#metabrainz

heh

alastairp: can you check that sql schema thing ?

agin I must stipulate: nn and nb are *not* languages, they're writingforms, the language for all is "norwegian"

CatQuest: oh? even more interesting!

so I write bokmål, but I speak norwegian

Serbian has also Cyrillic and Latin writing options IIRC?

eg kan prøva nynorsk men eg e ikkje flink i detta

(which is a similarish case)

yea might be similar :o

I guess you can do hindi with latin letters too

and there's pinyin

I know that mongolian in china is written in mongol letters but in mongolia mongolese is using latin

iirc

_lucifer: I don't know. I'll have a look shortly

that's interesting CatQuest you can open a ticket and for adding those to CB if you'd like to have them

sure thank!

<3

just as long as you don't remove "norwegian" becasue my recent review was not "bokmål"

:D

no reason to remove i guess, when the language is added you could reclassify it as well

well afai knb and nn and no is already in critique brainz (but ther is an erro when using no)

um CatQuest **no** is not in CB

yes. it is

i scohse it for the linked review in the ticket

oh, "no".

or rather it was pre-chosen

it was in the list

yes, CatQuest you are right actually

i was looking in the site language not review text language

funny i was going ot screenshot that evne before yo upsoted this list (wich is a lot shorter thna the dropdown too)

are you looking at website transaltions?

yeah

heh

my mistake 😓

no problem!

[listenbrainz-server] paramsingh merged pull request #1089 (master…ishaan/store-import-id): Make sure incremental dump import is not skipped if RC crashes https://github.com/metabrainz/listenbrainz-serv...

ishaanshah: hey

just to confirm

_lucifer: I believe that 16.sql has been applied, but not 17 or 19

I don't know where 18 is

i'm gonna import full dump 195 and then ask to import until the newest incremental dump

this will get the cluster back to valid data, right ishaanshah ?

alastairp: 19 will fix the errors in sentry

18.sql is in the recordings PR

iliekcomputers: Hi

hope you are feeling well

> this will get the cluster back to valid data, right ishaanshah ?

not exactly

some of the listens will still be lost

Why?

beacause new listens in full dump 202 and 218 are not present in incremental dump

They would be present in incremental dumps 202 and 218

oh

then yes

first full dump with id

and then newest incremental dump

one question, say i import listens in from the past month today will they be present in the today's dump?

Cool. I'll deploy this tomorrow.

_lucifer: yes

_lucifer: yes

👍

_lucifer: which error? The language one?

alastairp: no this https://sentry.metabrainz.org/metabrainz/critiq...

great

i was also able to track down language error to this https://github.com/python-babel/flask-babel/iss...

maybe upgrading flask babel fixes this?

otherwise, i'll try to add a try except and default to `en`

[musicbrainz-server] reosarevok opened pull request #1730 (master…MBS-11152): MBS-11152: Convert entity/ratings page to React https://github.com/metabrainz/musicbrainz-serve...

MBS-11152: Convert entity/ratings page to React https://tickets.metabrainz.org/browse/MBS-11152

[musicbrainz-server] yvanzo opened pull request #1731 (master…fix-github-release-format): Improve blog link format in releases on GitHub https://github.com/metabrainz/musicbrainz-serve...

oh wow ,i just thought of something in relation to individual instruments/individual brands we talked about on the summit: for individual gamelan orchestra, that's another example of "individual orchestra"

_lucifer: in this case we should probably consider a fallback "we don't know what language this was written in", and trigger an alert in sentry (without crashing)

wich is somewhat different thna orchestra in the western sense (i should ahve used the term "ensemble")

because the gamelan is about the instruments, whoever plays isn't that important, the important is the various ensemble's combination of instruments

alastairp: i am still taking a look and trying to figure why the language is shown at all if its not supported

iiuc, display a flash error on the top and log right?

reosarevok, bitmap: I fixed the message for tag v-2020-05-18-postgres12; make sure to delete your local tag to prevent accidental override: git tag -d v-2020-05-18-postgres12 # next fetch will grab the new tag

its an inconsistency, for instance people can write reviews in `no` but not view them afterwards

becasue the error is not triggered on writing the review in `no` but on trying to view them

yep it triggers ise both on saving and viewing

error on saving is because after the review is written, the user is redirected to the view page. from what i see locally, the review is saved in the db correctly

ahh that explains it indeed

(but why i got an error but it did save properly)

_lucifer: ideally we shouldn't allow people to choose an option that results in an error on save

yeah

But if it does happen, we shouldn't cause an error on read either

My suggestion for the text was the error to show on read

no meeting today ?

Because I think it shows "this review was written in [language]", right?

CatQuest: no

makes sense

no there isn't or no there is? :P

there is an ISE if the locale is not supported alastairp

it's in the topic

but it does show that language field in case it is supported

I guess CatQuest's error on save was because it tried to show the review after it was written?

yes right

it happens after the redirect, the review CatQuest wrote will show up correctly even on homepage but display an error when clicked on

yep

_lucifer: can we quickly discuss the search changes after the meeting please?

no meeting today :)

yvanzo: but we can discuss now or later if you want

(Just a reminder that we don’t have meeting today (yvanzo and others).)

Freso: Is that because it is Republic Day here in Portugal? ;-)

lol

_lucifer: ok, I was just checking which of these changes might break something or not

I like that we can review places in CB but MB doesn't even allow rating them

_lucifer: SEARCH-579 is breaking change as 'packaging' is moved from an object to a string

SEARCH-579: JSON format doesn’t use 'packaging-id' key for packaging MBID https://tickets.metabrainz.org/browse/SEARCH-579

yes right, i think that is the only one yvanzo. other changes only add new fields and do not change any existing one

Ok, I went through the other ones and found there are only additions too.

yup

So we don’t need to do an additional announcement, thank you.

Spending 11 hours in meetings on a bank holiday weekend is definitely above and beyond the call of duty.

we can leave out #38 and #42 yvanzo

the java 11 is obvious but the cleanup file. i do not know if there should be more comprehensive tests for that

i found that some entities are missing tests

works and event most prominently. there might be even more that i do not know of

_lucifer: For #43, it would be nice to test with more than one secondary type since it’s a list.

sure will add that!

i locally tested that but probably didn't commit it. will do

Thank you, I’m switching to MBS now (next release due for today).

can you merge the remaining ones ? i think some of these PRs may require rebasing

sure later then

Wooorking on it...

Does CB actually have any event or place review to test with?

reosarevok: mb test artist? mb sound team?

oh, event or place

is there an mb test event?

I meant "existing reviews that I can test on MB"

I found some though :)

MB is 100% hardcoded for RGs at the moment though

So it'll take a bit of work

that's one thing: I wish cb reviews showed on releases tbh, because it's fairly rare to g oto the RG isn't it

They do on the sidebar

But yes, I'd agree it would be nice to eventually have a reviews tab for releases that shows RG stuff maybe

Let's start with the less hard bit :D

UK using excel to store COVID-19 test results database...