Bonjour Michaël,
>> Except for the addition of a test case to skip empty results when
>> \aset is used, I think that we are pretty good here.
>
> While hacking on the patch more by myself, I found that mixing tests
> for \gset and \aset was rather messy. A test for an empty result
> leads also to a failure with the pgbench command as we want to make
> sure that the variable does not exist in this case using debug().
ISTM that I submitted a patch to test whether a variable exists in
pgbench, like available in psql (:{?var} I think), but AFAICS it did not
pass. Maybe I should resurect it as it would allow to test simply whether
an empty result was returned to aset, which could make sense in a bench
script (get something, if it does not exist skip remainder… I can see some
interesting use cases).
> So let's split the tests in three parts:
> - the set for \get is left alone.
> - addition of a new set for the valid cases of \aset.
> - addition of an invalid test for \aset (the empty set one).
Ok.
> Fabien, what do you think about the attached?
It does not need to create an UNLOGGED table, a mere "WHERE FALSE"
suffices.
I do not understand why you removed the comment about meta which makes it
false, so I added something minimal back.
> Perhaps we should also have a test where we return more than 1 row for
> \get? The last point is unrelated to this thread though.
Yes, but ISTM that it is not worth a dedicated patch… so I added a test
similar to the one about empty aset.
See v7 attached.
--
Fabien.
