OSDN Git Service
Bruce Momjian [Thu, 25 Oct 2001 19:22:05 +0000 (19:22 +0000)]
Add slash for comment spacing, for Tom.
Bruce Momjian [Thu, 25 Oct 2001 18:44:42 +0000 (18:44 +0000)]
Add comment spaces for trailing ) and comment.
Bruce Momjian [Thu, 25 Oct 2001 18:25:23 +0000 (18:25 +0000)]
Add spacing for single-line comments with trailing semicolon _and_
comma, per Tom.
Bruce Momjian [Thu, 25 Oct 2001 16:20:51 +0000 (16:20 +0000)]
Remove <> from FAQ_MSWIN, for Vince.
Tom Lane [Thu, 25 Oct 2001 14:10:07 +0000 (14:10 +0000)]
Add int2-to-int8 and int8-to-int2 conversion routines. Needed to avoid
breaking existing pg_dump scripts, which try to assign the result of
count(*) to an int2 variable. catversion bumped.
Tom Lane [Thu, 25 Oct 2001 14:08:11 +0000 (14:08 +0000)]
Add missing output routine for FkConstraint nodes.
PostgreSQL Daemon [Thu, 25 Oct 2001 13:02:01 +0000 (13:02 +0000)]
bmake sure that configure is tags as 7.2b1 now instead of 7.2devel
Bruce Momjian [Thu, 25 Oct 2001 06:27:56 +0000 (06:27 +0000)]
Code cleanup.
Bruce Momjian [Thu, 25 Oct 2001 06:00:05 +0000 (06:00 +0000)]
pgjindent jdbc files. First time jdbc files were formatted.
Bruce Momjian [Thu, 25 Oct 2001 05:50:21 +0000 (05:50 +0000)]
pgindent run on all C files. Java run to follow. initdb/regression
tests pass.
Bruce Momjian [Thu, 25 Oct 2001 05:07:56 +0000 (05:07 +0000)]
Update to prevent CATALOG() from wrapping.
Bruce Momjian [Thu, 25 Oct 2001 03:56:35 +0000 (03:56 +0000)]
Add current typedef symbols to pgindent.
Bruce Momjian [Thu, 25 Oct 2001 01:29:37 +0000 (01:29 +0000)]
Add do { ... } while (0) to more bad macros.
Bruce Momjian [Thu, 25 Oct 2001 00:55:48 +0000 (00:55 +0000)]
Add more missing 'do { ... } while (0)' in missing macros. Without it,
these macros fail in if/else cases:
#define X \
{ \
... \
}
{
if (...)
X;
else
...
}
with proper setup:
#define X \
do { \
... \
} while (0)
it works fine.
Bruce Momjian [Thu, 25 Oct 2001 00:50:50 +0000 (00:50 +0000)]
Add missing "do { ... } while(0)" in ODBC macros and add find_baddefs
script.
Peter Eisentraut [Wed, 24 Oct 2001 21:49:14 +0000 (21:49 +0000)]
Include locale.h before undefining gettext() to avoid compilation errors
on Solaris.
per report from Lee Kindness, 2001-10-23
Bruce Momjian [Wed, 24 Oct 2001 17:52:28 +0000 (17:52 +0000)]
All three *.po patches applied.
forth@pagic.net
Bruce Momjian [Wed, 24 Oct 2001 17:44:28 +0000 (17:44 +0000)]
Here is a patch for DatabaseMetaData to show precision properly. It is
from Mark Lillywhite. I am adding to the patch queue.
Hiroshi Inoue [Wed, 24 Oct 2001 09:28:31 +0000 (09:28 +0000)]
Allow concurrent index creation for the same table.
Tatsuo Ishii [Wed, 24 Oct 2001 08:07:22 +0000 (08:07 +0000)]
"time" is now a reserved keyword(changed to "mtime").
Barry Lind [Wed, 24 Oct 2001 04:31:50 +0000 (04:31 +0000)]
fix for a bug in DatabaseMetaData.getIndexInfo(). This fixes a bug reported by tom_falconer@lineone.net. On Sept 7th, he sent a test case to the list demonstrating the bug. His test case now works successfully with this patch
Bruce Momjian [Wed, 24 Oct 2001 01:01:36 +0000 (01:01 +0000)]
Update TODO list.
Bruce Momjian [Tue, 23 Oct 2001 23:27:18 +0000 (23:27 +0000)]
Update TODO list.
Tom Lane [Tue, 23 Oct 2001 22:11:22 +0000 (22:11 +0000)]
Editorial work on descriptions of options.
Tom Lane [Tue, 23 Oct 2001 21:26:44 +0000 (21:26 +0000)]
Forbid the switch combination --clean --create, which is pointless
(why bother dropping individual objects in a just-created database?)
as well as dangerous (as the code stands, the drops will be issued in
the wrong database, namely the one you were originally connected to).
Tom Lane [Tue, 23 Oct 2001 20:17:27 +0000 (20:17 +0000)]
Document range restriction on bit values used in INTERVAL typmod.
Tom Lane [Tue, 23 Oct 2001 20:12:54 +0000 (20:12 +0000)]
Add support for INTERVAL's new typmod values to format_type.
Tom Lane [Tue, 23 Oct 2001 17:39:03 +0000 (17:39 +0000)]
Fix foreign keys on system columns.
Tom Lane [Tue, 23 Oct 2001 17:38:25 +0000 (17:38 +0000)]
Make SPI's column-accessing functions work for system columns as well as
user columns. Needed for foreign keys on OID, but useful in general.
Bruce Momjian [Tue, 23 Oct 2001 16:42:36 +0000 (16:42 +0000)]
Move ALL out of UNION|INTERSECT|EXCEPT syntax.
Alvaro Herrera
Bruce Momjian [Tue, 23 Oct 2001 02:52:19 +0000 (02:52 +0000)]
Update TODO list.
Bruce Momjian [Tue, 23 Oct 2001 02:51:35 +0000 (02:51 +0000)]
Update TODO list.
Bruce Momjian [Tue, 23 Oct 2001 02:50:41 +0000 (02:50 +0000)]
Now I had them reversed. Wow that syntax is error-prone.
Bruce Momjian [Tue, 23 Oct 2001 02:47:14 +0000 (02:47 +0000)]
Syntax fix. Can't use #ifdef in gram.y.
Bruce Momjian [Tue, 23 Oct 2001 02:45:37 +0000 (02:45 +0000)]
Back out LIMIT #,# removal and mark it as to-be-removed in 7.3.
Tom Lane [Tue, 23 Oct 2001 02:20:15 +0000 (02:20 +0000)]
Add error checking to PageRepairFragmentation to ensure that it can
never overwrite adjacent pages with copied data, even if page header
and/or item pointers are already corrupt. Change inspired by trouble
report from Alvaro Herrera.
Tom Lane [Mon, 22 Oct 2001 23:48:11 +0000 (23:48 +0000)]
Add note explaining that multiple prompts for a password are normal
behavior when password auth is used.
Tom Lane [Mon, 22 Oct 2001 22:49:02 +0000 (22:49 +0000)]
Cause transformIndexConstraints() to do the right thing with requests
for indexes on system columns. Per complaint from Peter.
Tom Lane [Mon, 22 Oct 2001 22:47:57 +0000 (22:47 +0000)]
Consolidate tables of known system attributes into one table.
Tom Lane [Mon, 22 Oct 2001 19:52:41 +0000 (19:52 +0000)]
Update for latest fixes.
Tom Lane [Mon, 22 Oct 2001 19:41:38 +0000 (19:41 +0000)]
Further cleanup of ps_status setup code. On platforms where the
environment strings need to be moved around, do so when called from
initial startup (main.c), not in init_ps_status. This eliminates the
former risk of invalidating saved environment-string pointers, since
no code has yet had a chance to grab any such pointers when main.c
is running.
Tom Lane [Mon, 22 Oct 2001 19:34:13 +0000 (19:34 +0000)]
Add defenses against invalid operator names passed in CREATE OPERATOR
arguments (where the parser doesn't check them already). Minor code
cleanups too.
Tom Lane [Mon, 22 Oct 2001 19:32:27 +0000 (19:32 +0000)]
Rebuild cached function definition after CREATE OR REPLACE FUNCTION.
Fix typlen-vs-typmod errors inherited from pltcl.
Tom Lane [Mon, 22 Oct 2001 19:31:09 +0000 (19:31 +0000)]
Generate correct syntax for DROP OPERATOR with unary operators.
Bruce Momjian [Mon, 22 Oct 2001 18:42:44 +0000 (18:42 +0000)]
Add ecpg items to HISTORY file.
Peter Eisentraut [Mon, 22 Oct 2001 18:14:47 +0000 (18:14 +0000)]
Make the CREATE TABLE ref page more readable and update some information.
Michael Meskes [Mon, 22 Oct 2001 17:05:56 +0000 (17:05 +0000)]
synced preproc.y with gram.y
Tom Lane [Sun, 21 Oct 2001 03:43:54 +0000 (03:43 +0000)]
Fix FindExec() for case where executable is found via a relative path
in .:/home/postgres/testversion/bin:/opt/perl5.6.1/bin:/home/postgres/bin:/usr/local/bin:/usr/bin:/opt/ansic/bin:/usr/ccs/bin:/usr/contrib/bin:/opt/nettladm/bin:/opt/pd/bin:/usr/bin/X11:/usr/contrib/bin/X11:/opt/upgrade/bin:/opt/CC/bin:/opt/langtools/bin:/opt/graphics/phigs/bin:/opt/java/bin:/bin:/opt/imake/bin:/opt/hparray/bin:/opt/aCC/bin:/opt/lrom/bin:/usr/local/nmh/bin:. (I suppose the only common case for this is '.').
Tom Lane [Sun, 21 Oct 2001 03:25:36 +0000 (03:25 +0000)]
Fix getopt-vs-init_ps_display problem by copying original argv[] info,
per suggestion from Peter. Simplify several APIs by transmitting the
original argv location directly from main.c to ps_status.c, instead of
passing it down through several levels of subroutines.
Tom Lane [Sat, 20 Oct 2001 17:57:39 +0000 (17:57 +0000)]
Revert change to autodetect '__getopt_initialized' variable ... turns
out glibc doesn't actually export this variable, so we need a different
solution.
Bruce Momjian [Sat, 20 Oct 2001 16:51:02 +0000 (16:51 +0000)]
Change LIMIT #,# error message to be more generic.
Bruce Momjian [Sat, 20 Oct 2001 02:55:39 +0000 (02:55 +0000)]
Remove LIMIT #,# and suggest LIMIT # OFFSET #, per mailing list discussion.
Thomas G. Lockhart [Sat, 20 Oct 2001 01:02:22 +0000 (01:02 +0000)]
Fix transposed arguments for typmod for one INTERVAL production.
Mask both typmod subfields for INTERVAL to avoid setting the high bit,
per dire warning from Tom Lane.
Clear tmask for DTK_ISO_TIME case to avoid time zone troubles.
Symptom reported by Tom Lane.
Clean up checking for valid time zone info in output routine.
This should now work for both SQL99 and Unix-style time zones.
Put in explicit check for INTERVAL() typmod rounding to avoid accumulating
cruft in the lower bits. Not sure that this helps, but we'll need to do
something. The symptom is visible with a query like
select interval(2) '10000 days 01:02:03.040506';
Regression tests are patched to repair the Tom Lane symptom, and all pass.
Tom Lane [Fri, 19 Oct 2001 22:43:49 +0000 (22:43 +0000)]
Fix plperl to discard cached function definition after CREATE OR
REPLACE FUNCTION. Clean up typlen/typmod errors inherited from pltcl.
Tom Lane [Fri, 19 Oct 2001 20:47:09 +0000 (20:47 +0000)]
Proper fix for glibc getopt() botch. Surprising we did not see this
before.
Tom Lane [Fri, 19 Oct 2001 19:43:19 +0000 (19:43 +0000)]
Fix a lot of confusion between typlen and typmod. Didn't hurt too much
back when only varlena types paid any attention to typmod ...
Tom Lane [Fri, 19 Oct 2001 18:19:41 +0000 (18:19 +0000)]
Set optreset on platforms that have it before launching postmaster
subprocesses; perhaps this will fix portability problem just noted by
Lockhart. Also, move test for bad permissions of DataDir to a more
logical place.
Tom Lane [Fri, 19 Oct 2001 17:03:08 +0000 (17:03 +0000)]
Ensure that all startup paths (postmaster, standalone postgres, or
bootstrap) check for a valid PG_VERSION file before looking at anything
else in the data directory. This fixes confusing error report when
trying to start current sources in a pre-7.1 data directory.
Per trouble report from Rich Shepard 10/18/01.
Tom Lane [Fri, 19 Oct 2001 15:38:57 +0000 (15:38 +0000)]
Rename macro arg to avoid useless compiler warning when building plperl.
Bruce Momjian [Fri, 19 Oct 2001 15:06:48 +0000 (15:06 +0000)]
Without this patch I couldn't compile PostgreSQL on Solaris 8 x86 using
Sun's compiler. May be it will be usefull for someone else?
Denis Ustimenko
Bruce Momjian [Fri, 19 Oct 2001 15:00:26 +0000 (15:00 +0000)]
> This stops the interface from leaking the row tuples (and thus the
> results of every fetch).
Stephen Robert Norris
Bruce Momjian [Fri, 19 Oct 2001 14:59:04 +0000 (14:59 +0000)]
Update TODO list.
Michael Meskes [Fri, 19 Oct 2001 14:31:44 +0000 (14:31 +0000)]
removed "not fully implemented" warning for nullif
Hiroshi Inoue [Fri, 19 Oct 2001 05:39:14 +0000 (05:39 +0000)]
Change SC_fetch() to not retrieve data in case the
SQL_RETRIEVE_DATA option is SQL_RD_OFF. This would
avoid a crash when scrolling Grid controls.
Tom Lane [Fri, 19 Oct 2001 02:43:46 +0000 (02:43 +0000)]
Fix pltcl to update cached function def after
CREATE OR REPLACE FUNCTION.
Bruce Momjian [Fri, 19 Oct 2001 01:30:17 +0000 (01:30 +0000)]
Update TODO list.
Tom Lane [Fri, 19 Oct 2001 00:46:51 +0000 (00:46 +0000)]
Typo.
Tom Lane [Fri, 19 Oct 2001 00:44:08 +0000 (00:44 +0000)]
Move init_ps_display from postgres.c to postc, putting it
just after receipt of the startup packet. Now, postmaster children
that are waiting for client authentication response will show as
'postgres: user database host authentication'. Also, do an
init_ps_display for startup/shutdown/checkpoint subprocesses,
so that they are readily identifiable as well. Fix an obscure race
condition that could lead to Assert failure in the postmaster ---
attempting to start a checkpoint process before any connections have
been received led to calling PostmasterRandom before setting random_seed.
Tom Lane [Thu, 18 Oct 2001 23:16:09 +0000 (23:16 +0000)]
Spell 'between' correctly, clean up spacing in error messages.
Tom Lane [Thu, 18 Oct 2001 23:07:29 +0000 (23:07 +0000)]
Fix memory leakage when sending notice messages to client.
Tom Lane [Thu, 18 Oct 2001 22:44:37 +0000 (22:44 +0000)]
Fix authentication so that it doesn't record an extra 'Password
authentication failed' and a 'send() failed: Broken pipe' message
on every connection from psql in password auth mode. Problem is
that psql doesn't ask user for a password until it sees a password
challenge failure, and libpq just closes the connection unceremoniously
if it's challenged for a password when it hasn't got one to send.
Accordingly, EOF from the client after asking for a password is
normal behavior and should not result in postmaster log entries.
Tom Lane [Thu, 18 Oct 2001 22:40:52 +0000 (22:40 +0000)]
Add STATUS_EOF value needed for auth.c. Remove a bunch of unused
STATUS_xxx macros to reduce clutter.
Tom Lane [Thu, 18 Oct 2001 21:57:11 +0000 (21:57 +0000)]
Fix several problems with simple_prompt() --- the nastiest being that
the entered password would get echoed on some platforms, eg HPUX.
We have enough copies of this code that I'm thinking it ought to be
moved into libpq, but that's a task for another day.
Bruce Momjian [Thu, 18 Oct 2001 20:43:05 +0000 (20:43 +0000)]
Change error code docs to use colons.
Bruce Momjian [Thu, 18 Oct 2001 20:32:58 +0000 (20:32 +0000)]
Document all the ecpg error codes.
Tom Lane [Thu, 18 Oct 2001 20:10:31 +0000 (20:10 +0000)]
Update for new expected results.
Tom Lane [Thu, 18 Oct 2001 19:54:59 +0000 (19:54 +0000)]
Didn't compile on non-HAVE_TM_ZONE machines.
Tom Lane [Thu, 18 Oct 2001 19:52:03 +0000 (19:52 +0000)]
Repair missing brace in HAVE_INT_TIMEZONE case.
Thomas G. Lockhart [Thu, 18 Oct 2001 17:30:21 +0000 (17:30 +0000)]
Accept an INTERVAL argument for SET TIME ZONE per SQL99.
Modified the parser and the SET handlers to use full Node structures
rather than simply a character string argument.
Implement INTERVAL() YEAR TO MONTH (etc) syntax per SQL99.
Does not yet accept the goofy string format that goes along with, but
this should be fairly straight forward to fix now as a bug or later
as a feature.
Implement precision for the INTERVAL() type.
Use the typmod mechanism for both of INTERVAL features.
Fix the INTERVAL syntax in the parser:
opt_interval was in the wrong place.
INTERVAL is now a reserved word, otherwise we get reduce/reduce errors.
Implement an explicit date_part() function for TIMETZ.
Should fix coersion problem with INTERVAL reported by Peter E.
Fix up some error messages for date/time types.
Use all caps for type names within message.
Fix recently introduced side-effect bug disabling 'epoch' as a recognized
field for date_part() etc. Reported by Peter E. (??)
Bump catalog version number.
Rename "microseconds" current transaction time field
from ...Msec to ...Usec. Duh!
date/time regression tests updated for reference platform, but a few
changes will be necessary for others.
Tom Lane [Thu, 18 Oct 2001 16:11:42 +0000 (16:11 +0000)]
Extend code that deduces implied equality clauses to detect whether a
clause being added to a particular restriction-clause list is redundant
with those already in the list. This avoids useless work at runtime,
and (perhaps more importantly) keeps the selectivity estimation routines
from generating too-small estimates of numbers of output rows.
Also some minor improvements in OPTIMIZER_DEBUG displays.
Michael Meskes [Thu, 18 Oct 2001 11:01:35 +0000 (11:01 +0000)]
Added Lee Kindness' patches.
Bruce Momjian [Wed, 17 Oct 2001 00:48:24 +0000 (00:48 +0000)]
Update file for pgcrypto.
Tom Lane [Tue, 16 Oct 2001 23:57:06 +0000 (23:57 +0000)]
Add documentation about statistics collector and stats views & functions.
Tom Lane [Tue, 16 Oct 2001 22:35:27 +0000 (22:35 +0000)]
Make sure to clean out old activity string when recycling a backend entry.
Tom Lane [Tue, 16 Oct 2001 20:51:35 +0000 (20:51 +0000)]
Use LEFT JOIN, not FULL JOIN, in statistical views.
Barry Lind [Tue, 16 Oct 2001 20:07:17 +0000 (20:07 +0000)]
Updated the list of encodings supported to match what the backend now supports
Barry Lind [Tue, 16 Oct 2001 20:05:22 +0000 (20:05 +0000)]
Added some additional comments in the code
Bruce Momjian [Tue, 16 Oct 2001 13:42:57 +0000 (13:42 +0000)]
Python handle as string all int8 values from postgresql. This could be
view when using the aggregate function count() and function nextval
that returns an int8 value, but in python is represented like string:
>> db.query("select nextval('my_seq')").getresult()
[('2',)]
>> db.query("select count(*) from films").dictresult()
[{'count': '120'}]
Ricardo Caesar Lenzi
Tatsuo Ishii [Tue, 16 Oct 2001 10:09:17 +0000 (10:09 +0000)]
Ok, here is the modified encoding table (column1 is the standard name,
2 is our "official" name, and 3 is alias). If there's no objection, I
will change them.
ASCII SQL_ASCII
UTF-8 UNICODE UTF_8
MULE-INTERNAL MULE_INTERNAL
ISO-8859-1 LATIN1 ISO_8859_1
ISO-8859-2 LATIN2 ISO_8859_2
ISO-8859-3 LATIN3 ISO_8859_3
ISO-8859-4 LATIN4 ISO_8859_4
ISO-8859-5 ISO_8859_5
ISO-8859-6 ISO_8859_6
ISO-8859-7 ISO_8859_7
ISO-8859-8 ISO_8859_8
ISO-8859-9 LATIN5 ISO_8859_9
ISO-8859-10 LATIN6 ISO_8859_10
ISO-8859-13 LATIN7 ISO_8859_13
ISO-8859-14 LATIN8 ISO_8859_14
ISO-8859-15 LATIN9 ISO_8859_15
ISO-8859-16 LATIN10 ISO_8859_16
Tom Lane [Tue, 16 Oct 2001 01:13:44 +0000 (01:13 +0000)]
Add a section about the planner's statistics, including a description
of the new pg_stats view.
Tom Lane [Mon, 15 Oct 2001 22:47:47 +0000 (22:47 +0000)]
Add note about pg_stats to pg_statistic entry; add sections for
pg_largeobject, pg_listener, pg_rewrite, pg_trigger; miscellaneous
small improvements.
Tom Lane [Mon, 15 Oct 2001 20:15:09 +0000 (20:15 +0000)]
Doesn't anyone test stuff before they commit it?
Tom Lane [Mon, 15 Oct 2001 19:15:18 +0000 (19:15 +0000)]
Fix some portability problems (get it to compile, at least, on HP's cc)
Tom Lane [Mon, 15 Oct 2001 18:56:30 +0000 (18:56 +0000)]
Fix some portability problems (get it to compile, at least, on HP's cc)
Tom Lane [Mon, 15 Oct 2001 18:49:40 +0000 (18:49 +0000)]
Fix some portability problems (get it to compile, at least, on HP's cc)
Bruce Momjian [Mon, 15 Oct 2001 18:16:31 +0000 (18:16 +0000)]
> Uh, isn't the correct fix
> ! $$ = cat_str(8, make_str("grant"), $2, make_str("on"), $4, $5,
> make_str("to"), $7, $8);
> ISTM your patch loses the opt_with_grant clause. (Of course the
> backend doesn't currently accept that clause anyway, but that's no
> reason for ecpg to drop it.)
My patch doesn't loose the option, it's never been passed on anyway:
opt_with_grant: WITH GRANT OPTION
{
mmerror(ET_ERROR, "WITH GRANT OPTION is not supported. Only relation owners can
set privileges");
}
| /*EMPTY*/
;
The existing code in ecpg/preproc/preproc.y to handle the WITH option
simply throws an error and aborts the processing... The patch below
prevents the segfault and also passes on the WITH option to the
backend, probably a better fix.
Lee Kindness
Tom Lane [Mon, 15 Oct 2001 17:41:33 +0000 (17:41 +0000)]
Updated regression test for tsearch, from Teodor Sigaev.
Bruce Momjian [Mon, 15 Oct 2001 16:40:27 +0000 (16:40 +0000)]
Prompt for password from /dev/tty and fall back to stdin/stderr.
Tatsuo Ishii [Mon, 15 Oct 2001 04:52:59 +0000 (04:52 +0000)]
Forgot to add this file...
Bruce Momjian [Mon, 15 Oct 2001 02:50:16 +0000 (02:50 +0000)]
PG_DUMP NLS (Russian)
Here is another one :)
Another chunk of translated messages.
Please apply to the same file.
Serguei Mokhov