theory
WIP: Add `procedures are()`
code, test and doc overall ok, though there are some other references to functions_are() that I didn't explore yet.
Very thorough, but I think we should consider a deprecation period and a warning. Thoughts?
Yes I completely agree, though at the same time I'm facing another one with triggers, so maybe better keep functions_are as it is and just add a new set with an extra attribute (first), like ... mmmh a char ?! functions_are('f',....)
Oh! ISTR that I've been thinking about has_functions() as applying to all functions, and then one can do separate calls to is_procedure(), is_aggregate(), is_window() to determine each one specifically. But I also see that there is no is_function(), but there is has_function(). This is confusing!
This could use some normalization. I think, ideally, we'd have has_{thing}s() and is_{thing}() for each type of function. Seems do-able, though has_function() would be limited as you have it here. I think maybe that's okay? We can rename the current has_functions() to has_routines() (the docs say "routine" is the collective term). Maybe we can have has_functions() go with the 'f' as in this patch, but then, for some period, if it fails and it looks like there are routines other than functions, emit a warning/diagnostics that the behavior has changed and run the test again without the 'f'.
Too complicated?
