Re: [badalex@gmail.com: Re: [BUGS] Problem identifying constraints which should not be inherited]

On Tue, Apr 1, 2008 at 2:40 AM, NikhilS <nikkhils(at)gmail(dot)com> wrote:
> PFA, a revised version of Alex' patch. I have added the relevant pg_dump
> related changes too. As I mentioned earlier, I still don't know whether
> Alex' syscache related changes are necessary, but I will leave it to the
> patch reviewers to decide :)
>
> This combined patch meets the following TODOs:
>
> * Add logic to disallow ADD CONSTRAINT ONLY to parent of an inheritance
> hierarchy
>
> * Add logic to mark inherited constraints in the children:
> This is achieved by introducing bool "conislocal" and int4 "coninhcount"
> attributes in pg_constraint. The behaviour of these 2 attributes is pretty
> similar to attislocal and attinhcount attributes in pg_attribute.
>
> * Add logic to disallow dropping inherited constraints directly on children
> Obviously they will get dropped if a DROP CONSTRAINT is fired on the
> parent. with recurse set to true (this is the default behaviour)
>
> * Modify the pg_dump logic to use the new pg_constraint based attributes
> logic for versions above 80300 (or should it be PG_VERSION_NUM 80400?).
>
> Please direct comments/feedback towards both me and Alex.
>
>
> Regards,
> Nikhils
> --
> EnterpriseDB http://www.enterprisedb.com

Find attached v4 of the patch, changes since v3:

-updated/added catalog documentation for new coninhcount and conislocal columns.
-fixed some regression tests order by (causing them to fail randomly)
-some coding style fixes
-removed some unused vars

Please direct comments/feedback towards both me and NikhilS.