On 12 April 2016 at 08:54, Claudio Beccari wrote: > What do you mean with "2 for Greek unless you count Ancient Greek together > with modern Greek"?
You may ignore this. It wasn't meant absolutely seriously. The author of all three patterns is the same and the patterns are generated in a similar way and included in the same package "elhyphen". So one could argue that the three patterns could be packaged together, even if they don't officially represent the same language. There are similar "problems" with "Norwegian" which is itself a non-existent language (actually there are two languages, but the hyphenation patterns happen to be exactly the same). Or the "hell" of ex-YU languages where the same patterns serve everyone, but it might be offending to use the wrong name. We don't do anyone a favour by starting endless flamewars about how to properly name and merge the patterns. I repeat again that *we do not intend to remove any patterns* or change the names of any of patterns we currently ship. In particular we won't change anything related to Greek: neither for the modern nor the ancient one. > Definitely Ancient Greek and modern Greek are two different things for both > hyphenation and lexicon. Sure. We don't plan to change anything with that respect. In particular we *don't want to remove any patterns*. At all. > May be modern monotoniko and modern polutoniko may be merged in one for what > concerns hyphenation, May be, but unless the author of hyphenation patterns does that, we don't even have the skillset required to do so nor any special reason or need to do it. > for Greek there is another complication; for 8-bit aware typesetting engines > the hyphenation patterns are different; at the moment they are saved in a > directory different from hyph-utf8 I've been thinking of changing that for a while already. In the beginning we were somehow reluctant to include anything that wasn't "clean UTF-8 patterns" in the hyph-utf8, but in the meantime situation has changed, by now we ship 3 versions of each patterns already (initially we planned to use a single one) and I believe that we should include those three files (plus probably the sources and documentation). > and i do not see any means for getting > rid of them and of the 8-bit LGR encoding. What did you mean by that? Mojca
