On Tue, 2011-02-01 at 11:01 -0600, Kevin Grittner wrote:
> My compiler doesn't.
Strange. Maybe it requires -O2?
> Would it make sense to elog here, rather than
> Assert? I'm not clear on the rules for that.
elog looks fine there to me, assuming we have the default case. I'm not
100% clear on the rules, either. I think invalid input/corruption are
usually elog (so they can be caught in non-assert builds); but other
switch statements have them as well ("unrecognized node...").
> A small push dealing with all the above issues and adding a little
> to comments:
>
>
http://git.postgresql.org/gitweb?p=users/kgrittn/postgres.git;a=commitdiff;h=538ff57691256de0341e22513f59e9dc4dfd998f
>
> Let me know if any of that still needs work to avoid confusion and
> comply with PostgreSQL coding conventions. Like I said, I'm not
> totally clear whether elog is right here, but it seems to me a
> conceptually similar case to some I found elsewhere that elog was
> used.
Looks good. It also looks like it contains a bugfix for subtransactions,
right?
Regards,Jeff Davis