Re: attribute warn_unused_result

[Paul Eggert]

On 02/04/2011 12:41 AM, Eli Zaretskii wrote:
> The -Wno-unused-result compiler options looks easy to me, and is IMO
> cleaner than using ignore-value.h.  Heck, even the corresponding
> pragma in the affected source files looks cleaner.

Unfortunately those approaches shut off all instances
of the warnings, even the useful ones.  I wouldn't have
found the eight bugs fixed in BZR 102952 if I hadn't
had those warnings.

Like I said, though, I don't much like ignore_value,
and either of the above approaches would also be fine
with me.