In case we needed more reason to avoid doing our own copyright checkups like the plague: Commons, which is obviously a bigger org, has FIFTY DAYS of backlog for checkups
2017-10-10 28331, 2017
samj1912
ruaok: managed to get the complete search setup running on my localhost, even fed in the data :)
2017-10-10 28306, 2017
reosarevok
samj1912: did you use docker for all in the end or is it a 100% local install? (if the latter, any difficulties that we need to clarify in the docs?)
2017-10-10 28316, 2017
samj1912
100% local install
2017-10-10 28342, 2017
samj1912
reosarevok: yup, lots of little dependencies here and there, although I am not sure if thats due to my machine/os
2017-10-10 28359, 2017
reosarevok
If you're up for improving the docs a bit, that wouldn't hurt ;)
2017-10-10 28312, 2017
reosarevok
(if you aren't sure if it'll happen to everyone, you can always say "you might need X" :) )
2017-10-10 28317, 2017
samj1912
yeah
2017-10-10 28352, 2017
samj1912
hmm, work index is failing
2017-10-10 28335, 2017
lazka has quit
2017-10-10 28307, 2017
samj1912
umm where can I find the mbdata package?
2017-10-10 28330, 2017
samj1912
ah, nvm, found it
2017-10-10 28300, 2017
samj1912
Mineo: AttributeError: 'CustomWork' object has no attribute 'language' is this supposed to be present? I cant find the language attribute in either mbdata or sir models
2017-10-10 28308, 2017
samj1912
there is a language relationship list
2017-10-10 28313, 2017
samj1912
in CustomWork
2017-10-10 28307, 2017
reosarevok
samj1912: we recently changed work language to be work language*s*
2017-10-10 28312, 2017
reosarevok
So that probably just needs updating
2017-10-10 28317, 2017
samj1912
ah
2017-10-10 28318, 2017
samj1912
gotcha
2017-10-10 28335, 2017
samj1912
so how do we determine the work langauges then?
2017-10-10 28340, 2017
samj1912
we just return it as a list?
2017-10-10 28324, 2017
samj1912
also reosarevok can you link me to the schema change ticket?
samj1912 doesnit think it is as it adds a function to a list and returns nothing
2017-10-10 28302, 2017
samj1912
oh, the list comprehension is for just a throwaway
2017-10-10 28309, 2017
samj1912
it should just return then I guess
2017-10-10 28314, 2017
samj1912 will fix that
2017-10-10 28352, 2017
github joined the channel
2017-10-10 28352, 2017
github
[sir] samj1912 opened pull request #29: SOLR-51: Fix missing return type_list on function call (master...solr51) https://git.io/vdVJL
2017-10-10 28352, 2017
github has left the channel
2017-10-10 28308, 2017
Dalek_ joined the channel
2017-10-10 28301, 2017
ruaok looks at bb
2017-10-10 28309, 2017
ruaok
ping, no ssh. :(
2017-10-10 28322, 2017
Dalek_ has quit
2017-10-10 28345, 2017
ruaok
BB node is up, but didn't come up correctly.
2017-10-10 28330, 2017
ruaok
Leftmost: LordSputnik Leo_Verto : ^^
2017-10-10 28344, 2017
ruaok
now its up!
2017-10-10 28342, 2017
samj1912
ruaok: whom do I ask about mbrng?
2017-10-10 28306, 2017
ruaok
that mmd-schema, no?
2017-10-10 28344, 2017
samj1912
specifically, I am trying to build a language list, but it wont let me call to_etree since the language elements in the list arent an object but a stirng
samj1912: Using "mul" for multiple languages is, I think, mostly a legacy thing. Prior to this year's schema change, Works could only have one language assign, so this language would be "mul" in case it actually consisted of multiple languages.
2017-10-10 28357, 2017
reosarevok
Why would we have that in the first place :D
2017-10-10 28344, 2017
reosarevok
It is a legacy thing, but I guess we need to keep doing it?
2017-10-10 28354, 2017
reosarevok
Unless we're saying that SOLR + ws/3 drops stuff like this
2017-10-10 28319, 2017
reosarevok
(I'm not sure if ws/3 is expected to drop legacy stuff left in only for backwards-compatibility or not)
2017-10-10 28325, 2017
ruaok
SOLR needs to work for ws/2 as well.
2017-10-10 28326, 2017
reosarevok
re: "mul"
2017-10-10 28332, 2017
reosarevok
Ok :)
2017-10-10 28338, 2017
ruaok
is it a ws/1 thing?
2017-10-10 28344, 2017
reosarevok
No, it's a recent change
2017-10-10 28356, 2017
reosarevok
So I guess it stays
2017-10-10 28331, 2017
Freso
reosarevok: Well, ws/2 currently returns both "mul" but also an actual list of languages.
2017-10-10 28310, 2017
reosarevok
Sure
2017-10-10 28329, 2017
reosarevok
But we still need to keep returning mul anyway
2017-10-10 28332, 2017
Freso
Sure.
2017-10-10 28329, 2017
reosarevok
samj1912: I am guessing it already supports events?
2017-10-10 28344, 2017
samj1912
"it"?
2017-10-10 28349, 2017
reosarevok
the SOLR server
2017-10-10 28357, 2017
samj1912
let me check
2017-10-10 28307, 2017
reosarevok
(looking at schema change stuff)
2017-10-10 28310, 2017
samj1912
yup
2017-10-10 28314, 2017
reosarevok
Ok
2017-10-10 28331, 2017
reosarevok
That and instruments were the last entities added, unless I'm being stupid
2017-10-10 28344, 2017
reosarevok
So if it has both, it should only need small adjustments
MBS-8938: Schema changes to support alternative tracklists
2017-10-10 28329, 2017
samj1912
hmmm
2017-10-10 28355, 2017
reosarevok
I'm not going to claim I understand what https://tickets.metabrainz.org/browse/MBS-8800 means, but it might need some changes - it might also be that we're not using it at all so far, since I see all the depends-on-this tickets are open
2017-10-10 28355, 2017
BrainzBot
MBS-8800: Create the earthdistance extension and add a geodetic index for place coordinates
