]> granicus.if.org Git - postgresql/log
postgresql
18 years agoAllow Borland CC to compile libpq and psql.
Bruce Momjian [Thu, 11 Jan 2007 02:40:12 +0000 (02:40 +0000)]
Allow Borland CC to compile libpq and psql.

Backpatch to 8.2.X.

L Bayuk

18 years agoBackport patch.
Tatsuo Ishii [Wed, 10 Jan 2007 01:20:10 +0000 (01:20 +0000)]
Backport patch.

Call srandom() instead of srand().
pgbench calls random() later, so it should have called srandom().
On most platforms except Windows srandom() is actually identical
to srand(), so the bug only bites Windows users.
per bug report from Akio Ishida.

18 years agoTweak joinlist creation to avoid generating useless one-element subproblems
Tom Lane [Mon, 8 Jan 2007 16:47:35 +0000 (16:47 +0000)]
Tweak joinlist creation to avoid generating useless one-element subproblems
when collapsing of JOIN trees is stopped by join_collapse_limit.  For instance
a list of 11 LEFT JOINs with limit 8 now produces something like
((1 2 3 4 5 6 7 8) 9 10 11 12)
instead of
(((1 2 3 4 5 6 7 8) (9)) 10 11 12)
The latter structure is really only required for a FULL JOIN.
Noted while studying an example from Shane Ambler.

18 years agoRemove cost_hashjoin's very ancient hack to discourage (once, entirely forbid)
Tom Lane [Mon, 8 Jan 2007 16:09:31 +0000 (16:09 +0000)]
Remove cost_hashjoin's very ancient hack to discourage (once, entirely forbid)
hash joins with the estimated-larger relation on the inside.  There are
several cases where doing that makes perfect sense, and in cases where it
doesn't, the regular cost computation really ought to be able to figure that
out.  Make some marginal tweaks in said computation to try to get results
approximating reality a bit better.  Per an example from Shane Ambler.

Also, fix an oversight in the original patch to add seq_page_cost: the costs
of spilling a hash join to disk should be scaled by seq_page_cost.

18 years agoFix oversight in handling of row-comparison index keys: if the row comparison
Tom Lane [Sun, 7 Jan 2007 01:56:24 +0000 (01:56 +0000)]
Fix oversight in handling of row-comparison index keys: if the row comparison
doesn't exactly match the index, we may have to change our initial positioning
strategy.  For example, given an index on (f1,f2,f3) and a WHERE condition
"ROW(f1,f3) > ROW(2,3)", the code extracted the initial-positioning condition
"f1 > 2", which is wrong ... it has to be "f1 >= 2", else some rows matching
the WHERE condition may fail to be returned.

Applying patch to 8.2 only --- I'll fix it in HEAD later as part of the
planned index improvements (reverse-sort and NULLS FIRST/LAST work).

18 years agoFix filtered_base_yylex() to save and restore base_yylval and base_yylloc
Tom Lane [Sat, 6 Jan 2007 19:14:27 +0000 (19:14 +0000)]
Fix filtered_base_yylex() to save and restore base_yylval and base_yylloc
properly when doing a lookahead.  The lack of this was causing various
interesting misbehaviors when one tries to use "with" as a plain identifier.

18 years agotag configure REL8_2_1
PostgreSQL Daemon [Sat, 6 Jan 2007 05:02:26 +0000 (05:02 +0000)]
tag configure

18 years agoMinor copy-editing for release note updates.
Tom Lane [Fri, 5 Jan 2007 22:34:46 +0000 (22:34 +0000)]
Minor copy-editing for release note updates.

18 years agoUpdate Japanese FAQ.
Bruce Momjian [Fri, 5 Jan 2007 20:57:17 +0000 (20:57 +0000)]
Update Japanese FAQ.

18 years agoStamp release 8.2.1. Update FAQs.
Bruce Momjian [Fri, 5 Jan 2007 20:55:29 +0000 (20:55 +0000)]
Stamp release 8.2.1.  Update FAQs.

18 years agoCreate release notes for release 8.2.1.
Bruce Momjian [Fri, 5 Jan 2007 20:05:47 +0000 (20:05 +0000)]
Create release notes for release 8.2.1.

18 years agoAdd missing reference to pg_shdescription. Greg Mullane
Tom Lane [Fri, 5 Jan 2007 01:19:04 +0000 (01:19 +0000)]
Add missing reference to pg_shdescription.  Greg Mullane

18 years agoTweak pg_dumpall to add GRANT CONNECT ON DATABASE ... TO PUBLIC when dumping
Tom Lane [Thu, 4 Jan 2007 17:49:42 +0000 (17:49 +0000)]
Tweak pg_dumpall to add GRANT CONNECT ON DATABASE ... TO PUBLIC when dumping
database privileges from a pre-8.2 server.  This ensures that the reloaded
database will maintain the same behavior it had in the previous installation,
ie, everybody has connect privilege.  Per gripe from L Bayuk.

18 years agoFix erroneous implementation of -s in postmaster.c (the switch doesn't take
Tom Lane [Thu, 4 Jan 2007 00:58:01 +0000 (00:58 +0000)]
Fix erroneous implementation of -s in postmaster.c (the switch doesn't take
an optarg).  Add some comments noting that code in three different files has
to be kept in sync.  Fix erroneous description of -S switch (it sets work_mem
not silent_mode), and do some light copy-editing elsewhere in postgres-ref.

18 years agoFix regex_fixed_prefix() to cope reasonably well with regex patterns of the
Tom Lane [Wed, 3 Jan 2007 22:39:35 +0000 (22:39 +0000)]
Fix regex_fixed_prefix() to cope reasonably well with regex patterns of the
form '^(foo)$'.  Before, these could never be optimized into indexscans.
The recent changes to make psql and pg_dump generate such patterns (for \d
commands and -t and related switches, respectively) therefore represented
a big performance hit for people with large pg_class catalogs, as seen in
recent gripe from Erik Jones.  While at it, be more paranoid about
case-sensitivity checking in multibyte encodings, and fix some other
corner cases in which a regex might be interpreted too liberally.

18 years agoAdd a defense to prevent core dumps if 8.2 version of rank_cd() is used with
Tom Lane [Thu, 28 Dec 2006 01:09:04 +0000 (01:09 +0000)]
Add a defense to prevent core dumps if 8.2 version of rank_cd() is used with
the 8.1 SQL function definition for it.  Per report from Rajesh Kumar Mallah,
such a DBA error doesn't seem at all improbable, and the cost of checking for
it is not very high compared to the cost of running this function.  (It would
have been better to change the C name of the function so it wouldn't be called
by the old SQL definition, but it's too late for that now in the 8.2 branch.)

18 years agoRevert exports.list change pending closer study.
Tom Lane [Thu, 28 Dec 2006 00:01:12 +0000 (00:01 +0000)]
Revert exports.list change pending closer study.

18 years agoUse a more backward-compatible syntax for exports.list on Linux.
Tom Lane [Wed, 27 Dec 2006 23:53:21 +0000 (23:53 +0000)]
Use a more backward-compatible syntax for exports.list on Linux.
Per Thorkil Olesen.

18 years agoModify local buffer management to request memory for local buffers in blocks
Tom Lane [Wed, 27 Dec 2006 22:31:59 +0000 (22:31 +0000)]
Modify local buffer management to request memory for local buffers in blocks
of increasing size, instead of one at a time.  This reduces the memory
management overhead when num_temp_buffers is large: in the previous coding
we would actually waste 50% of the space used for temp buffers, because aset.c
would round the individual requests up to 16K.  Problem noted while studying
a performance issue reported by Steven Flatt.

Back-patch as far as 8.1 --- older versions used few enough local buffers
that the issue isn't significant for them.

18 years agoPrint combining characters (those reported as having zero width by
Tom Lane [Wed, 27 Dec 2006 19:45:51 +0000 (19:45 +0000)]
Print combining characters (those reported as having zero width by
PQdsplen()) normally, instead of replacing them by \uXXXX sequences.
Assume that they in fact occupy zero screen space for formatting purposes.
Per gripe from Michael Fuhr and ensuing discussion.

18 years agoUse FROM clause in example UPDATE commands where appropriate. Also
Tom Lane [Wed, 27 Dec 2006 16:07:42 +0000 (16:07 +0000)]
Use FROM clause in example UPDATE commands where appropriate.  Also
remove long-obsolete statement that there isn't a check for infinite
recursion in view rules.

18 years agoFix broken markup.
Tom Lane [Wed, 27 Dec 2006 16:04:47 +0000 (16:04 +0000)]
Fix broken markup.

18 years agoFix failure due to accessing an already-freed tuple descriptor in a plan
Tom Lane [Tue, 26 Dec 2006 21:37:28 +0000 (21:37 +0000)]
Fix failure due to accessing an already-freed tuple descriptor in a plan
involving HashAggregate over SubqueryScan (this is the known case, there
may well be more).  The bug is only latent in releases before 8.2 since they
didn't try to access tupletable slots' descriptors during ExecDropTupleTable.
The least bogus fix seems to be to make subqueries share the parent query's
memory context, so that tupdescs they create will have the same lifespan as
those of the parent query.  There are comments in the code envisioning going
even further by not having a separate child EState at all, but that will
require rethinking executor access to range tables, which I don't want to
tackle right now.  Per bug report from Jean-Pierre Pelletier.

18 years agoRepair bug #2839: the various ExecReScan functions need to reset
Tom Lane [Tue, 26 Dec 2006 19:26:56 +0000 (19:26 +0000)]
Repair bug #2839: the various ExecReScan functions need to reset
ps_TupFromTlist in plan nodes that make use of it.  This was being done
correctly in join nodes and Result nodes but not in any relation-scan nodes.
Bug would lead to bogus results if a set-returning function appeared in the
targetlist of a subquery that could be rescanned after partial execution,
for example a subquery within EXISTS().  Bug has been around forever :-(
... surprising it wasn't reported before.

18 years agoRepair bug #2836: SPI_execute_plan returned zero if none of the querytrees
Tom Lane [Tue, 26 Dec 2006 16:56:22 +0000 (16:56 +0000)]
Repair bug #2836: SPI_execute_plan returned zero if none of the querytrees
were marked canSetTag.  While it's certainly correct to return the result
of the last one that is marked canSetTag, it's less clear what to do when
none of them are.  Since plpgsql will complain if zero is returned, the
8.2.0 behavior isn't good.  I've fixed it to restore the prior behavior of
returning the physically last query's result code when there are no
canSetTag queries.

18 years agoFix memory reallocation condition
Teodor Sigaev [Tue, 26 Dec 2006 14:55:00 +0000 (14:55 +0000)]
Fix memory reallocation condition

18 years agoMake HISTCONTROL=ignoredups work again (broken by misordering of
Tom Lane [Sun, 24 Dec 2006 19:14:41 +0000 (19:14 +0000)]
Make HISTCONTROL=ignoredups work again (broken by misordering of
operations during recent code refactoring).  Per bug #2840 from Ned Crigler.

18 years agoFix convertion for 'PFX flag N num'
Teodor Sigaev [Thu, 21 Dec 2006 17:35:44 +0000 (17:35 +0000)]
Fix convertion for 'PFX flag N num'

18 years agoRemove completed TODO item about xid wrap-around from 8.2.X.
Bruce Momjian [Tue, 19 Dec 2006 21:58:43 +0000 (21:58 +0000)]
Remove completed TODO item about xid wrap-around from 8.2.X.

18 years agoFix some planner bugs exposed by reports from Arjen van der Meijden. These
Tom Lane [Fri, 15 Dec 2006 18:42:35 +0000 (18:42 +0000)]
Fix some planner bugs exposed by reports from Arjen van der Meijden.  These
are all in new-in-8.2 logic associated with indexability of ScalarArrayOpExpr
(IN-clauses) or amortization of indexscan costs across repeated indexscans
on the inside of a nestloop.  In particular:

Fix some logic errors in the estimation for multiple scans induced by a
ScalarArrayOpExpr indexqual.

Include a small cost component in bitmap index scans to reflect the costs of
manipulating the bitmap itself; this is mainly to prevent a bitmap scan from
appearing to have the same cost as a plain indexscan for fetching a single
tuple.

Also add a per-index-scan-startup CPU cost component; while prior releases
were clearly too pessimistic about the cost of repeated indexscans, the
original 8.2 coding allowed the cost of an indexscan to effectively go to zero
if repeated often enough, which is overly optimistic.

Pay some attention to index correlation when estimating costs for a nestloop
inner indexscan: this is significant when the plan fetches multiple heap
tuples per iteration, since high correlation means those tuples are probably
on the same or adjacent heap pages.

18 years agoPut JST back into the default set of timezone abbreviations;
Tom Lane [Fri, 15 Dec 2006 16:54:56 +0000 (16:54 +0000)]
Put JST back into the default set of timezone abbreviations;
was removed in an unexplainable moment of brain fade.

18 years agoBack out double-run of PDF/PS output. Requires building bookindex.sgml
Bruce Momjian [Fri, 15 Dec 2006 16:50:10 +0000 (16:50 +0000)]
Back out double-run of PDF/PS output.  Requires building bookindex.sgml
properly.

Remove SGML docs about openjade performance patch, and instead add
comment in style sheet where indenting code is commented out.

Backpatch to 8.2.X.

18 years agoDrop indentation of verbatim environments in print output. This increased
Peter Eisentraut [Fri, 15 Dec 2006 14:54:48 +0000 (14:54 +0000)]
Drop indentation of verbatim environments in print output.  This increased
the build time by what seemed like infinity.

18 years agoDocument patch needed to get PDF and PS output in a reasonable amount of
Bruce Momjian [Fri, 15 Dec 2006 02:44:33 +0000 (02:44 +0000)]
Document patch needed to get PDF and PS output in a reasonable amount of
time.

18 years agoRun Jade twice when outputting Postscript and PDF so the index is
Bruce Momjian [Fri, 15 Dec 2006 01:21:08 +0000 (01:21 +0000)]
Run Jade twice when outputting Postscript and PDF so the index is
correct, add comments about other multiple runs in the Makefile.

Backpatch to 8.2.X.

18 years agoMake --with-ldap build on Unixware, per Olivier Prenant.
Tom Lane [Thu, 14 Dec 2006 21:50:07 +0000 (21:50 +0000)]
Make --with-ldap build on Unixware, per Olivier Prenant.

18 years agoFix planner to do the right thing when a degenerate outer join (one whose
Tom Lane [Tue, 12 Dec 2006 21:31:09 +0000 (21:31 +0000)]
Fix planner to do the right thing when a degenerate outer join (one whose
joinclause doesn't use any outer-side vars) requires a "bushy" plan to be
created.  The normal heuristic to avoid joins with no joinclause has to be
overridden in that case.  Problem is new in 8.2; before that we forced the
outer join order anyway.  Per example from Teodor.

18 years agoDocument that log_line_prefix %t does not print timezone on Win32.
Bruce Momjian [Tue, 12 Dec 2006 21:30:41 +0000 (21:30 +0000)]
Document that log_line_prefix %t does not print timezone on Win32.

18 years agoCollect the bits of wisdom about dtrace installation in the installation
Peter Eisentraut [Tue, 12 Dec 2006 16:07:42 +0000 (16:07 +0000)]
Collect the bits of wisdom about dtrace installation in the installation
chapter rather than scattering them across several incomplete fragments.

(This makes the documentation consistent with the backported FAQ_Solaris.)

18 years agoBackpatch FAQ entry for null concatenation.
Bruce Momjian [Mon, 11 Dec 2006 22:48:08 +0000 (22:48 +0000)]
Backpatch FAQ entry for null concatenation.

18 years agoAvoid double free of _SPI_current->tuptable. AtEOSubXact_SPI() now tries to
Tom Lane [Fri, 8 Dec 2006 00:40:33 +0000 (00:40 +0000)]
Avoid double free of _SPI_current->tuptable.  AtEOSubXact_SPI() now tries to
release it in a subtransaction abort, but this neglects possibility that
someone outside SPI already did.  Fix is for spi.c to forget about a tuptable
as soon as it's handed it back to the caller.
Per bug #2817 from Michael Andreen.

18 years agoRepair incorrect placement of WHERE clauses when there are multiple,
Tom Lane [Thu, 7 Dec 2006 19:33:48 +0000 (19:33 +0000)]
Repair incorrect placement of WHERE clauses when there are multiple,
rearrangeable outer joins and the WHERE clause is non-strict and mentions
only nullable-side relations.  New bug in 8.2, caused by new logic to allow
rearranging outer joins.  Per bug #2807 from Ross Cohen; thanks to Jeff
Davis for producing a usable test case.

18 years agoFix planning of SubLinks to ensure that Vars generated from transformation of
Tom Lane [Wed, 6 Dec 2006 19:40:08 +0000 (19:40 +0000)]
Fix planning of SubLinks to ensure that Vars generated from transformation of
a sublink's test expression have the correct vartypmod, rather than defaulting
to -1.  There's at least one place where this is important because we're
expecting these Vars to be exactly equal() to those appearing in the subplan
itself.  This is a pretty klugy solution --- it would likely be cleaner to
change Param nodes to include a typmod field --- but we can't do that in the
already-released 8.2 branch.
Per bug report from Hubert Fongarnand.

18 years agoClean up psql -c documentation that uses \\.
Bruce Momjian [Wed, 6 Dec 2006 15:40:32 +0000 (15:40 +0000)]
Clean up psql -c documentation that uses \\.

18 years agoFix documentation example of using psql \x with a SELECT command, per
Bruce Momjian [Tue, 5 Dec 2006 17:41:02 +0000 (17:41 +0000)]
Fix documentation example of using psql \x with a SELECT command, per
Simon Riggs.  Backpatch to 8.2.X.

18 years agoFix pg_dump linking on Win32 with MSVS win32.mak:
Bruce Momjian [Mon, 4 Dec 2006 22:26:46 +0000 (22:26 +0000)]
Fix pg_dump linking on Win32 with MSVS win32.mak:

The module link is insufficient.:-(

---- Sorry, japanese message change to xxx ---
link.exe @C:\DOCUME~1\hi-saito\LOCALS~1\Temp\nmk03360.
common.obj : error LNK2001: xxxxxx "_pg_qsort" xxxxxx
pg_dump_sort.obj : error LNK2001: xxxxx "_pg_qsort" xxxxx
.\Release\pg_dump.exe : fatal error LNK1120: xxxxxxx
NMAKE : fatal error U1077: 'link.exe' : xxxxx '0x460'
Stop.

Hiroshi Saito

18 years agoPatch of Win32 Encoding problem for server messages using
Bruce Momjian [Mon, 4 Dec 2006 22:24:04 +0000 (22:24 +0000)]
Patch of Win32 Encoding problem for server messages using
FormatMessage() (This should have been in 8.2.0, patched to 8.2.X and
HEAD):

I think this problem to be complex....
http://archives.postgresql.org/pgsql-hackers/2006-11/msg00042.php

FormatMessage of windows cannot consider the encoding of the database.
However, I should try the solution now. It is necessary to clear the
problem.

Multi character-code exists together in message and log. It doesn't
consider
the data base encoding that the user intended....

The user in multi-byte country can try this.
http://inet.winpg.jp/~saito/pg_bug/MessageCheck.c

That is, it is likely to become it in this manner.(Japanese)
http://inet.winpg.jp/~saito/pg_bug/FormatMessage998.png

Hiroshi Saito

18 years agoFix core dump of ispell for case of non-successfull initialization.
Teodor Sigaev [Mon, 4 Dec 2006 09:27:45 +0000 (09:27 +0000)]
Fix core dump of ispell for case of non-successfull initialization.
Previous versions aren't affected.

Fix synonym dictionary init: string should be malloc'ed, not palloc'ed. Bug
introduced recently while fixing lowerstr().

18 years agoFix LIMIT/OFFSET for null limit values. This worked before 8.2 but was broken
Tom Lane [Sun, 3 Dec 2006 21:40:13 +0000 (21:40 +0000)]
Fix LIMIT/OFFSET for null limit values.  This worked before 8.2 but was broken
by the change to make limit values int8 instead of int4.  (Specifically, you
can do DatumGetInt32 safely on a null value, but not DatumGetInt64.)  Per
bug #2803 from Greg Johnson.

18 years agoUpdate 8.2.X release to say 8.2.0 instead of just 8.2.
Bruce Momjian [Sun, 3 Dec 2006 16:43:05 +0000 (16:43 +0000)]
Update 8.2.X release to say 8.2.0 instead of just 8.2.

18 years agov8.2.0 is now released ... REL8_2_0
PostgreSQL Daemon [Sat, 2 Dec 2006 08:36:41 +0000 (08:36 +0000)]
v8.2.0 is now released ...

18 years agoStamp 8.2, except configure.in.
Bruce Momjian [Sat, 2 Dec 2006 04:12:39 +0000 (04:12 +0000)]
Stamp 8.2, except configure.in.

18 years agoUpdate for release 8.2.
Bruce Momjian [Sat, 2 Dec 2006 04:12:11 +0000 (04:12 +0000)]
Update for release 8.2.

18 years agoWording refinement for external references in man pages.
Peter Eisentraut [Sat, 2 Dec 2006 01:26:19 +0000 (01:26 +0000)]
Wording refinement for external references in man pages.

18 years agoDon't ship spi_* man pages. (We don't do that anyway, but now it's automatic.)
Peter Eisentraut [Sat, 2 Dec 2006 01:25:50 +0000 (01:25 +0000)]
Don't ship spi_* man pages.  (We don't do that anyway, but now it's automatic.)

18 years agoTranslation updates
Peter Eisentraut [Sat, 2 Dec 2006 01:16:16 +0000 (01:16 +0000)]
Translation updates

18 years agoPut release date in release notes.
Tom Lane [Sat, 2 Dec 2006 00:46:46 +0000 (00:46 +0000)]
Put release date in release notes.

18 years agoAdd some documentation for DTrace support. Simon Riggs
Tom Lane [Sat, 2 Dec 2006 00:42:54 +0000 (00:42 +0000)]
Add some documentation for DTrace support.  Simon Riggs

18 years agoMake Options for Windows a second-level heading instead of first-level.
Peter Eisentraut [Sat, 2 Dec 2006 00:34:52 +0000 (00:34 +0000)]
Make Options for Windows a second-level heading instead of first-level.

18 years agoEditorial improvements for GIN documentation.
Tom Lane [Fri, 1 Dec 2006 23:46:46 +0000 (23:46 +0000)]
Editorial improvements for GIN documentation.

18 years agoSome more supported-platforms updates: buildfarm hare is alive again,
Tom Lane [Fri, 1 Dec 2006 21:17:51 +0000 (21:17 +0000)]
Some more supported-platforms updates: buildfarm hare is alive again,
and penguin reported in recently enough to justify the assumption that
we haven't broken ARM support in 8.2.

18 years agoDocument the recently-understood hazard that a rollback can release row-level
Tom Lane [Fri, 1 Dec 2006 20:49:53 +0000 (20:49 +0000)]
Document the recently-understood hazard that a rollback can release row-level
locks that logically should not be released, because when a subtransaction
overwrites XMAX all knowledge of the previous lock state is lost.  It seems
unlikely that we will be able to fix this before 8.3...

18 years agoMake the bgwriter's error recovery path do smgrcloseall(). On Windows this
Tom Lane [Fri, 1 Dec 2006 19:55:28 +0000 (19:55 +0000)]
Make the bgwriter's error recovery path do smgrcloseall().  On Windows this
should allow delete-pending files to actually go away, and thereby work
around the various complaints we've seen about 'permission denied'
errors in such cases.  Should be reasonably harmless in any case...

18 years agoEditorial improvements to backup and warm-standby documentation.
Tom Lane [Fri, 1 Dec 2006 03:29:15 +0000 (03:29 +0000)]
Editorial improvements to backup and warm-standby documentation.

18 years agoMinor wording improvements.
Tom Lane [Fri, 1 Dec 2006 03:19:48 +0000 (03:19 +0000)]
Minor wording improvements.

18 years agoAdjust the description of locking to clarify that locks held by a
Tom Lane [Fri, 1 Dec 2006 01:04:36 +0000 (01:04 +0000)]
Adjust the description of locking to clarify that locks held by a
subtransaction are released if the subtransaction aborts --- in user-level
terminology, this means either rolling back to a savepoint or escaping from
a plpgsql exception block.  Per recent suggestion from Simon.

18 years agoBetter solution to the tr problem: use sed instead. Per Martijn and Andrew.
Tom Lane [Thu, 30 Nov 2006 22:21:24 +0000 (22:21 +0000)]
Better solution to the tr problem: use sed instead.  Per Martijn and Andrew.

18 years agoImprove portability of 'tr' invocation in PGAC_ARG_CHECK. Reported by
Tom Lane [Thu, 30 Nov 2006 21:44:12 +0000 (21:44 +0000)]
Improve portability of 'tr' invocation in PGAC_ARG_CHECK.  Reported by
Olivier Prenant, fixed by Peter.

18 years agoUpdate supported-platforms list based on recent buildfarm results.
Tom Lane [Thu, 30 Nov 2006 21:30:22 +0000 (21:30 +0000)]
Update supported-platforms list based on recent buildfarm results.

18 years agoFix typos
Peter Eisentraut [Thu, 30 Nov 2006 20:50:44 +0000 (20:50 +0000)]
Fix typos

18 years agoMinor adjustments to make failures in startup/shutdown behave more cleanly.
Tom Lane [Thu, 30 Nov 2006 18:29:12 +0000 (18:29 +0000)]
Minor adjustments to make failures in startup/shutdown behave more cleanly.
StartupXLOG and ShutdownXLOG no longer need to be critical sections, because
in all contexts where they are invoked, elog(ERROR) would be translated to
elog(FATAL) anyway.  (One change in bgwriter.c is needed to make this true:
set ExitOnAnyError before trying to exit.  This is a good fix anyway since
the existing code would have gone into an infinite loop on elog(ERROR) during
shutdown.)  That avoids a misleading report of PANIC during semi-orderly
failures.  Modify the postmaster to include the startup process in the set of
processes that get SIGTERM when a fast shutdown is requested, and also fix it
to not try to restart the bgwriter if the bgwriter fails while trying to write
the shutdown checkpoint.  Net result is that "pg_ctl stop -m fast" does
something reasonable for a system in warm standby mode, and so should Unix
system shutdown (ie, universal SIGTERM).  Per gripe from Stephen Harris and
some corner-case testing of my own.

18 years agoFix bug with page deletion. If inner page is removed and it tries to
Teodor Sigaev [Thu, 30 Nov 2006 16:22:32 +0000 (16:22 +0000)]
Fix bug with page deletion. If inner page is removed and it tries to
remove page on next level linked from next inner page, ginScanToDelete()
wrongly sets parent page. Bug reveals when many item pointers from index
was deleted ( several hundred thousands).

Bug is discovered by hubert depesz lubaczewski <depesz@gmail.com>

Suppose, we need rc2 before release...

18 years agoFix Makefile problem which prevented installation on VPATH builds.
Alvaro Herrera [Wed, 29 Nov 2006 21:21:03 +0000 (21:21 +0000)]
Fix Makefile problem which prevented installation on VPATH builds.

18 years agoIgnore libedit/libreadline while probing for strlcpy and some other
Tom Lane [Wed, 29 Nov 2006 20:12:31 +0000 (20:12 +0000)]
Ignore libedit/libreadline while probing for strlcpy and some other
standard functions.  Per report from Stefan Kaltenbrunner.

18 years agoMore MSVC build support from Magnus.
Tom Lane [Wed, 29 Nov 2006 19:49:31 +0000 (19:49 +0000)]
More MSVC build support from Magnus.

18 years agoSpelling fix
Peter Eisentraut [Wed, 29 Nov 2006 14:50:07 +0000 (14:50 +0000)]
Spelling fix

18 years agoAdd an example showing how to cope with mixed-case names in pg_dump
Tom Lane [Tue, 28 Nov 2006 22:54:18 +0000 (22:54 +0000)]
Add an example showing how to cope with mixed-case names in pg_dump
switches.

18 years agoUpdate timezone data to tzdata2006p zic distribution. It seems Western
Tom Lane [Tue, 28 Nov 2006 19:37:03 +0000 (19:37 +0000)]
Update timezone data to tzdata2006p zic distribution.  It seems Western
Australia decided to institute DST with one month's notice ... way to go,
politicians.

18 years agoMark to_char(timestamp without timezone) as stable, not immutable, since its
Tom Lane [Tue, 28 Nov 2006 19:18:44 +0000 (19:18 +0000)]
Mark to_char(timestamp without timezone) as stable, not immutable, since its
result now depends on the lc_messages setting, as noted by Bruce.
Also, mark to_number() and the numeric-type variants of to_char() as stable,
because their results depend on lc_numeric; this is a longstanding oversight.
Also, mark to_date() and to_char(interval) as stable; although these appear
not to depend on any GUC variables as of CVS HEAD, that seems a property
unlikely to survive future improvements.  It seems best to mark all the
formatting functions stable and be done with it.
catversion not bumped, because this does not seem critical enough to force
a post-RC1 initdb, and anyway we cannot do so in the back branches.

18 years agoFix some translator comments so that xgettext finds them and pgindent does
Peter Eisentraut [Tue, 28 Nov 2006 12:54:42 +0000 (12:54 +0000)]
Fix some translator comments so that xgettext finds them and pgindent does
not destroy them.  Maybe we can adjust pgindent sometime.

18 years agoAdd workaround for localizing May and abbreviated May differently. Idea
Peter Eisentraut [Tue, 28 Nov 2006 12:53:44 +0000 (12:53 +0000)]
Add workaround for localizing May and abbreviated May differently.  Idea
of Dennis Björklund.

18 years agoAlso install ecpg_config.h
Michael Meskes [Tue, 28 Nov 2006 12:44:06 +0000 (12:44 +0000)]
Also install ecpg_config.h

18 years agoAdd $(CFLAGS) to the simplified build rule for .so libraries on Darwin.
Tom Lane [Tue, 28 Nov 2006 05:45:43 +0000 (05:45 +0000)]
Add $(CFLAGS) to the simplified build rule for .so libraries on Darwin.
Arguably we should do this on *all* platforms, but for the moment Ill

18 years agoprotect vfprintf from hijacking by Windows gettext just like other members of the...
Andrew Dunstan [Tue, 28 Nov 2006 01:12:34 +0000 (01:12 +0000)]
protect vfprintf from hijacking by Windows gettext just like other members of the *printf family.

18 years agoFix inheritance description to note that not-null constraints are
Tom Lane [Tue, 28 Nov 2006 01:09:01 +0000 (01:09 +0000)]
Fix inheritance description to note that not-null constraints are
inherited, per Taiki Yamaguchi.

18 years agoFix gratuitous message spelling differences
Peter Eisentraut [Mon, 27 Nov 2006 15:50:55 +0000 (15:50 +0000)]
Fix gratuitous message spelling differences

18 years agoMake pg_restore usage examples more useful: illustrate restoring into
Tom Lane [Sun, 26 Nov 2006 18:11:11 +0000 (18:11 +0000)]
Make pg_restore usage examples more useful: illustrate restoring into
both the same database name and a different one.

18 years agoFix misspellings of GB.
Peter Eisentraut [Sat, 25 Nov 2006 22:55:59 +0000 (22:55 +0000)]
Fix misspellings of GB.

18 years agoCorrect misspellings of kB.
Peter Eisentraut [Sat, 25 Nov 2006 22:44:48 +0000 (22:44 +0000)]
Correct misspellings of kB.

18 years agoCosmetic release note fix.
Neil Conway [Sat, 25 Nov 2006 07:03:57 +0000 (07:03 +0000)]
Cosmetic release note fix.

18 years agoupdate for rc1 REL8_2_RC1
PostgreSQL Daemon [Sat, 25 Nov 2006 03:34:13 +0000 (03:34 +0000)]
update for rc1

18 years agoCopy fsync() defines into test_fsync.c, someday place them in an
Bruce Momjian [Sat, 25 Nov 2006 01:22:28 +0000 (01:22 +0000)]
Copy fsync() defines into test_fsync.c, someday place them in an
include.

Propery align for O_DIRECT.

Check for write()/fsync() failures.

18 years agoDocument that to_char() "TM" is controlled by lc_messages.
Bruce Momjian [Sat, 25 Nov 2006 00:38:53 +0000 (00:38 +0000)]
Document that to_char() "TM" is controlled by lc_messages.

18 years agoUpdate release notes for RC1.
Tom Lane [Fri, 24 Nov 2006 23:31:55 +0000 (23:31 +0000)]
Update release notes for RC1.

18 years agoFix psql's \copy command to ensure that it cycles libpq back to the idle state
Tom Lane [Fri, 24 Nov 2006 23:06:50 +0000 (23:06 +0000)]
Fix psql's \copy command to ensure that it cycles libpq back to the idle state
(in particular, causing the ReadyForQuery message to be eaten) before
returning from do_copy.  The only known consequence of failing to do so is
that get_prompt might show a wrong result for the %x transaction status
escape, as reported by Bernd Helmle; but it's possible there are other issues.

Back-patch as far as 7.4, the oldest version supporting %x.

18 years agoRevert (too late in beta):
Bruce Momjian [Fri, 24 Nov 2006 22:25:56 +0000 (22:25 +0000)]
Revert (too late in beta):

Fix to_char() locale handling to honor LC_TIME, not LC_MESSAGES.

Euler Taveira de Oliveira

18 years agoChange pg_stat_all_tables and sister views to put the recently-added
Tom Lane [Fri, 24 Nov 2006 21:18:42 +0000 (21:18 +0000)]
Change pg_stat_all_tables and sister views to put the recently-added
vacuum/analyze timestamp columns at the end, rather than at a random
spot in the middle as in the original patch.  This was deemed more usable
as well as less likely to break existing application code.  initdb forced
accordingly.  In passing, remove former kluge for initializing
pg_stat_file()'s pg_proc entry --- bootstrap mode was fixed recently
so that this can be done without any hacks, but I overlooked this usage.

18 years agoMake contrib/isn pass the opr_sanity sanity checks: add missing
Tom Lane [Fri, 24 Nov 2006 18:44:37 +0000 (18:44 +0000)]
Make contrib/isn pass the opr_sanity sanity checks: add missing
commutator operators, and mark hash-opclass members as oprcanhash.
This is a pretty ugly, brute-force solution, but it seems that getting
rid of all these redundant-looking operators would require some tweaks
in the core operator-resolution code to behave nicely, and I'm not
willing to risk that just before RC1.

18 years agoTranslation updates
Peter Eisentraut [Fri, 24 Nov 2006 17:11:57 +0000 (17:11 +0000)]
Translation updates

18 years agoSeparate release preparation jobs for all releases and for major releases
Peter Eisentraut [Fri, 24 Nov 2006 17:03:50 +0000 (17:03 +0000)]
Separate release preparation jobs for all releases and for major releases