"A. Kretschmer" <andreas.kretschmer@schollglas.com> writes:
> there was a question on #postgresql (irc-channel) from CyberDuck without
> an answer. A little example:
Multiple SRFs in a targetlist behave, um, strangely. I think the rule
is that all are cycled until they all chance to report "done" at the
same time, so the number of result rows is the least common multiple
of the period lengths.
This is one of the reasons that SRF-in-targetlist is deprecated ...
regards, tom lane