broken

* maybe shape it as a daemon alone but with a (hopefully MeB-project-agnostic) API?

mayhem: do you have any suggestion on how to implement creator options for notification senders? also, we barely send out mails to users in listenbrainz but we do change their timeline (for example, if someone just got a track recommended) ,so should i add sending mails when we change their timeline in my project?

what do you mean by "creator options for notification senders". User settings wrt to their notifications preferences?

and we dont send a lot of emails because the notification system is missing. :)

mayhem[m]: no, im talking about settings for notifications like marking it important

fettuccinae[m]: I still dont understand

mayhem[m]: lets take yim notifs for example, when we send it out, we do need to set an expiry age and also mark it as important.

lucifer: " * The seed artist Radiohead has no similar artists, nor top recordings. Too niche?"

this is regarding the regression yesterday. has something happened to the similar artist data?

I'll fix the bug for when there are no similar artists, but we need to check the similar artist data.

fettuccinae[m]: this needs to done from maintainers side ,setting the notification as important or not, setting its expiry date and all.

fettuccinae[m]: and I still don't understand what you're asking.

mayhem: i'll take a look at similarity data and see if something is broken

fettuccinae[m]: ahhh, I finally understand. these options are provided to the notification system when sending a notifcaition.

mayhem: i think @fettuccinae:matrix.org is talking about notification priority

lucifer[m]: now I am extra confused, lol. :)

mayhem[m]: yess, how do u suggest i implement it

as parameters to the function call.

mayhem[m]: one of the option we choosein notification system is its priority

* one of the option we choose in notification system is its priority

@fettuccinae:matrix.org: yeah i think we'll have a common function that will receive those options in addition to the notification content and then take action appropriately.

and each option in particular can just be hard coded at the place you are calling the notification helper function.

bitmap: also, we don't do `NULLS FIRST` for `_slow`, not sure if there's a reason for that?

bitmap: updated https://github.com/metabrainz/musicbrainz-serve... a bit for now, anyway, please see if the changes so far + the listed tasks seem sensible and whether I seem to be forgetting something :)

<lucifer[m]> "and each option in particular..." <- thanks for handling that. :) #soconfused

<reosarevok[m]> "Two questions: 1) I assume not..." <- setting child orders makes sense to me, there's a ticket for that - MBS-13790. I don't think I bothered changing the slow function because it's only used outside of production, but we could make them consistent

MBS-13790: Give secondary release group types a meaningful order https://tickets.metabrainz.org/browse/MBS-13790

mayhem: I have added timeline for the project, can you review it?https://docs.google.com/document/d/1xHfzNdS8...

why do you have two lists now? combined both into one -- phases and dates, please.

two lists?

fettuccinae[m]: see comments in doc.

mayhem[m]: ohh, i went into more details on how im going to implement things in list 2 with dates.

fettuccinae[m]: yes, that is confusing. make a second with extra details on implementation, but keep the timeline clear

mayhem[m]: on it

lucifer: https://api.listenbrainz.org/1/lb-radio/artist/... returns {}.

similar artist data seems fine, but the data for lb-radio artist endpoint is clearly borked.

mayhem: https://docs.google.com/document/d/1Fv9ys0y8bxL...

<fettuccinae[m]> "mayhem: https://docs.google.com..."; <- I still see two lists.

mayhem[m]: can you verify if the document name is Project Timeline?

fettuccinae[m]: why the new document? now you lost the history from the old doc.

mayhem[m]: i still have the old doc, i just copy pasted the second list into a new doc to make it more clear.

I think you having exams all month of July is a real problem. when students have conflicts like this, they tend to fail.

at most I would consider you for a shorter projects, but even that is fraught with peril.

mayhem[m]: almost half of the work is done before july, only integration into listenbrainz and musicbraniz remain

fettuccinae[m]: after july, i get a break for in august where i can contribute more and i can start coding in may itself.

then almost half of the work is in danger of not being completed.

I've had so many of these "it will all work out, I promise" go bad, that I am not interesting mentoring you, I'm sorry.

fettuccinae[m]: im free whole of august as we get a month off after our exams

you can see if @lucifer is interested in taking you on.

<mayhem[m]> "you can see if @lucifer is..." <- can lucifer mentor me for this project?

that's up to him, but in theory he could.

monkey[m]: around?