On Tue, Jul 23, 2013 at 2:16 PM, Alvaro Herrera <alvhe...@2ndquadrant.com> wrote: > Alvaro Herrera wrote: >> Peter Eisentraut wrote: >> >> > I would suggest that these changes be undone, except that the old >> > "SELECT FOR ..." be replaced by a dynamic string that reverse-parses the >> > LockingClause to provide the actual clause that was used. >> >> Here's a patch for this. > > Pushed to 9.3 and master. Sample output: > > alvherre=# select * from foo, bar for update of foof for share of bar; > ERROR: relation "foof" in FOR UPDATE clause not found in FROM clause > > alvherre=# select * from foo, bar for update of foo for share of barf; > ERROR: relation "barf" in FOR SHARE clause not found in FROM clause > > Amusingly, the only test in which these error messages appeared, in > contrib/file_fdw, was removed after the two commits that changed the > wording. So there's not a single test which needed to be tweaked for > this change.
The fact that there are no tests of this functionality is probably not a good thing. We should add some. At the moment, the following test case crashes, and it looks like this commit is responsible: create table test_update2 (id integer); DECLARE test_update_cursor CURSOR FOR SELECT id, MIN(id) FROM test_update2 GROUP By id HAVING MIN(id) < 1 FOR UPDATE; -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers
