Re: Review: Extensions Patch
Dimitri Fontaine <dimitri@2ndquadrant.fr>
From: Dimitri Fontaine <dimitri@2ndQuadrant.fr>
To: Dimitri Fontaine <dimitri@2ndQuadrant.fr>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Robert Haas <robertmhaas@gmail.com>, "David E. Wheeler" <david@kineticode.com>, PostgreSQL-development Hackers <pgsql-hackers@postgresql.org>
Date: 2010-12-08T22:45:01Z
Lists: pgsql-hackers
Dimitri Fontaine <dimitri@2ndQuadrant.fr> writes: > Well it does not seem to be complex to code. It's about having a new > property in the control file, relocatable, boolean. This property is > required and controls the behavior of the CREATE EXTENSION ... WITH > SCHEMA command. When true we use the ALTER EXTENSION SET SCHEMA code > path and when false, the placeholder replacement code path. The ALTER > command has already been developed so I need to merge it into the main > patch. Ok I've done that on the git branch, for people interested into having a look or playing with it before the week-end, when I think I'll post the new patch revision. Well I've left alone the behavior change at CREATE EXTENSION time, and also, well, the necessary documentation. http://git.postgresql.org/gitweb?p=postgresql-extension.git;a=summary http://git.postgresql.org/gitweb?p=postgresql-extension.git;a=commitdiff;h=c31d8a7728706d539f50d764fe8f45db92869664 > The ALTER EXTENSION SET SCHEMA command needs to be adapted so that it > checks that all the extension's objects are currently in the same schema > and error out if that's not the case. Done in the commit above. WIP of course, but just so that commit fest manager notice things are moving here. Regards, -- Dimitri Fontaine http://2ndQuadrant.fr PostgreSQL : Expertise, Formation et Support