Hello, I'm trying to install gaia in MacOS (Catalina 10.15.4) with homebrew. When I'm running "brew install gaia --HEAD" the following error occurs:
2020-05-11 13244, 2020
jmp_music
Error: No available formula with the name "python@2" (dependency of mtg/essentia/gaia)
2020-05-11 13207, 2020
reosarevok
alastairp: ^
2020-05-11 13210, 2020
jmp_music
Any solution available?
2020-05-11 13238, 2020
ruaok
jmp_music: I'm on mac as well, but I avoid installing anything on the mac is at all possible.
2020-05-11 13214, 2020
jmp_music
Ok. Should I build the gaia on Ubuntu for example?
2020-05-11 13225, 2020
ruaok
I mostly use docker and have all of my code running in linux containers, not on mac itself -- it keeps consuming hours of time that doesn't happening when using linux
2020-05-11 13242, 2020
ruaok
if you have a linux install available, then absolutely yes.
2020-05-11 13203, 2020
jmp_music
ruaok: I should try then your solution. Thanks a lot
2020-05-11 13213, 2020
ruaok
good luck.
2020-05-11 13213, 2020
diru1100
Morning!! (lil late 😅)
2020-05-11 13258, 2020
ruaok
it is always morning somewhere. :)
2020-05-11 13233, 2020
diru1100
True...
2020-05-11 13259, 2020
diru1100
yvanzo : I have been doing the api part, the current one is too complex. As Leo_Verto said I am stripping it down to bare minimum keeping only necessary parts.
2020-05-11 13258, 2020
iliekcomputers
Mr_Monkey: hey
2020-05-11 13205, 2020
Mr_Monkey
Yo !
2020-05-11 13227, 2020
iliekcomputers
if you have time today, could you look at ishaanshah[m]'s graph and see if there's some low hanging fruit stuff we can fix to make it look better :D
otherwise, the acousticbrainz webserver image also has gaia installed, you could just `./develop.sh bash` and then you've got a container with ipython and gaia available
2020-05-11 13247, 2020
reosarevok
Probably no point working on that before it's converted to React? I know bitmap was supposed to be on it
2020-05-11 13256, 2020
reosarevok
Also: thanks yvanzo, that link helped
2020-05-11 13220, 2020
reosarevok
I now have a less hacky looking thing... that still has the same D problem though :D
2020-05-11 13229, 2020
yvanzo
Ok, then maybe we can temporarily allow linking to release here, will add a comment.
2020-05-11 13246, 2020
yvanzo
reosarevok: congrats! :D
2020-05-11 13225, 2020
Chinmay3199 joined the channel
2020-05-11 13236, 2020
reosarevok
yvanzo: looking at the comments, maybe we should only have them at release then
2020-05-11 13206, 2020
reosarevok
The last comment kinda makes sense
2020-05-11 13212, 2020
yvanzo
reosarevok: temporarily, yes.
2020-05-11 13223, 2020
yvanzo
(just replied)
2020-05-11 13225, 2020
reosarevok
I mean, if they don't really fit our RGs
2020-05-11 13240, 2020
yvanzo
They do fit our RGs.
2020-05-11 13242, 2020
reosarevok
Something that goes on the RG level should match all potential releases in the RG
2020-05-11 13257, 2020
reosarevok
Does that work here?
2020-05-11 13207, 2020
reosarevok
Last comment suggested no, but I'll read it all
2020-05-11 13233, 2020
yvanzo
I feel it is more a presentation problem right now.
2020-05-11 13244, 2020
jmp_music has quit
2020-05-11 13218, 2020
reosarevok
I mostly want to make sure there will only be one page per RG
2020-05-11 13246, 2020
reosarevok
I don't see a reason to open it on release though then
2020-05-11 13247, 2020
yvanzo
There won't, there is no reason to.
2020-05-11 13210, 2020
reosarevok
Why make a special case for this and not, say, Wikidata?
2020-05-11 13216, 2020
yvanzo
(There won't be one page per RG.)
2020-05-11 13223, 2020
reosarevok
Can there be more than one?
2020-05-11 13229, 2020
yvanzo
reosarevok: Is that required for Wikidata?
2020-05-11 13229, 2020
reosarevok
If so, it should go on all the relevant releases
2020-05-11 13213, 2020
reosarevok
We don't have Wikidata on releases at all, so not sure what the question is :D
2020-05-11 13221, 2020
reosarevok
A RG-level rel should be relevant for the whole of the RG, not just part of it
2020-05-11 13246, 2020
yvanzo
reosarevok: the issue is 45* uses catalog numbers as keys, we are not keys for releases.
2020-05-11 13255, 2020
reosarevok
If it only applies to some releases in the RG then it should be on all releases that apply
2020-05-11 13201, 2020
reosarevok
If that's 10 then it's 10
2020-05-11 13209, 2020
yvanzo
We should make clear there is no confusion about catalog numbers and releases here.
2020-05-11 13259, 2020
yvanzo
< reosarevok> We don't have Wikidata on releases at all, so not sure what the question is :D < reosarevok> Why make a special case for this and not, say, Wikidata?
2020-05-11 13208, 2020
reosarevok
We have Wikidata on RG
2020-05-11 13218, 2020
reosarevok
We don't allow it on releases just because it's annoying to set it on RG
2020-05-11 13227, 2020
jmp_music joined the channel
2020-05-11 13227, 2020
yvanzo
Some databases have a model that better fit our model, some others don't.
2020-05-11 13229, 2020
reosarevok
So if this belongs on RG I'm not sure why we would allow that there
2020-05-11 13253, 2020
reosarevok
(and Wikidata gets used a lot more than this site)
2020-05-11 13253, 2020
yvanzo
We don't allow it on releases because they don't have anything that match releases?
2020-05-11 13213, 2020
reosarevok
That's why I'm saying, if we allow this on releases, then it's because it matches releases
2020-05-11 13216, 2020
reosarevok
And then it shouldn't be on RGs
2020-05-11 13223, 2020
reosarevok
If you have a page that applies to, say, ten neighbourhoods of London but not the others, you wouldn't set it on London and say it applies to London :)
2020-05-11 13248, 2020
reosarevok
So if we have a page that applies to ten releases in an RG but not the others, we shouldn't set it to the RG and say it applies to the RG
2020-05-11 13205, 2020
yvanzo
Please write a clear comment to the issue, it just becomes impossible to follow it here.
2020-05-11 13219, 2020
yvanzo
You lost me :D
2020-05-11 13224, 2020
reosarevok
Ok
2020-05-11 13231, 2020
reosarevok
Commented
2020-05-11 13237, 2020
yvanzo
Thanks!
2020-05-11 13205, 2020
ruaok
I've updated with https://test.listenbrainz.org with the latest data from yesterday. It should now be running totally independent of beta or production.
2020-05-11 13232, 2020
ruaok
it should be reasonably up to date -- some listens in the past couple of day are going to be missing.
2020-05-11 13248, 2020
ruaok
My fixes so far didn't take care of the zastai duplicate issue, I'll have to dig again.
2020-05-11 13202, 2020
ruaok
Freso: did you look and see if your duplicates were taken care of?
2020-05-11 13243, 2020
reosarevok
yvanzo: updated https://github.com/metabrainz/musicbrainz-server/… a bit, it's less hacky (it follows the suggestions from the react-table examples). Still having a flow error though. But does the general idea seem fine to you?
2020-05-11 13254, 2020
Zastai
i'd be happy to reimport, so if wiping all my last.fm listens is easy, feel free to do that
2020-05-11 13257, 2020
reosarevok
If so I could continue with other entities and try to fix the flow issue for all at once once we figure it out
2020-05-11 13219, 2020
Zastai
(just not my spotify listens, because I can't recover those)
2020-05-11 13223, 2020
reosarevok
(bitmap: I've asked about this one before, but maybe you have any idea why D is not compatible with D or we can try to figure it out? ^)
2020-05-11 13251, 2020
jmp_music has quit
2020-05-11 13254, 2020
ruaok
Zastai: it would be better if I used your listens as test to see that those dups are gone. that way they go away for everyone.
2020-05-11 13203, 2020
Zastai
fair enough
2020-05-11 13205, 2020
ruaok
I'll grep your data out and do a separate import.
2020-05-11 13246, 2020
Zastai
there was also a separate import issue, presumed to be caused by rate limiting at the last.fm side, where 1 or 2 out of 4000+ pages did not end up imported
2020-05-11 13227, 2020
ruaok
is there a ticket for this issue? because that is a totally different area from what I am working on.
2020-05-11 13228, 2020
Zastai
i don't recall. just remembered it. I _think_ iliekcomputers was going to look at it. when i'm done with work I'll check and create a ticket if needed
2020-05-11 13246, 2020
ruaok
puuurfect.
2020-05-11 13206, 2020
Zastai
(there was also a cosmetic issue with the import dialog, but I know I reported a ticket for that one)
2020-05-11 13227, 2020
ishaanshah[m]
Zastai: That one has been fixed
2020-05-11 13233, 2020
shivam-kapila
Zastai: The overflow one was fixed
2020-05-11 13236, 2020
Zastai
excellent
2020-05-11 13218, 2020
Zastai
!m ListenBrainz team :)
2020-05-11 13218, 2020
BrainzBot
You're doing good work, ListenBrainz team :)!
2020-05-11 13208, 2020
ishaanshah[m]
ruaok: Hi. I am getting permission denied on the recommendation engine doc
2020-05-11 13221, 2020
shivam-kapila
ruaok: I will get the promised loading comparison times after I am done with the Payoneer thing. Apologies for not being active for some days
2020-05-11 13225, 2020
shivam-kapila
ishaanshah[m]: Me too
2020-05-11 13241, 2020
ruaok
shivam-kapila: all good. just stop apologizing, please.
2020-05-11 13244, 2020
ishaanshah[m]
Can you give me read access, I wanted to have look
2020-05-11 13215, 2020
ruaok
back to everyone having comment access.
2020-05-11 13240, 2020
ruaok
seems you can't give everyone @meb write access and the rest of the world comment access.
2020-05-11 13247, 2020
ishaanshah[m]
got it, thanks :)
2020-05-11 13201, 2020
v6lur joined the channel
2020-05-11 13222, 2020
thomasross joined the channel
2020-05-11 13216, 2020
yvanzo
reosarevok: IMHO, 'getRowProps' should get data from 'data' not from a separate 'scores' array.
2020-05-11 13247, 2020
yvanzo
That is, each 'row' should have both 'score' and 'artist' in it.
2020-05-11 13239, 2020
yvanzo
'getRowProps' would just pick 'row.original.score'
2020-05-11 13231, 2020
reosarevok
yvanzo: then we have different data for ArtistList for search results vs other tables, seems like a bigger hack
2020-05-11 13245, 2020
yvanzo
we will have different data for reports and collections too
2020-05-11 13239, 2020
reosarevok
I was expecting to also pass that as a map
2020-05-11 13240, 2020
yvanzo
utility/tableColumns.js functions should have a parameter for accessor if that is not already the case
2020-05-11 13247, 2020
reosarevok
Since that's how we pass other things
2020-05-11 13252, 2020
reosarevok
Elsewhere already
2020-05-11 13235, 2020
thomasross has quit
2020-05-11 13221, 2020
reosarevok
But I guess I can look into that
2020-05-11 13232, 2020
yvanzo
reosarevok: we could add an optional accessor parameter to ArtistList for use with tableColumns functions.
