I think the tradtional way as you described is fine. It is what readers
expect with language. That the encoding you are using such as Unicode
does not have a 1-1 match with those expectatins is of no significance
to the readers. They don't care, and they certainly don't suddenly start
sorting text mentally in Unicode order. They continue t sort the way
they were taught in school when they were 8 years old.
So it is up to you to come up with an algorithm that matches user
expectatins and needs and your data structure, not the user's
responsibility to relearn a large part of their language usage :)
Best,
Barry
Richard Jolly wrote:
Hi,
I have a we page that groups a list of items by splitting it into ranges
that match the first letter.
It not needs to be unicode capable, although the characters will most
likely be in the latin-1 range. Ideally I'd just like to use a unicode
equivalent of [A-Fa-f], [G-Kg-k], etc.
This may not make sense in a unicode world, however.
Any ideas?
Thanks, Richard
http://www.bbc.co.uk/