OSDN Git Service
Joe Conway [Fri, 19 Dec 2003 00:02:11 +0000 (00:02 +0000)]
Use a shutdown callback to ensure proper clean up when rescanning
partially-evaluated SRFs. Per report found here:
http://archives.postgresql.org/pgsql-general/2003-12/msg00851.php
Tom Lane [Thu, 18 Dec 2003 22:49:26 +0000 (22:49 +0000)]
Fix memory leak with SSL connections due to missing X509_free() calls.
Per Neil Conway.
Tom Lane [Thu, 18 Dec 2003 22:23:42 +0000 (22:23 +0000)]
Use a shutdown callback to clear setArgsValid in a FuncExprState that is
evaluating a set-valued function. This fixes some additional problems
with rescanning partially-evaluated SRFs.
Peter Eisentraut [Thu, 18 Dec 2003 20:25:58 +0000 (20:25 +0000)]
Forgot to change one compatlib.h.
Tom Lane [Thu, 18 Dec 2003 20:21:37 +0000 (20:21 +0000)]
Ensure set-returning functions in the targetlist of a plan node will be
shut down cleanly if the plan node is ReScanned before the SRFs are run
to completion. This fixes the problem for SQL-language functions, but
still need work on functions using the SRF_XXX() macros.
Teodor Sigaev [Thu, 18 Dec 2003 19:27:53 +0000 (19:27 +0000)]
Fix signed char in comparison and check memory allocation
Peter Eisentraut [Thu, 18 Dec 2003 18:55:09 +0000 (18:55 +0000)]
Move Informix compatibility include files out of the way. compatlib.h
was integrated into ecpg_informix.h, the other ones go into their own
subdirectory that is automatically considered by the embedded preprocessor
when in Informix mode.
Dave Cramer [Thu, 18 Dec 2003 04:17:17 +0000 (04:17 +0000)]
patch for new OID74Test
Dave Cramer [Thu, 18 Dec 2003 04:08:30 +0000 (04:08 +0000)]
fixed up OID74 test to conform with other tests, by Kris Jurka
Bruce Momjian [Thu, 18 Dec 2003 03:59:07 +0000 (03:59 +0000)]
Please find enclosed a patch exemplifying typical use of the ARE
Class-Shorthand Escapes<C2><AE>. I believe it will help intrepid regex
users. :)
David Fetter
Bruce Momjian [Thu, 18 Dec 2003 03:46:45 +0000 (03:46 +0000)]
Here is the definition of relation_byte_size() in optimizer/path/costsize.c:
----------------------------------------------------------------------
/*
* relation_byte_size
* Estimate the storage space in bytes for a given number of tuples
* of a given width (size in bytes).
*/
static double
relation_byte_size(double tuples, int width)
{
return tuples * (MAXALIGN(width) + MAXALIGN(sizeof(HeapTupleData)));
}
----------------------------------------------------------------------
Shouldn't this be HeapTupleHeaderData and not HeapTupleData ?
(Of course, from a costing perspective these shouldn't be very different but ...)
Sailesh Krishnamurthy
Dave Cramer [Thu, 18 Dec 2003 03:27:15 +0000 (03:27 +0000)]
patch by Kris Jurka to use the correct protocol based upon server information
Tom Lane [Thu, 18 Dec 2003 01:48:53 +0000 (01:48 +0000)]
Adjust rules output for unknown-vs-any change (affects expected contents
of pg_stats view definition).
Tom Lane [Thu, 18 Dec 2003 00:22:12 +0000 (00:22 +0000)]
Be a little smarter in group_clauses_by_indexkey_for_join: detect cases
where a joinclause is redundant with a restriction clause. Original coding
believed this was impossible and didn't need to be checked for, but that
was a thinko ...
Tom Lane [Wed, 17 Dec 2003 22:11:30 +0000 (22:11 +0000)]
information_schema.constraint_column_usage and key_column_usage should
not discriminate against system columns, since we support constraints on
system columns, and in fact constraints on OID are moderately useful.
Tom Lane [Wed, 17 Dec 2003 21:45:44 +0000 (21:45 +0000)]
Fix DecodeInterval to handle '-0.1' sanely, per gripe from Tilo Schwarz.
Tom Lane [Wed, 17 Dec 2003 20:15:41 +0000 (20:15 +0000)]
Repair badly broken estimation of output buffer size in lquery_out().
Tom Lane [Wed, 17 Dec 2003 19:49:39 +0000 (19:49 +0000)]
Reorder tests in parse_coerce so that ANY/ANYELEMENT/ANYARRAY coercion
does not affect UNKNOWN-type literals or Params. This fixes the recent
complaint about count('x') being broken, and improves consistency in
a few other respects too.
Peter Eisentraut [Wed, 17 Dec 2003 18:44:09 +0000 (18:44 +0000)]
Remove pg_id.
Tom Lane [Wed, 17 Dec 2003 17:07:48 +0000 (17:07 +0000)]
Repair planner failure when there are multiple IN clauses, each with
a join in its subselect. In this situation we *must* build a bushy
plan because there are no valid left-sided or right-sided join trees.
Accordingly, hoary sanity check needs an update. Per report from
Alessandro Depase.
Dave Cramer [Wed, 17 Dec 2003 15:45:05 +0000 (15:45 +0000)]
revoked patch from Kris Jurka to fix multiarguments, and changed test to create
a temp table
Dave Cramer [Wed, 17 Dec 2003 15:38:42 +0000 (15:38 +0000)]
patch from Kris Jurka to fix large object 7.1 compatible protocol issues
modified test case from Alexey Yudichev to be part of the testsuite
Michael Meskes [Wed, 17 Dec 2003 15:23:45 +0000 (15:23 +0000)]
- Added just another patch by Dave that fixes a reversed order in
variable listing for output variables in cursor definitions
- Fixed incorrect if call in long=>numeric conversion.
Dave Cramer [Wed, 17 Dec 2003 13:25:14 +0000 (13:25 +0000)]
added polish translation submitted by Piotr Maj
Bruce Momjian [Wed, 17 Dec 2003 07:07:00 +0000 (07:07 +0000)]
Restore information schema upgrade instructions of Peter.
Peter Eisentraut [Tue, 16 Dec 2003 15:27:58 +0000 (15:27 +0000)]
Add example for converting epoch back to timestamp.
Peter Eisentraut [Tue, 16 Dec 2003 14:57:20 +0000 (14:57 +0000)]
Fix constraint_column_usage for foreign keys.
Peter Eisentraut [Tue, 16 Dec 2003 09:47:55 +0000 (09:47 +0000)]
Repair name.
Bruce Momjian [Tue, 16 Dec 2003 01:35:40 +0000 (01:35 +0000)]
Add mention of non-standard extension:
< o Make SET CONNECTION thread-aware
> o Make SET CONNECTION thread-aware, non-standard?
Bruce Momjian [Tue, 16 Dec 2003 01:08:40 +0000 (01:08 +0000)]
Add for ecpg:
> o Make SET CONNECTION thread-aware
Bruce Momjian [Tue, 16 Dec 2003 01:06:39 +0000 (01:06 +0000)]
Mention ecpg SET CONNECTION is not thread-aware.
Peter Eisentraut [Mon, 15 Dec 2003 23:58:12 +0000 (23:58 +0000)]
Override some of the changes in DocBook DSSSL stylesheets 1.78 to restore
previous behavior of <literal>, <envar>, <acronym>, and others.
Bruce Momjian [Mon, 15 Dec 2003 22:56:44 +0000 (22:56 +0000)]
Add mention with might need to use cp -R someday for portability.
Bruce Momjian [Mon, 15 Dec 2003 22:24:59 +0000 (22:24 +0000)]
Update HISTORY and release notes for 7.4.1.
Neil Conway [Sun, 14 Dec 2003 00:55:46 +0000 (00:55 +0000)]
Fix two typos in the documentation for PREPARE.
Neil Conway [Sun, 14 Dec 2003 00:34:47 +0000 (00:34 +0000)]
I posted some bufmgr cleanup a few weeks ago, but it conflicted with
some concurrent changes Jan was making to the bufmgr. Here's an
updated version of the patch -- it should apply cleanly to CVS
HEAD and passes the regression tests.
This patch makes the following changes:
- remove the UnlockAndReleaseBuffer() and UnlockAndWriteBuffer()
macros, and replace uses of them with calls to the appropriate
functions.
- remove a bunch of #ifdef BMTRACE code: it is ugly & broken
(i.e. it doesn't compile)
- make BufferReplace() return a bool, not an int
- cleanup some logic in bufmgr.c; should be functionality
equivalent to the previous code, just cleaner now
- remove the BM_PRIVATE flag as it is unused
- improve a few comments, etc.
Neil Conway [Sun, 14 Dec 2003 00:15:03 +0000 (00:15 +0000)]
This patch fixes a few more uppercase GUC vars. I also removed an
example from the RESET reference page because it seemed completely
redundant.
Neil Conway [Sun, 14 Dec 2003 00:10:32 +0000 (00:10 +0000)]
This patch makes some improvements and adds some additional detail
to the documentation on routine database maintainence activities.
I also corrected a bunch of SGML markup.
Neil Conway [Sun, 14 Dec 2003 00:05:29 +0000 (00:05 +0000)]
This patch makes some SGML markup more consistent and makes a small
improvement to the SSL auth docs.
Neil Conway [Sat, 13 Dec 2003 23:59:07 +0000 (23:59 +0000)]
This patch makes some SGML markup more consistent and makes a small
improvement to the SSL auth docs.
Bruce Momjian [Sat, 13 Dec 2003 20:25:18 +0000 (20:25 +0000)]
Add /usr/local/sgml/docbook-dsssl to the default search patch for
docbook style sheets, as discussed with Peter.
Bruce Momjian [Sat, 13 Dec 2003 20:09:15 +0000 (20:09 +0000)]
SGML doc build instructions --- no more DTDDECL, fix double dash in path name.
Bruce Momjian [Sat, 13 Dec 2003 20:02:16 +0000 (20:02 +0000)]
Add fadvise TODO.detail.
Bruce Momjian [Sat, 13 Dec 2003 20:02:01 +0000 (20:02 +0000)]
Update TODO.detail syntax:
< * Add free-behind capability for large sequential scans (fadvise)
> * Add free-behind capability for large sequential scans [fadvise]
Bruce Momjian [Sat, 13 Dec 2003 16:56:00 +0000 (16:56 +0000)]
Add Czech language FAQ's
Pavel Stehule
Peter Eisentraut [Fri, 12 Dec 2003 18:45:10 +0000 (18:45 +0000)]
This patch properly sets the prototype for the on_shmem_exit and
on_proc_exit functions, and adjust all other related code to use
the proper types too.
by Kurt Roeckx
Dave Cramer [Fri, 12 Dec 2003 18:36:20 +0000 (18:36 +0000)]
add missing SQLState by Patrick Higgins
Dave Cramer [Fri, 12 Dec 2003 18:34:14 +0000 (18:34 +0000)]
cancel row updates sets values to null by Kris Jurka
Dave Cramer [Fri, 12 Dec 2003 18:30:27 +0000 (18:30 +0000)]
patch to indicate why test cases failed from Oliver Jowett
Dave Cramer [Fri, 12 Dec 2003 18:27:51 +0000 (18:27 +0000)]
patch to fix returning he information schema as a system object by Kris Jurka
Dave Cramer [Fri, 12 Dec 2003 18:05:34 +0000 (18:05 +0000)]
fix casting pooled connections to PGStatement problem patch by JariP
Dave Cramer [Fri, 12 Dec 2003 17:58:34 +0000 (17:58 +0000)]
patch to build under jdk1.1 from Richard Scranton
Dave Cramer [Fri, 12 Dec 2003 00:26:20 +0000 (00:26 +0000)]
binary stream patch by Kris Jurka fixes empty stream failure
Dave Cramer [Fri, 12 Dec 2003 00:21:36 +0000 (00:21 +0000)]
metadata encoding patch from Kris Jurka
Peter Eisentraut [Thu, 11 Dec 2003 22:10:51 +0000 (22:10 +0000)]
Fix for vpath build
Peter Eisentraut [Thu, 11 Dec 2003 22:03:05 +0000 (22:03 +0000)]
New translation
Tom Lane [Thu, 11 Dec 2003 21:21:55 +0000 (21:21 +0000)]
Fix thinko in comment.
Peter Eisentraut [Thu, 11 Dec 2003 20:13:53 +0000 (20:13 +0000)]
Make some wording consistent with DROP OPERATOR description.
Peter Eisentraut [Thu, 11 Dec 2003 19:56:54 +0000 (19:56 +0000)]
Fix instructions how to shut down postmaster.
Dave Cramer [Thu, 11 Dec 2003 18:10:40 +0000 (18:10 +0000)]
build number set to 300 to avoid conflict with 7.4 builds
threadsafe patch from Karl von Randow applied
Dave Cramer [Thu, 11 Dec 2003 15:50:20 +0000 (15:50 +0000)]
patch to deal with unique foreign keys in 7.4 from Kris Jurka
Dave Cramer [Thu, 11 Dec 2003 15:13:34 +0000 (15:13 +0000)]
tch to make sure PSQLState is Serializable and a test case to prove it
Dave Cramer [Thu, 11 Dec 2003 15:11:43 +0000 (15:11 +0000)]
patch to make sure PSQLState is Serializable and a test case to prove it
Dave Cramer [Wed, 10 Dec 2003 18:30:44 +0000 (18:30 +0000)]
incremented build to 211
Teodor Sigaev [Wed, 10 Dec 2003 15:54:58 +0000 (15:54 +0000)]
Fix integer types to use definition from c.h. Per bug report by Patrick Boulay <patrick.boulay@medrium.com>
Tom Lane [Tue, 9 Dec 2003 01:56:20 +0000 (01:56 +0000)]
query_tree_mutator should copy RangeTblEntry nodes even when it's not
planning to modify them itself. Otherwise we end up with shared RTE
substructure, which breaks inheritance_planner because the rte->inh
flag needs to be independent in each copied subquery. Per bug report
from Chris Piker.
Tom Lane [Mon, 8 Dec 2003 21:54:00 +0000 (21:54 +0000)]
Minor autovacuum fixes from Matthew O'Connor.
Tom Lane [Mon, 8 Dec 2003 21:51:31 +0000 (21:51 +0000)]
Add fflush() before sleeping, per Matthew O'Connor.
Tom Lane [Mon, 8 Dec 2003 18:19:58 +0000 (18:19 +0000)]
Whole-row references were broken for subqueries and functions, because
attr_needed/attr_widths optimization failed to allow for Vars with attno
zero in this case. Per report from Tatsuo Ishii.
Tom Lane [Mon, 8 Dec 2003 16:39:05 +0000 (16:39 +0000)]
Work around naming conflict between zlib and OpenSSL by tweaking inclusion
order. Remove some unnecessary #includes (that duplicate c.h).
Bruce Momjian [Mon, 8 Dec 2003 13:04:14 +0000 (13:04 +0000)]
Done:
< * COMMENT ON [ CAST | CONVERSION | OPERATOR CLASS | LARGE OBJECT | LANGUAGE ]
> * -COMMENT ON [ CAST | CONVERSION | OPERATOR CLASS | LARGE OBJECT | LANGUAGE ]
263c263
< * Allow pg_dump to dump CREATE CONVERSION (Christopher)
> * -Allow pg_dump to dump CREATE CONVERSION (Christopher)
Tom Lane [Sun, 7 Dec 2003 19:55:58 +0000 (19:55 +0000)]
Replace unportable and overflow-prone use of 'long long' with safer
'double' arithmetic, per recent discussion.
Tom Lane [Sun, 7 Dec 2003 19:43:02 +0000 (19:43 +0000)]
Remove test on c.relkind from check_constraints view; unnecessary and
prevents view from showing constraints on domains. This addresses the
other half of Claus Colloseus' bug report.
Peter Eisentraut [Sun, 7 Dec 2003 10:21:58 +0000 (10:21 +0000)]
Fix typmod interpretation for bit types. (It was erroneously assumed that
for bit(x), the typmod stores x+4, like for the character types.)
Tom Lane [Sun, 7 Dec 2003 05:44:50 +0000 (05:44 +0000)]
One more pass at reducing the cost of pg_dump's new implementation:
reduce the number of times TopoSort() has to be executed by trying to
extract multiple dependency loops from each pass, instead of only one.
This saves about another factor of ten on the regression database.
This could be considered as another exercise in grokking Fred Brooks'
maxim: Representation *is* the essence of programming.
Joe Conway [Sun, 7 Dec 2003 04:14:10 +0000 (04:14 +0000)]
Repair indexed bytea like operations, and related selectivity
functionality. Per bug report by Alvar Freude:
http://archives.postgresql.org/pgsql-bugs/2003-12/msg00022.php
Tom Lane [Sun, 7 Dec 2003 03:14:01 +0000 (03:14 +0000)]
Speed up findObjectByCatalogId() to get rid of the other salient
bottleneck in the new pg_dump code.
Bruce Momjian [Sun, 7 Dec 2003 00:54:05 +0000 (00:54 +0000)]
Done:
> * -Use dependency information to dump data in proper order
Joe Conway [Sat, 6 Dec 2003 23:10:23 +0000 (23:10 +0000)]
Documentation update for previously committed changes: read-only GUC variables
and pg_settings view.
Tom Lane [Sat, 6 Dec 2003 22:55:11 +0000 (22:55 +0000)]
Replace not-very-bright implementation of topological sort with a better
one (use a priority heap to keep track of items ready to output, instead
of searching the input array each time). This brings the runtime of
pg_dump back to about what it was in 7.4.
Tom Lane [Sat, 6 Dec 2003 03:00:16 +0000 (03:00 +0000)]
Massive overhaul of pg_dump: make use of dependency information from
pg_depend to determine a safe dump order. Defaults and check constraints
can be emitted either as part of a table or domain definition, or
separately if that's needed to break a dependency loop. Lots of old
half-baked code for controlling dump order removed.
Tom Lane [Fri, 5 Dec 2003 15:50:31 +0000 (15:50 +0000)]
Guard against bug in Solaris' bsearch(), per Michael Wildpaner.
Teodor Sigaev [Fri, 5 Dec 2003 15:37:51 +0000 (15:37 +0000)]
One more fix confusion
Teodor Sigaev [Fri, 5 Dec 2003 14:27:42 +0000 (14:27 +0000)]
Avoid confusion start_parse_str function with tsearch V1
Peter Eisentraut [Thu, 4 Dec 2003 21:24:56 +0000 (21:24 +0000)]
Markup improvements
Teodor Sigaev [Thu, 4 Dec 2003 12:21:11 +0000 (12:21 +0000)]
Avoid conflict strndup with glibc
Joe Conway [Wed, 3 Dec 2003 18:53:52 +0000 (18:53 +0000)]
catversion change needed by previous read-onlu GUC variables commit.
Joe Conway [Wed, 3 Dec 2003 18:52:00 +0000 (18:52 +0000)]
Added new group of read-only GUC variables to allow simple access
to certain compile-time options (FUNC_MAX_ARGS, INDEX_MAX_KEYS,
NAMEDATALEN, BLCKSZ, HAVE_INT64_TIMESTAMP). Also added "category",
"short_desc", and "extra_desc" to the pg_settings view. Per recent
discussion here:
http://archives.postgresql.org/pgsql-patches/2003-11/msg00363.php
Tom Lane [Wed, 3 Dec 2003 17:45:10 +0000 (17:45 +0000)]
Planner failed to be smart about binary-compatible expressions in pathkeys
and hash bucket-size estimation. Issue has been there awhile but is more
critical in 7.4 because it affects varchar columns. Per report from
Greg Stark.
Teodor Sigaev [Wed, 3 Dec 2003 16:07:48 +0000 (16:07 +0000)]
Fix for word with several infinitives
Michael Meskes [Wed, 3 Dec 2003 08:49:17 +0000 (08:49 +0000)]
Added patch by Dave Cramer for array handling in ecpglib.
Joe Conway [Tue, 2 Dec 2003 19:26:47 +0000 (19:26 +0000)]
Add a warning to AtEOXact_SPI() to catch cases where the current
transaction has been committed without SPI_finish() being called
first. Per recent discussion here:
http://archives.postgresql.org/pgsql-patches/2003-11/msg00286.php
Tom Lane [Tue, 2 Dec 2003 16:14:12 +0000 (16:14 +0000)]
Update release notes for 7.3.5.
Peter Eisentraut [Tue, 2 Dec 2003 10:24:37 +0000 (10:24 +0000)]
Fix some small errors.
Tom Lane [Tue, 2 Dec 2003 00:26:59 +0000 (00:26 +0000)]
Code and docs review for numeric-factorial patch.
Bruce Momjian [Mon, 1 Dec 2003 23:19:33 +0000 (23:19 +0000)]
I've run across a pretty serious problem with pg_autovacuum.
pg_autovacuum looses track of any table that's ever been truncated
(possibly other situations too). When i truncate a table it gets a
new relfilenode in pg_class. This is a problem because pg_autovacuum
assumes pg_class.relfilenode will join to pg_stats_all_tables.relid.
pg_stats_all_tables.relid is actallly the oid from pg_class, not the
relfilenode. These two values start out equal so pg_autovacuum works
initially, but it fails later on because of this incorrect assumption.
This patch fixes that problem. Applied to HEAD and 7.4.X.
Brian Hirt
Bruce Momjian [Mon, 1 Dec 2003 23:15:47 +0000 (23:15 +0000)]
Fix initdb use of mkdir_p().
Andrew Dunstan
Bruce Momjian [Mon, 1 Dec 2003 23:13:07 +0000 (23:13 +0000)]
I noticed in the 7.4 release that in contrib/pg_autovacuum, args->logfile is
not initialized if a log file is not specified on the command line. This
causes an immediate segfault on systems that fill allocated memory with some
value other than zero (my FreeBSD machine uses 0xD0).
Several crashes later I discovered that args->user, password, host, and port
are also used without being initialized.
This doesn't appear to be fixed in CVS and I came up empty on a mailing list
search -- hope it hasn't been reported already.
Craig Boston
Bruce Momjian [Mon, 1 Dec 2003 23:12:16 +0000 (23:12 +0000)]
The attached patch enables contrib/cube to build cleanly under Cygwin
(again). Please consider this patch for the 7.4.1 branch (if there will
be one) too.
Jason Tishler
Bruce Momjian [Mon, 1 Dec 2003 23:09:02 +0000 (23:09 +0000)]
This patch refactors execTuples.c in two ways.
Neil Conway