| From: | "Joe Conway" <joe(dot)conway(at)mail(dot)com> | 
|---|---|
| To: | <mdavis(at)sevainc(dot)com>, <pgsql-sql(at)postgresql(dot)org> | 
| Subject: | Re: RE: C function for use from PLpgSQL trigger | 
| Date: | 2001-02-05 18:42:45 | 
| Message-ID: | 00c301c08fa3$6edc0c90$c4d210ac@jecw2k1 | 
| Views: | Whole Thread | Raw Message | Download mbox | Resend email | 
| Thread: | |
| Lists: | pgsql-sql | 
> You could send the column name directly into your c function.  For
example:
> c_function_name(NEW.col1, NEW.col2, NEW.col3).  Otherwise I am not sure
how
> to send NEW into a C function.  You could try declaring NEW in your C
> function as a tuple.
Thanks for your reply. I was hoping that I could avoid hardcoding NEW.col1,
etc, so that the function could be used for multiple relations. I've also
tried to declare the input parameter to the function as a tuple, but PLpgSQL
never gets that far -- it doesn't seem to support passing NEW as a
parameter.
Oh, well. I will probably just write all of my logic into a C function and
skip PLpgSQL entirely. That's too bad because it would be far simpler (and
preferrable IMHO) to write a generic trigger function in PLpgSQL and call C
functions for only certain operations that PLpgSQL does not directly
support.
Joe
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2001-02-05 19:35:53 | Re: HELP: Scarey pl/pgsql problem | 
| Previous Message | Albert REINER | 2001-02-05 17:39:19 | Re: How to modify type in table? |