>>> >
>>> >A possible kluge answer is to make pg_dump's OID-ordering of views
>>> >depend on the OID of the view rule rather than the view relation.
>>> >I am not sure if that would break any cases that work now, however.
>>> >
>>>
>>> This seems good to me; it should be based on the 'oid of the view', and
>>> AFAICT, the rule OID should be it. Should I do this?
>>
>>The view oid is certainly better than the base relation oid.
>>
>
>Since I'm in pg_dump at the moment, I'll make the change...
>
Having now looked at pg_dump more closely, I'm not at all sure I want to
make the change directly in pg_dump. The reason is that I am trying to move
version-specific stuff from pg_dump, and I currently get a view definition
by doing 'select pg_getviewdef(<name>)' (rather than looking up the rule etc).
Would people mind me adding a 'pg_getviewoid(<name>)' for pg_dump's use?
(especially since 7.1 now seems to be out...)
----------------------------------------------------------------
Philip Warner | __---_____
Albatross Consulting Pty. Ltd. |----/ - \
(A.B.N. 75 008 659 498) | /(@) ______---_
Tel: (+61) 0500 83 82 81 | _________ \
Fax: (+61) 0500 83 82 82 | ___________ |
Http://www.rhyme.com.au | / \| | --________--
PGP key available upon request, | /
and from pgp5.ai.mit.edu:11371 |/
