Ok, but then a) we'd need to refine the issue to "enhance request to handle every non-ASCII character/digram that could occur in other Boggle languages":
- Spanish: ñ
- Czech: no extra characters (e.g. 'E' can be used in words as E, É and Ì)
- Danish/Norwegian: do Boggle dice use æ,å,ø ? I searched but couldn't find out.
- Filipino: ñ . And separately in some wordgames (e.g. Scrabble knockoffs), I see 'ng' treated as a single digraph. Apparently it's officially a separate digraph, so the Tagalog alphabet ('abakada') has 28 letters.
And b) if we have rare characters on a few dice (esp. the unofficial Bosnian one proposed), that imposes restrictions on your assumption that any cell can be any letter. So you'd have to postprocess to cull a few words that couldn't legally be made with that dice-set.
- Spanish: ñ
- Czech: no extra characters (e.g. 'E' can be used in words as E, É and Ì)
- Danish/Norwegian: do Boggle dice use æ,å,ø ? I searched but couldn't find out.
- Swedish: do Boggle dice use å,ä,ö ?
- Turkish: do Boggle dice use ç,ı,ğ,ş,ö,ü ?
- Bosnian: see https://boardgames.stackexchange.com/a/51289/2358
- Filipino: ñ . And separately in some wordgames (e.g. Scrabble knockoffs), I see 'ng' treated as a single digraph. Apparently it's officially a separate digraph, so the Tagalog alphabet ('abakada') has 28 letters.
And b) if we have rare characters on a few dice (esp. the unofficial Bosnian one proposed), that imposes restrictions on your assumption that any cell can be any letter. So you'd have to postprocess to cull a few words that couldn't legally be made with that dice-set.