On Mon, Nov 19, 2018 at 08:50:49PM +0100, Werner LEMBERG wrote: Dear Werner and others, > >> In this matter I defer entirely to the TUG hyphenation team > >> (Arthur, Mojca, ...) who know infinitely more about such things > >> than I ever did or will. > > > > I wrote a UTF-8 version of patgen a few years ago. It is very > > slow because I didn???t need it for production, but I can look into > > improving it. Documentation is scarce, for the same reason. > > > > https://github.com/hyphenation/hydra > > > > Mathias Nater also worked on his version, but I never tried it out. > > > > https://github.com/mnater/orthos > > > > I would suggest that if Prasad is allowed to publish his word > > list, he do so so that we have a better idea of how we can help him. > > This will certainly be more efficient than emailing back and forth. > > Another possibility for an experienced C++ user would be to convert > opatgen's (GPLed) source code to modern C++, then publishing it on > gitlab or something similar. > > > https://web.archive.org/web/20050225005508/http://www.fi.muni.cz/~xantos/patlib/src/ > > Please please please please :-)
Some work of another student of mine to follow David Antos thesis is already in progress: https://github.com/Haaxor1689/PATLIB It compiles now, but it yet has to be validated that is does the right thing. Petr
