Re: pg_restore --no-post-data and --post-data-only
Andrew Dunstan <andrew@dunslane.net>
From: Andrew Dunstan <andrew@dunslane.net>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Josh Berkus <josh@agliodbs.com>, Robert Haas <robertmhaas@gmail.com>, pgsql-hackers <pgsql-hackers@postgresql.org>, Alvaro Herrera <alvherre@commandprompt.com>, Jim Nasby <jim@nasby.net>
Date: 2011-12-07T17:48:26Z
Lists: pgsql-hackers
On 12/07/2011 11:31 AM, Tom Lane wrote: > Josh Berkus<josh@agliodbs.com> writes: >>>> Note that this feature has the odd effect that some constraints are loaded at the same time as the tables and some are loaded with the post-data. This is consistent with how text-mode pg_dump has always worked, but will seem odd to the user. This also raises the possibility of a future pg_dump/pg_restore optimization. >>> That does seem odd. Why do we do it that way? >> Beats me. > Performance, mostly --- we prefer to apply checks during the original > data load if possible, but for indexes and FK constraints it's faster to > apply them later. Also, we can separate constraints from the original > table declaration if it's necessary to break a reference circularity. > This isn't something that would be wise to whack around. > > Yeah, and if we did want to change it that should be a TODO and not hold up this feature. cheers andrew