Stylesheets and transforms for journals and publishers

[petermr]

This currently applies to HTML but may also apply to PDF and TXT transformations.

The HTML translation ( https://github.com/ContentMine/norma/blob/master/docs/TRANSFORM.md ) involves 2 or more stylesheets. Where do we put them and how do we organize it? Should there be a Github repo for publishers, separate from Norma (My current thinking is yes).

Currently we have (for publisher fooPub)

src/main/resources/org/xml-cml/norma/pubstyle/fooPub/

which contains toHtml.xsl, etc.

This means we have to rebuild the system for each alteration. If we use stylesheets this can be done independently from norma. It would be roughly similar to dictionary.

NOTE: we might want to lump together quickscrape scrapers and norma stylesheets for each journal.

[tarrow]

The quickscrape scrapers are currently in http://github.com/ContentMine/journal-scrapers

This may not be quite the right place for them. Probably we should do like we do with dictionaries and have yet another repository for them. Perhaps we should include both the XSLTs and the dictionaries in the repositories that use them (norma and ami) with a git subtree / submodule. I'm currently reading about which is probably a better option for us.

[petermr]

Good. I think we need to normalize out the publisher stuff and remove specific publishers from norma. Happy to use journal-scrapers as temporary home for "publisher transformers"