#metabrainz

reosarevok: the format is everything the function 'clean' does here: scheme, domain name, no subpath, no query, no fragment

I guess we can, I just assumed that's what tests are for :)

Tests are not about code, not live data :)

See other 'validate' functions, most check that already.

Yeah, but I thought they mostly did that as a way to get the path bit they want to check for the per-entity-type validation :)

<lucifer> "tandy: no, we just don't..." <- oh i see, my ci was failing so i was wondering why

<BANG>

It’s World Tourism Monday!

I received one mailed in review…

yyoung says:

"""

- Last week I worked on MBS-11901, it is now basically complete, but thetests are blocked by MBS-11990, so I started fixing that and submitted a patch.

- Also did regular editing on MB, added several releases.

- That's all, thank you.

MBS-11901: Add highlights to the links in external links editor https://tickets.metabrainz.org/browse/MBS-11901

MBS-11990: External links editor automatically merges auto-selected links even if it's a duplicate https://tickets.metabrainz.org/browse/MBS-11990

"""

Other for reviews: lucifer, Freso, zas, yvanzo, bitmap, monkey, ruaok, reosarevok, akshaaatt[m] , alastairp, CatQuest

lucifer: Go!

hi all!

i worked last on doing misc stats improvements. monkey had discovered some bugs in stats, which caused dupes. i fixed them temporarily. ruaok is working on getting better data in dumps for a permanent fix.

i also refactored some of stats code and worked on better aligning various stats with each other. plus also worked on de-duplicating some code in LB and spark to make adding newer stats easier.

also reviewed a few LB PRs.

that's it for me.

monkey: next?

Heyo !

Last week I finished three big PRs I was working on for ListenBrainz: user's Loved/Hated tracks page, refactoring the main component we use to display tracks, and submitting listens from BrainzPlayer when listen to music on LB.

The first two are deployed, the third one is in testing.

I also reviewed some PRs for LB and BB.

I took Friday mostly off but did find a bit of time to start working on the summit t-shirt. Apologies in advance if you don't get yours by summit time !

That was it for me last week! bitmap go !

last week I worked on MBS-11985 and MBS-11974 - the former is a pre-requisite to the newer relationship editor code

MBS-11985: Process seeded relationships on the server https://tickets.metabrainz.org/browse/MBS-11985

MBS-11974: Different dates of (P) relationships are wrongly grouped on release bottom credit display https://tickets.metabrainz.org/browse/MBS-11974

I also picked my PR for the rewritten timeline graph (https://github.com/metabrainz/musicbrainz-serve...) back up. started rebasing it and fixing issues + updating it to d3 7.0 which was released in the meantime. hope to push the changes there this week since I've neglected it for a while

other than that I mostly did code review for MBS

fin, go zas!

zas: Ping?

yvanzo: Go! in the meantime

Hi!

Others still up: Freso, zas, ruaok, reosarevok, akshaaatt[m] / akshat, alastairp, CatQuest – anyone else who want to give review, let me know. :)

Last week we released a new MB server and its docker-compose companion; See https://blog.metabrainz.org/2021/09/20/musicbra...

While addressing small issues, found a larger issue with errors in external links editor, the fix is under review.

Tried to limit disk I/O usage of MB dump containers to address disk latency alerts on aretha, didn’t change anything.

Then I worked on improving MB logs with timestamps to allow determining the exact cause of these alerts.

:(

(Spent time discussing logs handling with alastairp by the way.)

Also reviewed more PRs for next week’s release. (Decided to keep allowing music video links to either or release for now.)

(with reosarevok ^)

Fin, go reosarevok :)

Hi!

I worked on improving the style and clarity of our Perl code, plus a lot of small URL fixes, and looked into putting together an SQL style guide (first draft is at https://gist.github.com/reosarevok/2d8a2fa6310c...)

CatQuest: what about you?

Meow! `😺`

(oops, either +recording or release)

!m reosarevok (and others) for the style guide!

You're doing good work, reosarevok (and others) for the style guide!!

Working, adding releases, celebrating autumn equinox

did some testing of series on BB (will do more of this)

currently trying to figure out how to prepair herring (sild)

will be working on instruments later this week too

♍

fin, go alastairp ?

hullo

last week I was a few days in the office, which tends to turn into a lot of discussions, most of it useful

I spoke with monkey about brainzplayer listens reporting, ruaok about some data management and processing stuff, and with yvanzo about docker container logging once he arrived at the office too

I also continued work on upgrading the JS/TS/React in AB

one thing I didn't finish is reviews for lucifer, sorry lucifer :(

no hurry :)

Freso: up next?

o/

(Only zas, ruaok, and akshaaatt[m]/akshat left on my list. Last call for anyone else who want to give review!)

So.

I updated the staff on the forums, so I think everyone on https://metabrainz.org/team should have their title reflected there now.

I filed a couple of MBS spam related tickets (more on that later).

And other than that, dealt with various reports etc. and was around and about.

fin.

Still no sign of zas? :\

ruaok: Go!

hey

CatQuest: I didn’t change yours. :)

(zas is probably fine, he restarted sir earlier today)

last week was another week in the mbid mapping gulag, trying to make some minor improvements that are a lot of work.

I can hopefully finish that tomorrow, then return to the other task in the mapping.

and I took care of the usual background tasks, taking care of the foundation and stuff.

We've accepted into a short project with The ODI about open data stuff and we may end up getting paid 15k GBP, which is nice.

\=

:O

(that wasn't me)

was it a cat?

I'm kinda thinking about what we could do with that money, so if someone has a clear idea, let me know.

akshat: go!

Hi everyone!

This was an exciting week for me since I was focusing on making contributions.

I have done the proper setup for the dark mode of the MB interface and have quite a few major components ready. I love the design from my end so far.

(If no zas shows up, this is the last review, so last-last-last call for others who may want to give review!)

Thinking about a few more things and trying to follow the best practices as always.

I'm stuck in traffic, usual upgrades, maintenance, user support, Picard pr reviews, fin

lol

lol

don't text and drive, kids. :)

I also targeted some parts of LB this week and the design system for meb is also on its way for nice and clean usage.

XD

My college is finally slowing down with many things sorted and not much left in the semester to bug me. Hoping for refreshing work ahead!

Also, I will be available on IRC as akshat moving forward.

fin!

Already. zas phoned their review in, so I think that’s it for reviews. Thank you everyone who gave one!

[musicbrainz-server] 14reosarevok merged pull request #2257 (03master…MBS-11941): MBS-11941: Normalize Worldcat identities URLs https://github.com/metabrainz/musicbrainz-serve...

Oh, meh, sorry :)

We have three items on the agenda, so let’s get going!

reosarevok: dead link collection

Hi!

Well, I was asked to put together the steps for a bot dealing with broken links

What I have rn is https://gist.github.com/reosarevok/c155cef5d9a8...

Opinions welcome

Seems sensible to me.

oh there it is

reosarevok: thanks!

let me think about this tomorrow, when I am more awake.

looks pretty good at a glance

I think we should move on to a new topic -- giving feedback on this is beyond the moment.

Ok :)

Alright. I’ll leave in topic for now, but feel free to remove it if it gets resolved over the week. :)

It does seem sensible at a quick glance

Freso: MBS-11986

MBS-11986: New technical account flag: Spam(mer) https://tickets.metabrainz.org/browse/MBS-11986

o/

goals look good at least

I read it and I say I like it, freso

I’ve pretty much written my reasoning on the ticket, so I’m just going to copy/paste here.

I’m currently not deleting spam accounts since I’m hoping to feed them to SpamBrainz once SpamBrainz actually materialises. This means that currently I mark spam with "Editing/voting disabled", "Edit notes disabled", and "Untrusted" – the two former being dangerously close to "Banner message editor" (which probably deserves its own ticket, actually), so I’m wary of accidentally one day giving that permission to a

spammer.

Additionally, this approach means that the profile will still be viewable by users and search engines and they get to still use the account to log into forums, LB, CB, BB, or other sites using MB authentication.

This flag would ideally freeze a marked account so that only account admins are able to see the profile (I don’t know if we should return a 404 and pretend like it doesn’t exist, or a 403 and acknowledge it exists and just isn’t available currently; possibly greying out potential edit note links?) and that the account wouldn’t be able to log in (possibly giving an error message that their profile has been flagged as spam

and to mail support@ if they think this is a mistake?).

---

Does this seem reasonable to others or are there strong objections against this? Not sure how invasive new account flags are.

The idea sounds good to me, but not sure how we'd pass it further to LB, CB and whatnot too

it would probably need soem coding (making a n account invisible/etc?)

(since otherwise they'd still be logged in there, no?)

not a bad plan to collect training data for spambrainz. now we just nee someone to work on spambrainz

New flags are not necessarily hard IIRC. Making the code do different things for different flags, a bit more so, but should be doable

reosarevok: send the other projects a delete user message?

yes

ruaok: if we want to delete them from elsewhere, sure, why not

reosarevok: Yeah, but once logged out, they’d be prevented from logging back in. What I’m currently doing anyway is manually looking up whether the accounts exist on other sites and delete them there. I’d presumably still do this until we have a way to signal the other sites.