VnPenguin said this at Tue, 16 Nov 2004 15:20:21 +0100:
With current vnTeX (vntex.sf.net), we can use viscii, tcvn, vps and unicode (utf8) for input encoding.
Cool. If there are users of things other than viscii or utf8 who come to ConTeXt, then they should be able to be accommodated with a little work.
As I understand it, UTF-8 support will then require an update to ConTeXt's unic-* files (giving named glyphs for input) and then a slight update to the enco-x5 file. I guess what's really needed are canonical internal names (like 'ygrave' or 'ohornhook') for ConTeXt's use.
Don't be afraid: it's not difficult, just a bit tedious. So I should ask, Mr Penguin, how are you with sed, Perl, or Ruby?
I could use Perl and sed (for simple jobs) without problem :)
Okay. This first task is more a bit of human pattern recognition: Take the appropriate Unicode "page": http://www.unicode.org/charts/PDF/U1E00.pdf and fill in the attached unic-030 with the glyph names that go down the list. I've filled in the a's to set the convention (tone accent goes last, 'dot' gets called 'dotbelow'). (one step after this, and you can get started with your UTF-8 typesetting with x5 fonts, I think. Another step beyond, and we all can use it with Unicode fonts, and the final file to edit is to help viscii users hook into the global VN support we just created for ConTeXt. :)
Also, do you have any good references for character tables for the encoding(s?) that VnTeX uses? I don't have it installed on my machine (yet).
You can download vnTeX from http://vntex.sourceforge.net
Cheers, it looks like OSS is alive and well in VN, contrary to the impression I got in my recent project dealing with the GDVT. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Adam T. Lindsay, Computing Dept. atl@comp.lancs.ac.uk Lancaster University, InfoLab21 +44(0)1524/510.514 Lancaster, LA1 4WA, UK Fax:+44(0)1524/510.492 -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-