On Thursday 09 March 2006 04:18, Lumir Vanek wrote: > Agrree. This is clearly described here: > http://xml.apache.org/xerces-c/build-misc.html#XMLChInfo > Thank you. I had been lookind for that when I started this thread. I didn't realize it was under the build FAQ. It would appear that the C++ correct approach would be to create a C++ string class for XMLCh and provide corresponding codecvt functionality to go between the native encoding and UTF-16. I posted a message on comp.std.c++ suggesting that UTF support be specifically be required for UTF-8, UTF-16 and UTF-32 in the next C++.
I am confident that the goal of the C++ Standards Committee was to avoid restricting the implementation to any given character encoding. The goal of the W3C was certainly to ensure interoperability between systems. I'm still a bit unsure as to where the problem arrises. Xerces shouldn't care what the internal data representation is, so long as UTF-16 is read in correctly, and exported correctly. To my way of thinking, something is clearly wrong with the current state of affairs. Where exactly the problem lies is not exactly clear. I suspect, however, the problem is with the C++ Standard. Steven --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
