Robert Haas <robertmhaas@gmail.com> writes:
> On Mon, Jan 14, 2013 at 11:56 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> So far as I can tell, get_create_function_cmd (and lookup_function_oid
>> too) were intentionally designed to not show their queries, and for that
>> matter they go out of their way to produce terse error output if they
>> fail. I'm not sure why we should revisit that choice. In any case
>> it seems silly to change one and not the other.
> Agreed on the second point, but I think I worked on that patch, and I
> don't think that was intentional on my part. You worked on it, too,
> IIRC, so I guess you'll have to comment on your intentions.
> Personally I think -E is one of psql's finer features, so +1 from me
> for making it apply across the board.
Well, fine, but then it should fix both of them and remove
minimal_error_message altogether. I would however suggest eyeballing
what happens when you try "\ef nosuchfunction" (with or without -E).
I'm pretty sure that the reason for the special error handling in these
functions is that the default error reporting was unpleasant for this
use-case.
regards, tom lane