| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Ian Turner <iant(at)mail(dot)brainstorm(dot)net> |
| Cc: | Ian Turner <vectro(at)pipeline(dot)com>, pgsql-general(at)hub(dot)org |
| Subject: | Re: function language type? |
| Date: | 2000-07-26 21:29:16 |
| Message-ID: | 12339.964646956@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Ian Turner <iant(at)mail(dot)brainstorm(dot)net> writes:
> Looking at the source, I see the following parsenodes which are NOT
> supported by copyObject:
Uh, what version of the source are you looking at? Quite a few of
those *are* supported.
> Which of these is it worth supporting? I will implement the necessary
> _copy<type> functions.
The missing stuff is basically the 600-series node types; any XXXStmt
node that you want to be able to use in a plpgsql function needs to be
copiable. If you want to support CREATE TABLE you will likely find that
some more of the 700-series nodes are also needed for CREATE TABLE
infrastructure.
It is not worth your trouble to do this unless you are working from
current sources (CVS or a recent daily snapshot)...
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2000-07-26 21:42:09 | Re: performance help |
| Previous Message | Timothy H. Keitt | 2000-07-26 21:26:06 | Re: adding column constraint |