Hrm, well, proof's in the pudding I guess. I went back and rebuilt with
pg-hier after reading this and I'm seeing the results you expect, seems
to be working. I don't know what I did yesterday, but I wasn't paying
attention somewhere. Currently working on backing up a test db, then
will try the production if successful. I'll let you guys know.
Thanks again!
Gerald
Alvaro Herrera wrote:
>On Fri, Jun 03, 2005 at 05:55:36PM -0500, Gerald D. Anderson wrote:
>
>
>>Ok, you're over my head now so I'm just going to post the results in the
>>hope that something clicks. Here ya go:
>>
>>
>
>Ok, I diff'ed both and guess what:
>
>("a" is your output, "b" is mine)
>
>07:19 drizzt ~ 0$ diff a b
>38c38
>< :limitCount <> :hierClause <> :setOperations <> :resultRelations ()})
>---
>
>
>>:limitCount <> :setOperations <> :resultRelations ()})
>>
>>
>
>So we now have confirmed that the problem is the hierarchical queries
>patch.
>
>
>
>>That help? *cringe*
>>
>>
>
>I guess your build with the "use flag" wasn't successful. I think you
>have two choices:
>
>1. really build with the patch installed, and dump your data using that
>
>2. Make an update of pg_rewrite in order to get rid of that. I think
>this is easier, but you should be very careful not to corrupt the table.
>I'd start by doing a CREATE TABLE AS to copy the contents of pg_rewrite
>somewhere else, just in case you screw it up. Then come up with some
>clever UPDATE query to remove the extraneous ":hierClause <> " part.
>
>
>