#metabrainz

Last time I had a docker issue I ended up reading the source code to learn exactly what it was doing, because everything else was out of date

CatQuest he can be found on youtube

:)

huh, anyone here know what's up with http://flatpak.org/apps.html#picard - flatpak packaging for MusicBrainz Picard on linux?

If credentials is in .dockerignore, it shouldn't get added to the web container etc right?

Freso: me likes

But I see it in listenbrainz_web_1

CatQuest: https://www.youtube.com/watch?v=clTCMZIoVMQ

iliekcomputers: you may want to remove all running containers and then remove all the images.

the build from scratch.

;/ is mb slow for anyone else?

ok for me

yeait's better now

was very slow a little whie ago

is the musicbrainz-server torrent not working? or is it just me

ica: it should work

hmm

I think the tracker is down

it's always down. but DHT works

ah, not sure what that is

DHT?

distributed hash table

yeah

I'm getting it from ftp now but I guess I must have missed something on the download page

I'm not very good at this torrent thing I guess =P

bittorrent uses https://en.wikipedia.org/wiki/Kademlia as DHT protocol

there could a a few reasons why it wasn't working

¯\_(ツ)_/¯

not a big deal, there's still ftp ;)

yeah :)

[musicbrainz-server] mwiencek closed pull request #459: MBS-9135: Allow admins to modify usernames (master...mbs-9135) https://git.io/vDwQy

Yay :)

could I get some feedback on my initial rewrite of the search server README? https://github.com/zastai/search-server/blob/re... vs https://github.com/metabrainz/search-server/blo...

This still mentions the freedb indexes, but does make it clear that they're not required for MBS (saving people from wasting an hour if they're only setting up the search server for MBS)

!m Zastai

You're doing good work, Zastai!

:)

URLs can't seem to be searched? The search server indexes them, but the MBS URL controller doesn't handle search requests. Intentional or should I make a ticket?

yvanz: I think it's basically done (the spotify thing)

that sounds like something that would already be a ticket

reosarevok

if there's a ticket, I have a PR ready :)

:O

I know nextto nothing about these things, but already having read yours and the old readme I like yours better, it doesn't say explicitly "ubuntu" omg

well done Zastai, documentation work is such a underworked part of dev work. but such a needed thing

well i use debian, and i didn't want to just go s/ubuntu/debian/ :)

smart

and sounds more.. I dunno, what's the word...

professional

the big thing for me was marking the freedb stuff as optional. that download took a while and then turned out to be pointless. the rest was just cleanup (and markdownification)

tnha to go "well we use this or that system, if yo udon't blah blah"

:)

yes

I'd rather just dive in and rip out the freedb indexing entirely from the search server. but am not sure if it's intended for standalone use (which might mean there are people using the freedb search) or just as an MBS backend

Zastai: There is a ticket for URL searching, but I don't have it handy.

Zastai: Also, the current search server is pretty much abandonware, so don't expect to get any PRs merged. :/

(They *might* get merged, but please don't get your hopes up too high. :/)

Also, go ahead and rip out the freedb stuff from the search server. If people are using the freedb stuff outside of an MBS install, they've already wandered down their own path.

reosarevok: Does the BBC page need to be transcluded before the updates show up?

Uh...

Freso: yup. Sorry. Can you do it?

Or should I?

https://musicbrainz.org/admin/wikidoc shows "Error!" on the wiki versions.

"There was a problem accessing the wiki API"

legoktm: ^ maybe?

:)

(It still allowed updating the revision id, I just had to go to the wiki and fetch it myself.)

For CB, I want to use less files directly for developing , I linked main.less stylesheet and included less.js script.

But it didn't work. I tried it in firefox

how did you include the script?

and linked the stylesheet

href="../static/styles/main.less" and src="../static/scripts/less.js" in base.html

Zastai: Here is the ticket: https://tickets.metabrainz.org/browse/SEARCH-149

SEARCH-149: Allow searching for URLs

!m reosarevok

You're doing good work, reosarevok!

wut, why am i logged out of jira too :/

yvanz: how were you checking for what's left? nikki's searcher?

psolanki: that doesn't look right

gentlecat, anything wrong with the script part? I downloaded less.js and put it in scripts folder.

href="{{ url_for('static', filename='...') }}"

Freso: uhoh.

reosarevok: From my own server.

Freso: I don't see any API requests in the access.log, do you know where the code that makes API requests is?

yeah but that SEARCH-149 ticket was for url search in ws/1 - there was another one for url search in ws/2, which is closed (because the search server has implemented it). it's MBS that has a bug where it does not handle the requests and pass them on to the search server like it does for other entities. I made MBS-9261 for it now

SEARCH-149: Allow searching for URLs https://tickets.metabrainz.org/browse/SEARCH-149

MBS-9261: WS/2 does not allow a search for URLs https://tickets.metabrainz.org/browse/MBS-9261

also, wow - after 3 hours, the index creation for recordings is only at 77%

yerp, takes a long time.

if you don't need full indexes, you can run with -t flag where it builds 100,000 or so rows for each index

much faster, but only good for basic functionality testing

[musicbrainz-server] Zastai opened pull request #476: MBS-9261: Handle URL search requests. (master...mbs-9261) https://git.io/vyUXn

if the search server is indeed abandonware, what is the plan to replace it?

as it stands, the main issue I have with it is that its json output does not match that of other ws requests and I was planning to dig in and work on fixing that. but if there's basically no chance of PRs being merged, there's probably no point

Zastai: mbsssssssssssssssss is the planned sucesssssor.

live sql queries then?

Zastai: https://github.com/metabrainz/mbsssss

ah

SOLR-based

thought you'd gone al snake-y