Switch from HTML save files to Plain TXT

[HenrickTheBull]

Is your feature request related to a problem? Please describe.
It's odd to me that files are being saved as HTML with full HTML markup.

Describe the solution you'd like
I'd like to see the app migrate to using plain text files instead of HTML which I assume would be the goal of #64, I think that switching to plain text would also assist writers if they want to migrate their writings to another application or recovering their writings should something happen to this app.

[cgueret]

Thanks for the suggestion! The reason why things are that way is that I wanted to have plain text files to be able to use git as a versioning / backup back-end but I also needed a markup system to track italics etc. After having hesitated a lot between Markdown and HTML, I picked HTML as the primary goal of the app was/is to output Epub which are made of HTML files. That said, now that we have #61 in the pipeline the HTML files will have to be converted into other things at book construction time. So changing the base format on disk could make more sense

About #64, the idea there is more to remove an external dependency that is a bit tricky to use and blocks #21 . Again, because epub files are HTML documents I went for a WebView (which is also what applications like Foliate made too). It turns out that a full web view system may be a bit overkill anyway to render text with some styling so that could be changed

[HenrickTheBull]

I tend to drift to Markdown these days, though most of my novels are on Scrivener (which I'm trying to get away from), and I think they use .rtf to track styling.

Yes they do.

Pandoc's python library would solve a lot of the export issues without adding a lot of code complexity I think. I'm not sure as I've never worked with Pandoc's python library.

[cgueret]

Oh... RTF is an interesting choice but I would prefer Markdown. That way it would also make it easier to co-author the same book via git while using different tools. There are tons of markdown editors out there.

Never used pandoc in python either but I agree it could be very useful. I used it a few times as a command line tool it's pure magic :-)

[cgueret]

Just stumbled upon this and it's an interesting argument, even though Scriptorium would not have to deal with very complex Markdown: https://hackaday.com/2026/04/05/making-the-case-against-markdown/