Caleb Epstein wrote:
On 9/28/05, David Abrahams <dave@boost-consulting.com> wrote:
Hmm... Also, is the apparent dependency on ASCII encoding truly portable?
Doubtful. Wouldn't testing for std::isalnum || '-' || '_' be a better idea? Perhaps not quite as performant (once the lookup table was made static), but certainly more portable and simpler to read.
-- Caleb Epstein caleb dot epstein at gmail dot com _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
In most implementations, the is*()'s are implemented using exactly the same method. It's generally safe, as pretty much every code-page barring EBCDIC (or whatever it is) that isn't obviously non-latin (CJK codepages, espcially) uses the ASCII characters for values less than 128, including the Unicode code-points (8 is particularly good at this: all the surrogates are > 127). If it is a CJK code-page, you're screwed no matter what you do. (wide characters, and just what is an "alphabetic" ideograph?)