mojca.miklavec.lists at gmail.com
Tue Apr 27 08:50:51 CEST 2010
On Tue, Apr 27, 2010 at 02:30, Karl Berry wrote:
> to work OK with USenglish in the synonyms of english (which is not the
> current state)
> I don't really understand the questions about USenglish, but I'll blurt
> this out anyway: for luatex and lualatex, I think that
> english, usenglish, USenglish, and american
> should all refer to the original Knuth hyphen.tex patterns, and nothing else.
> The name "usenglishmax" can refer to Kuiken's full set of patterns.
> (As Mojca found in the current language.dat.)
> That is, I am thinking the new system you are devising should have
> exactly the same meaning for language names that we have now. Anything
> else seems headed for disaster to me. ConTeXt is a completely different
> scenario than (la)tex ...
For both Karl and Élie ...
When we were importing patterns into hyph-utf8, we had to rename and
change all the files.
In particular, we had to change all the non-ascii files. The ASCII
files were still perfect for loading with both 8-bit and Unicode
engines. (Italian, Indonesian, Interlingua and all the three variants
of English: British, Knuth's US and extended US.)
We still decided to rename all of the mentioned files apart from
hyphen.tex. I didn't want to mess up with that and there was no reason
for it either. We simply left "english hyphen.tex" in language.dat
instead of doing some unneeded black magic like "english
Consequently, we only had to take a single set of English patterns
into repository (at first we didn't even do that, but Hans needed the
extended set of patterns, so we took the extended US patterns and
named them "en-us", but didn't ever use that file for loading
So ... if you want to load Knuth's patterns using the "lua" way, then
we need to import both extended and Knuth's patterns into repository
and invent the proper name (en-us-WHATHERE?).
More information about the tex-hyphen