citation-style-language / schema

Citation Style Language schema
https://citationstyles.org/
MIT License
184 stars 61 forks source link

Restore titles allowing simple string values #396

Closed cormacrelf closed 2 years ago

cormacrelf commented 3 years ago

Targeting v1.1.

Description

Fixes title variables not allowing plain strings, but being documented being string or title-object. There was a title-string definition but it was unused, so title-variable is now an anyOf union.

See https://github.com/citation-style-language/schema/pull/393#event-4911844030 for general discussion of deprecating instead of outright removing.

Type of change

Checklist

bdarcus commented 3 years ago

Per my comment here, I think we need to settle a broader strategy before deciding particular changes.

https://github.com/citation-style-language/schema/pull/397#issuecomment-910093763

bdarcus commented 2 years ago

Do you want to drop the pre-commit commit here, now that I've separately merged #399 (though I'm unclear why github is indicating it's a change)?

Is there anything else we need to resolve?

bdarcus commented 2 years ago

While I'm hear, thought I'd circle back on this.

Where do we stand on this?

bwiernik commented 2 years ago

My original position was that I wanted to permit string values and have parsing main vs sub be a part of the processor steps in that case. That got pushback. I would like to have that option--a preferred structured approach and a string-parsing alternative

bdarcus commented 2 years ago

My original position was that I wanted to permit string values and have parsing main vs sub be a part of the processor steps in that case. That got pushback. I would like to have that option--a preferred structured approach and a string-parsing alternative.

I'm still strongly opposed to adding an entire title parsing section to the CSL spec, and new elements and/or attributes to the style schema, in order to support independent formatting of titles and subtitles.

Whether there's a way to do it without that I'm not sure (though don't think there is). Or maybe @cormacrelf has a practical way to do this that isn't as painful as I foresee.

But I think this needs input from @cormacrelf and the zotero team (cc @dstillman) with regard to their plans vis-a-vis CSL going forward. And of course, if there are other developers/implementers reading this, the more feedback the better.

I know @cormacrelf has done work which contributes towards supporting what we've been calling v1.1, and this (along with the ETDF date library) is actually part of it, but it's just hard to have these discussions in the abstract, without an understanding of concrete roadmaps and such.

For brief context for those new to this, the big changes in 1.1 are in title and dates.

The questions here are of course about WTD with the latter, particulary WRT to legacy 1.0 data. I just don't the problem with a breaking change here, when the files will version number.

bdarcus commented 2 years ago

Edit: actually, I think I misunderstood this. The schema commit actually just fixes a bug so that it does what is says it does.

I still need to check the pre-commit change.

bdarcus commented 2 years ago

Merging.

BTW, I've enabled rebase merging, which is useful in this case, and much better than merge commiting.