Here is a patch for the following construct, specified in the SQL
collation for (expr)
returns the collation of the argument. It's similar to pg_typeof.
The SQL standard is not clear on the exact format of the return value.
I went with what ruleutils.c produces, which is useful for human
inspection and also for plugging back into SQL statements.
pgsql-hackers by date
|Next:||From: Simon Riggs||Date: 2012-01-08 23:25:05|
|Subject: Re: Page Checksums|
|Previous:||From: Tom Lane||Date: 2012-01-08 22:09:25|
|Subject: Re: SP-GiST versus index-only scans |