ruaok: yup I agree with reo, I thought about raising a bad request but then wouldn't it mean bad request by user? But it was actually a bad request from our side
2020-09-10 25445, 2020
pristine___
Just thinking out loud.
2020-09-10 25437, 2020
alastairp
what's an example of a reqeust coming in from a user and then going to labs?
2020-09-10 25445, 2020
ruaok
yes, that is good, I think.
2020-09-10 25405, 2020
ruaok
if you want to aid debugging, then maybe take the error message received and add more context.
2020-09-10 25422, 2020
ruaok
lb-spark called to labs failed: <error message>
2020-09-10 25450, 2020
pristine___
lb-spark or lb-server?
2020-09-10 25411, 2020
ruaok
whichever, it was just an example
2020-09-10 25442, 2020
pristine___
Cool
2020-09-10 25457, 2020
reosarevok
yvanzo: what if anything should we set up for later today?
iliekcomputers: hey, do you plan to do a release today?
2020-09-10 25410, 2020
iliekcomputers
don't think so
2020-09-10 25424, 2020
iliekcomputers
i did one yesterday
2020-09-10 25456, 2020
yvanzo
reosarevok: because this page was not part of the discussion on yesterday.
2020-09-10 25412, 2020
reosarevok
The artist page works tab has the same issue as the search results :)
2020-09-10 25440, 2020
reosarevok
(they're not writers, but without the third column you wouldn't understand why the work is there in the orchestra artist from yesterday)
2020-09-10 25444, 2020
yvanzo
I understand that, but it's not documented neither in the ticket, nor in the comment we based our discussion on.
2020-09-10 25407, 2020
reosarevok
"In addition to inline search, the work writers are also shown in works lists (both search results and standard work tables), where the change to only have actual writers shown would also make sense to me." :)
2020-09-10 25424, 2020
reosarevok
That's a standard work table (that's what I meant with that anyway!)
2020-09-10 25434, 2020
yvanzo
reosarevok: the second screenshot looks ok too.
2020-09-10 25438, 2020
reosarevok
Yay
2020-09-10 25455, 2020
yvanzo
Just please document that damn ticket!
2020-09-10 25459, 2020
reosarevok
I'm adding these as a second commit on that PR, can also add more details to the ticket
2020-09-10 25402, 2020
reosarevok
Sure
2020-09-10 25403, 2020
yvanzo
:)
2020-09-10 25452, 2020
pristine___
iliekcomputers: ah! I wanted to monitor the recs while I am writing script to show them on site. Let me know if you change your mind, I will update weekly rec cronjob to daily.
2020-09-10 25433, 2020
iliekcomputers
what do you want to deploy, cron?
2020-09-10 25409, 2020
reosarevok
yvanzo: one thing: you proposed to add the third column also to the ISWC report
2020-09-10 25442, 2020
reosarevok
My view is that these aren't directly connected with ISWCs (not being related to writing) and since the report should probably avoid data not directly useful for the report, I'd skip it there
2020-09-10 25456, 2020
reosarevok
But if you prefer to keep it for parity with the ISWC index page, I can do that too :)
2020-09-10 25410, 2020
pristine___
iliekcomputers: yeah
2020-09-10 25440, 2020
pristine___
But change it to daily. Rn, cron for recs will only run on Tuesday.
2020-09-10 25442, 2020
yvanzo
reosarevok: good catch, I didn’t consider this specific case.
MBS-8328: Don't show dedication/publisher rels as "writers" for works
2020-09-10 25442, 2020
reosarevok
In "final choice"
2020-09-10 25438, 2020
yvanzo
Any reason to keep “previous attribution” apart?
2020-09-10 25430, 2020
reosarevok
Because that's actually "was thought to be a writer but now we know it's not"
2020-09-10 25432, 2020
yvanzo
Maybe it was originally suggested not to hide it, but now there is a 3rd column.
2020-09-10 25435, 2020
reosarevok
So it's directly related to writers :)
2020-09-10 25404, 2020
reosarevok
(if you have an old release, for example, before researchers figured out the attribution was wrong, it will be very useful to see it under "writers")
2020-09-10 25422, 2020
reosarevok
So, it's kind of a gray area that IMO belongs more with writers than not for ease of use
2020-09-10 25423, 2020
yvanzo
That artists did not nothing but letting rumors spread :D
2020-09-10 25438, 2020
reosarevok
Well, in many cases they died :p
2020-09-10 25453, 2020
yvanzo
That’s an excuse but still :)
2020-09-10 25408, 2020
reosarevok
But there are legitimate cases of "we thought this was by Bach, but it turns out Bach only copied it to his book from another manuscript and we assumed it was his!"
2020-09-10 25425, 2020
yvanzo
Maybe it should have a different parent then?
2020-09-10 25433, 2020
reosarevok
And then you'll get it in (mostly old) "complete recordings" by Bach :)
2020-09-10 25435, 2020
reosarevok
Hmm
2020-09-10 25417, 2020
reosarevok
So move it under composition? I mean, it's not necessarily a bad idea I guess
2020-09-10 25427, 2020
reosarevok
as a third option after arranger
2020-09-10 25456, 2020
reosarevok
I'm not sure I dare to do what you will propose next, which is to just put them in writers if they have composition as parent and misc otherwise :p
2020-09-10 25412, 2020
yvanzo
And what is your thought about renaming “composition” to “creation”?
2020-09-10 25425, 2020
reosarevok
Probably sensible, if we don't break anything by doing so
2020-09-10 25433, 2020
yvanzo
for everything that belongs to the creative process
2020-09-10 25446, 2020
reosarevok
I guess because this is a grouping rel, nobody should be using it via the ws or anything?
2020-09-10 25457, 2020
yvanzo
Should be done cautiously, sure, and there is no hurry about doing that now.
2020-09-10 25411, 2020
reosarevok
Feel free to add a style ticket, I think that makes sense
2020-09-10 25422, 2020
yvanzo
Ok, will do.
2020-09-10 25442, 2020
reosarevok
I moved the previous attribution one as composition child order 2
pristine___: the daily recommendations seems like a good idea to me. but, now its inconsistent with us saying that we will generate them once a week. what is your plan?
2020-09-10 25428, 2020
pristine___
Meanwhile I am writing the script to show recs on the website, I thought of monitoring the recs so i changed the cron job to daily. I will change it to weekly once the script is ready.
2020-09-10 25441, 2020
pristine___
Did not think about daily recs. Ummm... Should we?
2020-09-10 25454, 2020
ruaok
lets stick to the weekly plan after you're done monitoring. lets get feedback and then we'll see
2020-09-10 25406, 2020
zas
alastairp: overall looks good, but isn't this managed by transifex?
2020-09-10 25441, 2020
pristine___
ruaok: sounds good.
2020-09-10 25443, 2020
alastairp
zas: good question! I don't know. I guess we need a documentation day for CB too...
entity_type can be in the end point since it should always be present
2020-09-10 25450, 2020
_lucifer
if entity_id is missing, it will not match the endpoint (but i don't know flask that well so might be wrong)
2020-09-10 25457, 2020
alastairp
_lucifer: what do you mean?
2020-09-10 25405, 2020
_lucifer
i mean `/review/artist/` will not match `/review/entity_type/entity_id/` endpoint
2020-09-10 25411, 2020
alastairp
can you tell me an example of a url with entity_id as a request argument?
2020-09-10 25447, 2020
_lucifer
`/review/artist?id=<mbid>`
2020-09-10 25445, 2020
yvanzo
added BrainzGit to CB repo
2020-09-10 25454, 2020
alastairp
thanks yvanzo
2020-09-10 25426, 2020
alastairp
_lucifer: and why do you think that this is a better way of forming the url compared to `/review?artist=mbid` or `/review/artist/mbid`?
2020-09-10 25421, 2020
_lucifer
i do not think so
2020-09-10 25444, 2020
_lucifer
i do think `/review/artist/mbid` is the best way
2020-09-10 25411, 2020
_lucifer
in that case we should move the entity_id to elsewhere
2020-09-10 25423, 2020
_lucifer
because in the current form it won't work
2020-09-10 25444, 2020
_lucifer
typo above, i meant entity_id check to elsewhere
2020-09-10 25440, 2020
alastairp
if we have to move it elsewhere to make the check more easily then that's fine
2020-09-10 25453, 2020
_lucifer
👍
2020-09-10 25443, 2020
blinky42 has quit
2020-09-10 25454, 2020
nelgin
Morning peeps.
2020-09-10 25437, 2020
_lucifer
alastairp: i figured out what i was doing wrong.
2020-09-10 25406, 2020
nelgin
yvanzo I'm going to reinstall with virtualbox I'm going to use a fixed disk rather than dynamic, and I'm going to put it on a new ssd and see if that helps any.
