Ok, we got strnlen() some time back.  Shouldn't we have wcsnlen() too,
for completeness and orthogonality's sake?  Since wmemchr() exists, this
should I think be a straightforward derivation from the existing strnlen() code;
just change memchr() to wmemchr(), change the include files string.h and
sys/types.h to wchar.h, and change the comments appropriately.  Similar
modifications to testing code would probably also do the trick (I don't have
what I did for that right in front of me, but it certainly wasn't rocket 
science).
I haven't tried it yet, but it really looks almost too simple not to work!

For Austin Group and LSB status, google for something like:

wcsnlen site:opengroup.org OR site:linux-foundation.org

Note that without the site restrictions, one sees that others (e.g. Microsoft)
have implementations, too.  As long as it looks like there's no reason for
future POSIX standardization to jerk this around, I for one can't think of a
good reason why we shouldn't have as many as possible of these things that
are already in stable Austin Group proposals and also in the LSB; it would
probably make porting code a good deal easier for people.

Finally, is anyone thinking about functions (if any) that are in the LSB but 
_not_
in (nor incompatible with) SUSv3 or Austin Group proposals?  In a perfect
world, IMO there'd be a plan for all that, and we'd all know what it was.
 
 
This message posted from opensolaris.org
_______________________________________________
opensolaris-code mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/opensolaris-code

Reply via email to