| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Andreas Pflug <Andreas(dot)Pflug(at)web(dot)de> |
| Cc: | pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: pg_get_viewdef 7.4 parentheses |
| Date: | 2003-04-08 23:27:14 |
| Message-ID: | 2940.1049844434@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Andreas Pflug <Andreas(dot)Pflug(at)web(dot)de> writes:
> While implementing a function for making view definitions look prettier
> in pgadmin3, I had a look at ruleutils.c to examine why there are tons
> of parentheses coming out of pg_get_viewdef. I believe that some are
> really not needed.
The difficulty is to know which ones *are* needed in complex
expressions. If you think that any of them are always unnecessary,
you haven't thought hard enough. For example:
> - get_rule_expr, for CASES T_RelabelType and T_CoerceToDomain. The cast
> is always bound to the near left variable, so it can be left out.
What if the argument isn't just a variable?
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Shinji Teragaito | 2003-04-08 23:53:08 | Re: HP-UX 11.22 (IA64) IPF Patch for PostgreSQL 7.3.2 |
| Previous Message | Andrew Sullivan | 2003-04-08 23:05:18 | Re: Anyone working on better transaction locking? |