Allow markdown for generating XHTML files

[smhmd]

Right now, my process to generate files is: get an epub, covert it to a txt file, run it through syncabook, make it an epub again. By doing this, I lose a lot of information from the original epub like italicized and bold text, links, images, etc. I imagine we can instead of converting from epub to txt, convert to markdown to retain some of that info. Ofc, parsing txt for sentences isn't the same as parsing markdown, so this is probably not a straightforward feature, I imagine.

Would love to hear your thoughts on this.

[sjabsr]

+1 same situation here.
MD seems too limited though; there could be data we want to retain from the original ebook's xhtml that would be lost. A direct tagging of the epub's xhtml files would be an incredible functionality !

[sjabsr]

@smhmd Have a look at this issue, archived for people to check this solution of directly tagging xhtml files with a python script, instead of using syncabook to_xhtml. I haven't tried it yet, but I guess you can then put them in the /sync_text folder and execute the syncabook sync directly.
-> Use existing xhtml files for syncabook #21

[r4victor]

If you need an "epub + audiobook => epub with media overlays" conversion, you may consider a syncabook alternative: https://smoores.gitlab.io/storyteller/.

I haven't tested it. They claim syncing take significantly more time (compared to syncabook) but the quality should be good.

[sjabsr]

I have very good results with syncabook actually ! This issue is more about having to start over from the bare text.
But it's always interesting to see what other products are available in case they fit better our need. E.g. we'd ideally need a word level sync, not sentence level. Which might also be solved with "custom manual tagging" of the xhtml files, as described in issue #21, isn't it ?
Thanks for the info, I'll check storyteller out.