#metabrainz

reosarevok: got a sec?

mayhem: going to start making dinner soon, but if it's shortish I do

What is it?

doubt its shortish. :)

Well, ask and I'll see when I find the time? :)

Or if it's urgent, I can have dinner later 😅

I would like to have a query that fetches all of the artists and their origin country. skip all artists who do not have a country associated and resolve all areas that are more fine grained than country.

not urgent. but I suspect tricky.

Hmm

wait. dinner at 17h?

Dinner at 17:30 to 18 :)

still about 3 hours early.

Well, sure, we're still somewhat on an Estonian schedule :) But the EU lunch hours are actually 12:30 to 13:30, so it's almost Estonian

Anyway

Does your DB have the area_containment materialized table?

If so, it shouldn't actually be too hard

I'm working from an MB replica on wolf.

Assuming the replica has the materialized tables built it should be fine

it does

clue me what descendant, parent and depth are?

or write me a query.

wait.

sudo write me a query

lol

I can look into a query after dinner :)

ok.

thanks.

mayhem: yes, should be done soon. was interrupted by other stuff.

no rush, I just worked out a much better result format

>I would like to have a query that fetches all of the artists and their origin country. skip all artists who do not have a country associated and resolve all areas that are more fine grained than country.

hopefuly that won't screw you up too much.

we have this already no?

do we? where?

there's a labs api endpoint to get country by artist mbid

yea and that query is horrible and super slow. always times out.

I didn't know about that materialzed table. should make this much easier.

mayhem: ah cool, yes the table and we also fetch the entire data for spark so we have it cached there.

good good.

bitmap: hi! if i use `musicbrainz_ro` to access to the database on musicbrainz standby database that would ensure that i can't write to the database even by error, right?

lucifer: yes, it should only have select granted

great, thanks!

(you can also do SET TRANSACTION READ ONLY if you're worried about that)

makes sense

mayhem: do you still need the top level counts?

for how many recordings are there for a given tag.

[musicbrainz-server] 14yellowHatpro opened pull request #3238 (03master…MBS-13533-fixes-yellowhatpro): MBS-13533: Fix order of cover art in release group artwork https://github.com/metabrainz/musicbrainz-serve...

[musicbrainz-server] 14yvanzo opened pull request #3239 (03master…perl5dot38): MBS-13358 (II): Upgrade Perl version to 5.38 https://github.com/metabrainz/musicbrainz-serve...

mayhem: when you say "artist and country" do you mean names, MBIDs, both?

And when you say "resolve all areas that are more fine grained than country" it just means "give the country", right?

bitmap: does the consul service pgbouncer-slave exist under a new name?

*now

lucifer: no it should be the same

mayhem: also, when you say "origin country" do you mean "their country" or do you specifically care about "begin area"?

aerozol: hi! sorry i had figured too many issues last time to fix so i never scheduled the meeting for oauth stuff. but its working now, sort of and will you be available later today or tomorrow to meet and discuss some UI/UX stuff?

mayhem: I *think* this should do more or less what you want:

Although I guess if they have an artist area that is a country *and* part of a country, that'll be duplicated (Netherlands, Curaçao etc vs Kingdom of the Netherlands is probably the main one, but also Puerto Rico vs US IIRC)

bitmap: can you think of a better way to avoid that than just make the two queries separately and just say "if the artist has a direct country ignore the parent country"?

I guess you can left join on area again with type = 1 in the join condition, then use coalesce to prefer the direct country

bitmap, reosarevok: May I deploy perl 5.38 PR to the test branch for `test.mb.o`?

I'm not using test.mb - bitmap might be tho

bitmap: Please let me know if you think that it is needed to deploy it to test.mb.o because I already tested it locally, and I don’t expect much changes but emails are disabled on test.mb.o anyway.

*which are

yvanzo: I have the eaa branch running on test.db atm, but you may replace it if needed. (I don't see an urgent need if the tests all pass)

[musicbrainz-server] 14reosarevok opened pull request #3240 (03master…eslint-9): [WIP] Update to eslint 9.0.0 https://github.com/metabrainz/musicbrainz-serve...

reosarevok: should we have a refactoring ticket for https://github.com/metabrainz/musicbrainz-serve...

bitmap: I agree that test.mb.o may not be useful to this one.

I'll have to rebase the eaa branch on top of it once it's merged, anyway, so I could re-deploy things once I've done that.

[musicbrainz-server] 14yvanzo merged pull request #3239 (03master…perl5dot38): MBS-13358 (II), MBS-13546: Upgrade Perl version to 5.38 https://github.com/metabrainz/musicbrainz-serve...

mayhem: https://beta-api.listenbrainz.org/1/lb-radio/ta...

bitmap: possibly yes, especially given it's taking so many hours :)

lucifer: Hey! I am around today/tonight for a meeting. Just looking at a international planner tool atm...

Anytime between your 10.30 - 12.30am would work for me! Or feel free to ping me when you’re up and ready

Does anybody have strong feelings about reviews in CB that are the start of a review, and then ‘read more’ with a link to a review site? (looks like the same user wrote both, so no plagiarism)

It’s not against the rules, but we could consider adding something. I have no strong feelings on it, it seems fine - a bit annoying when links eventually die.

A user has started adding a bunch of them. Kind of nice that someone sees CB as visible enough to do so! Possible via LB