Originally Posted by user_none
Input or output support?
The base flow is: Input format -> intermediate (epub) format -> Output format. An input plugin takes the input format and converts it to html/CSS. The output format takes html/CSS and turns it into the output format.
TXT is a bit complex (lots of options) but it's a good one to look at for general flow. Mainly because it is a non-html format on both ends.
You'll probably end up subclassing a lot of the epub code since you really just need to translate the non-epub tags to something similar for input. You may or may not need to write an output plugin depending on how difficult it is to find patterns in an epub to turn into the kepub.
I want input and output support, but since a kepub is an epub, I'm hoping it'll be almost free. The only non-conforming part that I know of is extra attributes in the span tags, which I assume the parser will happily ignore.
When you say the intermediate format is epub, do you mean it's basically an exploded (unzipped) epub?