]> granicus.if.org Git - postgresql/log
postgresql
21 years agoRevise platform FAQs.
Peter Eisentraut [Mon, 11 Nov 2002 20:04:05 +0000 (20:04 +0000)]
Revise platform FAQs.

21 years agoImprove wording.
Peter Eisentraut [Mon, 11 Nov 2002 20:03:40 +0000 (20:03 +0000)]
Improve wording.

21 years agoUpdate jdbc doc to reflect correct name of the PGConnection class.
Barry Lind [Mon, 11 Nov 2002 07:31:28 +0000 (07:31 +0000)]
Update jdbc doc to reflect correct name of the PGConnection class.

21 years agoFixes bug where join to pg_description was incorrect. Also modifies the
Barry Lind [Mon, 11 Nov 2002 07:11:12 +0000 (07:11 +0000)]
Fixes bug where join to pg_description was incorrect.  Also modifies the
regression test to test for this case.  Patch submitted by Kris Jurka.

 Modified Files:
  jdbc/org/postgresql/jdbc1/AbstractJdbc1DatabaseMetaData.java
  jdbc/org/postgresql/test/jdbc2/DatabaseMetaDataTest.java

21 years agoAdd cast to suppress compile warning on Alphas.
Tom Lane [Mon, 11 Nov 2002 03:33:38 +0000 (03:33 +0000)]
Add cast to suppress compile warning on Alphas.

21 years agoBack out use of palloc0 in place if palloc/MemSet. Seems constant len
Bruce Momjian [Mon, 11 Nov 2002 03:02:20 +0000 (03:02 +0000)]
Back out use of palloc0 in place if palloc/MemSet.  Seems constant len
to MemSet is a performance boost.

21 years agoUpdate russian FAQ, from Viktor Vislobokov.
Bruce Momjian [Sun, 10 Nov 2002 19:48:41 +0000 (19:48 +0000)]
Update russian FAQ, from Viktor Vislobokov.

21 years agoMerge palloc()/MemSet(0) calls into a single palloc0() call.
Bruce Momjian [Sun, 10 Nov 2002 07:25:14 +0000 (07:25 +0000)]
Merge palloc()/MemSet(0) calls into a single palloc0() call.

21 years agoAdd palloc0 function to inline MemSet for newNode call.
Bruce Momjian [Sun, 10 Nov 2002 02:17:25 +0000 (02:17 +0000)]
Add palloc0 function to inline MemSet for newNode call.

21 years agoThis trivial patch removes 2 unused defines from pg_config.h.in:
Bruce Momjian [Sun, 10 Nov 2002 00:38:21 +0000 (00:38 +0000)]
This trivial patch removes 2 unused defines from pg_config.h.in:
FASTBUILD and DATEDEBUG weren't referenced by any other code.

Neil Conway

21 years agoI am attaching a patch file that will replace the previous one. It only
Bruce Momjian [Sun, 10 Nov 2002 00:37:28 +0000 (00:37 +0000)]
I am attaching a patch file that will replace the previous one.  It only
contains the patches to Makefile.global.in and Makefile.unixware.  The
Makefile.unixware patch has been updated to include the contents of
LD_LIBRARY_PATH, if present, to the -rpath (-R) option.  This change
will  simplify configuring and building PostgreSQL on systems that
support  LD_LIBRARY_PATH.  You can set LD_LIBRARY_PATH to include all
the directorys  you want to have searched for additional libraries, run
configure, then run  make.  The paths in LD_LIBRARY_PATH will then be
embedded in the executables  via the -rpath (-R) option to the linker,
and so will not require  LD_LIBRARY_PATH in order to run.

Billy G. Allie

21 years agoThis patch makes a minor cleanup to the implementation of PERFORM in
Bruce Momjian [Sun, 10 Nov 2002 00:35:58 +0000 (00:35 +0000)]
This patch makes a minor cleanup to the implementation of PERFORM in
PL/PgSQL. Previously, it had been bundled together with the assign
statement implementation, for some reason that wasn't clear to me
(they certainly don't share any code with one another). So I separated
them and made PERFORM a statement like any other. No changes in
functionality.

Along the way, I added some regression tests for PERFORM, added a
bunch more SGML tags to the PL/PgSQL docs, and removed an obsolete
comment relating to the implementation of RETURN NEXT.

Neil Conway

21 years ago> > I'll re-check that with the ppc architecture guy here.
Bruce Momjian [Sun, 10 Nov 2002 00:33:43 +0000 (00:33 +0000)]
> > I'll re-check that with the ppc architecture guy here.
>
> ... he is now about to write an inlined version that can go into
> s_lock.h . I'll send the new patch later on...

OK, here it comes:

An inlined version of tas(), that works for both, powerpc and
powerpc64. The patch is against 7.3b5 and passes the test suite on
both architectures.

Reinhard Max

21 years agoAdd documentation for array_upper/lower().
Bruce Momjian [Sun, 10 Nov 2002 00:32:16 +0000 (00:32 +0000)]
Add documentation for array_upper/lower().

21 years agoI just discovered, that there is missing a const when passing a buffer
Bruce Momjian [Sun, 10 Nov 2002 00:14:22 +0000 (00:14 +0000)]
I just discovered, that there is missing a const when passing a buffer
to  PQescapeBytea and PQunescapeBytea. I fixed it and tried to create a
usable  diff (I'm not so familar to diff).

Tommi M?kitalo

21 years agoTweak CREATE SEQUENCE grammar to be more SQL1999 standards compliant.
Bruce Momjian [Sun, 10 Nov 2002 00:10:20 +0000 (00:10 +0000)]
Tweak CREATE SEQUENCE grammar to be more SQL1999 standards compliant.

Neil Conway

21 years agoAdd code to handle [ON COMMIT { PRESERVE ROWS | DELETE ROWS | DROP }]
Bruce Momjian [Sat, 9 Nov 2002 23:56:39 +0000 (23:56 +0000)]
Add code to handle [ON COMMIT { PRESERVE ROWS | DELETE ROWS | DROP }]
for temp tables.

Gavin Sherry

22 years agoUpdate discussion of floating-point problems in regression tests.
Tom Lane [Fri, 8 Nov 2002 20:26:12 +0000 (20:26 +0000)]
Update discussion of floating-point problems in regression tests.

22 years agoThis patch removes a bunch of superfluous #include directives: if
Bruce Momjian [Fri, 8 Nov 2002 20:23:57 +0000 (20:23 +0000)]
This patch removes a bunch of superfluous #include directives: if
postgres.h or c.h includes a system header (such as stdio.h or
stdlib.h), there's no need to specifically include it in any of the .c
files in the backend.

Neil Conway

22 years agoClean up format of SQL.
Bruce Momjian [Fri, 8 Nov 2002 20:22:12 +0000 (20:22 +0000)]
Clean up format of SQL.

22 years agoThe attached patch defines functions for getting distances between
Bruce Momjian [Fri, 8 Nov 2002 20:20:22 +0000 (20:20 +0000)]
The attached patch defines functions for getting distances between
points on the surface of the earth and locating points within a
specified distance using an index based on the contrib/cube package. The
new functions are all of language type sql. A couple of bugs in the old
earthdistance function based on the point datatype are fixed. A
regression test has been added for both sets of functions. The README
file has been updated to include documentation on the new stuff. There
are comments about how this package is also useful for Astronomers.

Bruno Wolff III

22 years agoRevise geometry regression testing to eliminate most cross-platform
Tom Lane [Fri, 8 Nov 2002 20:09:55 +0000 (20:09 +0000)]
Revise geometry regression testing to eliminate most cross-platform
variation.  To do this, set extra_float_digits to -3 in the geometry
test, and tweak the CIRCLE_TBL dataset to avoid values that suffer
from severe cancellation error (eg, circles that just touch an axis).
We still need two geometry 'expected' files to account for the
difference between platforms that display minus zero as '-0' and those
that just say '0', but with luck that's all we'll need.

22 years agoHere is a patch that does just that, while maintaining the
Bruce Momjian [Fri, 8 Nov 2002 19:12:21 +0000 (19:12 +0000)]
Here is a patch that does just that, while maintaining the
"traditional" behavior, so the change should be transparent. Use the
command "\pset pager always" to turn it on. Anything else does the
normal toggle between "on" and "off"

Greg Sabino Mullane

22 years agoReplace imprecise value of PI with a better one, and tweak circle_poly
Tom Lane [Fri, 8 Nov 2002 18:32:47 +0000 (18:32 +0000)]
Replace imprecise value of PI with a better one, and tweak circle_poly
in hopes of reducing platform-to-platform variations in its results.
This will cause the geometry regression test to start failing on some
platforms.  I plan to update the test later today.

22 years agoAdd name:
Bruce Momjian [Fri, 8 Nov 2002 17:42:18 +0000 (17:42 +0000)]
Add name:

> * -Add GUC variables to control floating number output digits (Pedro Ferreira)

22 years agoMark as done:
Bruce Momjian [Fri, 8 Nov 2002 17:41:05 +0000 (17:41 +0000)]
Mark as done:

> * -Add GUC variables to control floating number output digits

22 years agoAdd extra_float_digits GUC parameter to allow adjustment of displayed
Tom Lane [Fri, 8 Nov 2002 17:37:52 +0000 (17:37 +0000)]
Add extra_float_digits GUC parameter to allow adjustment of displayed
precision for float4, float8, and geometric types.  Set it in pg_dump
so that float data can be dumped/reloaded exactly (at least on platforms
where the float I/O support is properly implemented).  Initial patch by
Pedro Ferreira, some additional work by Tom Lane.

22 years agoThe "Allow easy display of usernames in a group (pg_hba.conf uses groups
Bruce Momjian [Fri, 8 Nov 2002 17:27:03 +0000 (17:27 +0000)]
The "Allow easy display of usernames in a group (pg_hba.conf uses groups
now)" item on the open items, and subsequent plpgsql function I sent in,
made me realize it was too hard to get the upper and lower bound of an
array. The attached creates two functions that I think will be very
useful when combined with the ability of plpgsql to return sets.

array_lower(array, dim_num)
- and -
array_upper(array, dim_num)

They return the value (as an int) of the upper and lower bound of the
requested dim in the provided array.

Joe Conway

22 years agoUpdate FAQ_SCO from Larry Rosenman, correction from Tom Lane.
Bruce Momjian [Fri, 8 Nov 2002 16:49:15 +0000 (16:49 +0000)]
Update FAQ_SCO from Larry Rosenman, correction from Tom Lane.

22 years agoUpdate libpq PQcmdTuples() and PQcmdStatus() docs, remove const param
Bruce Momjian [Fri, 8 Nov 2002 05:03:11 +0000 (05:03 +0000)]
Update libpq PQcmdTuples() and PQcmdStatus() docs, remove const param
doc mention to match code and include files.

22 years agoI'm using libpgtcl with PostgreSQL-7.2.3. There is a discrepancy between
Bruce Momjian [Fri, 8 Nov 2002 04:56:40 +0000 (04:56 +0000)]
I'm using libpgtcl with PostgreSQL-7.2.3. There is a discrepancy between
the documentation and behavior of "pg_result $res -oid" when $res is not
from an INSERT.  The documentation says it should return an empty
string, but in fact it returns 0. I think it used to return an empty
string around PostgreSQL-7.1.3 when PQoidStatus() was used, but now it
uses PQoidValue() which returns InvalidOid, 0 in this case.

Assuming the current behavior is desired, here is a patch to the
documentation doc/src/sgml/libpgtcl.sgml to match what really happens:

ljb

22 years agoAdd:
Bruce Momjian [Fri, 8 Nov 2002 04:31:14 +0000 (04:31 +0000)]
Add:

> * Add GUC variables extra_float_digits and extra_double_digits to
>   control output digits
419a422
> * Research interaction of setitimer() and sleep() used by statement_timeout

22 years agoUpdate release docs for 7.3.
Bruce Momjian [Fri, 8 Nov 2002 03:44:15 +0000 (03:44 +0000)]
Update release docs for 7.3.

22 years agoAdd #define _GNU_SOURCE to work around what seems to be Perl 5.8.0's
Tom Lane [Thu, 7 Nov 2002 22:09:51 +0000 (22:09 +0000)]
Add #define _GNU_SOURCE to work around what seems to be Perl 5.8.0's
problem.  Per recent discussions about plperl failing to build on Linux.

22 years agoMove substitute extern declarations to end of file, so that they are
Tom Lane [Thu, 7 Nov 2002 22:08:30 +0000 (22:08 +0000)]
Move substitute extern declarations to end of file, so that they are
not read until after we've read the port-specific header file.  In
particular this should make it safer to #include system headers for
inet_aton; in general it seems that the port header file ought to be
in a position to set definitions before we do stuff based on having
a definition or not.

22 years agoRemove inappropriate inclusions of OpenSSL internal header e_os.h,
Tom Lane [Thu, 7 Nov 2002 18:45:51 +0000 (18:45 +0000)]
Remove inappropriate inclusions of OpenSSL internal header e_os.h,
as well as unnecessary (and incorrect on Windows) assignments to
errno/SOCK_ERRNO.

22 years agoSynced parser one more time.
Michael Meskes [Thu, 7 Nov 2002 09:48:09 +0000 (09:48 +0000)]
Synced parser one more time.

22 years agoRemove no-longer-needed inclusions to improve backward compatibility
Tom Lane [Thu, 7 Nov 2002 06:06:17 +0000 (06:06 +0000)]
Remove no-longer-needed inclusions to improve backward compatibility
with older bison versions.

22 years agoPhase 2 of hashed-aggregation project. nodeAgg.c now knows how to do
Tom Lane [Wed, 6 Nov 2002 22:31:24 +0000 (22:31 +0000)]
Phase 2 of hashed-aggregation project.  nodeAgg.c now knows how to do
hashed aggregation, but there's not yet planner support for it.

22 years agoRemove GPL file pursuant to relicensing.
Tom Lane [Wed, 6 Nov 2002 22:23:08 +0000 (22:23 +0000)]
Remove GPL file pursuant to relicensing.

22 years agoRemove:
Bruce Momjian [Wed, 6 Nov 2002 17:51:03 +0000 (17:51 +0000)]
Remove:

< * Move dbmirror and rserv from /contrib to gborg.postgresql.org

22 years agoChange dbmirror lincense to BSD.
Bruce Momjian [Wed, 6 Nov 2002 17:50:53 +0000 (17:50 +0000)]
Change dbmirror lincense to BSD.

22 years agoNot done:
Bruce Momjian [Wed, 6 Nov 2002 06:21:00 +0000 (06:21 +0000)]
Not done:

> * Add hash for evaluating GROUP BY aggregates (Tom)

22 years agoCompleted:
Bruce Momjian [Wed, 6 Nov 2002 05:56:36 +0000 (05:56 +0000)]
Completed:

> * -Add hash for evaluating GROUP BY aggregates (Tom)

22 years agoRemove completed items:
Bruce Momjian [Wed, 6 Nov 2002 05:55:14 +0000 (05:55 +0000)]
Remove completed items:

< A dash (-) marks changes that will appear in the upcoming 7.3 release.
> A dash (-) marks changes that will appear in the upcoming 7.4 release.
38d37
< * -Change DEBUG startup tag to LOG (Bruce)
40,51d38
< * -Add pg_backend_pid() function to backend
< * -Allow logging of query durations
<
<
< Permissions
< ===========
<
< * -Improve control over user privileges, including table creation
< * -Allow user/group names to be specified directly in pg_hba.conf (Bruce)
< * -Add ~/.pgpass to store passwords with user/host/password combinations
< * -Allow permissions for functions (Peter E)
< * -Allow object creation to be disabled for specific users
58,60d44
< * -Make it easier to create a database owned by someone who can't createdb,
<   perhaps CREATE DATABASE dbname WITH OWNER = "user" (Gavin)
< * -Make equals sign optional in CREATE DATABASE WITH param = 'val'
65,66d48
< * -Prevent SIGHUP and 'pg_ctl reload' from changing command line
<   specified parameters to postgresql.conf defaults (Peter E)
69,70d50
< * -Reserve last few process slots for super-user if max_connections reached
< * -Add GUC parameter to print queries that generate errors
79d58
< * -Add domain capability (Rod Taylor)
82,84d60
< * -SELECT cash_out(2) crashes because of opaque
< * -Declare typein/out functions in pg_proc with a special "C string" data type
< * -Functions returning sets do not totally work
89d64
< * -Add GUC parameter for DATESTYLE
91,93d65
< * -Allow bytea to handle LIKE with non-TEXT patterns
< * -to_char(0,'FM999.99') returns a period, to_char(1,'FM999.99') doesn't (Karel)
< * -Add floor(float8) and other missing functions
97d68
<  o -Store binary-compatible type information in the system
104d74
<  o -Ensure we have array-eq operators for every built-in array type
139,140d108
< * -Remove brackets as multi-statement rule grouping, must use parens (Bruce)
< * -Prevent aggregates from being used in rule WHERE clauses
154d121
< * -Allow UPDATE/DELETE on inherited table
166d132
< * -Add deleted bit to index tuples to reduce heap access
176d141
< * -Test hash index performance and discourage usage
182d146
< * -Add SIMILAR TO to allow character classes, 'pg_[a-c]%'
184d147
< * -Remove LIMIT #,# and force use LIMIT and OFFSET clauses in 7.3 (Bruce)
186,187d148
< * -Disallow TRUNCATE on tables that are involved in referential constraints
< * -Add OR REPLACE clauses to non-FUNCTION object creation
190d150
< * -Prevent create/drop scripts from allowing extra args (Bruce)
201,205d160
<  o -Add ALTER TABLE DROP COLUMN feature
<  o -Add ALTER TABLE DROP non-CHECK CONSTRAINT
<  o -ALTER TABLE ADD PRIMARY KEY (Tom)
<  o -ALTER TABLE ADD UNIQUE (Tom)
<  o -ALTER TABLE ALTER COLUMN SET/DROP NOT NULL (Christopher)
210d164
<  o -Have ALTER TABLE OWNER change all dependant objects like indexes
214,216d167
<  o -Cluster all tables at once using pg_index.indisclustered set during
<           previous CLUSTER
<  o -Prevent loss of indexes, permissions, inheritance
221d171
<  o -Allow specification of column names
224d173
<  o -Change syntax to WITH DELIMITER, (keep old syntax around?)
228d176
<  o -Generate failure on short COPY lines rather than pad NULLs
242,243d189
<  o -Allow INSERT INTO my_table VALUES (a, b, c, DEFAULT, x, y, z, ...)
<  o -Disallow missing columns in INSERT ... (col) VALUES, per ANSI
248,249d193
<  o -Add command to display locks
<  o -Add SET or BEGIN timeout parameter to cancel query
251d194
<  o -Remove SET KSQO option now that OR processing is improved (Bruce)
254,256d196
<  o -Add SHOW command to see locale
<  o -Allow SHOW to output as a query result, like EXPLAIN
<  o -Abort all SET changes made in an aborted transaction
264d203
<  o -Fix PL/PgSQL to handle quoted mixed-case identifiers
287,291d225
< * -Have pg_dump use LEFT OUTER JOIN in multi-table SELECTs
<   or multiple SELECTS to avoid bad system catalog entries
< * -Have pg_dump -C dump database location and encoding information
< * -Allow psql \d to show foreign keys
< * -Allow psql \d to show temporary table structure (Tom)
294d227
< * -Have pg_dump use ADD PRIMARY KEY after COPY, for performance (Neil)
302d234
<  o -Updateable resultSet
307d238
<  o -Implement cancel() method on Statement
309d239
<         o -Add support for CallableStatements
311d240
<  o -Compile under jdk 1.4
334d262
< * -Allow oid to act as a foreign key
337,338d264
< * -Allow user to control trigger firing order (Tom)
< * -Add ALTER TRIGGER ... RENAME
341d266
< * -Fix foreign key constraints to not error on intermediate db states (Stephan)
350,359d274
< * -Add pg_depend table for dependency recording; use sysrelid, oid,
<   depend_sysrelid, depend_oid, name
< * -Auto-destroy sequence on DROP of table with SERIAL; perhaps a separate
<   SERIAL type
< * -Prevent column dropping if column is used by foreign key
< * -Propagate column or table renaming to foreign key constraints
< * -Automatically drop constraints/functions when object is dropped
< * -Make foreign key constraints clearer in dump file
< * -Make other constraints clearer in dump file
< * -Make foreign keys easier to identify
367d281
< * -Allow autocommit so always in a transaction block
377d290
< * -Add SQL92 schemas (Tom)
400d312
< * -Cache most recent query plan(s) (Neil) [prepare]
439d350
< * -Cache system catalog information in per-database files (Tom)
450,451d360
< * -Reorder postgresql.conf WAL items in order of importance (Bruce)
< * -Remove wal_files postgresql.conf option because WAL files are now recycled
465,466d373
< * -Improve dynamic memory allocation by introducing tuple-context memory
<   allocation (Tom)
468d374
< * -Nested FULL OUTER JOINs don't work (Tom)
470,472d375
< * -Add new pg_proc cachable settings to specify whether function can be
<   evaluated only once or once per query
< * -Change FIXED_CHAR_SEL to 0.20 from 0.04 to give better selectivity (Bruce)
494,496d396
< * -Add Intimate Shared Memory(ISM) for Solaris
< * -Use faster flex flags for performance improvement (Peter E)
< * -Add BSD-licensed qsort() for Solaris
503,507d402
< * -Fix problems with libpq non-blocking/async code
< * -Make sure all block numbers are unsigned to increase maximum table size
< * -Merge LockMethodCtl and LockMethodTable into one shared structure (Bruce)
< * -HOLDER/HOLDERTAB rename to PROCLOCK/PROCLOCKTAG (Bruce)
< * -Remove LockMethodTable.prio field, not used (Bruce)
512,513d406
< * -Make one version of simple_prompt() in code (Bruce, Tom)
< * -Compile in syslog functionaility by default (Tatsuo)
517d409
< * -Report failure to find readline or zlib at end of configure run
519,520d410
< * -Increase identifier length (NAMEDATALEN) if small performance hit,
< * -Increase maximum number of function parameters if little wasted space
529,530d418
< * -Fix glibc's mktime() to handle pre-1970's dates
< * -Move /contrib/retep to gborg.postgresql.org

22 years agoNone.
Bruce Momjian [Wed, 6 Nov 2002 05:13:02 +0000 (05:13 +0000)]
None.

22 years agoFirst phase of implementing hash-based grouping/aggregation. An AGG plan
Tom Lane [Wed, 6 Nov 2002 00:00:45 +0000 (00:00 +0000)]
First phase of implementing hash-based grouping/aggregation.  An AGG plan
node now does its own grouping of the input rows, and has no need for a
preceding GROUP node in the plan pipeline.  This allows elimination of
the misnamed tuplePerGroup option for GROUP, and actually saves more code
in nodeGroup.c than it costs in nodeAgg.c, as well as being presumably
faster.  Restructure the API of query_planner so that we do not commit to
using a sorted or unsorted plan in query_planner; instead grouping_planner
makes the decision.  (Right now it isn't any smarter than query_planner
was, but that will change as soon as it has the option to select a hash-
based aggregation step.)  Despite all the hackery, no initdb needed since
only in-memory node types changed.

22 years agoAdd OS/X kernel tuning section.
Bruce Momjian [Tue, 5 Nov 2002 23:16:56 +0000 (23:16 +0000)]
Add OS/X kernel tuning section.

22 years agoAdd:
Bruce Momjian [Tue, 5 Nov 2002 22:16:57 +0000 (22:16 +0000)]
Add:

> * Move dbmirror and rserv from /contrib to gborg.postgresql.org

22 years agoUpdate FAQ_DEV.
Bruce Momjian [Tue, 5 Nov 2002 21:53:24 +0000 (21:53 +0000)]
Update FAQ_DEV.

22 years agoUpdate FAQ_DEV.
Bruce Momjian [Tue, 5 Nov 2002 21:49:52 +0000 (21:49 +0000)]
Update FAQ_DEV.

22 years agoUpdate ports list.
Bruce Momjian [Tue, 5 Nov 2002 19:01:07 +0000 (19:01 +0000)]
Update ports list.

22 years agoUpdate ports list.
Bruce Momjian [Tue, 5 Nov 2002 17:31:36 +0000 (17:31 +0000)]
Update ports list.

22 years agoUpdate ports list.
Bruce Momjian [Tue, 5 Nov 2002 17:11:46 +0000 (17:11 +0000)]
Update ports list.

22 years agoUpdate ports list.
Bruce Momjian [Tue, 5 Nov 2002 17:09:50 +0000 (17:09 +0000)]
Update ports list.

22 years agoFix broken markup.
Tom Lane [Tue, 5 Nov 2002 04:34:51 +0000 (04:34 +0000)]
Fix broken markup.

22 years agoUpdate ports list.
Bruce Momjian [Tue, 5 Nov 2002 03:18:09 +0000 (03:18 +0000)]
Update ports list.

22 years agoUpdate ports list.
Bruce Momjian [Mon, 4 Nov 2002 22:40:43 +0000 (22:40 +0000)]
Update ports list.

22 years agoBrand CVS tip as 7.4devel.
Tom Lane [Mon, 4 Nov 2002 21:36:13 +0000 (21:36 +0000)]
Brand CVS tip as 7.4devel.

22 years agoRemove extraneous semicolons after routine bodies. These don't bother
Tom Lane [Mon, 4 Nov 2002 17:14:29 +0000 (17:14 +0000)]
Remove extraneous semicolons after routine bodies.  These don't bother
gcc, but some other compilers don't like 'em.

22 years agoFix inclusion order, per Andreas.
Tom Lane [Mon, 4 Nov 2002 14:27:21 +0000 (14:27 +0000)]
Fix inclusion order, per Andreas.

22 years agoRemove unnecessary inclusion, per Andreas.
Tom Lane [Mon, 4 Nov 2002 14:25:42 +0000 (14:25 +0000)]
Remove unnecessary inclusion, per Andreas.

22 years agoRemove no-longer-needed inclusion of bootstrap_tokens.h, per patch
Tom Lane [Mon, 4 Nov 2002 14:22:32 +0000 (14:22 +0000)]
Remove no-longer-needed inclusion of bootstrap_tokens.h, per patch
from Andreas.

22 years agoApplied two patches from Kris Jurka.
Barry Lind [Mon, 4 Nov 2002 06:42:33 +0000 (06:42 +0000)]
Applied two patches from Kris Jurka.
  - First fixes a problem with a recent patch allowing setNull on updateable
    resultsets
  - Second removed toLower() calls on database object names.  Leave it to
    the caller to correctly pass lower, upper or mixed case.  The driver
    already has methods that the caller can use to determine that postgres
    stores identifiers in lowercase. (unless the identifier was quoted when
    created).

 Modified Files:
  jdbc/org/postgresql/jdbc1/AbstractJdbc1DatabaseMetaData.java
  jdbc/org/postgresql/jdbc2/AbstractJdbc2ResultSet.java

22 years agoFix dblink regression test so that it works when repeated: don't
Tom Lane [Sun, 3 Nov 2002 04:52:09 +0000 (04:52 +0000)]
Fix dblink regression test so that it works when repeated: don't
create a second database, just loop back to 'regression'.  Patch
from Joe Conway 10/22/02.

22 years agoUpdate FAQ_DEV.
Bruce Momjian [Sun, 3 Nov 2002 04:02:32 +0000 (04:02 +0000)]
Update FAQ_DEV.

22 years agoStamp beta4.
Bruce Momjian [Sun, 3 Nov 2002 01:35:06 +0000 (01:35 +0000)]
Stamp beta4.

22 years agoThis patch adds a couple of extra index entries and removes some
Bruce Momjian [Sun, 3 Nov 2002 01:31:32 +0000 (01:31 +0000)]
This patch adds a couple of extra index entries and removes some
out-of-date text about multibyte encoding (which is now always
included).

Oliver Elphick

22 years agoThis patch adds information to the documentation on .pgpass and creates
Bruce Momjian [Sun, 3 Nov 2002 01:30:46 +0000 (01:30 +0000)]
This patch adds information to the documentation on .pgpass and creates
additional index entries for it.

Oliver Elphick

22 years agoUse sed rather than perl for reindexdb.
Bruce Momjian [Sun, 3 Nov 2002 01:20:06 +0000 (01:20 +0000)]
Use sed rather than perl for reindexdb.

22 years agoFix some misstatements in WAL parameter discussion.
Tom Lane [Sat, 2 Nov 2002 22:23:01 +0000 (22:23 +0000)]
Fix some misstatements in WAL parameter discussion.

22 years agoCode review for recent patch to allow ALTER TABLE ADD COLUMN when
Tom Lane [Sat, 2 Nov 2002 22:02:08 +0000 (22:02 +0000)]
Code review for recent patch to allow ALTER TABLE ADD COLUMN when
a child table already has a matching column.  Acquire appropriate
lock on child table; do the right thing with any CHECK constraints
attached to the new parent column.

22 years agoDuring swap_relfilenodes, swap relation size statistic fields along with
Tom Lane [Sat, 2 Nov 2002 21:20:40 +0000 (21:20 +0000)]
During swap_relfilenodes, swap relation size statistic fields along with
the relfilenode and toast fields.  This ensures that the newly-computed
statistics will be available on completion of CLUSTER.

22 years agoRemove encoding lookups from grammar stage, push them back to places
Tom Lane [Sat, 2 Nov 2002 18:41:22 +0000 (18:41 +0000)]
Remove encoding lookups from grammar stage, push them back to places
where it's safe to do database access.  Along the way, fix core dump
for 'DEFAULT' parameters to CREATE DATABASE.  initdb forced due to
change in pg_proc entry.

22 years agoClean up a few fprintf(stderr)'s that should be elog's.
Tom Lane [Sat, 2 Nov 2002 15:54:13 +0000 (15:54 +0000)]
Clean up a few fprintf(stderr)'s that should be elog's.

22 years agoAdd:
Bruce Momjian [Sat, 2 Nov 2002 05:42:01 +0000 (05:42 +0000)]
Add:

> * Acquire lock on a relation before building a relcache entry for it

22 years agoUpdate:
Bruce Momjian [Sat, 2 Nov 2002 02:44:16 +0000 (02:44 +0000)]
Update:

>  o MOVE 0 should not move to end of cursor (Bruce)

22 years agoAdd:
Bruce Momjian [Sat, 2 Nov 2002 02:43:03 +0000 (02:43 +0000)]
Add:

> * Support statement-level triggers and triggers on columns (Neil)

22 years agoFix permissions-checking bugs and namespace-search-path bugs in
Tom Lane [Sat, 2 Nov 2002 02:33:03 +0000 (02:33 +0000)]
Fix permissions-checking bugs and namespace-search-path bugs in
CONVERSION code.  Still need to figure out what to do about inappropriate
coding in parsing.

22 years agoAdd gcj discussion.
Bruce Momjian [Sat, 2 Nov 2002 02:27:49 +0000 (02:27 +0000)]
Add gcj discussion.

22 years agoRe-add Win32 missing files, I think.
Bruce Momjian [Sat, 2 Nov 2002 02:00:35 +0000 (02:00 +0000)]
Re-add Win32 missing files, I think.

22 years agoRe-add Win32 files.
Bruce Momjian [Sat, 2 Nov 2002 01:58:27 +0000 (01:58 +0000)]
Re-add Win32 files.

22 years agoUpdate port list.
Bruce Momjian [Sat, 2 Nov 2002 01:29:37 +0000 (01:29 +0000)]
Update port list.

22 years agoUpdates from Giles Lean.
Tom Lane [Sat, 2 Nov 2002 00:46:23 +0000 (00:46 +0000)]
Updates from Giles Lean.

22 years agoMake it possible to run 'make installcheck' in contrib when building
Tom Lane [Sat, 2 Nov 2002 00:16:21 +0000 (00:16 +0000)]
Make it possible to run 'make installcheck' in contrib when building
outside the source tree.

22 years agoAdd:
Bruce Momjian [Sat, 2 Nov 2002 00:15:38 +0000 (00:15 +0000)]
Add:

> * Allow psql to do table completion for SELECT * FROM schema_part and
>   table completion for SELECT * FROM schema_name.

22 years agoFix contrib/rserv to install successfully when built outside source tree.
Tom Lane [Fri, 1 Nov 2002 23:45:37 +0000 (23:45 +0000)]
Fix contrib/rserv to install successfully when built outside source tree.
(Not that it will actually work with 7.3, I fear.)

22 years agoOoops, forgot to fix contrib regression tests for WARNING/NOTICE change.
Tom Lane [Fri, 1 Nov 2002 23:44:34 +0000 (23:44 +0000)]
Ooops, forgot to fix contrib regression tests for WARNING/NOTICE change.

22 years agoNeed -I. to build outside source tree.
Tom Lane [Fri, 1 Nov 2002 23:23:29 +0000 (23:23 +0000)]
Need -I. to build outside source tree.

22 years agoFix incorrect subdir value.
Tom Lane [Fri, 1 Nov 2002 23:16:52 +0000 (23:16 +0000)]
Fix incorrect subdir value.

22 years agoAdd:
Bruce Momjian [Fri, 1 Nov 2002 22:53:05 +0000 (22:53 +0000)]
Add:

> * Have standalone backend read postgresql.conf

22 years agoArrange to compile flex output files as inclusions into other files
Tom Lane [Fri, 1 Nov 2002 22:52:34 +0000 (22:52 +0000)]
Arrange to compile flex output files as inclusions into other files
(usually bison output files), not as standalone files.  This hack
works around flex's insistence on including <stdio.h> before we are
able to include postgres.h; postgres.h will already be read before
the compiler starts to read the flex output file.  Needed for largefile
support on some platforms.

22 years agoOoops, forgot to fix contrib regression tests for WARNING/NOTICE change.
Tom Lane [Fri, 1 Nov 2002 22:46:04 +0000 (22:46 +0000)]
Ooops, forgot to fix contrib regression tests for WARNING/NOTICE change.

22 years agoReduce a couple of debugging messages from LOG to DEBUG1 category.
Tom Lane [Fri, 1 Nov 2002 19:33:09 +0000 (19:33 +0000)]
Reduce a couple of debugging messages from LOG to DEBUG1 category.

22 years agoReduce messages associated with shell-type function arguments/results
Tom Lane [Fri, 1 Nov 2002 19:19:58 +0000 (19:19 +0000)]
Reduce messages associated with shell-type function arguments/results
from WARNING to NOTICE, since they are expected messages in common cases.

22 years agoUpdate to match expected results on OS X 10.2.
Tom Lane [Fri, 1 Nov 2002 18:15:34 +0000 (18:15 +0000)]
Update to match expected results on OS X 10.2.

22 years agoAfter elog(PANIC), exit with abort() not proc_exit(). This allows a
Tom Lane [Fri, 1 Nov 2002 17:55:23 +0000 (17:55 +0000)]
After elog(PANIC), exit with abort() not proc_exit().  This allows a
core file to be produced for debugging, and avoids trying to run the
normal proc-exit cleanup hooks, which are likely to cause additional
problems if the system is hosed.

22 years agoCorrect ordering of geometry/freebsd entries, add one for freebsd5.
Tom Lane [Fri, 1 Nov 2002 16:32:32 +0000 (16:32 +0000)]
Correct ordering of geometry/freebsd entries, add one for freebsd5.

22 years agoUpdate for test changes.
Tom Lane [Fri, 1 Nov 2002 15:20:09 +0000 (15:20 +0000)]
Update for test changes.

22 years agoAvoid malloc(0) when printing a table of no columns. On some platforms
Tom Lane [Fri, 1 Nov 2002 15:12:19 +0000 (15:12 +0000)]
Avoid malloc(0) when printing a table of no columns.  On some platforms
this returns NULL, which confuses the code.