Also, it seems that everyone ignored the original announcement of the new file. There are some (minor?) issues with it.
If you don't know what I'm talking about, or if you want to help with the issues, I'm reprinting the original notice:
On Monday, June 23, 2003, at 4:43 AM, Daryle Walker wrote:
I added another test file for the I/O-state saving classes. It's in the main Boost CVS as "boost/libs/io/test/ios_state_unit_test.cpp".
I have two compilers installed (GCC 3.1 and CodeWarrior 8). I wrote and tested the new file with CodeWarrior. But my GCC doesn't support the wchar_t part of the libraries, so I can't test with that compiler. Can other people test the file with other compilers to make sure everything is OK?
I want the new file to replace the current test file there, but I want to make sure everything works.
For locale experts and Metrowerks engineers: the unit test for the locale saving class has big comments about a problem with use_facet. Some book I got on IOStreams says that use_facet should do an ID check and a dynamic_cast check, but the CodeWarrior implementation seems to only do the ID check. Should CodeWarrior change to also do the dynamic-check? The standard is vague on this question. (It could just be a QoI issue; maybe there's a DR on it.)
Daryle
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
