Hello all, I'm currently working on a new aspell dictionary for Frisian. We're having a bit of a problem getting capitalization rules to work correctly - the issue is that when you have words like "ôfgrûn" in the word list that have a diacritical mark on the first letter, the proper uppercase form drops the mark: "Ofgrûn". So "Ofgrûn" should be accepted and "Öfgrûn" rejected.
My attempted solution was to create a file l-fy.txt with these lines: include iso-8859-1.txt == iso-8859-1 case A â case E ê case O ô case U û Then use mkchardata from the aspell-lang package, copy the .cmap and .cset files into the tarball, etc. The problem with this of course is that words that begin with plain old "u" aren't then recognized when they are written in uppercase, since aspell is looking in the word list for a lowercased version that begins with "û". My only workaround so far is ugly; it uses this l-fy.txt file: include iso-8859-1.txt == iso-8859-1 case  x case Ê x case Ô x case Û x since no words in our Frisian word list start with "x". So no words with capitalized diacriticals will be accepted, as desired Then I add the correct uppercase forms to the raw word list with a li'l Perl script. Sorry if I'm missing something obvious. -Kevin
_______________________________________________ Aspell-devel mailing list [email protected] http://lists.gnu.org/mailman/listinfo/aspell-devel
