OSDN Git Service

pg-rex/syncrep.git
20 years agoBack out:
Bruce Momjian [Sun, 21 Dec 2003 04:34:36 +0000 (04:34 +0000)]
Back out:

>  Attached is a patch that addressed all the discussed issues
>  that did not break backward compatability, including the
>  ability to output ISO-8601 compliant intervals by setting
>  datestyle to iso8601basic.

20 years ago[ This description should have been on the earlier fork/exec
Bruce Momjian [Sun, 21 Dec 2003 04:30:10 +0000 (04:30 +0000)]
[ This description should have been on the earlier fork/exec
  commit, but I am adding it now so it is in CVS.]

The patch basically is a slight rearrangement of the code to allow
fork/exec on Unix, with the ultimate goal of doing CreateProcess on
Win32.  The changes are:

        o  Write out postmaster global variables and per-backend
variables to be read by the exec'ed backend

        o  Mark some static variables as global when exec is used so
then can be dumped from postmaster.c, marked NON_EXEC_STATIC

        o  Remove value passing with -p now that we have per-backend
file

        o  Move some pointer storage out of shared memory for easier
dumping.

        o  Modified pgsql_temp directory cleanup to handle per-database
directories and the backend exec directory under datadir.

Claudio Natoli

20 years agoPrevious change exposed some opportunities for further simplification
Tom Lane [Sun, 21 Dec 2003 03:00:04 +0000 (03:00 +0000)]
Previous change exposed some opportunities for further simplification
in _bt_first().

20 years agoImprove btree's initial-positioning-strategy code so that we never need
Tom Lane [Sun, 21 Dec 2003 01:23:06 +0000 (01:23 +0000)]
Improve btree's initial-positioning-strategy code so that we never need
to step more than one entry after descending the search tree to arrive at
the correct place to start the scan.  This can improve the behavior
substantially when there are many entries equal to the chosen boundary
value.  Per suggestion from Dmitry Tkach, 14-Jul-03.

20 years agoThe recent DUMMY_PROCS patch broke accounting for the number of semaphores
Tom Lane [Sun, 21 Dec 2003 00:33:33 +0000 (00:33 +0000)]
The recent DUMMY_PROCS patch broke accounting for the number of semaphores
needed.  This caused us to fail all the time on Darwin, and we'd fail for
some values of maxBackends on SysV-sema platforms, too.

20 years agobufmgr.c failed to compile on Darwin, because it didn't include
Tom Lane [Sat, 20 Dec 2003 22:18:02 +0000 (22:18 +0000)]
bufmgr.c failed to compile on Darwin, because it didn't include
<sys/time.h> where struct timeval is defined.

20 years agoDept. of third thoughts: in fact, libpq should support SCM_CREDS challenge
Tom Lane [Sat, 20 Dec 2003 18:45:49 +0000 (18:45 +0000)]
Dept. of third thoughts: in fact, libpq should support SCM_CREDS challenge
even when HAVE_GETPEEREID is defined, else it will be unable to connect to
pre-7.4 backends that are using IDENT authentication.

20 years agoFix broken IDENT support for FreeBSD (appears to have been broken by
Tom Lane [Sat, 20 Dec 2003 18:24:52 +0000 (18:24 +0000)]
Fix broken IDENT support for FreeBSD (appears to have been broken by
ill-considered conditional logic in getpeereid patch of 3-Dec-2002).
Per bug #1021.

20 years agoThis patch is the next step towards (re)allowing fork/exec.
Bruce Momjian [Sat, 20 Dec 2003 17:31:21 +0000 (17:31 +0000)]
This patch is the next step towards (re)allowing fork/exec.

Claudio Natoli

20 years agoAnother markup fix for recent patch.
Bruce Momjian [Sat, 20 Dec 2003 16:06:54 +0000 (16:06 +0000)]
Another markup fix for recent patch.

20 years agoFix broken markup.
Bruce Momjian [Sat, 20 Dec 2003 15:59:25 +0000 (15:59 +0000)]
Fix broken markup.

20 years agoIn my mind there were two categories of open issues
Bruce Momjian [Sat, 20 Dec 2003 15:32:55 +0000 (15:32 +0000)]
In my mind there were two categories of open issues
  a) ones that are 100% backward (such as the comment about
     outputting this format)
and
  b) ones that aren't (such as deprecating the current
     postgresql shorthand of
         '1Y1M'::interval = 1 year 1 minute
     in favor of the ISO-8601
         'P1Y1M'::interval = 1 year 1 month.

Attached is a patch that addressed all the discussed issues that
did not break backward compatability, including the ability to
output ISO-8601 compliant intervals by setting datestyle to
iso8601basic.

Interval values can now be written as  ISO 8601 time intervals, using
the "Format with time-unit designators". This format always starts with
the character 'P', followed  by a string of values followed
by single character time-unit designators. A 'T' separates the date and
time parts of the interval.

Ron Mayer

20 years agoUpdate HISTORY file for 7.4.1.
Bruce Momjian [Sat, 20 Dec 2003 00:57:22 +0000 (00:57 +0000)]
Update HISTORY file for 7.4.1.

20 years agoSupress ecpg thread test if configure didn't enable threads. Fix
Bruce Momjian [Fri, 19 Dec 2003 23:29:15 +0000 (23:29 +0000)]
Supress ecpg thread test if configure didn't enable threads.  Fix
tools/thread to run even if configure didn't enable threads because this
test is used before enabling threads for the OS.

20 years agoPrevent service dbname from defaulting to service name, per bug report
Bruce Momjian [Fri, 19 Dec 2003 21:50:54 +0000 (21:50 +0000)]
Prevent service dbname from defaulting to service name, per bug report
from Michael Fuhr

20 years ago7.4.1 release note improvements
Peter Eisentraut [Fri, 19 Dec 2003 20:06:20 +0000 (20:06 +0000)]
7.4.1 release note improvements

20 years agoForbid REVOKE on untrusted languages, and don't dump privileges of
Peter Eisentraut [Fri, 19 Dec 2003 14:21:56 +0000 (14:21 +0000)]
Forbid REVOKE on untrusted languages, and don't dump privileges of
untrusted languages (in case they sneak in).

20 years agoAdd missing $(X).
Peter Eisentraut [Fri, 19 Dec 2003 11:54:25 +0000 (11:54 +0000)]
Add missing $(X).

20 years agoMake to_hex() behave portably on negative input values (treat them as
Tom Lane [Fri, 19 Dec 2003 04:56:41 +0000 (04:56 +0000)]
Make to_hex() behave portably on negative input values (treat them as
unsigned integers).  Per report from Jim Crate.

20 years agoUse a shutdown callback to ensure proper clean up when rescanning
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

20 years agoFix memory leak with SSL connections due to missing X509_free() calls.
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.

20 years agoUse a shutdown callback to clear setArgsValid in a FuncExprState that is
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.

20 years agoForgot to change one compatlib.h.
Peter Eisentraut [Thu, 18 Dec 2003 20:25:58 +0000 (20:25 +0000)]
Forgot to change one compatlib.h.

20 years agoEnsure set-returning functions in the targetlist of a plan node will be
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.

20 years agoFix signed char in comparison and check memory allocation
Teodor Sigaev [Thu, 18 Dec 2003 19:27:53 +0000 (19:27 +0000)]
Fix signed char in comparison and check memory allocation

20 years agoMove Informix compatibility include files out of the way. compatlib.h
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.

20 years agopatch for new OID74Test
Dave Cramer [Thu, 18 Dec 2003 04:17:17 +0000 (04:17 +0000)]
patch for new OID74Test

20 years agofixed up OID74 test to conform with other tests, by Kris Jurka
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

20 years agoPlease find enclosed a patch exemplifying typical use of the ARE
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

20 years agoHere is the definition of relation_byte_size() in optimizer/path/costsize.c:
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

20 years agopatch by Kris Jurka to use the correct protocol based upon server information
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

20 years agoAdjust rules output for unknown-vs-any change (affects expected contents
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).

20 years agoBe a little smarter in group_clauses_by_indexkey_for_join: detect cases
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 ...

20 years agoinformation_schema.constraint_column_usage and key_column_usage should
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.

20 years agoFix DecodeInterval to handle '-0.1' sanely, per gripe from Tilo Schwarz.
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.

20 years agoRepair badly broken estimation of output buffer size in lquery_out().
Tom Lane [Wed, 17 Dec 2003 20:15:41 +0000 (20:15 +0000)]
Repair badly broken estimation of output buffer size in lquery_out().

20 years agoReorder tests in parse_coerce so that ANY/ANYELEMENT/ANYARRAY coercion
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.

20 years agoRemove pg_id.
Peter Eisentraut [Wed, 17 Dec 2003 18:44:09 +0000 (18:44 +0000)]
Remove pg_id.

20 years agoRepair planner failure when there are multiple IN clauses, each with
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.

20 years agorevoked patch from Kris Jurka to fix multiarguments, and changed test to create
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

20 years agopatch from Kris Jurka to fix large object 7.1 compatible protocol issues
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

20 years ago- Added just another patch by Dave that fixes a reversed order in
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.

20 years agoadded polish translation submitted by Piotr Maj
Dave Cramer [Wed, 17 Dec 2003 13:25:14 +0000 (13:25 +0000)]
added polish translation submitted by Piotr Maj

20 years agoRestore information schema upgrade instructions of Peter.
Bruce Momjian [Wed, 17 Dec 2003 07:07:00 +0000 (07:07 +0000)]
Restore information schema upgrade instructions of Peter.

20 years agoAdd example for converting epoch back to timestamp.
Peter Eisentraut [Tue, 16 Dec 2003 15:27:58 +0000 (15:27 +0000)]
Add example for converting epoch back to timestamp.

20 years agoFix constraint_column_usage for foreign keys.
Peter Eisentraut [Tue, 16 Dec 2003 14:57:20 +0000 (14:57 +0000)]
Fix constraint_column_usage for foreign keys.

20 years agoRepair name.
Peter Eisentraut [Tue, 16 Dec 2003 09:47:55 +0000 (09:47 +0000)]
Repair name.

20 years agoAdd mention of non-standard extension:
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?

20 years agoAdd for ecpg:
Bruce Momjian [Tue, 16 Dec 2003 01:08:40 +0000 (01:08 +0000)]
Add for ecpg:

>  o Make SET CONNECTION thread-aware

20 years agoMention ecpg SET CONNECTION is not thread-aware.
Bruce Momjian [Tue, 16 Dec 2003 01:06:39 +0000 (01:06 +0000)]
Mention ecpg SET CONNECTION is not thread-aware.

20 years agoOverride some of the changes in DocBook DSSSL stylesheets 1.78 to restore
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.

20 years agoAdd mention with might need to use cp -R someday for portability.
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.

20 years agoUpdate HISTORY and release notes for 7.4.1.
Bruce Momjian [Mon, 15 Dec 2003 22:24:59 +0000 (22:24 +0000)]
Update HISTORY and release notes for 7.4.1.

20 years agoFix two typos in the documentation for PREPARE.
Neil Conway [Sun, 14 Dec 2003 00:55:46 +0000 (00:55 +0000)]
Fix two typos in the documentation for PREPARE.

20 years agoI posted some bufmgr cleanup a few weeks ago, but it conflicted with
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.

20 years agoThis patch fixes a few more uppercase GUC vars. I also removed an
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.

20 years agoThis patch makes some improvements and adds some additional detail
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.

20 years agoThis patch makes some SGML markup more consistent and makes a small
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.

20 years agoThis patch makes some SGML markup more consistent and makes a small
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.

20 years agoAdd /usr/local/sgml/docbook-dsssl to the default search patch for
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.

20 years agoSGML doc build instructions --- no more DTDDECL, fix double dash in path name.
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.

20 years agoAdd fadvise TODO.detail.
Bruce Momjian [Sat, 13 Dec 2003 20:02:16 +0000 (20:02 +0000)]
Add fadvise TODO.detail.

20 years agoUpdate TODO.detail syntax:
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]

20 years agoAdd Czech language FAQ's
Bruce Momjian [Sat, 13 Dec 2003 16:56:00 +0000 (16:56 +0000)]
Add Czech language FAQ's

Pavel Stehule

20 years agoThis patch properly sets the prototype for the on_shmem_exit and
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

20 years agoadd missing SQLState by Patrick Higgins
Dave Cramer [Fri, 12 Dec 2003 18:36:20 +0000 (18:36 +0000)]
add missing SQLState by Patrick Higgins

20 years agocancel row updates sets values to null by Kris Jurka
Dave Cramer [Fri, 12 Dec 2003 18:34:14 +0000 (18:34 +0000)]
cancel row updates sets values to null by Kris Jurka

20 years agopatch to indicate why test cases failed from Oliver Jowett
Dave Cramer [Fri, 12 Dec 2003 18:30:27 +0000 (18:30 +0000)]
patch to indicate why test cases failed from Oliver Jowett

20 years agopatch to fix returning he information schema as a system object by Kris Jurka
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

20 years agofix casting pooled connections to PGStatement problem patch by JariP
Dave Cramer [Fri, 12 Dec 2003 18:05:34 +0000 (18:05 +0000)]
fix casting pooled connections to PGStatement problem patch by JariP

20 years agopatch to build under jdk1.1 from Richard Scranton
Dave Cramer [Fri, 12 Dec 2003 17:58:34 +0000 (17:58 +0000)]
patch to build under jdk1.1 from Richard Scranton

20 years agobinary stream patch by Kris Jurka fixes empty stream failure
Dave Cramer [Fri, 12 Dec 2003 00:26:20 +0000 (00:26 +0000)]
binary stream patch by Kris Jurka fixes empty stream failure

20 years agometadata encoding patch from Kris Jurka
Dave Cramer [Fri, 12 Dec 2003 00:21:36 +0000 (00:21 +0000)]
metadata encoding patch from Kris Jurka

20 years agoFix for vpath build
Peter Eisentraut [Thu, 11 Dec 2003 22:10:51 +0000 (22:10 +0000)]
Fix for vpath build

20 years agoNew translation
Peter Eisentraut [Thu, 11 Dec 2003 22:03:05 +0000 (22:03 +0000)]
New translation

20 years agoFix thinko in comment.
Tom Lane [Thu, 11 Dec 2003 21:21:55 +0000 (21:21 +0000)]
Fix thinko in comment.

20 years agoMake some wording consistent with DROP OPERATOR description.
Peter Eisentraut [Thu, 11 Dec 2003 20:13:53 +0000 (20:13 +0000)]
Make some wording consistent with DROP OPERATOR description.

20 years agoFix instructions how to shut down postmaster.
Peter Eisentraut [Thu, 11 Dec 2003 19:56:54 +0000 (19:56 +0000)]
Fix instructions how to shut down postmaster.

20 years agobuild number set to 300 to avoid conflict with 7.4 builds
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

20 years agopatch to deal with unique foreign keys in 7.4 from Kris Jurka
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

20 years agotch to make sure PSQLState is Serializable and a test case to prove it
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

20 years agopatch 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

20 years agoincremented build to 211
Dave Cramer [Wed, 10 Dec 2003 18:30:44 +0000 (18:30 +0000)]
incremented build to 211

20 years agoFix integer types to use definition from c.h. Per bug report by Patrick Boulay <patri...
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>

20 years agoquery_tree_mutator should copy RangeTblEntry nodes even when it's not
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.

20 years agoMinor autovacuum fixes from Matthew O'Connor.
Tom Lane [Mon, 8 Dec 2003 21:54:00 +0000 (21:54 +0000)]
Minor autovacuum fixes from Matthew O'Connor.

20 years agoAdd fflush() before sleeping, per Matthew O'Connor.
Tom Lane [Mon, 8 Dec 2003 21:51:31 +0000 (21:51 +0000)]
Add fflush() before sleeping, per Matthew O'Connor.

20 years agoWhole-row references were broken for subqueries and functions, because
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.

20 years agoWork around naming conflict between zlib and OpenSSL by tweaking inclusion
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).

20 years agoDone:
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)

20 years agoReplace unportable and overflow-prone use of 'long long' with safer
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.

20 years agoRemove test on c.relkind from check_constraints view; unnecessary and
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.

20 years agoFix typmod interpretation for bit types. (It was erroneously assumed that
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.)

20 years agoOne more pass at reducing the cost of pg_dump's new implementation:
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.

20 years agoRepair indexed bytea like operations, and related selectivity
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

20 years agoSpeed up findObjectByCatalogId() to get rid of the other salient
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.

20 years agoDone:
Bruce Momjian [Sun, 7 Dec 2003 00:54:05 +0000 (00:54 +0000)]
Done:
> * -Use dependency information to dump data in proper order

20 years agoDocumentation update for previously committed changes: read-only GUC variables
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.

20 years agoReplace not-very-bright implementation of topological sort with a better
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.

20 years agoMassive overhaul of pg_dump: make use of dependency information from
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.