Hello Ian, Wednesday, October 31, 2007, 11:43:47 PM, you wrote: > Currently the lexer looks for them as regexps, e.g. > "{-#" $whitechar* (LANGUAGE|language) { token ITlanguage_prag } standard solution: [l|L][a|A]... -- Best regards, Bulat mailto:Bulat.Ziganshin at gmail.com