Philippe Verdy wrote: > So you can have a wchar_t datatype in C/C++ that stores UCS-4, but > your strings will most often not be arrays of wchar_t but of an > intermediate 16-bit size which gets parsed to 32-bit wchar_t by > very simple run-time scanners.
Gee, I want to add run-time converters embedded into the heart of my application. It makes things so much simpler. > APIs that really use 32-bit chars to represent strings are quite > rare and in fact not needed, as UTF-16 strings will perform better. All of his examples used 32-bit words to represent strings, despite your hand-wavings. There is no evidence that gunichar or wchar_t aren't meant to store strings, general old multi-purpose strings. To boot, you've haven't given a single example of a UTF-16 API. So why should we think is one is rare, when we have ten examples of it and none of the other? Lastly, how have you conducted your tests to find if UTF-16 strings perform better? On what platforms, using what code, doing what? -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm
