I mean depending on how PG plans the query, it could be quicker to do it per-edit, since we have more information (a specific editor ID and whether the flag is set)
reosarevok
Hmm
Ok
I thought something like "UPDATE editor SET privs = privs - $BEGINNER_FLAG WHERE member_since > NOW() - INTERVAL '2 weeks' AND EXISTS (SELECT 1 FROM edit WHERE edit.editor = editor.id AND edit.autoedit = 0 AND edit.status = ' . $STATUS_APPLIED . ' OFFSET 9)" would do them all in one go
And be a lot faster than sending a ton of queries
But maybe not?
(I just realized we're doing "EXISTS (SELECT 1 FROM edit WHERE edit.editor = editor.id AND edit.autoedit = 0 AND edit.status = ' . $STATUS_APPLIED . ' OFFSET 9) AS has_ten_accepted_edits" every time we load an editor, which seems... eh)
bitmap
I don't think it will be slow whether it's done in one query or many
just saying that even if the batch query only takes 1s, the edit check on its own (for a single editor ID) might take sub-1ms, and we probably aren't handling 1,000 beginners per hour
reosarevok
Well, we can try that then :)
bitmap
(the privs and member_since would already be loaded as part of the editor, so could be checked in Perl)
lucifer: I just logged into beta.lb without trouble, if that helps
lucifer
aerozol[m]: if you could do it on test.lb now, that would certainly help
actually wait a sec
yup ready now
aerozol[m]
No issues here
(just logged in again)
lucifer
thanks
aerozol[m]
lucifer: ansh: mayhem: monkey: isabelxxx posted some interesting background info and a github link to an existing ‘genre explorer’/map project that looks cool. If you’re interesting it’s on https://tickets.metabrainz.org/browse/LB-1509?
BrainzBot
LB-1509: Genre explorer
aerozol[m]
preeeettty
aerozol[m] uploaded an image: (888KiB) < https://matrix.moviebrainz.org/_matrix/media/v3/download/matrix.org/LmMQraihGDaMLhyqQtPGcexr/image.png >
aerozol[m] uploaded an image: (132KiB) < https://matrix.moviebrainz.org/_matrix/media/v3/download/matrix.org/qKjerqrgFjoZHpxxQXvSwZfB/image.png >
Also very cool to compare different databases (thanks @UltimateRiff for the find and post)
discordbrainz
<02UltimateRiff> one of the devs pinged me about it when they saw I created the ticket for the genre explorer~
i have set the threshold to 10 at the moment, i can do a run to generate the data tomorrow.
technmad
I have been going through issue tacker, I was looking for frontend related issue but had a harder time finding those. Is there a way to segregate those issues, any filter criteria in particular
bobbyflips joined the channel
bobbyflips has quit
ssam has quit
technmad has quit
aerozol[m]
lucifer: awesome, thanks for tackling this!
ssam joined the channel
technmad: did you have a project in mind? e.g. MusicBrainz, ListenBrainz, Picard, BookBrainz
discordbrainz
<09technmad> I'm currently looking into ListenBrainz