On 18/09/14 06:56, Reinhard Tartler wrote: > On Thu, Sep 18, 2014 at 12:54 AM, Luca Barbato <[email protected]> wrote: >> On 18/09/14 06:36, Reinhard Tartler wrote: >>> What would be a better reporting mechanism in this case? >> >> Options: >> >> (void) cast to make apparent we do not care about the value >> >> pass the context to avpriv_open and use av_streerror to report _why_ it >> failed. >> >> and whatever would make this call fail would make the open above fail as >> well. >> >> Courmish, which system does support FD_CLOEXEC and does not support >> O_CLOEXEC ? >> >> I'd drop that additional syscall and just use the O_ option. >> >> Making coverity happy is nice, but I'd like to take the reports as a >> mean to improve not just by its statistics. > > sure, feel free to take over this patch then.
Probably I can add (void) casts everywhere it is needed in a single patch, before I spend time on that, is anybody against that? The other viable option is suppress such reports as they come. Why do: - make coverity happier - make compilers reporting it happier Why not: - it adds 6 chars - it is mainly noise. lu _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
