Re: vacuum and row type
Tom Lane <tgl@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Teodor Sigaev <teodor@sigaev.ru>
Cc: Pgsql Hackers <pgsql-hackers@postgresql.org>
Date: 2011-06-02T16:39:54Z
Lists: pgsql-hackers
Teodor Sigaev <teodor@sigaev.ru> writes: >> I think we could just let this code assume success for type RECORD. It >> won't affect VACUUM/ANALYZE, since there are (for reasons that should >> now be obvious) no table or index columns of anonymous composite types. > Of course, it's impossible to store anonymous composite type anywhere, but > we still have possibility to use it in ORDER BY at least, following query works > on HEAD but fails with patch: > select ROW(1, n) as r from generate_series(1,5) as n order by r; Right, so for type RECORD we should let the parser assume that comparisons will work. If the anonymous composite type isn't actually sortable, it'll fail at runtime, same as now. regards, tom lane