]> git.ipfire.org Git - thirdparty/postgresql.git/log
thirdparty/postgresql.git
19 years agoBackpatch to 8.0.X:
Bruce Momjian [Fri, 23 Sep 2005 21:02:37 +0000 (21:02 +0000)] 
Backpatch to 8.0.X:

In several places PL/Python was calling PyObject_Str() and then
PyString_AsString() without checking if the former had returned
NULL to indicate an error.  PyString_AsString() doesn't expect a
NULL argument, so passing one causes a segmentation fault.  This
patch adds checks for NULL and raises errors via PLy_elog(), which
prints details of the underlying Python exception.  The patch also
adds regression tests for these checks.  All tests pass on my
Solaris 9 box running HEAD and Python 2.4.1.

19 years agoUpdate FAQ's in 8.0.X branch.
Bruce Momjian [Thu, 22 Sep 2005 22:14:11 +0000 (22:14 +0000)] 
Update FAQ's in 8.0.X branch.

19 years agoReturn proper value for psql -f filename failure if filename open fails.
Bruce Momjian [Tue, 20 Sep 2005 18:59:15 +0000 (18:59 +0000)] 
Return proper value for psql -f filename failure if filename open fails.

Backpatch to 8.0.X.

19 years agoUpdate Snowball. I have to update it because of
Teodor Sigaev [Thu, 15 Sep 2005 11:56:58 +0000 (11:56 +0000)] 
Update Snowball. I have to update it because of
old version doesn't available on Snowball's site and new version
of stemmers can't be compiled with old interface.

19 years agoEnsure that any memory leaked during an error inside the bgwriter is
Tom Lane [Mon, 12 Sep 2005 22:20:30 +0000 (22:20 +0000)] 
Ensure that any memory leaked during an error inside the bgwriter is
recovered.  I did not see any actual leak while testing this in CVS tip,
but 8.0 definitely has a problem with leaking the space temporarily
palloc'd by BufferSync().  In any case this seems a good idea to forestall
similar problems in future.  Per report from Arjen van der Meijden.

19 years agoFix a mistake in the documentation for SPI_getbinval(), per Michael Fuhr.
Neil Conway [Mon, 12 Sep 2005 18:49:08 +0000 (18:49 +0000)] 
Fix a mistake in the documentation for SPI_getbinval(), per Michael Fuhr.

19 years agoFixed transaction command handling to not ignore savepoints and to correctly check...
Michael Meskes [Mon, 12 Sep 2005 11:58:33 +0000 (11:58 +0000)] 
Fixed transaction command handling to not ignore savepoints and to correctly check for errors.

19 years agoAvoid changing stdin/stdout to binary mode on Windows unless that is
Tom Lane [Sun, 11 Sep 2005 00:36:35 +0000 (00:36 +0000)] 
Avoid changing stdin/stdout to binary mode on Windows unless that is
really the source or destination of the archive.  I think this will
resolve recent complaints that password prompting is broken in pg_restore
on Windows.  Note that password prompting and reading from stdin is an
unworkable combination on Windows ... but that was true anyway.

19 years agoUpdate regression tests for new USA timezone data. Mea culpa for not
Tom Lane [Thu, 8 Sep 2005 16:49:18 +0000 (16:49 +0000)] 
Update regression tests for new USA timezone data.  Mea culpa for not
realizing that the regression tests could be affected.

19 years agoUpdate timezone data files to release 2005m of the zic database.
Tom Lane [Wed, 7 Sep 2005 21:39:41 +0000 (21:39 +0000)] 
Update timezone data files to release 2005m of the zic database.
Among other changes, this reflects the recently passed change in USA
daylight savings rules.

19 years agoTranslation update
Peter Eisentraut [Sat, 3 Sep 2005 07:41:58 +0000 (07:41 +0000)] 
Translation update

19 years agoFix missing rows in query
Teodor Sigaev [Tue, 30 Aug 2005 07:57:48 +0000 (07:57 +0000)] 
Fix missing rows in query
update a=.. where a... with GiST index on column 'a'

19 years agoSigh, looks like you need '.set mips2' before you can access MIPS
Tom Lane [Mon, 29 Aug 2005 00:41:44 +0000 (00:41 +0000)] 
Sigh, looks like you need '.set mips2' before you can access MIPS
SYNC instruction.

19 years agoAdd a SYNC instruction to the S_UNLOCK sequence for MIPS.
Tom Lane [Sun, 28 Aug 2005 18:26:07 +0000 (18:26 +0000)] 
Add a SYNC instruction to the S_UNLOCK sequence for MIPS.

19 years agoGet the MIPS assembler syntax right. Also add a separate sync command;
Tom Lane [Sat, 27 Aug 2005 16:22:58 +0000 (16:22 +0000)] 
Get the MIPS assembler syntax right.  Also add a separate sync command;
the reference I consulted yesterday said SC does a SYNC, but apparently
this is not true on newer MIPS processors, so be safe.

19 years agoFix typo:
Bruce Momjian [Sat, 27 Aug 2005 14:13:31 +0000 (14:13 +0000)] 
Fix typo:

sql_information_info -> sql_implementation_info

Robert Treat

19 years agoAnother try at the inlined MIPS spinlock code. Can't test this myself,
Tom Lane [Fri, 26 Aug 2005 22:04:53 +0000 (22:04 +0000)] 
Another try at the inlined MIPS spinlock code.  Can't test this myself,
but for sure it's not any more broken than the prior version.

19 years agoBack-port recent MIPS and M68K spinlock improvements to 8.0 branch.
Tom Lane [Fri, 26 Aug 2005 14:48:13 +0000 (14:48 +0000)] 
Back-port recent MIPS and M68K spinlock improvements to 8.0 branch.

19 years agoBack-patch fixes for problems with VACUUM destroying t_ctid chains too soon,
Tom Lane [Thu, 25 Aug 2005 19:45:06 +0000 (19:45 +0000)] 
Back-patch fixes for problems with VACUUM destroying t_ctid chains too soon,
and with insufficient paranoia in code that follows t_ctid links.
This patch covers the 8.0 branch.

19 years ago - Check for NULL before checking whether argument is an array.
Michael Meskes [Wed, 24 Aug 2005 10:35:12 +0000 (10:35 +0000)] 
    - Check for NULL before checking whether argument is an array.
- Removed stray character from string quoting.
- Fixed check to report missing varchar pointer implementation.

19 years agoFix broken lrand48() implementation, per Merlin Moncure.
Tom Lane [Tue, 23 Aug 2005 13:31:19 +0000 (13:31 +0000)] 
Fix broken lrand48() implementation, per Merlin Moncure.

19 years agoInvoke mksafefunc and mkunsafefunc with :: decoration. This seems a good
Tom Lane [Sat, 20 Aug 2005 19:19:31 +0000 (19:19 +0000)] 
Invoke mksafefunc and mkunsafefunc with :: decoration.  This seems a good
idea on consistency grounds, whether or not it really fixes bug #1831.
Michael Fuhr

19 years agoRemove unnecessary configure test for inet_ntop(), per Andrew Dunstan.
Tom Lane [Wed, 17 Aug 2005 20:20:22 +0000 (20:20 +0000)] 
Remove unnecessary configure test for inet_ntop(), per Andrew Dunstan.

19 years agoReject operator names >= NAMEDATALEN characters. These will not work
Tom Lane [Tue, 16 Aug 2005 00:48:29 +0000 (00:48 +0000)] 
Reject operator names >= NAMEDATALEN characters.  These will not work
anyway, and in assert-enabled builds you are likely to get an assertion
failure.  Backpatch as far as 7.3; 7.2 seems not to have the problem.

19 years agoarray_in() and array_recv() need to be more paranoid about validating
Tom Lane [Mon, 15 Aug 2005 19:40:43 +0000 (19:40 +0000)] 
array_in() and array_recv() need to be more paranoid about validating
their OID parameter.  It was possible to crash the backend with
select array_in('{123}',0,0); because that would bypass the needed step
of initializing the workspace.  These seem to be the only two places
with a problem, though (record_in and record_recv don't have the issue,
and the other array functions aren't depending on user-supplied input).
Back-patch as far as 7.4; 7.3 does not have the bug.

19 years agoint_array_enum function should be using fcinfo->flinfo->fn_extra for
Tom Lane [Mon, 15 Aug 2005 19:05:30 +0000 (19:05 +0000)] 
int_array_enum function should be using fcinfo->flinfo->fn_extra for
working state, not fcinfo->context.  Silly oversight on my part in last
go-round of fixes.

19 years agoThis patch fixes the event type used to log output from the
Bruce Momjian [Fri, 12 Aug 2005 21:38:00 +0000 (21:38 +0000)] 
This patch fixes the event type used to log output from the
stderr-in-service or output-from-syslogger-in-service code. Previously
everything was flagged as ERRORs there, which caused all instances to
log "LOG: logger shutting down" as error...

Please apply for 8.1. I'd also like it considered for 8.0 since logging
non-errors as errors can be cause for alarm amongst people who actually
look at their logs...

Magnus Hagander

19 years ago[ backpatched to 8.0.X.]
Bruce Momjian [Fri, 12 Aug 2005 21:23:18 +0000 (21:23 +0000)] 
[ backpatched to 8.0.X.]

> >> 3) I restarted the postmaster both times. I got this error
> both times.
> >> :25: ERROR:  could not load library "C:/Program
> >> Files/PostgreSQL/8.0/lib/testtrigfuncs.dll": dynamic load error
>
> > Yes. We really need to look at fixing that error message. I had
> > forgotten it completely :-(
>
> > Bruce, you think we can sneak that in after feature freeze? I would
> > call it a bugfix :-)
>
> Me too.  That's been on the radar for awhile --- please do
> send in a patch.

Here we go, that wasn't too hard :-)

Apart from adding the error handling, it does one more thing: it changes
the errormode when loading the DLLs. Previously if a DLL was broken, or
referenced other DLLs that couldn't be found, a popup dialog box would
appear on the screen. Which had to be clicked before the backend could
continue. This patch also disables the popup error message for DLL
loads.

I think this is something we should consider doing for the entire
backend - disable those popups, and say we deal with it ourselves. What
do you other win32 hackers thinnk about this?

In the meantime, this patch fixes the error msgs. Please apply for 8.1
and please consider a backpatch to 8.0.

Magnus Hagander

19 years agoBackpatch fseeko fix for seeking from end of file.
Bruce Momjian [Fri, 12 Aug 2005 00:29:22 +0000 (00:29 +0000)] 
Backpatch fseeko fix for seeking from end of file.

19 years agoBackpatch new CVS home page URL.
Bruce Momjian [Thu, 11 Aug 2005 13:54:10 +0000 (13:54 +0000)] 
Backpatch new CVS home page URL.

19 years agoFix crash when reading 'timezone = unknown' from postgresql.conf during
Tom Lane [Mon, 8 Aug 2005 23:39:14 +0000 (23:39 +0000)] 
Fix crash when reading 'timezone = unknown' from postgresql.conf during
SIGHUP; it's not OK for an assign_hook to return a non-malloc'd string.
Problem was introduced during timezone library rewrite.

19 years agoFix count_usable_fds() to stop trying to open files once it reaches
Tom Lane [Sun, 7 Aug 2005 18:47:38 +0000 (18:47 +0000)] 
Fix count_usable_fds() to stop trying to open files once it reaches
max_files_per_process.  Going further than that is just a waste of
cycles, and it seems that current Cygwin does not cope gracefully
with deliberately running the system out of FDs.  Per Andrew Dunstan.

19 years agormtree() reported the wrong pathname if final rmdir failed.
Tom Lane [Tue, 2 Aug 2005 15:14:56 +0000 (15:14 +0000)] 
rmtree() reported the wrong pathname if final rmdir failed.

19 years agoBack-port several small portability fixes to get contrib building
Tom Lane [Tue, 2 Aug 2005 01:35:23 +0000 (01:35 +0000)] 
Back-port several small portability fixes to get contrib building
cleanly on AIX in the 8.0 branch.  Rocco Altier

19 years agoUpdate AIX FAQ.
Bruce Momjian [Sat, 30 Jul 2005 03:39:40 +0000 (03:39 +0000)] 
Update AIX FAQ.

Chris Browne

19 years agoBack-patch fix for NOT-below-a-NOT case.
Tom Lane [Fri, 29 Jul 2005 21:40:26 +0000 (21:40 +0000)] 
Back-patch fix for NOT-below-a-NOT case.

19 years agoAvoid crashing pg_dump if we can't connect to the database server, and
Neil Conway [Wed, 27 Jul 2005 05:15:03 +0000 (05:15 +0000)] 
Avoid crashing pg_dump if we can't connect to the database server, and
no database has been explicitly specified. Per gripe from Omar Kilani.

19 years agoFix compare_fuzzy_path_costs() to behave a bit more sanely. The original
Tom Lane [Fri, 22 Jul 2005 19:12:33 +0000 (19:12 +0000)] 
Fix compare_fuzzy_path_costs() to behave a bit more sanely.  The original
coding would ignore startup cost differences of less than 1% of the
estimated total cost; which was OK for normal planning but highly not OK
if a very small LIMIT was applied afterwards, so that startup cost becomes
the name of the game.  Instead, compare startup and total costs fuzzily
but independently.  This changes the plan selected for two queries in the
regression tests; adjust expected-output files for resulting changes in
row order.  Per reports from Dawid Kuroczko and Sam Mason.

20 years agoIt appears that Darwin (OS X) does not cope well with C functions that
Tom Lane [Mon, 18 Jul 2005 21:37:23 +0000 (21:37 +0000)] 
It appears that Darwin (OS X) does not cope well with C functions that
have the same name as the containing shared library --- as best I can
tell, the compiler internally creates a function of that name, and does
not warn you about the conflict.  Fix buildfarm failure in back branches
by renaming tsearch() trigger function at the C level.

20 years agoBack-patch fix for erroneous backslashing of LIKE pattern.
Tom Lane [Mon, 18 Jul 2005 19:18:19 +0000 (19:18 +0000)] 
Back-patch fix for erroneous backslashing of LIKE pattern.

20 years agoMemSet() must not cast its pointer argument to int32* until after it has
Tom Lane [Mon, 18 Jul 2005 15:53:46 +0000 (15:53 +0000)] 
MemSet() must not cast its pointer argument to int32* until after it has
checked that the pointer is actually word-aligned.  Casting a non-aligned
pointer to int32* is technically illegal per the C spec, and some recent
versions of gcc actually generate bad code for the memset() when given
such a pointer.  Per report from Andrew Morrow.

20 years agoMake pg_regress accept a command-line option for the temporary installation's
Tom Lane [Sun, 17 Jul 2005 18:29:13 +0000 (18:29 +0000)] 
Make pg_regress accept a command-line option for the temporary installation's
port number, and use a default value for it that is dependent on the
configuration-time DEF_PGPORT.  Should make the world safe for running
parallel 'make check' in different branches.  Back-patch as far as 7.4
so that this actually is useful.

20 years agoBack-patch recent changes to alter the order of -L flags inserted from
Tom Lane [Sun, 17 Jul 2005 04:05:49 +0000 (04:05 +0000)] 
Back-patch recent changes to alter the order of -L flags inserted from
LDFLAGS versus those built into the Makefiles.  This looks like it will
fix several buildfarm failures in the back branches.

20 years agoCheck for out-of-range varoattno in deparse_context_for_subplan.
Tom Lane [Fri, 15 Jul 2005 18:40:20 +0000 (18:40 +0000)] 
Check for out-of-range varoattno in deparse_context_for_subplan.
I have seen this case in CVS tip due to new "physical tlist" optimization
for subqueries.  I believe it probably can't happen in existing releases,
but the check is not going to hurt anything, so backpatch to 8.0 just
in case.

20 years agoFix overenthusiastic optimization of 'x IN (SELECT DISTINCT ...)' and related
Tom Lane [Fri, 15 Jul 2005 17:09:50 +0000 (17:09 +0000)] 
Fix overenthusiastic optimization of 'x IN (SELECT DISTINCT ...)' and related
cases: we can't just consider whether the subquery's output is unique on its
own terms, we have to check whether the set of output columns we are going to
use will be unique.  Per complaint from Luca Pireddu and test case from
Michael Fuhr.

20 years agoFix libpq memory leak during PQreset() --- closePGconn() was not
Tom Lane [Wed, 13 Jul 2005 15:26:06 +0000 (15:26 +0000)] 
Fix libpq memory leak during PQreset() --- closePGconn() was not
freeing all transient state of the PGconn object.

20 years agoFix config file lexer to not barf if postgresql.conf ends with a comment
Tom Lane [Fri, 8 Jul 2005 18:41:55 +0000 (18:41 +0000)] 
Fix config file lexer to not barf if postgresql.conf ends with a comment
that has no terminating newline.  Per report from maps.on at gmx.net.

20 years agoMake libpq_gettext save and restore errno in a Windows-compatible way.
Tom Lane [Fri, 8 Jul 2005 15:24:53 +0000 (15:24 +0000)] 
Make libpq_gettext save and restore errno in a Windows-compatible way.
Also, back-patch fix into back branches.

20 years agoBackpatch to 8.0.X openssl portability fixes to pgcrypto.
Bruce Momjian [Mon, 4 Jul 2005 14:42:39 +0000 (14:42 +0000)] 
Backpatch to 8.0.X openssl portability fixes to pgcrypto.

20 years agoFix date_trunct for December dates that are in the next year, e.g.:
Bruce Momjian [Mon, 4 Jul 2005 14:12:45 +0000 (14:12 +0000)] 
Fix date_trunct for December dates that are in the next year, e.g.:

SELECT date_trunc('week', '2002-12-31'::date);

Backpatch to 8.0.X.

Per report from Nick Johnson.

20 years agoFix memory leak in plperl_hash_from_tuple(), per report from Jean-Max Reymond.
Tom Lane [Sun, 3 Jul 2005 21:56:27 +0000 (21:56 +0000)] 
Fix memory leak in plperl_hash_from_tuple(), per report from Jean-Max Reymond.

20 years agoModify pg_dump to assume that a check constraint is inherited if its
Tom Lane [Mon, 27 Jun 2005 02:18:14 +0000 (02:18 +0000)] 
Modify pg_dump to assume that a check constraint is inherited if its
name matches the name of any parent-table constraint, without looking
at the constraint text.  This is a not-very-bulletproof workaround for
the problem exhibited by Berend Tober last month.  We really ought to
record constraint inheritance status in pg_constraint, but it's looking
like that may not get done for 8.1 --- and even if it does, we will
need this kluge for dumping from older servers.

20 years agoForce a checkpoint before committing a CREATE DATABASE command. This
Tom Lane [Sat, 25 Jun 2005 22:47:49 +0000 (22:47 +0000)] 
Force a checkpoint before committing a CREATE DATABASE command.  This
should fix the recent reports of "index is not a btree" failures,
as well as preventing a more obscure race condition involving changes
to a template database just after copying it with CREATE DATABASE.

20 years agoFix ancient memory leak in index_create(): RelationInitIndexAccessInfo
Tom Lane [Sat, 25 Jun 2005 16:54:00 +0000 (16:54 +0000)] 
Fix ancient memory leak in index_create(): RelationInitIndexAccessInfo
was being called twice in normal operation, leading to a leak of one set
of relcache subsidiary info.  Per report from Jeff Gold.

20 years agoCorrect some code in pg_restore when reading the header of a tar archive:
Neil Conway [Wed, 22 Jun 2005 02:02:09 +0000 (02:02 +0000)] 
Correct some code in pg_restore when reading the header of a tar archive:

(1) The code doesn't initialize `sum', so the initial "does the checksum
    match?" test is wrong.

(2) The loop that is intended to check for a "null block" just checks
    the first byte of the tar block 512 times, rather than each of the
    512 bytes one time (!), which I'm guessing was the intent.

It was only through sheer luck that this worked in the first place.

Per Coverity static analysis performed by EnterpriseDB.

20 years agoFix a potential backend crash during authentication when parsing a
Neil Conway [Tue, 21 Jun 2005 01:23:25 +0000 (01:23 +0000)] 
Fix a potential backend crash during authentication when parsing a
malformed ident map file.  This was introduced by the linked list
rewrite in 8.0 -- mea maxima culpa.

Per Coverity static analysis performed by EnterpriseDB.

20 years agoexec_eval_datum leaks memory when dealing with ROW or REC values.
Tom Lane [Mon, 20 Jun 2005 22:51:49 +0000 (22:51 +0000)] 
exec_eval_datum leaks memory when dealing with ROW or REC values.
It never leaked memory before PG 8.0, so none of the callers are
expecting this.  Cleanest fix seems to be to make it allocate the needed
memory in estate->eval_econtext, where it will be cleaned up by
the next exec_eval_cleanup.  Per report from Bill Rugolsky.

20 years agoplpgsql's exec_assign_value() freed the old value of a variable before
Tom Lane [Mon, 20 Jun 2005 20:44:50 +0000 (20:44 +0000)] 
plpgsql's exec_assign_value() freed the old value of a variable before
copying/converting the new value, which meant that it failed badly on
"var := var" if var is of pass-by-reference type.  Fix this and a similar
hazard in exec_move_row(); not sure that the latter can manifest before
8.0, but patch it all the way back anyway.  Per report from Dave Chapeskie.

20 years agoWhen using C-string lookup keys in a dynahash.c hash table, use strncpy()
Tom Lane [Sat, 18 Jun 2005 20:51:44 +0000 (20:51 +0000)] 
When using C-string lookup keys in a dynahash.c hash table, use strncpy()
not memcpy() to copy the offered key into the hash table during HASH_ENTER.
This avoids possible core dump if the passed key is located very near the
end of memory.  Per report from Stefan Kaltenbrunner.

20 years agoNetBSD uses "options" not "option".
Bruce Momjian [Fri, 17 Jun 2005 16:46:45 +0000 (16:46 +0000)] 
NetBSD uses "options" not "option".

20 years agoTranslation updates
Peter Eisentraut [Fri, 17 Jun 2005 11:42:02 +0000 (11:42 +0000)] 
Translation updates

20 years agoThe random selection in function linear() could deliver a value equal to max
Tom Lane [Tue, 14 Jun 2005 14:21:23 +0000 (14:21 +0000)] 
The random selection in function linear() could deliver a value equal to max
if geqo_rand() returns exactly 1.0, resulting in failure due to indexing
off the end of the pool array.  Also, since this is using inexact float math,
it seems wise to guard against roundoff error producing values slightly
outside the expected range.  Per report from bug@zedware.org.

20 years agoFix bug in MIC -> EUC_JP conversion. Per Atsushi Ogawa.
Tatsuo Ishii [Sat, 11 Jun 2005 02:46:16 +0000 (02:46 +0000)] 
Fix bug in MIC -> EUC_JP conversion. Per Atsushi Ogawa.

20 years agoRepair error in description of nonblocking usage of PQgetCopyData().
Tom Lane [Thu, 9 Jun 2005 19:08:36 +0000 (19:08 +0000)] 
Repair error in description of nonblocking usage of PQgetCopyData().
Per Volkan Yazici.

20 years agoUse just NULL not NULL::TEXT --- the latter coding is unnecessary and
Tom Lane [Tue, 7 Jun 2005 14:05:01 +0000 (14:05 +0000)] 
Use just NULL not NULL::TEXT --- the latter coding is unnecessary and
not schema-safe.  Per report from Jochem van Dieten.

20 years agoCode for SET/SHOW TIME ZONE with a fixed-interval timezone was not
Tom Lane [Sun, 5 Jun 2005 01:48:45 +0000 (01:48 +0000)] 
Code for SET/SHOW TIME ZONE with a fixed-interval timezone was not
prepared for HAVE_INT64_TIMESTAMP.  Per report from Guillaume Beaudoin.

20 years agoUpdate Chinese FAQ to fix XHTML format.
Bruce Momjian [Sat, 4 Jun 2005 03:33:25 +0000 (03:33 +0000)] 
Update Chinese FAQ to fix XHTML format.

Weiping (Laser)

20 years agoUpdate Russian FAQ.
Bruce Momjian [Fri, 3 Jun 2005 18:17:54 +0000 (18:17 +0000)] 
Update Russian FAQ.

Viktor Vislobokov

20 years agoPush enable/disable of notify and catchup interrupts all the way down
Tom Lane [Thu, 2 Jun 2005 21:03:46 +0000 (21:03 +0000)] 
Push enable/disable of notify and catchup interrupts all the way down
to just around the bare recv() call that gets a command from the client.
The former placement in PostgresMain was unsafe because the intermediate
processing layers (especially SSL) use facilities such as malloc that are
not necessarily re-entrant.  Per report from counterstorm.com.

20 years agoAdded patch by Gavin Scott <gavin@planetacetech.com> for Intel 64bit hardware.
Michael Meskes [Thu, 2 Jun 2005 12:49:26 +0000 (12:49 +0000)] 
Added patch by Gavin Scott <gavin@planetacetech.com> for Intel 64bit hardware.
[One half already was committed with the last commit.]

20 years agoFixed memory leak in ecpglib by adding some missing free() commands.
Michael Meskes [Thu, 2 Jun 2005 12:37:25 +0000 (12:37 +0000)] 
Fixed memory leak in ecpglib by adding some missing free() commands.

20 years agoFix log_statement to properly recognize SELECT INTO and CREATE TABLE AS
Bruce Momjian [Wed, 1 Jun 2005 23:27:12 +0000 (23:27 +0000)] 
Fix log_statement to properly recognize SELECT INTO and CREATE TABLE AS
and DDL statements.

Backpatch fix to 8.0.X.

Per report from Murthy Kambhampaty

20 years agopatternsel() was improperly stripping RelabelType from the derived
Tom Lane [Wed, 1 Jun 2005 17:05:25 +0000 (17:05 +0000)] 
patternsel() was improperly stripping RelabelType from the derived
expressions it constructed, causing scalarineqsel to become confused
if the underlying variable was of a domain type.  Per report from
Kevin Grittner.

20 years agoPrevent to divide by zero and range out of 0..1
Teodor Sigaev [Wed, 1 Jun 2005 11:45:42 +0000 (11:45 +0000)] 
Prevent to divide by zero and range out of 0..1

20 years agoAdd test to WAL replay to verify that xl_prev points back to the previous
Tom Lane [Tue, 31 May 2005 19:10:39 +0000 (19:10 +0000)] 
Add test to WAL replay to verify that xl_prev points back to the previous
WAL record; this is necessary to be sure we recognize stale WAL records
when a WAL page was only partially written during a system crash.

20 years agoexpandRTE and get_rte_attribute_type mistakenly always imputed typmod -1
Tom Lane [Sun, 29 May 2005 17:10:35 +0000 (17:10 +0000)] 
expandRTE and get_rte_attribute_type mistakenly always imputed typmod -1
to columns of an RTE that was a function returning RECORD with a column
definition list.  Apparently no one has tried to use non-default typmod
with a function returning RECORD before.

20 years agoUpdate Farsi FAQ.
Bruce Momjian [Fri, 27 May 2005 22:10:20 +0000 (22:10 +0000)] 
Update Farsi FAQ.

Mahmoud Taghizadeh

20 years agoAdjust datetime parsing to be more robust. We now pass the length of the
Neil Conway [Thu, 26 May 2005 02:10:03 +0000 (02:10 +0000)] 
Adjust datetime parsing to be more robust. We now pass the length of the
working buffer into ParseDateTime() and reject too-long input there,
rather than checking the length of the input string before calling
ParseDateTime(). The old method was bogus because ParseDateTime() can use
a variable amount of working space, depending on the content of the
input string (e.g. how many fields need to be NUL terminated). This fixes
a minor stack overrun -- I don't _think_ it's exploitable, although I
won't claim to be an expert.

Along the way, fix a bug reported by Mark Dilger: the working buffer
allocated by interval_in() was too short, which resulted in rejecting
some perfectly valid interval input values. I added a regression test for
this fix.

20 years agoTranslation updates
Peter Eisentraut [Wed, 25 May 2005 08:37:22 +0000 (08:37 +0000)] 
Translation updates

20 years agoFix previous patch to exprTypmod.
Tatsuo Ishii [Wed, 25 May 2005 02:17:55 +0000 (02:17 +0000)] 
Fix previous patch to exprTypmod.

20 years agoInserting 5 characters into char(10) does not produce 5 padding spaces
Tatsuo Ishii [Tue, 24 May 2005 23:02:54 +0000 (23:02 +0000)] 
Inserting 5 characters into char(10) does not produce 5 padding spaces
if they are two-byte multibyte characters. Same thing can be happen
if octet_length(multibyte_chars) == n where n is char(n).
Long standing bug since 7.3 days. Per report and fix from Yoshiyuki Asaba.

20 years agoPrevious fix for "x FULL JOIN y ON true" failed to handle the case
Tom Lane [Tue, 24 May 2005 18:02:55 +0000 (18:02 +0000)] 
Previous fix for "x FULL JOIN y ON true" failed to handle the case
where there was also a WHERE-clause restriction that applied to the
join.  The check on restrictlist == NIL is really unnecessary anyway,
because select_mergejoin_clauses already checked for and complained
about any unmergejoinable join clauses.  So just take it out.

20 years agoFix typo in PL/Perl Safe.pm initialization that prevented the proper
Neil Conway [Mon, 23 May 2005 02:02:52 +0000 (02:02 +0000)] 
Fix typo in PL/Perl Safe.pm initialization that prevented the proper
sharing of %_SHARED. From Andrew Dunstan.

20 years agoUpdate Japanese FAQ.
Bruce Momjian [Fri, 20 May 2005 15:53:52 +0000 (15:53 +0000)] 
Update Japanese FAQ.

Jun Kuwamura

20 years agoFix typo in ECPG docs, per Andreas Seltenreich.
Neil Conway [Fri, 20 May 2005 12:37:51 +0000 (12:37 +0000)] 
Fix typo in ECPG docs, per Andreas Seltenreich.

20 years agoGuard against duplicate IDs in input file in SortTocFromFile().
Tom Lane [Tue, 17 May 2005 17:30:41 +0000 (17:30 +0000)] 
Guard against duplicate IDs in input file in SortTocFromFile().
Per report from Brian Hackett.

20 years agoConvert Chinese FAQ to valid XHTML, finally.
Bruce Momjian [Tue, 17 May 2005 04:18:30 +0000 (04:18 +0000)] 
Convert Chinese FAQ to valid XHTML, finally.

20 years agoFix Chinese markup some more.
Bruce Momjian [Tue, 17 May 2005 03:41:14 +0000 (03:41 +0000)] 
Fix Chinese markup some more.

20 years agoUpdate Chinese FAQ to xhtml.
Bruce Momjian [Mon, 16 May 2005 02:50:58 +0000 (02:50 +0000)] 
Update Chinese FAQ to xhtml.

20 years agoAdd -N make flag to bcc builds from /src dir.
Bruce Momjian [Fri, 13 May 2005 18:13:16 +0000 (18:13 +0000)] 
Add -N make flag to bcc builds from /src dir.

20 years agoAdd missing quote, per Dave Page.
Tom Lane [Fri, 13 May 2005 16:48:22 +0000 (16:48 +0000)] 
Add missing quote, per Dave Page.

20 years agoFix broken markup.
Tom Lane [Fri, 13 May 2005 16:47:03 +0000 (16:47 +0000)] 
Fix broken markup.

20 years agoUpdate createuser examples to match the current program behavior,
Tom Lane [Fri, 13 May 2005 16:31:50 +0000 (16:31 +0000)] 
Update createuser examples to match the current program behavior,
and add an example showing assignment of a password.  Per suggestion
from Jari Aalto (via Martin Pitt).

20 years agoUpdate chinese encoding specification.
Bruce Momjian [Fri, 13 May 2005 13:48:05 +0000 (13:48 +0000)] 
Update chinese encoding specification.

20 years agoFix bug in COPY CSV mode: handle consecutive embedded newlines in COPY
Neil Conway [Fri, 13 May 2005 06:35:25 +0000 (06:35 +0000)] 
Fix bug in COPY CSV mode: handle consecutive embedded newlines in COPY
input. Also add a regression test for this bug. From Andrew Dunstan.

20 years agoFix pg_autovacuum -s flag to handle values > 2000 by using sleep()
Bruce Momjian [Wed, 11 May 2005 17:58:32 +0000 (17:58 +0000)] 
Fix pg_autovacuum -s flag to handle values > 2000 by using sleep()
instead of pg_usleep.

Backpatch to 8.0.X.

20 years agoUpdate FAQ URLs.
Bruce Momjian [Wed, 11 May 2005 16:13:58 +0000 (16:13 +0000)] 
Update FAQ URLs.

Robert Treat

20 years agoDocument where to download free Microsoft and Borland development tools.
Bruce Momjian [Wed, 11 May 2005 16:02:59 +0000 (16:02 +0000)] 
Document where to download free Microsoft and Borland development tools.

20 years agoDocument that bcc compiles now need the -N flag, backpatch to 8.0.X.
Bruce Momjian [Wed, 11 May 2005 15:10:56 +0000 (15:10 +0000)] 
Document that bcc compiles now need the -N flag, backpatch to 8.0.X.

20 years agoUpdate Chinese FAQ, per Magnus.
Bruce Momjian [Wed, 11 May 2005 14:57:26 +0000 (14:57 +0000)] 
Update Chinese FAQ, per Magnus.