whedon
whedon copied to clipboard
openjournals
[WIP]: Adding initial code for generating JATS output
This is a work in progress. A few known issues/things to do:
- The Markdown figure conversion seems to produce invalid XML: https://github.com/jgm/pandoc/issues/5290
- Not sure whether we need to pass the csl file here as an option
- We might need to figure out what to do with images here too, i.e. how do we bundle these with the JATS?
- It's going to be a not insignificant amount of work to generate the JATS XML for the past ~500 submissions...
/ cc https://github.com/openjournals/joss/issues/495
Hey, not sure that this is the place for it, but we got some discussion around producing jats from markdown in different applications (and about including more information about authors in the yaml). I think we the discussion goes toward extending pandoc, it started with a discussion you were involved in @arfon ...
see https://github.com/manubot/manubot/issues/187 What would you think is could be the best course of action?
Ultimately I'd like to complete this work :soon:, I'm in discussions with @tarleb to see how he might be able to assist us get this across the line.
I'm working to get jgm/pandoc#6067 into the next pandoc release. That should get us half-way. I guess the next step after that is could be to try and see if the output is sufficient for what's required. I found pandoc's current handling of citations in JATS a bit lacking, but wrote a partial work-around in pandoc-scholar. The combination might already be enough – otherwise we might have to do some larger changes in pandoc-citeproc.
FWIW, https://github.com/stencila/encoda supports pandoc-jats and a custom jats encoder/decoder and can also read Markdown and various other formats. I have been adding support for a few missing pieces in the JATS decoder recently. Using Pandoc JATS directly is of course also a great and fast option.
(Just saw https://twitter.com/arfon/status/1415723609671311365 and was curious how this was going.)
