The only alternatives I've seen to po4a is very naive parsing of Markdown, e.g. just breaking up documents based on paragraphs like what Transifex does. That leaves out the possibility of removing pure syntax from the translator's view, like the ```bash marks.
We're specing it out here: https://github.com/translate/translate/issues/3956 .hc Antonin Delpeuch (lists): > Hi Hans-Christoph, > > I am very interested in this subject too. I think po4a is a fantastic > project, and it seems to work relatively well on markdown. I am > considering using it to set up translation for OpenRefine's new > documentation. > > One thing that worries me is that we might soon run into the limitations > of the markdown parser (since it is in fact a generic parser for text > files with some markdown support). If we run into that, we could > potentially contribute patches, but: > - po4a is written in perl, so I will struggle to contribute myself; > - isn't it a bit dangerous to change the way po4a splits a file into > translated blocks, when people already rely on the tool to translate > things? If they start using a newer version of the tool which splits > markdown differently, can they convert their existing translations > easily? I guess by extracting them from the generated markdown files, > perhaps? It sounds like a pretty dangerous exercise. > > Are there any alternatives to po4a? I would prefer to investigate all > our options before settling on a particular choice, since I don't want > to waste translating effort. > > Best, > Antonin > > On 23/03/2020 12:26, Hans-Christoph Steiner wrote: >> >> Hey all, >> >> I've been diving into po4a with the goal of making Markdown a lot easier >> to use with Weblate. I heard mention that there has been some >> discussion of this already in Weblate, so I'm writing to sync up with >> any Weblate efforts here. >> >> The first specific question is: will Weblate implement Markdown parsing >> in Python, or use po4a? Or do something else? >> >> Also, any thoughts on how Weblate can automatically detect Markdown in >> PO/POT files so that it can automatically add md-text and other Markdown >> checks and modes? >> >> FYI, two small but notable improvements were just merged in po4a: >> >> * type: Code fence info string should not be translatable >> https://github.com/mquinson/po4a/issues/194 >> >> * text inside Markdown code fences is marked as "#. type: Plain text" >> https://github.com/mquinson/po4a/issues/195 >> >> And I'm currently working on handling YAML Front Matter in po4a: >> https://github.com/mquinson/po4a/issues/196 >> >> .hc >> > _______________________________________________ > Weblate mailing list -- [email protected] > To unsubscribe send an email to [email protected] > -- PGP fingerprint: EE66 20C7 136B 0D2C 456C 0A4D E9E2 8DEA 00AA 5556 https://pgp.mit.edu/pks/lookup?op=vindex&search=0xE9E28DEA00AA5556 _______________________________________________ Weblate mailing list -- [email protected] To unsubscribe send an email to [email protected]
