* Bruce Momjian (pgman@candle.pha.pa.us) wrote:
> > It's been asserted that Coverity can be taught to understand about
> > elog/ereport without this sort of hack, so I'd rather take that tack.
>
> Agreed. The idea of modifying our binary in any way to help a sanity
> tool not complain is totally backwards.
This is very amusing. I have to agree w/ Tom in general, the code in
this case does the right thing and the Coverity tool should be able to
be told about that. However, for areas where the tool is actually right
and there's some bug in Postgres, well, I'd hope we'd modify Postgres to
fix the bug... ;)
Enjoy,
Stephen