#metabrainz

yes, thanks!

lgtm, added shivam-kapila to the contributors

oh, right. thanks!

wait, adding pristine__ too

update it when done.

👍🏽

done.

[listenbrainz-server] release v-2020-03-28.0 has been published by paramsingh: https://github.com/metabrainz/listenbrainz-serv...

[listenbrainz-server] release v-2020-03-28.1 has been published by paramsingh: https://github.com/metabrainz/listenbrainz-serv...

blog post twatted.

nice.

let's try to keep master and production as close as possible from now on.

yes, please.

thank you iliekcomputers !

!m

!m iliekcomputers

You're doing good work, iliekcomputers!

sigh. :)

related to that, is there public docs about the branching method used? :)

not yet, but I'm glad you'll be writing them up for us. :)

you're welcome

someone needs to tell me what that is first

and then, I'll just copy that

we don't plan to merge PRs until right before release, essentially.

get all PRs ready to merge. then on release, merge to master, merge master to production, deploy production?

yep. right iliekcomputers ?

which means that it's either a good idea to release often, or make sure PRs are small, or both?

critical part of this would be to have regular releases (possibly weekly)

especially when 2 PRs may be overlappin

otherwise it'll just be a bunch of prs based on each other

if the idea is to release regularly, maybe merging to master isn't a bad idea - at least it means there'll be a centralised branch which has _all_ changes on it at once, for testing etc

something to deploy to beta, etc

(just throwing ideas out)

alastairp: yeah, i think that makes more sense

but agreed that it depends strongly on keeping discipline for regular releases

iliekcomputers: Thanks for that :D.

shivam-kapila: you pinged last night?

New release. Yay! 🎉

alastairp: Hi. Yes. Actually regarding that PSQL login. I found a fix. Can you have a look at it? I will share it as a pastebin and then push the commit

shivam-kapila: use the PR, that's what we have it for

it's much easier to keep the discussion all in one place

shivam-kapila: can you please change this PR https://github.com/metabrainz/listenbrainz-serv... to not merge into master, but into timescale ?

alastairp: Sure. Please allow me a sec.

shivam-kapila: I'm in no hurry, you'll have to ask iliekcomputers to explain how I review PRs on listenbrainz ;)

ruaok: Ok I will. But the `init_msb_db` and `innt_db` exist in master too.

alastairp:

alastairp: I will surely like to know :p

ruaok: https://github.com/metabrainz/listenbrainz-serv...

ruaok: Base modified

shivam-kapila: yes, but for now just the timescale branch should be sufficient.

alastairp: The PR is updated. Its waiting for you. :)

iliekcomputers: uhm.

ruaok: Modified to timescale branch :)

th

thx

shivam-kapila: it's easy. first I ignore it completely, and wait until it's been approved and merged. and then I review it, asking for a number of changes, and so the author has to do more work

iliekcomputers: I dont think we've moved that yet. we moved the android package.

apparently it's become a running joke now

:D

s/become/been/

ruaok: we can merge the pr then, right?

alastairp: Well perhaps GitHub will like to honour you for your approach :p

iliekcomputers: sure.

cool cool cool

[listenbrainz-server] paramsingh merged pull request #753 (master…master): added vlc-listenbrainz-plugin to https://listenbrainz.org/api-docs https://github.com/metabrainz/listenbrainz-serv...

I can fix that.

thank you. :)

fixed.

ruaok: Welcome :)

ruaok: Just out curiousity when will you review the proposals for a final push?

when I feel well enough to do so.

please stop pestering me, ok?

oohh. is the new lb update on prod or beta?

prod

CatQuest: prod

:OO

the last.fm importer definitely needs some tester love

ruaok: Oh sorry for that. Take care of your health.

ok first thing: it now shows in the middle of the page and not the bottom \o/

it's responisive to resising the window, good good

aand the import worked!

!m well done ListenBrainz devs! 👏

You're doing good work, well done ListenBrainz devs! 👏!

oh. guess what i'm 33106 listens off 1,5M! so then I thought about stuff and here's an idea: how about a timeline for listens? it could easily be calculated with play count-numbers. so it could be like "months" and a graph how many accumulated listens. or you could do "year" view also

what do you mean by timeline?

so then like, I could see when likly I'll have listened to 33k tracks

oh i meant a grap with x time and y listens

sorry, english

that does sound cool

:D

although i guess it would be relatively linear for most people

i think ishaanshah[m] has something similar-ish in his proposal

possibly linear yes. but the "angle" of that linear is interesting to me. tbh

ooohhhh

CatQuest: https://www.figma.com/proto/qXJTaGsrjmzBFVztScy...

the first graph in this

and of course you could have non-accumulated grap as well. wich would not be linear

.. could you do a screen shot or soemthing? it wont show here. not evne chrome

I can't update chrome any more, it wont let me :(

it just redirects to some "unsuported browser thing

ishaanshah[m]: yeh, that looks like a "how many listens this day" (non accumulated) grap!

it'd be possible ot do accumulated too, rght?

ishaanshah[m]: thank you :)

> it'd be possible ot do accumulated too, rght?

it is, but I am not sure how usefull will it be

personally I'd love that. it'd show me how fast my "listen" count increases

and, sure possibly it'll be lineary

but maybe not, and i'd care about the liklyhood of it letting me predict certain lsiten milestones

iliekcomputers: interesting error on sentry

listen*

!m iliekcomputers

You're doing good work, iliekcomputers!

ishaanshah[m]: for your modeling work, you are free to use my user to test and model your stat data: https://listenbrainz.org/user/CatCat ()

CatQuest: Thanks a lot, I wanted a user with a lot of listens

🙋 :D

ruaok: https://sentry.metabrainz.org/metabrainz/listen... ?

oh the spotify reader one?

ruaok: Hi, this might be a silly question but can we get musicbrainz database on lb

we can get read access, yes

Oh direct read access, not through api

it's hard to develop for because you have to set up an entire MB environment locally

what's your use case here?

In my proposal I have stats which require data to be obtained from MB

oh jeez. it i devide the number of listens minus my listens on the number of users minus me it's 37 730 listens

Artist Origin and Top Genres specifically

> here's another stat idea: amount listens leaderboard

That's actually really cool

ishaanshah[m]: i'd prefer to do it via the API.

i don't want to add musicbrainz as a dependency unless absolutely necessary.

iliekcomputers: about that, I was writing the code for the trackMBID issue that I described before

However api requests slow down the process significantly

Not to mention rate limiting

hmmm

As we have get such kind of data for all users, I thought direct access would be better

we can't really add an exception for the rate limiting because the importer works on the client.

>trackMBID

is that a fix to the problem with lastfm imported listens using a bugged recording mbid? (it uses a track mbid but website assumes recording mbid, so every single link is error)

CatQuest: yep

ishaanshah[m]: AWESOME

CatQuest: is every single link incorrect?

iliekcomputers: yes

👏 👏 👏

iliekcomputers: yes

test any noe https://listenbrainz.org/user/CatCat

one*

the quickfix on the frontend would be to change the links