pgsql@mohawksoft.com wrote:
> >
> > Where are we on this?
>
> That's a good question.
>
> Tom doesn't like the syntax of "include" and there are a couple bugs he is
> concered it.
>
> I'm pretty agnostic about the syntax, but I wouldn't get overly worried
> about the metaphor presented either.
>
> "include='...'" doesn't bother me at all, but some people have a problem
> with it.
>
> Then there is the design of using a callable function for a configuration
> parameter, personally, I think this feature is useful for the future, Tom
> seems to have a problem it it.
>
> After that, the discussion sort of ends.
>
>
> I'm willing to adress the bugs.
> I don't think the syntax is a huge deal, IMHO at most it is a
> documentation problem.
Well, it seems pretty clear were we need to go on this. First, we could
just add the documentation to the non-include part of the patch based on
the version I posted and apply that to be sure it gets into 7.5.
Then, for "include", it needs to be an operation and not a variable,
probably something in guc-file.l.
It should not use an equals and not be something you can say SHOW with.
-- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610)
359-1001+ If your life is a hard drive, | 13 Roberts Road + Christ can be your backup. | Newtown Square,
Pennsylvania19073