#bookbrainz

Interesting! Well done on figuring it out

Hello monkey I want to participate in the gsoc and the registration have just started I want to ask how should fill my name in it from the bookbrainz I really loved the work and would want to continue it here only

Hello maddy007[m] ! I'm not sure what question you are asking about filling your name

monkey: I think I'm mostly done with BB-530, caching is working now :)

BB-530: Wikidata Blurbs https://tickets.metabrainz.org/browse/BB-530

Awesome !

!m kellnerd

You're doing good work, kellnerd!

Some of the enforced linter rules were rather annoying, to be honest.

Are you open to improvement suggestions which disable rules which were once specifically enabled?

Yes, we can definitely discuss it

Great, then I won't fix the remaining linter issues in this PR... some of them are not even reported by CI, I only saw them locally.

So I will definitely add linter improvements to my backlog, but first I want to focus on finishing the Wikipedia PR and preparing my GSoC proposal.

Definitely

Remaining TODO items: "Collapse extract if it is too long"

I have not seen an extarct that was long enough to warrant collapsing, but I can still do this, either in this PR or later in a separate PR if we find a longer extract.

I think you can probably reuse some of the logic from https://github.com/metabrainz/bookbrainz-site/b...

Yes, that's the idea.

We might have discussed that already, but I haven't written this down anywhere.

I think I remember reading a mention of it

now that I think of it

So you would do this in the current PR?

I don't think that's absolutely necessary. Can definitely be in another PR

Ok :)

DOn't let this beast grow to large

Exactly

Next item: https://www.mediawiki.org/wiki/API:Etiquette

I think we should be fine once we've provided a User-Agent header.

Yes, I think so too

Is there already an existing MeB example how it could be composed?

The use of the pipe character to fetch multiple pages is interesting, but certainly not something we need to worry at the moment

*worry about

Or should I just use "BookBrainz (https://bookbrainz.org)?

Oh, actually there is a version number on the website... I thought there was none.

Multiple pages might be interesting if we consider to use both Wikidata items of e.g. https://bookbrainz.org/author/37e2bdfe-28b1-456...

Currently I just use the first Wikidata identifier

Hm. How would we display multiple extracts?

Do you mind asking the MB team in the main channel about the user agent string? They'll have a suitable format I'm sure, and better to harmonize with it

But probably that's the 1% case, while the current solution works for 99%.

Let's ask the MB team.

(So my intention was never to display two extracts for one entity page.)

Unless I hear objections in #metabrainz, the user agent will be "bookbrainz-site/latest (bookbrainz.org)" in my dev setup.

In production it is supposed to use the current version instead of "latest", which I hope to obtain from $GIT_COMMIT_SHA

Does this env variable or the .git-version file contain anything in your dev setup, monkey? For me both are empty...