Thread: proposed todo: use insert/update returning anywhere a table is allowed
It would be great to be able to join to update, returning, etc. It looks like the code was deliberately tied off as-is and I was surprised not to see a todo for this. the basic idea is to be able to do things like: select * from (update foo set id = 1 returning *) q; which currently syntax errors out on 'set' which is a bit odd. here are tom's brief notes on it and also some context on how select combined with insert/update might influence triggers: http://archives.postgresql.org/pgsql-hackers/2006-10/msg00693.php merlin
On Tue, 2007-02-06 at 16:54 -0500, Merlin Moncure wrote: > It would be great to be able to join to update, returning, etc. It > looks like the code was deliberately tied off as-is and I was > surprised not to see a todo for this. > > the basic idea is to be able to do things like: > select * from (update foo set id = 1 returning *) q; > > which currently syntax errors out on 'set' which is a bit odd. > > here are tom's brief notes on it and also some context on how select > combined with insert/update might influence triggers: > > http://archives.postgresql.org/pgsql-hackers/2006-10/msg00693.php > Also a relevent thread: http://archives.postgresql.org/pgsql-general/2006-09/msg00803.php Regards,Jeff Davis
Added to TODO: * Allow INSERT/UPDATE ... RETURNING inside a SELECT 'FROM' clause http://archives.postgresql.org/pgsql-general/2006-09/msg00803.php http://archives.postgresql.org/pgsql-hackers/2006-10/msg00693.php --------------------------------------------------------------------------- Jeff Davis wrote: > On Tue, 2007-02-06 at 16:54 -0500, Merlin Moncure wrote: > > It would be great to be able to join to update, returning, etc. It > > looks like the code was deliberately tied off as-is and I was > > surprised not to see a todo for this. > > > > the basic idea is to be able to do things like: > > select * from (update foo set id = 1 returning *) q; > > > > which currently syntax errors out on 'set' which is a bit odd. > > > > here are tom's brief notes on it and also some context on how select > > combined with insert/update might influence triggers: > > > > http://archives.postgresql.org/pgsql-hackers/2006-10/msg00693.php > > > > Also a relevent thread: > > http://archives.postgresql.org/pgsql-general/2006-09/msg00803.php > > Regards, > Jeff Davis > > > ---------------------------(end of broadcast)--------------------------- > TIP 9: In versions below 8.0, the planner will ignore your desire to > choose an index scan if your joining column's datatypes do not > match -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://www.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +