Re: [HACKERS] Warnings in objectaddress.c

On Tue, Oct 3, 2017 at 9:41 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Robert Haas <robertmhaas@gmail.com> writes:
>> Perhaps we should apply some glorified version of this:
>
>> +    if (list_length(object) < 2)
>> +        elog(ERROR, "fail");
>
>> However, I'm not 100% sure that would be sufficient to suppress these
>> warnings, because the compiler has got to be smart enough to know that
>> elog() doesn't return and that i >= 2 is sufficient to guarantee that
>> everything is initialized.
>
> I'm betting it wouldn't help.  I was considering something along the line
> of unrolling the loop:
>
>         Assert(list_length(object) == 2);
>
>         assign typenames[0] and typeoids[0] from linitial(object)
>
>         assign typenames[1] and typeoids[1] from lsecond(object)
>
> This would involve duplicating the loop body, but that's only 3 lines,
> so I think it wouldn't even net out as more code.

Yeah, that's an idea, too.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company


-- 
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers