#metabrainz

Hey reosarevok, finally found the time to look at the changes to https://wiki.musicbrainz.org/About

It look good to me, though I made some further edits to simplify some of the language. I don't think all of it is, but some of the sentences started getting a bit long and "impressive" in a chatGPT-style (for instance, I removed a reference to all the "incredible music" in the database, since I take pride in MB containing a lot of really shit music as well!). But with those tweaks I think it's a great improvement.

Though this para needs a quick fact check from the team, please, to make sure nothing incorrect was added: "Most of the data in our database — any factual elements describing the who, what, when and where — is published using the Creative Commons Zero 1.0 Universal license, which puts it into the worldwide public domain and allows anyone to download the data and use it however they like. Remaining data is released under the terms

of the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 license, which makes it easy for everyone to access while ensuring that all our hard work doesn't become the story of yet another Gracenote."

lucifer: Hey, I was working to build a `test_get_era_activity` at `/listenbrainz_spark/stats/user/tests/test_user_stats.py` but I tried printing the messages from `messages = list(get_era_activity("all_time"))` but nothing is being printed for some reason, could you help me with the same

will take a look.

bitmap: left a question for you in MBS-14111 (not urgent)

MBS-14111: Show area anthems on area's overview https://tickets.metabrainz.org/browse/MBS-14111

<aerozol[m]> "Though this para needs a quick..." <- mayhem: can you double check if you're happy with this?

I like the disgracenote diss, but I am not sure we should put that on our home page.

and explaining what CC licenses are are fraught with peril. If our description is wrong, that could have legal consequences.

I would stick to the classic "released under CC licenses (links, links)" and be done

[listenbrainz-server] 14fettuccinae opened pull request #3333 (03master…new): Add notification_sender.py to use notification/send endpoint https://github.com/metabrainz/listenbrainz-server…

[metabrainz.org] 14mayhem merged pull request #523 (03metabrainz-notifications…bug-fix): Fix a server error bug when all notifications are marked as important. https://github.com/metabrainz/metabrainz.org/pull…

<mayhem[m]> "I would stick to the classic "..." <- aerozol: ^ can you reword it a bit? :)

holycow23: have you pushed the latest changes to the era activity PR?

lucifer monkey

The UI is also functional now

date select is also working and listens are being successfully imported!!!

i had merged the backend branch to frontend locally

now pushing on github

Just have to make sure the errors are shown as toast messages

and display import status and previous imports as well

<rayyan_seliya123> "hey lucifer just wanted to..." <- Hey lucifer: thx for merging! : I have seen you have reviewed this pr https://github.com/metabrainz/listenbrainz-server…

and suggested some changes that we have to keep it separate right ? Like wanted to ask in another pr or the one which is left for the review this one :>https://github.com/metabrainz/listenbrainz-server/pull/3316

And what about next how should I proceed further as queried you earlier !

rayyan_seliya123: the BP integration won't be merged until monkey is around. do you have any questions about it or is the PR pending review?

<lucifer[m]> "rayyan_seliya123: the BP..." <- Yeah no prblm about merging but wanted a review from u fir the api part and the store.py fiel I have made on that pr I am doing it properly or not and talking about further implementation wanted to ask where should I put my search bar Integration for IA metadata cache like we have to have a feature like other players like soundcloud apple or youtube like any recordings user want to

listen he will just search the track name or artist name or something on the official manual addition search bar and will have successfully submit the listen and can hear it am I right ?? So for this I am asking how to proceed

can you point me to what search bar you are referring to?

* Yeah no prblm lucifer: about merging but wanted a review from u for the api part and the store.py file I have made on that pr I am doing it properly or not and talking about further implementation wanted to ask where should I put my search bar Integration for IA metadata cache like we have to have a feature like other players like soundcloud apple or youtube like any recordings user want to listen he will just search the

track name or artist name or something on the official manual addition search bar and will have successfully submit the listen and can hear it am I right ?? So for this I am asking how to proceed

lucifer[m]: Wait

like i am asking where to place ia search logic in which folder like https://github.com/metabrainz/listenbrainz-server… or we have to add IA as a search source so i am confused where to put that logic

* lucifer: like i am asking where to place ia search logic in which folder like https://github.com/metabrainz/listenbrainz-server… and we have to add IA as a search source so nah , so i am confused where to put that logic

yes but where is the search bar in the UI that you are referring to?

lucifer[m]: like for other songs we are manually adding by searching thier name i was referring about that search box !

do you mean manual addition of listens?

lucifer[m]: Yeah it is the part of the project nah ?

that is completely different.

Ooh I think I am have misunderstanding then

that searches MB for tracks and not apple music/spotify etc.

What's the remaining part of my project then

lucifer[m]: Ooh !

when you play a song using BrainzPlayer, it searches for the track in various music sources. Spotify/Apple Music/Soundcloud etc.

You need to create an Internet Archive data source for BrainzPlayer and implement searching for tracks in that.

lucifer[m]: That part is done like I can now play recordings which matches the data logic of ia ! But I have done it manually using submit_listens.py !!

submit_listens is only for submitting the listens to ListenBrainz. it doesn't matter how the listen is submitted, if your brainzplayer logic is correct it should work for all listens that are available in IA metadata cache.

so sounds fine to me.

in that case, you can work on the backend side of things again.

add logic for searching more tracks for the given artists/creators using IA api and add those results to process queue.

the spotify/apple music cache already do this. so take a look at it on how to implement it.

So talking about the frontend thing it's this much only !! What I have done so far ??

i'll check the PR but seems to me like yes

lucifer[m]: Sure thx !

lucifer[m]: Like I should focus on getting more stream_urls for the respective artist/ creators ?

yes.

you need to find more metadata items from the artist and creator of existing items.

you should use a cache/dict to not query the same artists again and again.

lucifer[m]: Sure and what about the limit we have set for the seeding logic should I have to increase that also ?

the spotify cache does it automatically.

lucifer[m]: Like the notes or other fields we have done it earlier or only focus on extraction of stream_urls ?

lucifer[m]: Sure will follow this approach!

no.

use the IA search API with queries filtering on creator/author field to find new metadata items. add the items ids to the processing queue. then the crawler will process them normally.

for the seeding logic, we should change it to using the collections' rss feed.

or instead of a numerical limit we can use a datetime limit (last 7 days example).

read through the spotify metadata cache code, it handles all of these thing and will probably clear things up.

So for these backend refactor I jave to make new pr ??

yes

lucifer[m]: Sure will do it ! And yeah if u get some moments help me review that front-end pr too ! And asking about this page we have to add a text for internetarchive also ? like youtube has !

sure makes sense.

lucifer[m]: Okk will do this also

[musicbrainz-server] 14reosarevok opened pull request #3602 (03master…MBS-14084): MBS-14084: Sort event lists by newest first https://github.com/metabrainz/musicbrainz-server/…

MBS-14084: Sort event lists by newest first https://tickets.metabrainz.org/browse/MBS-14084

lucifer: Should i create new pr for fw test or continue in that only! i think except test its ready to merge! for matching of tracks was on my list now its also done and working fine!

one more thing db_conn parameter explicitly passed in all others services /db/.py i think i also have to pass will be used in test?

how to run the specific test file? ./test.sh and then filename?

same PR.

Yes that's the right way to run specific tests. And yes you need to pass db_conn in the same way.

Thanks!

Hello ! I try to to create an artist using seeding. I have my URL, i use GET for artist but i receive a login page. How to manage to go thru ?

in header, i wrote my application, the version and a mail but i'm still stuck with the login page when i send my request

kojitsu: you're required to login so that you can review and submit the changes

how to do it with in my code ?

s/with//

with seeding you're supposed to log in yourself and review each change before submission to ensure data quality

we don't allow fully-automated submissions except for approved [bot accounts](https://wiki.musicbrainz.org/Code_of_Co…

thanks, i'll working on bots

what kind of bot are you working on? i.e. what data are you planning to submit and where is it sourced from

artists and releases from discogs that doesnt exist in musicbrainz

ah, those are the kinds of submissions seeding is designed for. we wouldn't allow bot submissions for those, except perhaps in very narrow circumstances where the data is pre-vetted, since a human is required to check for duplicates and ensure the data follows our guidelines

lucifer: we dont need `write` scope for funkwhale? `read` only we need. i'm updating scoped to `read` which includes all like `read:profile` etc! removed manual addition in the beginning i was getting some error so used specific scopes now its working!