OSDN Git Service

pg-rex/syncrep.git
14 years agoDon't allow walsender to send WAL data until it's been safely fsync'd on the
Tom Lane [Thu, 17 Jun 2010 16:41:25 +0000 (16:41 +0000)]
Don't allow walsender to send WAL data until it's been safely fsync'd on the
master.  Otherwise a subsequent crash could cause the master to lose WAL that
has already been applied on the slave, resulting in the slave being out of
sync and soon corrupt.  Per recent discussion and an example from Robert Haas.

Fujii Masao

14 years agoFix typo, init => int, per KOIZUMI Satoru.
Tom Lane [Thu, 17 Jun 2010 16:03:30 +0000 (16:03 +0000)]
Fix typo, init => int, per KOIZUMI Satoru.

14 years agoDocument that receive location can rewind if replication restarts.
Robert Haas [Thu, 17 Jun 2010 01:32:09 +0000 (01:32 +0000)]
Document that receive location can rewind if replication restarts.

Fujii Masao, with some further wordsmithing by me.

14 years agoRemove prototype of GetOldestWALSendPointer(), that is marked as NOT_USED.
Itagaki Takahiro [Thu, 17 Jun 2010 00:06:34 +0000 (00:06 +0000)]
Remove prototype of GetOldestWALSendPointer(), that is marked as NOT_USED.

14 years agoFix pg_upgrade to remove malloc(0) call.
Bruce Momjian [Wed, 16 Jun 2010 19:43:11 +0000 (19:43 +0000)]
Fix pg_upgrade to remove malloc(0) call.

14 years agoUpdate doc description for 9.0 release note item:
Bruce Momjian [Wed, 16 Jun 2010 18:46:18 +0000 (18:46 +0000)]
Update doc description for 9.0 release note item:

      Have <command>SELECT</> and <command>CREATE TABLE AS</> return
      row counts to the client

14 years ago9.0 release notes updates.
Bruce Momjian [Wed, 16 Jun 2010 18:31:39 +0000 (18:31 +0000)]
9.0 release notes updates.

Josh Berkus

14 years agoRemove perl symbol table additions for plperl functions, and mention of it
Andrew Dunstan [Wed, 16 Jun 2010 14:50:34 +0000 (14:50 +0000)]
Remove perl symbol table additions for plperl functions, and mention of it
in the release notes, as it is not apparently providing anything useful.

14 years agoAdd notes that CREATE/DROP CONVERSION is similar to CREATE/DROP
Peter Eisentraut [Wed, 16 Jun 2010 02:12:51 +0000 (02:12 +0000)]
Add notes that CREATE/DROP CONVERSION is similar to CREATE/DROP
TRANSLATION in the SQL standard.

14 years agoRefactor sprintf calls with computed format strings into multiple calls with
Peter Eisentraut [Wed, 16 Jun 2010 00:54:16 +0000 (00:54 +0000)]
Refactor sprintf calls with computed format strings into multiple calls with
constant format strings, so that the compiler can more easily check the
formats for correctness.

14 years agoFix storage of getopt() return, should be 'int', for pg_upgrade.
Bruce Momjian [Tue, 15 Jun 2010 23:25:01 +0000 (23:25 +0000)]
Fix storage of getopt() return, should be 'int', for pg_upgrade.

Steve Singer

14 years agoChange the interpretation of the primary_key_attnums parameter of
Tom Lane [Tue, 15 Jun 2010 20:29:01 +0000 (20:29 +0000)]
Change the interpretation of the primary_key_attnums parameter of
dblink_build_sql_insert() and related functions.  Now the column numbers
are treated as logical not physical column numbers.  This will provide saner
behavior in the presence of dropped columns; furthermore, if we ever get
around to allowing rearrangement of logical column ordering, the original
definition would become nearly untenable from a usability standpoint.
Per recent discussion of dblink's handling of dropped columns.
Not back-patched for fear of breaking existing applications.

14 years agoClarify SELECT FOR UPDATE behavior in docs.
Bruce Momjian [Tue, 15 Jun 2010 20:04:53 +0000 (20:04 +0000)]
Clarify SELECT FOR UPDATE behavior in docs.

14 years agoRemove hstore's text[] => text[] operator.
Robert Haas [Tue, 15 Jun 2010 19:48:30 +0000 (19:48 +0000)]
Remove hstore's text[] => text[] operator.

This is not yet in any released version, so we still have the option to
backtrack.  Instead, document hstore(text[], text[]).  Per discussion.

14 years agoFix dblink_build_sql_insert() and related functions to handle dropped
Tom Lane [Tue, 15 Jun 2010 19:04:15 +0000 (19:04 +0000)]
Fix dblink_build_sql_insert() and related functions to handle dropped
columns correctly.  In passing, get rid of some dead logic in the
underlying get_sql_insert() etc functions --- there is no caller that
will pass null value-arrays to them.

Per bug report from Robert Voinea.

14 years agoDocument new 9.0 behavior of ANALYZE on inheritance hierarchies.
Robert Haas [Tue, 15 Jun 2010 18:43:35 +0000 (18:43 +0000)]
Document new 9.0 behavior of ANALYZE on inheritance hierarchies.

In particular, note that autovacuum does not yet understand that it might
need to vacuum inheritance parents as a result of changes to the child
tables.

14 years agoConsolidate and improve checking of key-column-attnum arguments for
Tom Lane [Tue, 15 Jun 2010 16:22:19 +0000 (16:22 +0000)]
Consolidate and improve checking of key-column-attnum arguments for
dblink_build_sql_insert() and related functions.  In particular, be sure to
reject references to dropped and out-of-range column numbers.  The numbers
are still interpreted as physical column numbers, though, for backward
compatibility.

This patch replaces Joe's patch of 2010-02-03, which handled only some aspects
of the problem.

14 years agoUnbreak MSVC builds for pg_archivecleanup by linking with libpgport
Andrew Dunstan [Tue, 15 Jun 2010 12:48:36 +0000 (12:48 +0000)]
Unbreak MSVC builds for pg_archivecleanup by linking with libpgport

14 years agoAdd new GUC categories corresponding to sections in docs, and move
Itagaki Takahiro [Tue, 15 Jun 2010 07:52:11 +0000 (07:52 +0000)]
Add new GUC categories corresponding to sections in docs, and move
description for vacuum_defer_cleanup_age to the correct category.
Sections in postgresql.conf are also sorted in the same order with docs.

Per gripe by Fujii Masao, suggestion by Heikki Linnakangas, and patch by me.

14 years agoProperly define pg_upgrade global variable, per bug report from Chris
Bruce Momjian [Tue, 15 Jun 2010 02:08:01 +0000 (02:08 +0000)]
Properly define pg_upgrade global variable, per bug report from Chris
Ruprecht on Mac (64 bit).

14 years agoRearrange dblink's dblink_build_sql_insert() and related routines to open and
Tom Lane [Mon, 14 Jun 2010 20:49:33 +0000 (20:49 +0000)]
Rearrange dblink's dblink_build_sql_insert() and related routines to open and
lock the target relation just once per SQL function call.  The original coding
obtained and released lock several times per call.  Aside from saving a
not-insignificant number of cycles, this eliminates possible race conditions
if someone tries to modify the relation's schema concurrently.  Also
centralize locking and permission-checking logic.

Problem noted while investigating a trouble report from Robert Voinea --- his
problem is still to be fixed, though.

14 years agoFix doc plperl doc with is -> are change.
Bruce Momjian [Mon, 14 Jun 2010 18:47:05 +0000 (18:47 +0000)]
Fix doc plperl doc with is -> are change.

14 years agoDocs for pg_archivecleanup
Simon Riggs [Mon, 14 Jun 2010 17:25:24 +0000 (17:25 +0000)]
Docs for pg_archivecleanup

14 years agoFiles for pg_archivecleanup
Simon Riggs [Mon, 14 Jun 2010 16:19:24 +0000 (16:19 +0000)]
Files for pg_archivecleanup

14 years agoNew contrib module for use as an archive_cleanup_command, or as a
Simon Riggs [Mon, 14 Jun 2010 16:17:56 +0000 (16:17 +0000)]
New contrib module for use as an archive_cleanup_command, or as a
standalone utility for removing files from archive.

14 years agoFix typo in Japanese translation for psql "Use \d+ to list them."
Itagaki Takahiro [Mon, 14 Jun 2010 08:54:35 +0000 (08:54 +0000)]
Fix typo in Japanese translation for psql "Use \d+ to list them."

14 years agoIf a corrupt WAL record is received by streaming replication, disconnect
Heikki Linnakangas [Mon, 14 Jun 2010 06:04:21 +0000 (06:04 +0000)]
If a corrupt WAL record is received by streaming replication, disconnect
and retry. If the record is genuinely corrupt in the master database,
there's little hope of recovering, but it's better than simply retrying
to apply the corrupt WAL record in a tight loop without even trying to
retransmit it, which is what we used to do.

14 years agoFix misplaced modifier.
Robert Haas [Mon, 14 Jun 2010 02:18:43 +0000 (02:18 +0000)]
Fix misplaced modifier.

As suggested by Ian Barwick.

14 years agoFix typo.
Robert Haas [Mon, 14 Jun 2010 01:07:41 +0000 (01:07 +0000)]
Fix typo.

14 years agoRemove max_standby_delay message from ps display of recovery process
Itagaki Takahiro [Mon, 14 Jun 2010 00:49:24 +0000 (00:49 +0000)]
Remove max_standby_delay message from ps display of recovery process
in waiting status. The parameter is not so interesting in ps display
because it is referable in postgresql.conf.

14 years agoFix ALTER LARGE OBJECT and GRANT ... ON LARGE OBJECT for large OIDs.
Robert Haas [Sun, 13 Jun 2010 17:43:13 +0000 (17:43 +0000)]
Fix ALTER LARGE OBJECT and GRANT ... ON LARGE OBJECT for large OIDs.

The previous coding failed for OIDs too large to be represented by
a signed integer.

14 years agopostgres.txt should get cleaned by 'make clean'.
Tom Lane [Sat, 12 Jun 2010 21:40:31 +0000 (21:40 +0000)]
postgres.txt should get cleaned by 'make clean'.

14 years agoUpdate pg_upgrade C comment about cwd.
Bruce Momjian [Sat, 12 Jun 2010 17:45:28 +0000 (17:45 +0000)]
Update pg_upgrade C comment about cwd.

14 years agoAdd missing --use-existing --help mention from regression binary.
Bruce Momjian [Sat, 12 Jun 2010 17:21:29 +0000 (17:21 +0000)]
Add missing --use-existing --help mention from regression binary.

Jan Urba?ski

14 years agoRemove lynx -stdin flag for postgres.text.
Bruce Momjian [Sat, 12 Jun 2010 17:17:25 +0000 (17:17 +0000)]
Remove lynx -stdin flag for postgres.text.

14 years agoHave pg_upgrade create its output files in the current directory, rather
Bruce Momjian [Sat, 12 Jun 2010 17:05:29 +0000 (17:05 +0000)]
Have pg_upgrade create its output files in the current directory, rather
than in a subdirectory of the $HOME directory, or $TMP in Windows.

14 years agoAdd SGML Makefile rule for single-page text, postgres.txt.
Bruce Momjian [Sat, 12 Jun 2010 15:58:38 +0000 (15:58 +0000)]
Add SGML Makefile rule for single-page text, postgres.txt.

14 years agoAdd space between after ">" in ">$@" in SGML Makefile, for clarity.
Bruce Momjian [Sat, 12 Jun 2010 15:42:44 +0000 (15:42 +0000)]
Add space between after ">" in ">$@" in SGML Makefile, for clarity.

14 years agoAdd target to build HTML documentation as single page
Peter Eisentraut [Sat, 12 Jun 2010 09:55:12 +0000 (09:55 +0000)]
Add target to build HTML documentation as single page

14 years agoFix typo/bug, found by Clang compiler
Peter Eisentraut [Sat, 12 Jun 2010 09:14:52 +0000 (09:14 +0000)]
Fix typo/bug, found by Clang compiler

14 years agoUpdate Python version information
Peter Eisentraut [Sat, 12 Jun 2010 06:05:48 +0000 (06:05 +0000)]
Update Python version information

14 years agoAdd a regression test case for bug #5497
Peter Eisentraut [Sat, 12 Jun 2010 06:05:20 +0000 (06:05 +0000)]
Add a regression test case for bug #5497

14 years agoAdd missing 'Z' letter to getopt_long call --- the newly added
Tom Lane [Fri, 11 Jun 2010 23:58:24 +0000 (23:58 +0000)]
Add missing 'Z' letter to getopt_long call --- the newly added
--analyze-only switch did not work in its short form -Z.

Josh Berkus

14 years agoUse "replication" as the database name when constructing a connection
Heikki Linnakangas [Fri, 11 Jun 2010 10:13:09 +0000 (10:13 +0000)]
Use "replication" as the database name when constructing a connection
string for a streaming replication connection. It's ignored by the
server, but allows libpq to pick up the password from .pgpass where
"replication" is specified as the database name.

Patch by Fujii Masao per Tom's suggestion, with some wording changes by me.

14 years agoUpdate text of 9.0 release notes
Bruce Momjian [Thu, 10 Jun 2010 21:48:28 +0000 (21:48 +0000)]
Update text of 9.0 release notes

Josh Berkus

14 years agoRename restartpoint_command to archive_cleanup_command.
Itagaki Takahiro [Thu, 10 Jun 2010 08:13:50 +0000 (08:13 +0000)]
Rename restartpoint_command to archive_cleanup_command.

14 years agoMake TriggerFile variable static. It's not used outside xlog.c.
Heikki Linnakangas [Thu, 10 Jun 2010 07:49:23 +0000 (07:49 +0000)]
Make TriggerFile variable static. It's not used outside xlog.c.

Fujii Masao

14 years agoReturn NULL instead of 0/0 in pg_last_xlog_receive_location() and
Heikki Linnakangas [Thu, 10 Jun 2010 07:00:27 +0000 (07:00 +0000)]
Return NULL instead of 0/0 in pg_last_xlog_receive_location() and
pg_last_xlog_replay_location(). Per Robert Haas's suggestion, after
Itagaki Takahiro pointed out an issue in the docs. Also, some wording
changes in the docs by me.

14 years agoFix quite-bogus handling of arrays in plpython datum-to-PyObject
Tom Lane [Thu, 10 Jun 2010 04:05:01 +0000 (04:05 +0000)]
Fix quite-bogus handling of arrays in plpython datum-to-PyObject
conversion.  Per bug #5497 from David Gardner.

14 years agoQuote all string values in EXPLAIN (FORMAT YAML) output.
Robert Haas [Thu, 10 Jun 2010 01:26:30 +0000 (01:26 +0000)]
Quote all string values in EXPLAIN (FORMAT YAML) output.

While my previous attempt seems to always produce valid YAML, it
doesn't always produce YAML that means what it appears to mean,
because of tokens like "0xa" and "true", which without quotes will
be interpreted as integer or Boolean literals.  So, instead, just
quote everything that's not known to be a number, as we do for
JSON.

Dean Rasheed, with some changes to the comments by me.

14 years agoRemove stray word from comment.
Robert Haas [Wed, 9 Jun 2010 21:14:28 +0000 (21:14 +0000)]
Remove stray word from comment.

14 years agoUpdate ALTER TABLE docs to account for exclusion and deferrable uniqueness
Alvaro Herrera [Wed, 9 Jun 2010 17:48:10 +0000 (17:48 +0000)]
Update ALTER TABLE docs to account for exclusion and deferrable uniqueness
constraints

Dean Rasheed

14 years agoAdd index entry for ::, per complaint from John Gage.
Alvaro Herrera [Wed, 9 Jun 2010 16:43:47 +0000 (16:43 +0000)]
Add index entry for ::, per complaint from John Gage.

14 years agoIn standby mode, respect checkpoint_segments in addition to
Heikki Linnakangas [Wed, 9 Jun 2010 15:04:07 +0000 (15:04 +0000)]
In standby mode, respect checkpoint_segments in addition to
checkpoint_timeout to trigger restartpoints. We used to deliberately only
do time-based restartpoints, because if checkpoint_segments is small we
would spend time doing restartpoints more often than really necessary.
But now that restartpoints are done in bgwriter, they're not as
disruptive as they used to be. Secondly, because streaming replication
stores the streamed WAL files in pg_xlog, we want to clean it up more
often to avoid running out of disk space when checkpoint_timeout is large
and checkpoint_segments small.

Patch by Fujii Masao, with some minor changes by me.

14 years agoMake the walwriter close it's handle to an old xlog segment if it's no longer
Magnus Hagander [Wed, 9 Jun 2010 10:54:45 +0000 (10:54 +0000)]
Make the walwriter close it's handle to an old xlog segment if it's no longer
the current one. Not doing this would leave the walwriter with a handle to a
deleted file if there was nothing for it to do for a long period of time,
preventing the file from  being completely removed.

Reported by Tollef Fog Heen, and thanks to Heikki for some hand-holding with
the patch.

14 years agoAvoid "identifier will be truncated" warning in dblink
Itagaki Takahiro [Wed, 9 Jun 2010 03:39:26 +0000 (03:39 +0000)]
Avoid "identifier will be truncated" warning in dblink
when connection string is longer than NAMEDATALEN.
The previous fix for long connection name broke the behavior.

14 years agoAttempt to fix EXPLAIN (FORMAT YAML) quoting to behave sanely.
Robert Haas [Wed, 9 Jun 2010 02:39:34 +0000 (02:39 +0000)]
Attempt to fix EXPLAIN (FORMAT YAML) quoting to behave sanely.

The previous code failed to quote in many cases where quoting was necessary -
YAML has loads of special characters, including -:[]{},"'|*& - so quote much
more aggressively, and only refrain from quoting things where it seems fairly
clear that it isn't necessary.

Per report from Dean Rasheed.

14 years agoFix connection leak in dblink when dblink_connect() or dblink_connect_u()
Itagaki Takahiro [Wed, 9 Jun 2010 00:56:02 +0000 (00:56 +0000)]
Fix connection leak in dblink when dblink_connect() or dblink_connect_u()
end with "duplicate connection name" errors.

Backported to release 7.4.

14 years agoFix typo in the header comment. Per request from Masao Fujii.
Tatsuo Ishii [Wed, 9 Jun 2010 00:54:39 +0000 (00:54 +0000)]
Fix typo in the header comment. Per request from Masao Fujii.

14 years agoMake procedural language handler reference C-language function docs.
Robert Haas [Tue, 8 Jun 2010 20:12:59 +0000 (20:12 +0000)]
Make procedural language handler reference C-language function docs.

Based on suggestions from Jonathan Leto and Joshua Tolley.

14 years agoAvoid useless snprintf() call when update_process_title is turned off.
Tom Lane [Mon, 7 Jun 2010 15:49:30 +0000 (15:49 +0000)]
Avoid useless snprintf() call when update_process_title is turned off.
Fujii Masao

14 years agoAdd missed function dblink_connect_u(text[,text]) to uninstall script
Teodor Sigaev [Mon, 7 Jun 2010 15:14:36 +0000 (15:14 +0000)]
Add missed function dblink_connect_u(text[,text]) to uninstall script

14 years agoEnsure default-only storage parameters for TOAST relations
Itagaki Takahiro [Mon, 7 Jun 2010 02:59:02 +0000 (02:59 +0000)]
Ensure default-only storage parameters for TOAST relations
to be initialized with proper values. Affected parameters are
fillfactor, analyze_threshold, and analyze_scale_factor.

Especially uninitialized fillfactor caused inefficient page usage
because we built a StdRdOptions struct in which fillfactor is zero
if any reloption is set for the toast table.

In addition, we disallow toast.autovacuum_analyze_threshold and
toast.autovacuum_analyze_scale_factor because we didn't actually
support them; they are always ignored.

Report by Rumko on pgsql-bugs on 12 May 2010.
Analysis by Tom Lane and Alvaro Herrera. Patch by me.

Backpatch to 8.4.

14 years agoReplace "slave" to "standby" in documentation for consistent terminology.
Itagaki Takahiro [Mon, 7 Jun 2010 02:01:09 +0000 (02:01 +0000)]
Replace "slave" to "standby" in documentation for consistent terminology.
Almost all of the terms in docs and messages were replaced, but still
remains in a few comments and README files in codes.

14 years agoImprove our explanation of the OVERLAPS operator.
Tom Lane [Sat, 5 Jun 2010 14:56:36 +0000 (14:56 +0000)]
Improve our explanation of the OVERLAPS operator.
Per gripe from Frank van Vugt.

14 years agoAdded variable handling for RETURNING clause to ecpg.
Michael Meskes [Fri, 4 Jun 2010 10:09:58 +0000 (10:09 +0000)]
Added variable handling for RETURNING clause to ecpg.

While the values were correctly returned they were not moved into C variables
as they should be.

Closes: #5489

14 years agotag 9.0beta2
Marc G. Fournier [Fri, 4 Jun 2010 07:28:30 +0000 (07:28 +0000)]
tag 9.0beta2

14 years agoAdjust misleading comment in walsender.c. We try to send all WAL data that's
Tom Lane [Thu, 3 Jun 2010 23:00:14 +0000 (23:00 +0000)]
Adjust misleading comment in walsender.c.  We try to send all WAL data that's
been written out from shared memory, but the previous phrasing might be read
to say that we send only what's been fsync'd.

14 years agoAdd current WAL end (as seen by walsender, ie, GetWriteRecPtr() result)
Tom Lane [Thu, 3 Jun 2010 22:17:32 +0000 (22:17 +0000)]
Add current WAL end (as seen by walsender, ie, GetWriteRecPtr() result)
and current server clock time to SR data messages.  These are not currently
used on the slave side but seem likely to be useful in future, and it'd be
better not to change the SR protocol after release.  Per discussion.
Also do some minor code review and cleanup on walsender.c, and improve the
protocol documentation.

14 years agoRemove link that breaks HISTORY file generation.
Tom Lane [Thu, 3 Jun 2010 21:23:02 +0000 (21:23 +0000)]
Remove link that breaks HISTORY file generation.

14 years agoTranslation updates for 9.0beta2
Peter Eisentraut [Thu, 3 Jun 2010 21:12:05 +0000 (21:12 +0000)]
Translation updates for 9.0beta2

14 years agoFix some inconsistent quoting of wal_level values in messages
Peter Eisentraut [Thu, 3 Jun 2010 21:02:12 +0000 (21:02 +0000)]
Fix some inconsistent quoting of wal_level values in messages

When referring to postgresql.conf syntax, then it's without quotes
(wal_level=archive); in narrative it's with double quotes.  But never
single quotes.

14 years agoAdd comments about definitions that may affect PG_CONTROL_VERSION,
Alvaro Herrera [Thu, 3 Jun 2010 20:37:13 +0000 (20:37 +0000)]
Add comments about definitions that may affect PG_CONTROL_VERSION,
per recent unintended-initdb-forcing fiasco

14 years agoFix CREATE TRIGGER release mention, WHERE -> WHEN.
Bruce Momjian [Thu, 3 Jun 2010 19:38:39 +0000 (19:38 +0000)]
Fix CREATE TRIGGER release mention, WHERE -> WHEN.

14 years agoFix reference to nonexistent configure option
Peter Eisentraut [Thu, 3 Jun 2010 19:29:38 +0000 (19:29 +0000)]
Fix reference to nonexistent configure option

--enable-ssl -> --with-openssl

14 years agoUpdate 9.0 release notes to current.
Bruce Momjian [Thu, 3 Jun 2010 16:33:36 +0000 (16:33 +0000)]
Update 9.0 release notes to current.

14 years agoBump PG_CONTROL_VERSION to account for the incompatible change committed earlier.
Tom Lane [Thu, 3 Jun 2010 14:50:30 +0000 (14:50 +0000)]
Bump PG_CONTROL_VERSION to account for the incompatible change committed earlier.

14 years agoMarkup fix.
Bruce Momjian [Thu, 3 Jun 2010 14:41:25 +0000 (14:41 +0000)]
Markup fix.

14 years agoDocument regexp_matches() better and show example of single-row usage.
Bruce Momjian [Thu, 3 Jun 2010 14:40:42 +0000 (14:40 +0000)]
Document regexp_matches() better and show example of single-row usage.

14 years agoDocument use of C++ for extension use.
Bruce Momjian [Thu, 3 Jun 2010 14:39:58 +0000 (14:39 +0000)]
Document use of C++ for extension use.

14 years agoFix dblink to treat connection names longer than NAMEDATALEN-2 (62 bytes).
Itagaki Takahiro [Thu, 3 Jun 2010 09:38:33 +0000 (09:38 +0000)]
Fix dblink to treat connection names longer than NAMEDATALEN-2 (62 bytes).
Now long names are adjusted with truncate_identifier() and NOTICE messages
are raised if names are actually truncated.

Backported to release 8.0.

14 years agoOn clean shutdown during recovery, don't warn about possible corruption.
Robert Haas [Thu, 3 Jun 2010 03:20:00 +0000 (03:20 +0000)]
On clean shutdown during recovery, don't warn about possible corruption.

Fujii Masao.  Review by Heikki Linnakangas and myself.

14 years agoDocument that citext operators must be in the current search path.
Bruce Momjian [Thu, 3 Jun 2010 03:04:55 +0000 (03:04 +0000)]
Document that citext operators must be in the current search path.

14 years agoClarify array generate_subscripts() documentation example.
Bruce Momjian [Thu, 3 Jun 2010 02:06:10 +0000 (02:06 +0000)]
Clarify array generate_subscripts() documentation example.

Tim Landscheidt

14 years agoDocument that && geometric operator is true even if only a point
Bruce Momjian [Thu, 3 Jun 2010 01:34:02 +0000 (01:34 +0000)]
Document that && geometric operator is true even if only a point
overlaps.

David Fetter

14 years agoFix obsolete comments that I neglected to update in a previous patch.
Heikki Linnakangas [Wed, 2 Jun 2010 09:28:44 +0000 (09:28 +0000)]
Fix obsolete comments that I neglected to update in a previous patch.

Fujii Masao

14 years agoUpdate C++ release note item wording.
Bruce Momjian [Tue, 1 Jun 2010 13:43:39 +0000 (13:43 +0000)]
Update C++ release note item wording.

14 years agoComment out C++ docs for later user. Tone down C++ compatibility in 9.0
Bruce Momjian [Tue, 1 Jun 2010 03:19:36 +0000 (03:19 +0000)]
Comment out C++ docs for later user.  Tone down C++ compatibility in 9.0
release notes.

14 years agoMention palloc/pfree for C++ memory allocation in docs.
Bruce Momjian [Tue, 1 Jun 2010 02:54:37 +0000 (02:54 +0000)]
Mention palloc/pfree for C++ memory allocation in docs.

14 years agoFix SGML markup for tag title.
Bruce Momjian [Tue, 1 Jun 2010 02:35:37 +0000 (02:35 +0000)]
Fix SGML markup for tag title.

14 years agoAdd documentation section "Using C++ for Extensibility".
Bruce Momjian [Tue, 1 Jun 2010 02:31:36 +0000 (02:31 +0000)]
Add documentation section "Using C++ for Extensibility".

Craig Ringer

14 years agoShow schema name for REINDEX.
Bruce Momjian [Tue, 1 Jun 2010 00:33:23 +0000 (00:33 +0000)]
Show schema name for REINDEX.

Greg Sabino Mullane

14 years agoOverflow box is on page after listed page number, per Alvaro.
Bruce Momjian [Mon, 31 May 2010 22:42:59 +0000 (22:42 +0000)]
Overflow box is on page after listed page number, per Alvaro.

14 years agoUpdate TeX overflow documentation page pattern mask, in docs.
Bruce Momjian [Mon, 31 May 2010 22:15:25 +0000 (22:15 +0000)]
Update TeX overflow documentation page pattern mask, in docs.

14 years agoDocument how to find TeX overflow boxes in our documentation build, per
Bruce Momjian [Mon, 31 May 2010 22:14:20 +0000 (22:14 +0000)]
Document how to find TeX overflow boxes in our documentation build, per
details from Alvaro Herrera.

14 years agoAdd error hint that PL/pgSQL "EXECUTE of SELECT ... INTO" can be
Bruce Momjian [Mon, 31 May 2010 20:02:30 +0000 (20:02 +0000)]
Add error hint that PL/pgSQL "EXECUTE of SELECT ... INTO" can be
performed by "EXECUTE ... INTO".

Jaime Casanova

14 years agoAdd note about database ownership to REASSIGN OWNED command documentation,
Alvaro Herrera [Mon, 31 May 2010 18:47:35 +0000 (18:47 +0000)]
Add note about database ownership to REASSIGN OWNED command documentation,
per Josh Berkus.  Add ALTER DATABASE to the "see also" section, too.

14 years agoReword fsync and full_page_writes docs to be clearer about when to turn
Bruce Momjian [Mon, 31 May 2010 15:50:48 +0000 (15:50 +0000)]
Reword fsync and full_page_writes docs to be clearer about when to turn
them off.

Josh Berkus, with slight wording changes by me.

14 years agoSend all outstanding WAL before exiting when smart shutdown is requested.
Heikki Linnakangas [Mon, 31 May 2010 10:44:37 +0000 (10:44 +0000)]
Send all outstanding WAL before exiting when smart shutdown is requested.
This was broken by my previous patch to send WAL in smaller batches.

Patch by Fujii Masao.

14 years agoFix misuse of Lossy Counting (LC) algorithm in compute_tsvector_stats().
Tom Lane [Sun, 30 May 2010 21:59:02 +0000 (21:59 +0000)]
Fix misuse of Lossy Counting (LC) algorithm in compute_tsvector_stats().

We must filter out hashtable entries with frequencies less than those
specified by the algorithm, else we risk emitting junk entries whose
actual frequency is much less than other lexemes that did not get
tabulated.  This is bad enough by itself, but even worse is that
tsquerysel() believes that the minimum frequency seen in pg_statistic is a
hard upper bound for lexemes not included, and was thus underestimating
the frequency of non-MCEs.

Also, set the threshold frequency to something with a little bit of theory
behind it, to wit assume that the input distribution is approximately
Zipfian.  This might need adjustment in future, but some preliminary
experiments suggest that it's not too unreasonable.

Back-patch to 8.4, where this code was introduced.

Jan Urbanski, with some editorialization by Tom