Tom Lane [Wed, 6 Sep 2006 20:40:48 +0000 (20:40 +0000)]
Change processing of extended-Query mode so that an unnamed statement
that has parameters is always planned afresh for each Bind command,
treating the parameter values as constants in the planner. This removes
the performance penalty formerly often paid for using out-of-line
parameters --- with this definition, the planner can do constant folding,
LIKE optimization, etc. After a suggestion by Andrew@supernews.
Fix compiler warnings on 64-bit boxes: difference between
pointers are int64, but warnings are emitted for position info in
error messages in parser, so, just cast it to int32
Tom Lane [Tue, 5 Sep 2006 21:08:36 +0000 (21:08 +0000)]
Get rid of the separate RULE privilege for tables: now only a table's owner
can create or modify rules for the table. Do setRuleCheckAsUser() while
loading rules into the relcache, rather than when defining a rule. This
ensures that permission checks for tables referenced in a rule are done with
respect to the current owner of the rule's table, whereas formerly ALTER TABLE
OWNER would fail to update the permission checking for associated rules.
Removal of separate RULE privilege is needed to prevent various scenarios
in which a grantee of RULE privilege could effectively have any privilege
of the table owner. For backwards compatibility, GRANT/REVOKE RULE is still
accepted, but it doesn't do anything. Per discussion here:
http://archives.postgresql.org/pgsql-hackers/2006-04/msg01138.php
Fix Intel compiler bug. Per discussion
'GIN FailedAssertions on Itanium2 with Intel compiler' in pgsql-hackers,
http://archives.postgresql.org/pgsql-hackers/2006-08/msg01914.php
Tom Lane [Tue, 5 Sep 2006 17:20:29 +0000 (17:20 +0000)]
Remove contrib modules that have been migrated to pgfoundry: adddepend,
dbase, dbmirror, fulltextindex, mac, userlock; or abandoned: mSQL-interface,
tips.
Tom Lane [Mon, 4 Sep 2006 21:15:56 +0000 (21:15 +0000)]
Disallow TRUNCATE when there are any pending after-trigger events for
the target relation(s). There might be some cases where we could discard
the pending event instead, but for the moment a conservative approach
seems sufficient. Per report from Markus Schiltknecht and subsequent
discussion.
Bruce Momjian [Mon, 4 Sep 2006 21:03:18 +0000 (21:03 +0000)]
Sequences were not being shown due to the use of lowercase 's' instead
of 'S', and the views were not checking for table visibility with
regards to temporary tables and sequences.
Tom Lane [Mon, 4 Sep 2006 18:32:55 +0000 (18:32 +0000)]
Change ILIKE to invoke lower() and then do plain LIKE comparison when
working in a multibyte encoding. This fixes the problems exhibited in
bug #1931 and other reports of ILIKE misbehavior in UTF8 encoding.
It's a pretty grotty solution though --- should rethink how to do it
after we install better locale support, someday.
Tom Lane [Mon, 4 Sep 2006 01:26:28 +0000 (01:26 +0000)]
Fix interval input parser so that fractional weeks and months are
cascaded first to days and only what is leftover into seconds. This
seems to satisfy the principle of least surprise given the general
conversion to three-part interval values --- it was an oversight that
these cases weren't dealt with in 8.1. Michael Glaesemann
Tom Lane [Sun, 3 Sep 2006 22:37:06 +0000 (22:37 +0000)]
Code review for UPDATE SET (columnlist) patch. Make it handle as much
of the syntax as this fundamentally dead-end approach can, in particular
combinations of single and multi column assignments. Improve rather
inadequate documentation and provide some regression tests.
Tom Lane [Sun, 3 Sep 2006 19:06:15 +0000 (19:06 +0000)]
Fix non-improvement of description of archive_timeout. archive_command
is only invoked on completed WAL segments, period --- there's no 'by
default' about it.
Tom Lane [Sun, 3 Sep 2006 15:59:39 +0000 (15:59 +0000)]
Arrange for GetSnapshotData to copy live-subtransaction XIDs from the
PGPROC array into snapshots, and use this information to avoid visits
to pg_subtrans in HeapTupleSatisfiesSnapshot. This appears to solve
the pg_subtrans-related context swap storm problem that's been reported
by several people for 8.1. While at it, modify GetSnapshotData to not
take an exclusive lock on ProcArrayLock, as closer analysis shows that
shared lock is always sufficient.
Itagaki Takahiro and Tom Lane
Bruce Momjian [Sat, 2 Sep 2006 23:47:39 +0000 (23:47 +0000)]
Add:
> * Simplify ability to create partitioned tables
>
> This would allow creation of partitioned tables without requiring
> creation of rules for INSERT/UPDATE/DELETE, and constraints for
> rapid partition selection. Options could include range and hash
> partition selection.
>
> * Allow auto-selection of partitioned tables for min/max() operations
Bruce Momjian [Sat, 2 Sep 2006 23:12:16 +0000 (23:12 +0000)]
Make autovacuum behavior more agressive, per discussion on hackers list
--- was part of autovacuum default 'on' patch that was reverted, but we
want this part.
Bruce Momjian [Sat, 2 Sep 2006 20:31:50 +0000 (20:31 +0000)]
Update:
< Last updated: Sat Sep 2 08:31:04 EDT 2006
> Last updated: Sat Sep 2 16:31:46 EDT 2006
< o Allow UPDATE tab SET ROW (col, ...) = (...) for updating multiple
< columns
> o -Allow UPDATE tab SET ROW (col, ...) = (val, ...) for updating
> multiple columns
> o Allow UPDATE tab SET ROW (col, ...) = (SELECT...)
< A subselect can also be used as the value source.
Tom Lane [Sat, 2 Sep 2006 17:06:52 +0000 (17:06 +0000)]
Apply a simple solution to the problem of making INSERT/UPDATE/DELETE
RETURNING play nice with views/rules. To wit, have the rule rewriter
rewrite any RETURNING clause found in a rule to produce what the rule's
triggering query asked for in its RETURNING clause, in particular drop
the RETURNING clause if no RETURNING in the triggering query. This
leaves the responsibility for knowing how to produce the view's output
columns on the rule author, without requiring any fundamental changes
in rule semantics such as adding new rule event types would do. The
initial implementation constrains things to ensure that there is
exactly one, unconditionally invoked RETURNING clause among the rules
for an event --- later we might be able to relax that, but for a post
feature freeze fix it seems better to minimize how much invention we do.
Per gripe from Jaime Casanova.
Bruce Momjian [Sat, 2 Sep 2006 13:12:50 +0000 (13:12 +0000)]
Add new variable "server_version_num", which is almost the same as
"server_version" but uses the handy PG_VERSION_NUM which allows apps to
do things like if ($version >= 80200) without having to parse apart the
value of server_version themselves.
Tom Lane [Thu, 31 Aug 2006 17:31:33 +0000 (17:31 +0000)]
Clean up rather sloppy fix in HEAD for the ancient bug that CREATE CONVERSION
didn't create a dependency from the new conversion to its schema. Back-patch
to all supported releases.
Bruce Momjian [Thu, 31 Aug 2006 00:34:34 +0000 (00:34 +0000)]
Done:
< Last updated: Tue Aug 29 12:21:52 EDT 2006
> Last updated: Wed Aug 30 20:34:28 EDT 2006
< o Allow COPY (SELECT ...) TO 'filename'
<
< COPY should also be able to output views using COPY (SELECT
< * FROM view) TO 'filename' internally.
< http://archives.postgresql.org/pgsql-patches/2005-09/msg00148.php
> o -Allow COPY (SELECT ...) TO 'filename'
Tom Lane [Wed, 30 Aug 2006 18:06:27 +0000 (18:06 +0000)]
Fix things so that fopen's, not only open's, pass FILE_SHARE_DELETE
and other special flags on Windows. May fix intermittent 'Permission
denied' errors. Magnus Hagander
Tom Lane [Tue, 29 Aug 2006 22:25:08 +0000 (22:25 +0000)]
Create a FETCH_COUNT parameter that causes psql to execute SELECT-like
queries via a cursor, fetching a limited number of rows at a time and
therefore not risking exhausting memory. A disadvantage of the scheme
is that 'aligned' output mode will align each group of rows independently
leading to odd-looking output, but all the other output formats work
reasonably well. Chris Mair, with some additional hacking by moi.
Tom Lane [Tue, 29 Aug 2006 15:19:51 +0000 (15:19 +0000)]
Invent an assign-hook mechanism for psql variables similar to the one
existing for backend GUC variables, and use this to eliminate repeated
fetching/parsing of psql variables in psql's inner loops. In a trivial
test with lots of 'select 1;' commands, psql's CPU time went down almost
10%, although of course the effect on total elapsed time was much less.
Per discussion about how to ensure the upcoming FETCH_COUNT patch doesn't
cost any performance when not being used.
Bruce Momjian [Tue, 29 Aug 2006 02:11:30 +0000 (02:11 +0000)]
Now bind displays prepare as detail, and execute displays prepare and
optionally bind. I re-added the "statement:" label so people will
understand why the line is being printed (it is log_*statement
behavior).
Use single quotes for bind values, instead of double quotes, and double
literal single quotes in bind values (and document that). I also made
use of the DETAIL line to have much cleaner output.
Bruce Momjian [Mon, 28 Aug 2006 23:21:46 +0000 (23:21 +0000)]
Move to referential integrity section:
> o Allow DEFERRABLE and end-of-statement UNIQUE constraints?
>
> This would allow UPDATE tab SET col = col + 1 to work if col has
> a unique index. Currently, uniqueness checks are done while the
> command is being executed, rather than at the end of the statement
> or transaction.
>
<
< * Allow DEFERRABLE and end-of-statement UNIQUE constraints?
<
< This would allow UPDATE tab SET col = col + 1 to work if col has
< a unique index. Currently, uniqueness checks are done while the
< command is being executed, rather than at the end of the statement
< or transaction.