]> granicus.if.org Git - postgresql/log
postgresql
22 years agoAdd INSERT(..., DEFAULT, ).
Bruce Momjian [Fri, 5 Apr 2002 11:56:55 +0000 (11:56 +0000)]
Add INSERT(..., DEFAULT, ).

Rod Taylor

22 years agoI was tinkering with creating rules on views (so, for instance, one could
Bruce Momjian [Fri, 5 Apr 2002 11:52:38 +0000 (11:52 +0000)]
I was tinkering with creating rules on views (so, for instance, one could
insert on a view), and noticed that psql wouldn't show the list of rules
set up on a view, like it does for tables.

The fix was extremely simple, so I figured I'd share it.  Not sure what
the standard is for communicating these things, so I've attached the diff
file for /src/bin/psql/describe.c.

Paul (?)

22 years agoAdds domain dumping support to pg_dump.
Bruce Momjian [Fri, 5 Apr 2002 11:51:13 +0000 (11:51 +0000)]
Adds domain dumping support to pg_dump.

Rod Taylor

22 years agoWhen a macro is replaced by the preprocessor, pgc.l reaches a end of
Bruce Momjian [Fri, 5 Apr 2002 11:39:47 +0000 (11:39 +0000)]
When a macro is replaced by the preprocessor, pgc.l reaches a end of
file, which is not the actual end of the file. One side effect of that
is that if you are i n a ifdef block, you get a wrong error telling you
that a endif is missing.

This patch corrects pgc.l and also adds a test of this problem to
test1.pgc. To  convince you apply the patch to test1.pgc first then try
to compile the test the n apply the patch to pgc.l.

The patch moves the test of the scope of an ifdef block to the end of
the file b eeing parsed, including all includes files, ... .

Nicolas Bazin

22 years agoLooks like a small patch is needed as well to do the right thing on Linux.
Bruce Momjian [Fri, 5 Apr 2002 11:38:13 +0000 (11:38 +0000)]
Looks like a small patch is needed as well to do the right thing on Linux.

The patch enables the mips2 ISA for the ll/sc operations, and then restores
it when done.  The kernel/libc emulation code will take over on CPUs without
ll/sc, and on CPUs with it, it'll use the operations provided by the CPU.

Combined with the earlier fix (removing -mips2), postgresql builds again on
mips and mipsel.  The patch is against 7.2-7.

Oliver Elphick

22 years agoUndo not-so-hot decision to postpone insertion of default values into
Tom Lane [Fri, 5 Apr 2002 05:47:05 +0000 (05:47 +0000)]
Undo not-so-hot decision to postpone insertion of default values into
INSERT statements to the planner.  Taking it out of the parser was right
(so that defaults don't get into stored rules), but it has to happen
before rewrite rule expansion, else references to NEW.field behave
incorrectly.  Accordingly, add a step to the rewriter to insert defaults
just before rewrite-rule expansion.

22 years agoMark as done:
Bruce Momjian [Fri, 5 Apr 2002 03:06:10 +0000 (03:06 +0000)]
Mark as done:

> * -Add new pg_proc cachable settings to specify whether function can be

22 years agoDivide functions into three volatility classes (immutable, stable, and
Tom Lane [Fri, 5 Apr 2002 00:31:36 +0000 (00:31 +0000)]
Divide functions into three volatility classes (immutable, stable, and
volatile), rather than the old cachable/noncachable distinction.  This
allows indexscan optimizations in many places where we formerly didn't.
Also, add a pronamespace column to pg_proc (it doesn't do anything yet,
however).

22 years agoFix a bug in SQLError().
Hiroshi Inoue [Thu, 4 Apr 2002 23:29:16 +0000 (23:29 +0000)]
Fix a bug in SQLError().

22 years agoDuplicate removed:
Bruce Momjian [Thu, 4 Apr 2002 07:12:36 +0000 (07:12 +0000)]
Duplicate removed:

<  o Allow cursors to be DECLAREd/OPENed/CLOSEed outside transactions

22 years agoMark as done:
Bruce Momjian [Thu, 4 Apr 2002 07:03:55 +0000 (07:03 +0000)]
Mark as done:

* -Allow user/group names to be specified directly in pg_hba.conf (Bruce)

22 years agoThis patch against 0.98.7 lib/tables.tcl will allow PGAccess to create new
Bruce Momjian [Thu, 4 Apr 2002 06:27:45 +0000 (06:27 +0000)]
This patch against 0.98.7 lib/tables.tcl will allow PGAccess to create new
records containing apostrophes in text fields without altering the appearance
of the entry in the GUI interface (by copying the fldval to fldvalfixed).

This will alleviate the need for users to create a record and then go back to
edit apostrophes into the text they entered.

Ryan Grange

22 years agoUpdate to new version of Oracle conversion utility, version 1.8.
Bruce Momjian [Thu, 4 Apr 2002 05:55:57 +0000 (05:55 +0000)]
Update to new version of Oracle conversion utility, version 1.8.

Gilles DAROLD

22 years agoCleanup of level ordering and add missing LOG entry.
Bruce Momjian [Thu, 4 Apr 2002 04:43:44 +0000 (04:43 +0000)]
Cleanup of level ordering and add missing LOG entry.

22 years agoAdd missing mention of 'log' option for client_min_messages.
Bruce Momjian [Thu, 4 Apr 2002 04:33:58 +0000 (04:33 +0000)]
Add missing mention of 'log' option for client_min_messages.

22 years agoAuthentication improvements:
Bruce Momjian [Thu, 4 Apr 2002 04:25:54 +0000 (04:25 +0000)]
Authentication improvements:

A new pg_hba.conf column, USER
Allow specifiction of lists of users separated by commas
Allow group names specified by +
Allow include files containing lists of users specified by @
Allow lists of databases, and database files
Allow samegroup in database column to match group name matching dbname
Removal of secondary password files
Remove pg_passwd utility
Lots of code cleanup in user.c and hba.c
New data/global/pg_pwd format
New data/global/pg_group file

22 years agoFix a bug in multibyte_strchr().
Hiroshi Inoue [Thu, 4 Apr 2002 01:36:17 +0000 (01:36 +0000)]
Fix a bug in multibyte_strchr().

22 years agoLocale support is on by default. The choice of locale is done in initdb
Peter Eisentraut [Wed, 3 Apr 2002 05:39:33 +0000 (05:39 +0000)]
Locale support is on by default.  The choice of locale is done in initdb
and/or with GUC variables.

22 years agoReplace perror() calls by elog()s, so that messages can be routed to
Tom Lane [Wed, 3 Apr 2002 00:44:27 +0000 (00:44 +0000)]
Replace perror() calls by elog()s, so that messages can be routed to
syslog when appropriate.  These were the last perror() calls remaining
in the backend; let's not reintroduce any...

22 years agoAllow postmaster to start up anyway when PGSTAT code fails to initialize,
Tom Lane [Wed, 3 Apr 2002 00:27:25 +0000 (00:27 +0000)]
Allow postmaster to start up anyway when PGSTAT code fails to initialize,
per recent discussion on pghackers.  Also, fix PGSTAT code to report
errors via elog, not scribbling directly on stderr.

22 years agoFix compiler warnings.
Hiroshi Inoue [Tue, 2 Apr 2002 10:57:25 +0000 (10:57 +0000)]
Fix compiler warnings.

22 years ago[HACKERS] Proposed patch for ODBC driver w/ C-a-n-c-e-l
Hiroshi Inoue [Tue, 2 Apr 2002 10:50:50 +0000 (10:50 +0000)]
[HACKERS] Proposed patch for ODBC driver w/ C-a-n-c-e-l
    From: Bradley McLean <brad@bradm.net>

Patch against 7,2 submitted for comment.

This seems to work just fine; Now, when our users submit a 2 hour
query with four million row sorts by accident, then cancel it 30 seconds
later, it doesn't bog down the server ...

22 years agoRemoved obsolete DROP_COLUMN_HACK stuff.
Hiroshi Inoue [Tue, 2 Apr 2002 08:51:52 +0000 (08:51 +0000)]
Removed obsolete DROP_COLUMN_HACK stuff.

22 years agoTweak SERIAL column creation to emit a fully qualified sequence name
Tom Lane [Tue, 2 Apr 2002 06:30:34 +0000 (06:30 +0000)]
Tweak SERIAL column creation to emit a fully qualified sequence name
as argument for nextval().

22 years agoRemoved error message that was incorectly being issued. This fixes a problem reporte...
Barry Lind [Tue, 2 Apr 2002 06:24:10 +0000 (06:24 +0000)]
Removed error message that was incorectly being issued.  This fixes a problem reported a few months ago where a select in a rule was causing an insert statement to return a result set which the code was explicitly prohibiting.

22 years agoFix CLOG truncation code to not do the Wrong Thing when there are already
Tom Lane [Tue, 2 Apr 2002 05:11:55 +0000 (05:11 +0000)]
Fix CLOG truncation code to not do the Wrong Thing when there are already
wrapped-around databases.  The unvacuumed databases might be fine, or
they might not, but things will definitely not be fine if we remove the
wrong CLOG segments.  Per trouble report from Gary Wolfe, 1-Apr-2002.

22 years agoFix contrib/dbsize for schema-qualified table names.
Tom Lane [Tue, 2 Apr 2002 01:17:28 +0000 (01:17 +0000)]
Fix contrib/dbsize for schema-qualified table names.

22 years agoMake VACUUM handle schema-qualified relation names properly.
Tom Lane [Tue, 2 Apr 2002 01:03:07 +0000 (01:03 +0000)]
Make VACUUM handle schema-qualified relation names properly.

22 years agoAdd tgconstrrelid to stored Trigger structures, make RI trigger functions
Tom Lane [Mon, 1 Apr 2002 22:36:13 +0000 (22:36 +0000)]
Add tgconstrrelid to stored Trigger structures, make RI trigger functions
depend on this rather than the trigger argument strings to locate the
other relation to test.  This makes RI triggers function properly in
the presence of schemas and temp tables.  Along the way, fix bogus lack
of locking in RI triggers, handle quoting of names fully correctly,
compute required sizes of query buffers with some semblance of accuracy.

22 years agoAttached is a patch which adds 2 missing semi-colons to
Bruce Momjian [Mon, 1 Apr 2002 14:22:41 +0000 (14:22 +0000)]
Attached is a patch which adds 2 missing semi-colons to
bootstrap/bootparse.y, so that recent versions of bison don't emit a
warning.

Neil Conway

22 years agoUpdate for SET/DROP NOT NULL.
Bruce Momjian [Mon, 1 Apr 2002 14:12:23 +0000 (14:12 +0000)]
Update for SET/DROP NOT NULL.

22 years agoALTER TABLE SET/DROP NOT NULL, from Christopher Kings-Lynne.
Tom Lane [Mon, 1 Apr 2002 04:35:40 +0000 (04:35 +0000)]
ALTER TABLE SET/DROP NOT NULL, from Christopher Kings-Lynne.

22 years agoCreate a new GUC variable search_path to control the namespace search
Tom Lane [Mon, 1 Apr 2002 03:34:27 +0000 (03:34 +0000)]
Create a new GUC variable search_path to control the namespace search
path.  The default behavior if no per-user schemas are created is that
all users share a 'public' namespace, thus providing behavior backwards
compatible with 7.2 and earlier releases.  Probably the semantics and
default setting will need to be fine-tuned, but this is a start.

22 years ago1) Add rollback functionality to updatable cursors.
Hiroshi Inoue [Mon, 1 Apr 2002 03:01:15 +0000 (03:01 +0000)]
1) Add rollback functionality to updatable cursors.
2) Implement some options for SQLGetDescField().
3) Handle *Inifinity* timestamp for SQL_C_CHAR type output.
4) Separate Unicode conversions from common implementations.
5) Improve internal parse_statement() function.

22 years agoTemporary kluge to let RI triggers work on tables outside the system
Tom Lane [Mon, 1 Apr 2002 02:02:34 +0000 (02:02 +0000)]
Temporary kluge to let RI triggers work on tables outside the system
catalog namespace.  This will not do as a production solution because
the interpretation of RI trigger arguments will vary depending on the
current namespace search path.  I'm just putting it in so that the RI
regression tests don't fail while schema development proceeds.  We
must find a better answer before 7.3 can be released.

22 years agoOops, not sure these are done.
Bruce Momjian [Mon, 1 Apr 2002 01:56:38 +0000 (01:56 +0000)]
Oops, not sure these are done.

> * Remove unreferenced table files and temp tables during database vacuum
> * Allow psql \d to show temporary table structure

22 years agoFix broken PG_VERSION_STR.
Tom Lane [Mon, 1 Apr 2002 01:11:00 +0000 (01:11 +0000)]
Fix broken PG_VERSION_STR.

22 years agoFixed now, I think:
Bruce Momjian [Sun, 31 Mar 2002 12:56:16 +0000 (12:56 +0000)]
Fixed now, I think:

> * -Remove unreferenced table files and temp tables during database vacuum
> * -Allow psql \d to show temporary table structure

22 years agoMake renamerel take an OID, not a RangeVar, to identify the relation
Tom Lane [Sun, 31 Mar 2002 07:49:30 +0000 (07:49 +0000)]
Make renamerel take an OID, not a RangeVar, to identify the relation
to rename.  Avoids some corner-case bugs in cluster.c, improves
consistency with renameatt.

22 years agoReimplement temp tables using schemas. The temp table map is history;
Tom Lane [Sun, 31 Mar 2002 06:26:32 +0000 (06:26 +0000)]
Reimplement temp tables using schemas.  The temp table map is history;
temp table entries in pg_class have the names the user would expect.

22 years agoFurther cleanups for relations in schemas: teach nextval and other
Tom Lane [Sat, 30 Mar 2002 01:02:42 +0000 (01:02 +0000)]
Further cleanups for relations in schemas: teach nextval and other
sequence functions how to cope with qualified names.  Same code is
also used for int4notin, currtid_byrelname, pgstattuple.  Also,
move TOAST tables into special pg_toast namespace.

22 years agoFix more random breakage manifesting on FreeBSD.
Peter Eisentraut [Sat, 30 Mar 2002 00:59:52 +0000 (00:59 +0000)]
Fix more random breakage manifesting on FreeBSD.

22 years agoReorder some things so prerequisite macros don't get expanded implicitly
Peter Eisentraut [Sat, 30 Mar 2002 00:20:15 +0000 (00:20 +0000)]
Reorder some things so prerequisite macros don't get expanded implicitly
within a shell conditional.

22 years agoMop-up some infelicities in new relation lookup handling.
Tom Lane [Fri, 29 Mar 2002 22:10:34 +0000 (22:10 +0000)]
Mop-up some infelicities in new relation lookup handling.

22 years agoAdd missing comma.
Peter Eisentraut [Fri, 29 Mar 2002 20:54:33 +0000 (20:54 +0000)]
Add missing comma.

22 years agopg_type has a typnamespace column; system now supports creating types
Tom Lane [Fri, 29 Mar 2002 19:06:29 +0000 (19:06 +0000)]
pg_type has a typnamespace column; system now supports creating types
in different namespaces.  Also, cleanup work on relation namespace
support: drop, alter, rename commands work for tables in non-default
namespaces.

22 years agoUpgrade to Autoconf version 2.53. Replaced many custom macro
Peter Eisentraut [Fri, 29 Mar 2002 17:32:55 +0000 (17:32 +0000)]
Upgrade to Autoconf version 2.53.  Replaced many custom macro
calls with new or now-built-in versions.  Make sure that all
calls to AC_DEFINE have a third argument, for possible use of
autoheader in the future.

22 years agoFix ODBC warnings about missing parens and pg_ismb not used (added
Bruce Momjian [Fri, 29 Mar 2002 07:48:56 +0000 (07:48 +0000)]
Fix ODBC warnings about missing parens and pg_ismb not used (added
#ifdef NOT_USED).

22 years agoFox for old python PyMem_DEL, was PyMem_Del,
Bruce Momjian [Fri, 29 Mar 2002 07:45:39 +0000 (07:45 +0000)]
Fox for old python PyMem_DEL, was PyMem_Del,

22 years agoFix '#' commant to be proper C comment.
Bruce Momjian [Fri, 29 Mar 2002 07:21:25 +0000 (07:21 +0000)]
Fix '#' commant to be proper C comment.

22 years ago[2002-03-28]
Hiroshi Inoue [Thu, 28 Mar 2002 08:08:07 +0000 (08:08 +0000)]
[2002-03-28]
1) Prepare to separate 4 kinds of Descriptor handles.
2) Detect the transaction status more naturally.
3) Improve Parse Statement functionality for the use
   of updatable cursors.
4) Improve updatable cursors.
5) Implement SQLGetDescField() and improve SQLColAttribute().
6) etc.

22 years agoDon't mark up as acronym that which is not an acronym.
Peter Eisentraut [Wed, 27 Mar 2002 19:19:23 +0000 (19:19 +0000)]
Don't mark up as acronym that which is not an acronym.

22 years agoReset parenthesis level counter upon \r.
Peter Eisentraut [Wed, 27 Mar 2002 19:16:13 +0000 (19:16 +0000)]
Reset parenthesis level counter upon \r.

22 years agoapplied patch from Liam Stewart fixing a message in the properties file
Barry Lind [Wed, 27 Mar 2002 05:33:27 +0000 (05:33 +0000)]
applied patch from Liam Stewart fixing a message in the properties file
also fixed a NPE when calling the next() method on a result set after the
connection or resultset has been closed. (bug reported by Hans Deragon)

22 years agoImprove spelling of lock names, and some other editing.
Peter Eisentraut [Wed, 27 Mar 2002 02:36:51 +0000 (02:36 +0000)]
Improve spelling of lock names, and some other editing.

22 years agopg_class has a relnamespace column. You can create and access tables
Tom Lane [Tue, 26 Mar 2002 19:17:02 +0000 (19:17 +0000)]
pg_class has a relnamespace column.  You can create and access tables
in schemas other than the system namespace; however, there's no search
path yet, and not all operations work yet on tables outside the system
namespace.

22 years agoapplied patch submitted by Florian (mailing-list@urbanet.ch) for BigDecimal support
Barry Lind [Tue, 26 Mar 2002 06:33:21 +0000 (06:33 +0000)]
applied patch submitted by Florian (mailing-list@urbanet.ch) for BigDecimal support

22 years agoChanged QueryExecutor.java to correctly read responses from the backend.
Barry Lind [Tue, 26 Mar 2002 05:52:50 +0000 (05:52 +0000)]
Changed QueryExecutor.java to correctly read responses from the backend.
Fixed NPE when database name was not passed on the jdbc connection URL
Fixed Connection.isClosed() to not hit the DB for every call

22 years agoUpdate release list.
Bruce Momjian [Tue, 26 Mar 2002 05:34:33 +0000 (05:34 +0000)]
Update release list.

22 years agoUpdate sgml version properly.
Bruce Momjian [Tue, 26 Mar 2002 05:33:23 +0000 (05:33 +0000)]
Update sgml version properly.

22 years agoFurther changes following discussion on INTERFACES mailing list:
D'Arcy J.M. Cain [Tue, 26 Mar 2002 01:50:14 +0000 (01:50 +0000)]
Further changes following discussion on INTERFACES mailing list:

  - Use PyObject_Del() rather than macro version
  - Check version and drop back to PyMem_Del() for older systems.

22 years agoNew wording:
Bruce Momjian [Mon, 25 Mar 2002 21:24:08 +0000 (21:24 +0000)]
New wording:

 Allow EXECUTE of "CREATE TABLE AS ... SELECT" in PL/PgSQL (Tom)

22 years agoAdd to HISTORY:
Bruce Momjian [Mon, 25 Mar 2002 20:58:05 +0000 (20:58 +0000)]
Add to HISTORY:

 Allow CREATE TABLE AS ... SELECT in PL/PgSQL (Tom)

22 years agoUpdate for temp views:
Bruce Momjian [Mon, 25 Mar 2002 20:56:08 +0000 (20:56 +0000)]
Update for temp views:

< * Automatically make a view temporary if it uses a temporary table
> * Require view using temporary tables to be temporary views

22 years agoRe-allow CREATE AS (but not SELECT INTO) in EXECUTE.
Tom Lane [Mon, 25 Mar 2002 07:41:10 +0000 (07:41 +0000)]
Re-allow CREATE AS (but not SELECT INTO) in EXECUTE.

22 years agoAdd:
Bruce Momjian [Mon, 25 Mar 2002 03:37:01 +0000 (03:37 +0000)]
Add:
>  o Fix problems with complex temporary table creation/destruction
>    without using PL/PgSQL EXECUTE

22 years agoFix for ODBC compile warnings.
Bruce Momjian [Mon, 25 Mar 2002 01:53:55 +0000 (01:53 +0000)]
Fix for ODBC compile warnings.

22 years agoAdd:
Bruce Momjian [Sun, 24 Mar 2002 22:37:20 +0000 (22:37 +0000)]
Add:

> * Automatically make a view temporary if it uses a temporary table

22 years agoIndenting #if commands is not portable, is it?
Tom Lane [Sun, 24 Mar 2002 18:22:21 +0000 (18:22 +0000)]
Indenting #if commands is not portable, is it?

22 years agoTweak labeling of plan qual conditions for more consistency.
Tom Lane [Sun, 24 Mar 2002 17:11:37 +0000 (17:11 +0000)]
Tweak labeling of plan qual conditions for more consistency.

22 years agoFix cut-and-paste-o.
Tom Lane [Sun, 24 Mar 2002 16:57:29 +0000 (16:57 +0000)]
Fix cut-and-paste-o.

22 years agoEXPLAIN output now comes out as a query result, not a NOTICE message.
Tom Lane [Sun, 24 Mar 2002 04:31:09 +0000 (04:31 +0000)]
EXPLAIN output now comes out as a query result, not a NOTICE message.
Also, fix debug logging of parse/plan trees so that the messages actually
go through elog(), not directly to stdout.

22 years agoCreate the pg_namespace system catalog. Doesn't do much yet, but it's
Tom Lane [Fri, 22 Mar 2002 21:34:44 +0000 (21:34 +0000)]
Create the pg_namespace system catalog.  Doesn't do much yet, but it's
there and CREATE SCHEMA will make entries in it...

22 years agoImprove catalog commentary.
Tom Lane [Fri, 22 Mar 2002 20:14:42 +0000 (20:14 +0000)]
Improve catalog commentary.

22 years agoSpell checking and markup additions
Peter Eisentraut [Fri, 22 Mar 2002 19:20:45 +0000 (19:20 +0000)]
Spell checking and markup additions

22 years agoA little further progress on schemas: push down RangeVars into
Tom Lane [Fri, 22 Mar 2002 02:56:37 +0000 (02:56 +0000)]
A little further progress on schemas: push down RangeVars into
addRangeTableEntry calls.  Remove relname field from RTEs, since
it will no longer be a useful unique identifier of relations;
we want to encourage people to rely on the relation OID instead.
Further work on dumping qual expressions in EXPLAIN, too.

22 years agoChange the aclchk.c routines to uniformly use OIDs to identify the
Tom Lane [Thu, 21 Mar 2002 23:27:25 +0000 (23:27 +0000)]
Change the aclchk.c routines to uniformly use OIDs to identify the
objects to be privilege-checked.  Some change in their APIs would be
necessary no matter what in the schema environment, and simply getting
rid of the name-based interface entirely seems like the best way.

22 years agoFix minor syntax error.
Tom Lane [Thu, 21 Mar 2002 18:28:07 +0000 (18:28 +0000)]
Fix minor syntax error.

22 years agoFirst phase of SCHEMA changes, concentrating on fixing the grammar and
Tom Lane [Thu, 21 Mar 2002 16:02:16 +0000 (16:02 +0000)]
First phase of SCHEMA changes, concentrating on fixing the grammar and
the parsetree representation.  As yet we don't *do* anything with schema
names, just drop 'em on the floor; but you can enter schema-compatible
command syntax, and there's even a primitive CREATE SCHEMA command.
No doc updates yet, except to note that you can now extract a field
from a function-returning-row's result with (foo(...)).fieldname.

22 years agoRemove -mips2 flag, which breaks the build on some MIPS systems.
Peter Eisentraut [Thu, 21 Mar 2002 15:21:41 +0000 (15:21 +0000)]
Remove -mips2 flag, which breaks the build on some MIPS systems.

Report from Ryan Murray via Debian

22 years agoAdded typedef patches and a new option '-c' to automatically create C typedefs from...
Michael Meskes [Thu, 21 Mar 2002 09:42:52 +0000 (09:42 +0000)]
Added typedef patches and a new option '-c' to automatically create C typedefs from SQL ones.

22 years agoRemove long-dead 'fix for SELECT NULL' to stop current coredump.
Tom Lane [Thu, 21 Mar 2002 06:21:04 +0000 (06:21 +0000)]
Remove long-dead 'fix for SELECT NULL' to stop current coredump.

22 years agoFix pg_dump help option alignment.
Bruce Momjian [Thu, 21 Mar 2002 05:47:14 +0000 (05:47 +0000)]
Fix pg_dump help option alignment.

22 years agoTwo versions of QueryExecutor, currently only version 2 works 100%
Dave Cramer [Thu, 21 Mar 2002 03:20:30 +0000 (03:20 +0000)]
Two versions of QueryExecutor, currently only version 2 works 100%
these versions adhere to the backend protocol better than previous version
fixes problem when an error occurs on the backend, and the connection is still used
previous versions were throwing an exception half way through the protocol, leaving it
indeterminate.
also removes empty query code, should speed things up a bit

22 years agomodifications to the way the protocol is handled to be consistent with
Dave Cramer [Thu, 21 Mar 2002 02:52:37 +0000 (02:52 +0000)]
modifications to the way the protocol is handled to be consistent with
QueryExecutor. This includes:
1) only exit after we receive a 'Z' packet
2) append error messages to a buffer and throw the exception at the end

22 years agoPart of Anders Bengtsson's patch to clean up Connection.java
Dave Cramer [Thu, 21 Mar 2002 02:40:03 +0000 (02:40 +0000)]
Part of Anders Bengtsson's patch to clean up Connection.java

22 years agoApplied Anders patch to move the startup code out of Connection into StartupPacket
Dave Cramer [Thu, 21 Mar 2002 02:39:06 +0000 (02:39 +0000)]
Applied Anders patch to move the startup code out of Connection into StartupPacket
* Introduces a new class, StartupPacket.
* Moves a lot of constants from Connection to StartupPacket.
* Makes two instance variables in Connection into locals.

22 years agoJust one more test case.
Michael Meskes [Wed, 20 Mar 2002 21:00:24 +0000 (21:00 +0000)]
Just one more test case.

22 years agoCode review for DOMAIN patch.
Tom Lane [Wed, 20 Mar 2002 19:45:13 +0000 (19:45 +0000)]
Code review for DOMAIN patch.

22 years agoTweak behavior of array slicing operations: seems like it ought to be
Tom Lane [Wed, 20 Mar 2002 19:41:47 +0000 (19:41 +0000)]
Tweak behavior of array slicing operations: seems like it ought to be
okay to omit low-order dimensions when accessing an array slice.

22 years agoPyGreSQL causes a segfault when used with a Python executable that was
D'Arcy J.M. Cain [Wed, 20 Mar 2002 14:36:06 +0000 (14:36 +0000)]
PyGreSQL causes a segfault when used with a Python executable that was
compiled with --with-pymalloc.  This change fixes that.  Thanks to
Dave Wallace <dwallace@udel.edu>

22 years agoPut back development status section overwritten by Bruce.
Tom Lane [Tue, 19 Mar 2002 18:53:01 +0000 (18:53 +0000)]
Put back development status section overwritten by Bruce.

22 years agoMake CREATE DOMAIN emit CREATE DOMAIN on completion, per Peter E.
Bruce Momjian [Tue, 19 Mar 2002 16:10:48 +0000 (16:10 +0000)]
Make CREATE DOMAIN emit CREATE DOMAIN on completion, per Peter E.

22 years ago> > Users of contrib/tsearch needs after upgrading of module (compiling, installing)
Bruce Momjian [Tue, 19 Mar 2002 14:15:55 +0000 (14:15 +0000)]
> > Users of contrib/tsearch needs after upgrading of module (compiling, installing)
> > to perform sql command:
> >  update pg_amop set amopreqcheck = true where amopclaid =
> >       (select oid from pg_opclass where opcname = 'gist_txtidx_ops');
>
> Oleg, sorry, I don't understand where this should appear.  In the README
> file, and if so, where?  Is this something only for people upgrading
> from 7.2?

Sorry Bruce, I was unclear. I have attached patch to Readme.tsearch
Also, It'd be worth to mention in Changes to point users of tsearch
about importang upgrade notices.

Oleg Bartunov

22 years ago> > Users of contrib/tsearch needs after upgrading of module (compiling, installing)
Bruce Momjian [Tue, 19 Mar 2002 14:14:44 +0000 (14:14 +0000)]
> > Users of contrib/tsearch needs after upgrading of module (compiling, installing)
> > to perform sql command:
> >  update pg_amop set amopreqcheck = true where amopclaid =
> >       (select oid from pg_opclass where opcname = 'gist_txtidx_ops');
>
> Oleg, sorry, I don't understand where this should appear.  In the README
> file, and if so, where?  Is this something only for people upgrading
> from 7.2?

Sorry Bruce, I was unclear. I have attached patch to Readme.tsearch
Also, It'd be worth to mention in Changes to point users of tsearch
about importang upgrade notices.

Oleg Bartunov

22 years agoBump version number to match tree.
D'Arcy J.M. Cain [Tue, 19 Mar 2002 13:21:14 +0000 (13:21 +0000)]
Bump version number to match tree.

22 years agoFix name in comments and add info about this module's place in PyGreSQL.
D'Arcy J.M. Cain [Tue, 19 Mar 2002 13:20:52 +0000 (13:20 +0000)]
Fix name in comments and add info about this module's place in PyGreSQL.

22 years agoJust adds a regressions test suite for the ALTER TABLE/ADD PRIMARY KEY
Bruce Momjian [Tue, 19 Mar 2002 12:59:08 +0000 (12:59 +0000)]
Just adds a regressions test suite for the ALTER TABLE/ADD PRIMARY KEY
feature.

I'll do ALTER TABLE / DROP CONSTRAINT next...

Christopher Kings-Lynne

22 years agoAdd missing trailing semicolons to ecpg rules.
Bruce Momjian [Tue, 19 Mar 2002 12:56:38 +0000 (12:56 +0000)]
Add missing trailing semicolons to ecpg rules.

Neil Conway

22 years agoFix DOMAIN breakage.
Bruce Momjian [Tue, 19 Mar 2002 12:52:22 +0000 (12:52 +0000)]
Fix DOMAIN breakage.