The "Dataflow Iterators" section of the serialization library contains iterators for this purpose. Its been part of boost since last November. Robert Ramey Daryle Walker wrote:
On 8/19/05 9:15 AM, "Vladimir Prus" <ghost@cs.msu.su> wrote:
Daryle Walker wrote:
Nothing fancy, just something like:
int_fast32_t char_to_Unicode( char c ); int_fast32_t wchar_to_Unicode( wchar_t c );
that converts a native character to a Unicode value.
Maybe, but it's hard to comment as you haven't even explained what those function will do. What's a "native character" and what a "Unicode value" and how the conversion will be done? If the first function does conversion from local 8 bit encoding to unicode then:
"Native characters" are the character set a particular platform uses. Before the Unicode era, a platform (could) assume that all text files used the platform's character set. (e.g. MacRoman for pre-X Macs, Cp-1251 for Windows, Latin-1 for UNIX) My functions assume a one-to-one mapping from a native character to a Unicode code-point, because Phase 1 of C++ translation (see section 2.1 of the standard) assumes that.
- do you have a working implementation?
No. I'm just requesting for comments.
- isn't dealing with individual characters too slow?
Probably. Maybe we could add an iterator-copying version.