small project

[Ulrich Drepper <drepper@cygnus.com>]

Related to the wide character streams there is a small project
somebody could tackle.

Currently, the only way to determine the character set associated with
a wide char stream is to set a locate.  This might not be possible all
the time, e.g., in multi-threaded applications.  Therefore there
should be another possiblity.

What I have in mind is an extension to fopen.  The second parameter is
designed with extensions in might.  ISO C allows this.  Therefore I'd
like to see something like this:

	fp = fopen ("somefile", "r,ccs=XYZ");

This would open the file as usual, but also orient the string to be
wide and instead of using the conversion functions from the current
locale, load the functions and create the data structures so that
charset XYZ is used.


Any volunteers?  The needed code should all be there already, you
simply would have to take it from the different places.  Please let me
know before starting on this.

Thanks,

-- 
---------------.      drepper at gnu.org  ,-.   1325 Chesapeake Terrace
Ulrich Drepper  \    ,-------------------'   \  Sunnyvale, CA 94089 USA
Cygnus Solutions `--' drepper at cygnus.com   `------------------------