Sat, 7 Jul 2001 14:02:33 -0700
Sorry for the combo reply. :)
> Thank you Chris I was starting to understand how "tribulation worketh
I needed it myself for the web front end, where I'm only allowed one encoding format, and I didn't want to parse the UTF-8 into ampersand codes for Unicode glyphs. Necessity is the mother of invention. :) It was either write a filter for the API that would be fast and useful to others to do the same function in the CGI. Pretty easy decision.
> > But do you think it would be possible to convert the remaining latin1
> > to UTF-8?
Yes we will, but not until UTF-8 issues are worked out completely. Until then, I'd like to only do UTF-8 versions of modules that would otherwise be impractical in a regional encoding or impossible without a font-specific encoding.
I'll look into IBM's ICU this weekend and see how difficult it would be to harness some of its power for Sword before we hit 1.6.0.