Hi Eric, On 19 Dec 2006, at 12:22, Eric Blake-1 wrote:
That sounds really tricky. Much simpler is to say: if you care about the return value of a function, do the function before you call 'exit'. Once you're exiting, don't issue any diagnostics: at thatpoint your only goal is to exit cleanly. This is the general strategythat coreutils uses.Here's the patch I'm applying. In non-interactive mode, the error message actually improves - it now tracks the line number of the m4exit that triggered the detection of the disk full write error. But in interactive mode, stdout is unbuffered, hence the write error is detected during the write rather than the fflush, you've already lost the errno long before. But I don't have any problems with that [[snip]]
Nor me. Nice job with the patch!
Cheers,
Gary
--
Gary V. Vaughan ())_. [EMAIL PROTECTED]
Research Scientist ( '/ http://blog.azazil.net
GNU Hacker / )= http://www.gnu.org/software/{libtool,m4}
Technical Author `(_~)_ http://sources.redhat.com/autobook
PGP.sig
Description: This is a digitally signed message part
_______________________________________________ M4-patches mailing list [email protected] http://lists.gnu.org/mailman/listinfo/m4-patches
