Software: beastie

[Karl Berry]

Hi Norman,

    $ beastie extract-bib.scm -O crossref test.aux <citation_list><citation
    $ beastie extract-bib.scm -O yaml test.aux

Excellent. Thanks! I belatedly realized that I should have just asked
about outputting Crossref <citation> info in the first place. Then
there's no need for YAML :).

    was that the sort of thing you were meaning?

Yes.  

    (hmm: it would do no harm for me to chuck in a few line-breaks there...)

Yes please. The output needs to be readable by humans (me) as well as
computers, for the sake of checking the results (and debugging).

So the -O crossref is essentially extracting the fields from the
selected bib records and mapping them onto the available crossref
elements? That seems ideal! Thank you.

    since it doesn't split page ranges so that it produces <first_page>
    correctly, 

That niglet doesn't sound too hard :).

    and it should be type-sensitive enough to detect when a
    BibTeX 'title' is a crossref article_title or book_title,

I don't see book_title in the crossref <citation> element?
Am I missing it? It looks to me like their structured citations are
mainly intended for articles. E.g., there is no publisher field.
https://data.crossref.org/reports/help/schema_doc/5.3.1/common5_3_1_xsd.html#citation

I'm thinking that for TUGboat what I should do is upload both the
unstructured citation info for everything, as I'm doing now, as well as
the structured citation.

One question: what happens with BibTeX entries that use the (bibtex)
crossref field?  Are they resolved? (I hope/trust.)

Anyway, I think we should take the further production details offline.
Can you tell me when/where to get the code? Is the dev repo online?
I only saw the releases on your page at a quick glance. --thanks, karl.