Re: Bug? Function with side effects not evaluated in CTE

[Moshe Jacobson <moshe@xxxxxxxxxxxx>]

On Fri, Oct 18, 2013 at 5:08 PM, David Johnston <polobo@xxxxxxxxx> wrote:

this thread hasn't really provided a compelling use-case for making a change: the example
provided is too contrived.

It seems contrived because I distilled it down from what it originally was. 

There is an actual use case in which this was failing:

We use stored procs to provide us the functionality of an UPSERT, which PostgreSQL lacks.

We are using this in the first CTE to create new entries in a table, and we are using the DELETE to delete the entries that already existed that we didn't just create.
I am of the belief that if the function in a CTE is volatile, that it should be executed unconditionally.

Moshe Jacobson

Manager of Systems Engineering, Nead Werx Inc.

2323 Cumberland Parkway · Suite 201 · Atlanta, GA 30339

"Quality is not an act, it is a habit." -- Aristotle