is there a way to find exactly what object/table/column creates this
circular dependency?
pg_dumb even in verbose mode does not warn/error of this situation. I
managed to find this by testing copy of production db(this should be
reported by pg_dump). also there is nothing fancy about this 'view' of mine.
in db it looks like this:
t1, t2, t3.......t10
create view1 from t1-t10
create view2 from view1 (only agregation with group by).
pg_dump creates table from view1.
best regards,
maciej lopuszanski
W dniu 31.01.2013 20:27, Tom Lane pisze:
lopuszan...@oleofarm.com writes:
1. after using pg_dump to dump WHOLE database to file, 1 of views 'turned'
into a table.
so there is no 'create or replace VIEW XXXX .......' with definition, but
instead:
its scripted as 'create TABLE XXXX......' and definition.(in file that
pg_dump created)
There's probably a CREATE RULE command further down that turns it into a
view again.
2. but after using pg_dump to dump only this 1 problematic 'view' (with -t
option) it stores it correcly as 'create or replace VIEW XXXX .......'
That's not terribly surprising. The separate table and rule entries
would only be used when it's necessary to work around a circular
dependency with some other object.
it looks like pg_dump has problem with dependancies of this view.....
I see no reason to think there's a bug here. If you still think it's
a bug, you need to provide enough information for someone else to
reproduce the case.
http://www.postgresql.org/docs/9.1/static/bug-reporting.html
regards, tom lane
--
Sent via pgsql-bugs mailing list (pgsql-bugs@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-bugs