This is the mail archive of the glibc-bugs@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug stdio/12836] fmemopen(buf,size,"w+b") isn't binary


https://sourceware.org/bugzilla/show_bug.cgi?id=12836

--- Comment #6 from Eric Blake <eblake at redhat dot com> ---
(In reply to paxdiablo from comment #5)
> The POSIX standard calls for fmemopen() to accept all mode strings as
> allowed in fopen() but notes that use of "b" results in
> implementation-defined behaviour.

POSIX has been amended to cater to glibc's behavior of having different
behavior for 'b':
http://austingroupbugs.net/view.php?id=396

And in fact, consensus was that the next version of POSIX should require
glibc's behavior of having saner behavior for 'b':

http://austingroupbugs.net/view.php?id=456

> 
> So, technically, I think an implementation _could_ treat w+b and wb+ as
> different if it so desired but it would have to document that.
> 
> Of course, it would probably be easier to fix the code to make them
> identical :-)

No, we WANT them to be different, per the adjustments to POSIX.

-- 
You are receiving this mail because:
You are on the CC list for the bug.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]