Alexey Pechnikov <[email protected]> wrote: > Yes, the BOM is on the original string. But with ICU collation we can > see that 17 symbols string is equal to 16 symbols string. I think > this result is not right.
What's the basis for this belief? It's not at all uncommon for two Unicode strings of different length (in codepoints) to collate equal - for example, they could be canonically equivalent but in different normalization forms, or contain weightless characters such as a zero-width non-breaking space (U+FEFF), also known as BOM. > May be automatically dropping the BOM for > ICU collated fields is more correct way. Why don't you do just that in your application? Igor Tandetnik _______________________________________________ sqlite-users mailing list [email protected] http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
