Re: using isalpha/isalnum in locale-independent code

On Thu, 25 Nov 2004, [UTF-8] Branko Ä^Libej wrote:

> No, you're not being paranoid. Ideally, we'd need a set of
> classification functions that are Unicode-aware. For example, the code
> that parses a "-rN[:M]" option doesn't know that ASCII 0-9 aren't the
> only digits in Unicode. Of course it's debatable whether or not
> Subversion should accept such non-ASCII numbers.
>
> I think it would be best if we created our own set of svn_utf_is*
> functions that would only work on the ASCII subset and always use those
> when dealing with UTF-8 strings.
>
Maybe svn_ascii_is*, since svn_utf_is would imply that they work for the
whole UTF-8 range. But for the property ame validation, I'd like
svn_xml_is_name, which I pinted out in the other thread.

//Peter

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Nov 25 20:01:43 2004