#bookbrainz

everything looking good, docker image build with few warnings but much faster than npm .

i think we should also get rid of npm run in package.json and use yarn run instead ..

although that should not make any difference but still

Hey CatQuest

hm?

Would it better if we suggest language based on starting characters , here BB-407

BB-407: Simplify "No linguistic content" selection https://tickets.metabrainz.org/browse/BB-407

and is it okay if i push the " No Linguistic content " up the options List (which passed down to react-select) manually like below Multi Language

uh

I'd just want to see it while typing "no"

also [multi] should maybe be removed (since we already *can* select several languages) (tho not fortitle, so thatmiht need fixing)

also i'd like a listof them ost comonly used (like on mb) and under this [no linguistic], [multiple] (if we keep it) as well as [artificial (other)] since title or language could be in some language that doesn't exist/was created only for this book, and we can't add a one-book language

Hmm , we can do custom language thing where user can create one if he didn't find desired language in options

hahaha no.

when user type no wouldn't it be good if we suggest him option like north .. , user generally won't search for "no linguistic"

of course.

usually i type "no" for "norwegian" :D

my point is that [no linguistic] should *also* be in the list whne writing "n"

erh whne writing "no"

tnot require to be at top?

I'd assume it'll sort the way everything else does

so if i type "no li" it shows first

but if i type "no" it sorts alphabetically with "north lumbago" "norwegian" and all the other languages with "no"

and that happens when we suggest user using starting character of word

(I made up north lumbago, i don't recall any languages with "north" in it right now)

i don't know what to say. have you tried using the interface?

ha there are some north something

Yep !

by default it suggest based on "if that language contain that word" but if we do "if that language starts from that word" then no linguistic would appear on the top

again ,i suggest checking how it already works on the site /talk with monkey

you wil lsee how it works. it works fine

my issue is that "no linguistic" doenst actually show whne it is expected that it should

it is showing when i type no l..

Hey CatQuest , then what would be best way to show langauge suggestion and that no linguistic thing ..

monkey ...

Hi Shubh

about BB-407 , your thoughts please ..

BB-407: Simplify "No linguistic content" selection https://tickets.metabrainz.org/browse/BB-407

So not sure about this one. For starters, "No linguistic content" does show in the search results dropdown when typing "no", so I think that part of the ticket has been fixed.

Agh, messages got lost.

I do think adding brackets would help make it stand out as a special case language like for [Multiple languages]

That's a change that I would have to do in the database

In fact, let me try it out on test.BB.org

Ah, it's a bit more complicated I guess, since we're using MusicBrainz' languages table

is it better if we put [No linguistic] option below the [Multi .] option

The issue is that we're not really sorting those anywhere in the code

So they get sorted automatically, and "[" comes before "a"

We can sort like there is custom filter optio for react-select

I'd rather not if we can avoid it

But I can add brackets; this is what it looks like: https://usercontent.irccloud-cdn.com/file/UwWev...

CatQuest: Satisfied with this?

I lied, I think there is some form of sorting

but should we rather suggest Language based on starting character ?

I don't understand

So there is minimal sorting on "frequency" which is defined in the database (basically = most used languages) and then alphabetically.

I don't think "no linguistic content" would make it to the list of most used languages, so I think just adding the brackets is a good solution to mark it out as different.

In short, I think this is probably resolved

i didn't know that there was some sort of sorting mechanism based in options , now it make sense to not use any custom filter ..

anyway , thanks for clearing the doubts monkey !

No worries !

By the way, I wanted to say you're doing great for your introduction in open source projects ! :)

not sure about that ...

Well look, you came to us when you had questions setting up, you've been polite and nice, helped me debug what was broken in our setup, opened 5 PRs and helped me review one of mine, all that in the space of a week ! 🚀

but i couldn't have done this without your help !

monkey about that input field , showing invalid input for other fields in PR for BB-513

BB-513: Improve "Are we missing an entry?" section. https://tickets.metabrainz.org/browse/BB-513

Do you mean "sort name", language, etc?

Yes

I think that's fine. We can't fully assume any of those other fields.

but i did send default values with name field in namesection object , so why it is not validating correctly?

If you mean an empty string, that's not a valid value

i.e. user need to fill in the sort name and languages in order for it to be valid

but isn't that true when user create new entity or there some kinda check to verify ? btw i made the changes in that PR

I didn't understand the question

If you go to create a new entity on the website now and enter a name, you'll see the same validation mechanism because the other fields are missing: https://usercontent.irccloud-cdn.com/file/Ipt9d...

I saw the same thing in your PR, which is what I would expect to happen. We prefill the name, which is the same as if the user typed it. Validation stays the same

Ohh so validation trigger if user type something , hmmm

sort of. I think validation happens when you load the page, and subsequently when any field is modified

I have one more question ..

(Validation happens here: https://github.com/bookbrainz/bookbrainz-site/b... )

So whenever you load the page or change the state validation is run

Shoot your question :)

why we use react proptypes for type checking rather than TS

They just haven't been updated :)

We only recently moved to using Typescript

So conversion is either a very big job for someone, or modifying little by little

Hmm TS tooling makes life easier

Agreed

Hmm baby steps

That :)

Shubh, monkey : this is what I see https://usercontent.irccloud-cdn.com/file/R3ks0...

Same , as monkey said issue is already fixed

KassOtsimine: The result is in there, you have to scroll further down in the list. However if you now try on test.LB you'll see I adde [brackets] and that will put no linguistic content at the top of the not-the-most-used languages

oic

yea i noticed that it was lower down

jup i see it.

Yeah, we do sort more frequent languages above the rest.

👍 now it behves as i expect. i'd consider this closed when on beta :D

All righty

good good

Wanna try again in a new window in prod? :)

oh

wait why would you directly to prod? you sohuld only directly to beta :P

It's only a modification of the name of the language in the database

test DB first, and then prod DB (which is also used by beta.BB).

:|

?

There's no code change so I can't deploy anything (whether it's on beta or prod). But we did test the database change on test.BB which uses a test database;

i don't really get it :D

erh ok? i dunno :)

I change the name of the language from "No linguistic content" to "[No linguistic content]" in the database, that's the only thing I changed

No worries :)