Re: [HACKERS] 9.0b1: ERROR: btree index keys must be ordered by attribute

2010-06-13 Thread Ian Barwick
2010/6/1 Bruce Momjian br...@momjian.us: Tom Lane wrote: (...) The index-based-max code is throwing in the IS NOT NULL condition without thought for where it has to go in the index condition order. Will look into fixing this tomorrow. FYI, this no longer throws an error in current CVS so was

Re: [HACKERS] 9.0b1: ERROR: btree index keys must be ordered by attribute

2010-05-31 Thread Bruce Momjian
Tom Lane wrote: Ian Barwick barw...@gmail.com writes: Apologies, slight cp error; correct version of query: SELECT ov.object_id FROM object_version ov WHERE ov.object_id = 1 AND ov.version =0 AND ov.object_status_id = ( SELECT MAX(ov1.object_status_id)

[HACKERS] 9.0b1: ERROR: btree index keys must be ordered by attribute

2010-05-09 Thread Ian Barwick
Hi I've just compiled the 9.0 beta1 source tarball and am testing my custom application against it (which has been running on PostgreSQL since 7.3 or so). The below statement results in the following error message: ERROR: btree index keys must be ordered by attribute evidently in relation

Re: [HACKERS] 9.0b1: ERROR: btree index keys must be ordered by attribute

2010-05-09 Thread David Fetter
On Sun, May 09, 2010 at 05:48:27PM +0900, Ian Barwick wrote: Hi I've just compiled the 9.0 beta1 source tarball and am testing my custom application against it (which has been running on PostgreSQL since 7.3 or so). The below statement results in the following error message: ERROR:

Re: [HACKERS] 9.0b1: ERROR: btree index keys must be ordered by attribute

2010-05-09 Thread Ian Barwick
2010/5/9 David Fetter da...@fetter.org: On Sun, May 09, 2010 at 05:48:27PM +0900, Ian Barwick wrote: Hi I've just compiled the 9.0 beta1 source tarball and am testing my custom application against it (which has been running on PostgreSQL since 7.3 or so). The below statement results in the

Re: [HACKERS] 9.0b1: ERROR: btree index keys must be ordered by attribute

2010-05-09 Thread Tom Lane
Ian Barwick barw...@gmail.com writes: 2010/5/9 David Fetter da...@fetter.org: A self-contained way to reproduce this, ideally small, would be fantastic :) s/fantastic/absolutely required to do anything with this report/ I will see if I can knock together a reproducible test case, might take

Re: [HACKERS] 9.0b1: ERROR: btree index keys must be ordered by attribute

2010-05-09 Thread Ian Barwick
Hi 2010/5/10 Tom Lane t...@sss.pgh.pa.us: Ian Barwick barw...@gmail.com writes: 2010/5/9 David Fetter da...@fetter.org: A self-contained way to reproduce this, ideally small, would be fantastic :) s/fantastic/absolutely required to do anything with this report/ Yes, I appreciate that :) I

Re: [HACKERS] 9.0b1: ERROR: btree index keys must be ordered by attribute

2010-05-09 Thread Ian Barwick
2010/5/10 Ian Barwick barw...@gmail.com: SELECT ov.object_id    FROM object_version ov   WHERE ov.object_id = 1     AND ov.version ='0     AND ov.object_status_id = (     SELECT MAX(ov1.object_status_id)       FROM object_version ov1      WHERE ov1.object_id=ov.object_id        AND

Re: [HACKERS] 9.0b1: ERROR: btree index keys must be ordered by attribute

2010-05-09 Thread Tom Lane
Ian Barwick barw...@gmail.com writes: Apologies, slight cp error; correct version of query: SELECT ov.object_id FROM object_version ov WHERE ov.object_id = 1 AND ov.version =0 AND ov.object_status_id = ( SELECT MAX(ov1.object_status_id) FROM object_version ov1