Hi,
No Tidy, so no playing with your classes, no need to remove duplicate classes, just a really robust parser that tries its best to come up with something useful out of any html soup. Plus we get the inherent benefit of recognizing all html5 tags.
Still lots and lots to do before any release as we also replaced the strict Xml processor Xerces with Gumbo node trees and parsing for all xhtml files.
We are replacing the remainder of Xerces use with pure xml (opf and ncx) with python and lxml. So lots of code had to be rewritten and redesigned and still needs some work. I'm sure lots of new bugs will have to be tracked down and fixed. But if we want to support both epub2 and 3 there was no other way then overhaul the tools Sigil used. This is the first step. Once this is complete we can start adding in support for epub3.
KevinH
Quote:
Originally Posted by JSWolf
Thanks for the information. It is indeed the latter (Sigil without Tidy).
So with the next Sigil, will it be possible to not have the structure changed?
|