Re: Re: [BUGS] BUG #4027: backslash escapingnotdisabledinplpgsql

Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov> writes:
>> a change to CREATE FUNCTION such that there is an implied SET
>> standard_compliant_strings FROM CURRENT

Hopefully obvious, I meant standard_conforming_strings.

> it seems like a really bad idea.

Then perhaps a note in the PL/pgSQL docs about the importance of
specifying that clause if the function contains any character string
literals which include a backslash? Such a note should probably point
out that without this clause, the runtime value of any such literal
will be dependent on the value of standard_conforming_strings when the
plan is generated.

I think that many will find that behavior surprising; so if it's not
feasible to change it, we should at least document it.

-Kevin