It just occurred to me that another option, for my specific example, would be to record/cache FOUND instead of testing the RECORD variable for its NULLness. Unless of course assigning FOUND to a variable is a pass-by-reference assignment, which in the actual code that I'm writing would be problematic because FOUND is set many times because there are at least 4 SQL commands that my function executes.
AFAIK, functions don't really do pass by reference, except for INOUT parameters. In any case, FOUND is definitely reset when you do things like SELECT INTO.
Anytime parameters are passed by value. OUT variables are emulated via dynamic composite types. After execution of any SQL statement is FOUND variable refreshed.
Regards
Pavel
-- Jim Nasby, Data Architect, Blue Treble Consulting, Austin TX Experts in Analytics, Data Architecture and PostgreSQL Data in Trouble? Get it in Treble! http://BlueTreble.com
