#metabrainz

[musicbrainz-server] 14reosarevok opened pull request #3192 (03master…MBS-12359): MBS-12359: Refactor _columns implementation in data model https://github.com/metabrainz/musicbrainz-serve...

[listenbrainz-server] 14rimma-kubanova opened pull request #2800 (03master…playlist): added import from Spotify https://github.com/metabrainz/listenbrainz-serv...

<BrainzGit> "[listenbrainz-server] rimma-..." <- Hi! I have finished the part with importing playlists (yay!) So there is no problem with functionality, but when it came to design I kinda gave up... I will work on it later

zas: the github actions issue is indeed because of the latest git for windows release. it no longer provides gettext binaries (which where considered internal anyway). it's a pitty, because it was a good source of up-to-date gettext binary builds for windows.

broke my windows dev machine now also by upgrading git :) I'll get this fixed. likely will use https://github.com/vslavik/gettext-tools-windows

[listenbrainz-server] 14lzzy12 opened pull request #2801 (03master…areapage): LB-1538: Added ability to add area from settings page https://github.com/metabrainz/listenbrainz-serv...

zas: https://github.com/metabrainz/picard/pull/2371

lucifer: monkey: I have pushed a PR implementing LB-1538. Could you please check when you get time

LB-1538: Set user location/area (e.g. for events) https://tickets.metabrainz.org/browse/LB-1538

Sure thing

Hello guys, nice to meet you all here!

I'd like to apply for GSoC program. According to the "Application Template" page, I should submit my application under "GSoCAppliction" tag, yet I don't see any new post from this year there. Can anyone point me to the correct place?

Thanks in advance!

akshaaatt: sounds good, so this means the bug will get fixed when the API gets updated?

Meanwhile is there anything else I can work on

ykc: you might be in the right place, I dont think we've received any apps yet.

yvanzo, bitmap: we should probably put a beta out on Monday to at least update translations, it's been a bit :) If there's anything I should review for that, let me know

reosarevok: sorry, I'll try to review stuff today. got a bit sidetracked by JSON dumps

I guess the AC editor still needs a bit more reviewing - yvanzo, did you have any time to check it? (I left some comments on https://github.com/metabrainz/musicbrainz-serve... but they're basically nitpicks)

Hi! No worries :)

maybe I can look into your comments there today too

How's the artwork indexer?

I think it's ready to deploy, as long as both of you are okay with the documentation and everything

Was that the last thing missing for EAA beta?

yes, though one other thing I was concerned about is that the musicbrainz-website-test contain kept getting OOM killed while it was on the eaa branch

container*

I thought perhaps it was the ssssss proxy I added (since that serves the images through Perl), but disabling it didn't seem to resolve things...

Any way to find out more?

I guess redeploy it and add some additional logging (how much memory Perl is using before & after each request that is served) and see if it happens again

What do we have on test now?

I guess the AC editor

If you could be bothered to let it do the logging while you look at other stuff anyway, that might not hurt?

yea, it will take a bit to set up the logging but I'll look into it today or tomorrow

Perfect

I'm guessing that means neither of the two big things will make beta on Monday but that's fine :)

At least they're both slowly progressing

well, I'll see how much work your AC editor comments, maybe we can put that in

https://github.com/metabrainz/artwork-indexer/b... this section is a bit confusing

So we need to send an insert and everything is default except for the specific release mbid, and it will index whatever is there?

Ideally we'd have a script to do this given an MBID, then :)

How common is this?

Similarly, for the failed events, I understand there's no retry option, each one needs to be retried by hand from psql?

My comments for the AC probably shouldn't be a lot, but yvanzo probably would want to review that before beta since it's big :)

Oh, yeah, ok, those things are specified in more detail under Hacking :)

"Failed events (any that encounter an exception during their execution) are tried up to 5 times" in succession or do we wait for a while in between to ensure the IA is not down? :)

mayhem: thank you very much! It feels great knowing I'm in the right place. :)

<07.zima_blue> thanks!

<07.zima_blue> Hello. I'm setting up my own MusicBrainz server and wondering about its Redis use for caching.

<07.zima_blue> If my app's server will call this MusicBrainz server, should I also implement my own caching, or is relying on MusicBrainz's Redis cache generally enough? Thanks!

reosarevok: we don't have to manually reindex stuff very often (the old CAA-indexer uses a similar command)

for retrying stuff it looks like it waits 30 minutes * 2 * the current number of attempts so far

Hi mayhem , monkey , aerozol , please update the LB app to 2.6.1 and help test listens submission and now listening!

Oh, that sounds fairly good then

Working perfectly for me mow after extensive testing on my side

the current CAA-indexer just waits 4 hours each time AFAIK

updating. wlll test on the way home

Thanks!

found a bug in one of the comments, heh

mayhem: you are absolutely going to LOVE the update if it works correctly for you. I have been on it for the last 1hr or so and it is sexy as hell

Listening now is dot on point

Plus everything including offline submissions after turning on the internet have been dealt with

hi! nice to see you

akshaaatt (IRC): The new version is working great for me so far. No dropped listens so far, and listening now is updating, and fast <3

Will keep an eye on it in the coming week to see if there's any issue, but it's looking very promising, thanks a lot to the whole Android team !

Ikr!!!

Kudos to jasje for this fix 🙌🔥

jasje: 🍪

fingers crossed

android shits when there’s constipation

so im still on lookout for bugs

feed’s swipe animation should be smoother as well now

so far, looks good!

reosarevok: not yet for AC editor, agreed on Beta Monday, will review some more PRs

Hey, I'm getting into the MJML-based email renderer project, and I'm confused between choosing mjml-react and the MRML Rust implementation. Could you guide me the right direction on which to go for?

soniii_: are you sure you have the right IRC channel/open source project?

soniii_: hey, basically mjml-react would let us use our existing React components to render some things in the emails, e.g., entity links. (I'm not sure what else, tbh.)

mayhem: it's regading https://wiki.musicbrainz.org/Development/Summer...

well, look there. lol

regarding*

:)

the ideas page notes that the rust version would be faster, but I don't think rendering emails will be a bottleneck for us, either, so...

yvanzo might have some more thoughts about the pros and cons of these since he proposed the idea

at the same time it would be nice if you could research them yourself and try to understand how they might fit in with our project

<07mostlyk> Can anyone help me setup a local env? So I created apps in both musicbrainz and crique, client id and everything is perfecct but when I go to localhost:9099/cb it first redirects me to the login page and now it's not even loading?

@ bitmap Thanks for the insights!

The point about leveraging existing React components does sound like a big plus for mjml-react.

The main cons I see with MRML (Rust) will be integrating it with our existing tech stack.

I will dive into both mjml-react and MRML a bit more to understand more about it!

Also, would love to get Yvanzo's view. Any tips on the best way to reach out? Also, is this right channel to discuss this project?

<07mostlyk> I think MRML would make it faster and more secure as well ig

Hi soniii_, it’s the right channel :)

Hey yvanzo!

Hi mostlyk, 9099 instead of 8200? do you follow? https://critiquebrainz.readthedocs.io/intro.html

hi .zima_blue, we also use nginx through openresty for caching.

bitmap: True that rendering isn’t the bottleneck, but processing subscriptions takes ~4h as a whole every day so even small improvements are welcome :)

[bookbrainz-site] 14aabbi15 opened pull request #1073 (03master…add_attributes): Add begin date and end date attribute to relationship https://github.com/metabrainz/bookbrainz-site/p...

bitmap: True also that there might not be so many React components to reuse with mjml-react since all the layout has to be written in MJML for proper rendering in mail clients.

<07.zima_blue> yvanzo: thanks for the info. i'm going to run some experiments to see what works for me

soniii_: Both options seemed to be doable, this is intentionally open to Rust which we would like to explore a bit but this isn’t mandatory.

yvanzo: true, I was wondering how much the react would actually help us (and how much time would be spent in rendering). the rust version could also be a good fit

the Atikamekw language has ISO 639-3 code atj 😁

mōrena!

Sorry to keep coming back to you with these outsidecontext. As you answer I am also learning! https://www.reddit.com/r/MusicBrainz/comments/1...

despite this, atj is easily understandable 😉

😂