Commit Graph

13880 Commits

Author SHA1 Message Date
Tom Lane 193a97c2d3 Fix plperl's elog() function to convert elog(ERROR) into Perl croak(),
rather than longjmp'ing clear out of Perl and thereby leaving Perl in
a broken state.  Also some minor prettification of error messages.
Still need to do something with spi_exec_query() error handling.
2004-11-20 19:07:40 +00:00
Tom Lane d5013ab50f Fix one more place where we were expecting lcons() to be nondestructive
to the original List; per report from Sebastian BÎck.  I think this is
the last such bug --- I examined every lcons() call in the backend and
the rest seem OK --- but it's nervous-making that we're still finding
'em so many months after the List rewrite went in.
2004-11-20 17:59:31 +00:00
Tom Lane 0b5c72d9ba On some HPUX versions/hardware, it seems we need to include libgcc.a in
the regression test shared libraries.
2004-11-20 03:52:00 +00:00
Tom Lane 83fea34b5b Fix unportable isdigit() call --- must cast arg to unsigned char. 2004-11-20 02:09:47 +00:00
Tom Lane 8a7025f0bb Move pgstat_report_tabstat() call so that stats are not reported to the
collector until the transaction commits.  Per recent discussion, this
should avoid confusing autovacuum when an updating transaction runs for
a long time.
2004-11-20 00:48:58 +00:00
Tom Lane e719591ea3 Make sure we produce a useful error message for the case where the
postgresql.crt file simply isn't there, too.
2004-11-20 00:35:13 +00:00
Tom Lane 7506677b62 Improve error reporting for SSL connection failures. Remove redundant
free operations in client_cert_cb --- openssl will also attempt to free
these structures, resulting in core dumps.
2004-11-20 00:18:18 +00:00
Tom Lane 8e57975219 Clean up rpath handling for HPUX --- we can't use the cc-style rpath
switch syntax when calling ld directly.
2004-11-19 21:27:42 +00:00
Tom Lane 466fb06739 Set rpath for plperl and plpython to ensure we can find the .so files
for the languages even when not installed in a standard directory.
pltcl may need this treatment as well, but we don't have the right path
conveniently available, so I'll leave it alone as long as there aren't
actual reports of trouble.
2004-11-19 19:23:01 +00:00
Tom Lane a9ddd649df Modify the platform-specific makefiles so that macro 'rpath' is defined
in terms of macro 'rpathdir', as I proposed a few weeks ago.  In itself
this commit shouldn't change the behavior at all, but it opens the door
to using special rpaths for the PL shared libraries, as seems to be
needed for plperl in particular.
2004-11-19 00:41:39 +00:00
Tom Lane b6038484f8 Fix memory clobber problem reported by John Hansen: plperl_safe_init()
may expand the Perl stack, therefore we must SPAGAIN to reload the local
stack pointer after calling it.  Also a couple other marginal readability
improvements.
2004-11-18 21:35:42 +00:00
Tom Lane da1c19aa57 Whoops, missed converting the other sleep() call to pg_usleep(). 2004-11-18 17:13:38 +00:00
Peter Eisentraut 050a0074c2 Translation update 2004-11-18 10:09:59 +00:00
Tom Lane b2a2f4cef7 Force pg_database updates out to disk immediately after ALTER DATABASE;
this is to avoid scenarios where incoming backends find no live copies
of a database's row because the only live copy is in an as-yet-unwritten
shared buffer, which they can't see.  Also, use FlushRelationBuffers()
for forcing out pg_database, instead of the much more expensive BufferSync().
There's no need to write out pages belonging to other relations.
2004-11-18 01:14:26 +00:00
Tom Lane c36496a101 Include <getopt.h> if available, to avoid warnings on Windows. 2004-11-17 21:37:47 +00:00
Tom Lane 1c1f2f5b96 Remove ill-considered suppression of gcc warnings in plperl, and fix
some of the bugs exposed thereby.  The remaining 'might be used uninitialized'
warnings look like live bugs, but I am not familiar enough with Perl/C hacking
to tell how to fix them.
2004-11-17 21:23:36 +00:00
Tom Lane edcaa8f691 Fix off-by-one memory allocation, as reported by Rod Taylor. Also
avoid repalloc'ing twice when once is sufficient.
2004-11-17 19:54:24 +00:00
Tom Lane 6444bc3535 Set DEFAULT_EDITOR to notepad.exe on Windows, as per recent discussion. 2004-11-17 18:32:31 +00:00
Tom Lane 4a6fd46950 Force LANG=en on Windows, since system may fail to default to English. 2004-11-17 18:06:04 +00:00
Tom Lane 98640c3130 Add installcheck-parallel regression test target. Magnus Hagander 2004-11-17 18:05:06 +00:00
Tom Lane 77fe4fd656 Use pg_usleep() not sleep(), per Andrew Dunstan. 2004-11-17 17:50:20 +00:00
Tom Lane 5d72ef83fd Miscellaneous Cygwin build fixes from Reini Urban. 2004-11-17 17:46:24 +00:00
Tom Lane ede8f4e311 Add variant regression file to support BSDen that underflow to plus zero instead
of minus zero.  Per Andrew Dunstan.
2004-11-17 17:25:32 +00:00
Tom Lane e5f7a9f404 Install include/port header files, and fix PGXS build to use them.
Fabien COELHO
2004-11-17 17:16:17 +00:00
Tom Lane 86b5545260 Adjust SHLIB_LINK for cygwin case.
Reini Urban
2004-11-17 17:08:15 +00:00
Tom Lane 18b58350db Be sure length limit passed to snprintf matches what malloc was given.
Just paranoia ...
2004-11-17 16:34:42 +00:00
Tom Lane 37d693033d Minor adjustment of message style. 2004-11-17 16:26:59 +00:00
Neil Conway 2fa36d7e41 Win32 build cleanups, from Andrew Dunstan. 2004-11-17 08:30:11 +00:00
Neil Conway ffe130f52e Remove debugging printf from #ifdef WIN32 section. 2004-11-17 04:05:42 +00:00
Neil Conway 5d1dd2bc55 Micro-optimization of markpos() and restrpos() in btree and hash indexes.
Rather than using ReadBuffer() to increment the reference count on an
already-pinned buffer, we should use IncrBufferRefCount() as it is
faster and does not require acquiring the BufMgrLock.
2004-11-17 03:13:38 +00:00
Neil Conway b25d23e1e6 Don't allow pg_start_backup() to be invoked if archive_command has not
been defined. Patch from Gavin Sherry, editorializing by Neil Conway.
2004-11-17 02:22:54 +00:00
Tom Lane 0021ae06be Fix Win32 problems with signals and sockets, by making the forkexec code
even uglier than it was already :-(.  Also, on Windows only, use temporary
shared memory segments instead of ordinary files to pass over critical
variable values from postmaster to child processes.  Magnus Hagander
2004-11-17 00:14:14 +00:00
Neil Conway e1bf6527f6 Prevent a backend crash when processing CREATE TABLE commands with
more than 65K columns, or when the created table has more than 65K columns
due to adding inherited columns from parent relations. Fix a similar
crash when processing SELECT queries with more than 65K target list
entries. In all three cases we would eventually detect the error and
elog, but the check was being made too late.
2004-11-16 23:34:26 +00:00
Peter Eisentraut 8a1821ab5b Translation update 2004-11-16 22:58:44 +00:00
Tom Lane 4a0bee976c Adjust safety restrictions for plperl functions. Andrew Dunstan's
patch, but allow srand and disallow sprintf as per subsequent discussion.
2004-11-16 22:05:22 +00:00
Tom Lane bb85f1b9b1 Suppress duplicate rules for lib$(NAME).a on WIN32 and Cygwin.
Andrew Dunstan
2004-11-16 21:51:13 +00:00
Tom Lane 6beb6fa495 Use dynamically-sized buffers in pgwin32_is_service().
Magnus Hagander
2004-11-16 19:52:22 +00:00
Tom Lane 7efa8411cc Rethink plpgsql's way of handling SPI execution during an exception block.
We don't really want to start a new SPI connection, just keep using the old
one; otherwise we have memory management problems as illustrated by
John Kennedy's bug report of today.  This requires a bit of a hack to
ensure the SPI stack state is properly restored, but then again what we
were doing before was a hack too, strictly speaking.  Add a regression
test to cover this case.
2004-11-16 18:10:16 +00:00
Tom Lane 3d1e269e7e Don't quote the value of EDITOR on Unix, only on Windows. Per discussion. 2004-11-15 23:15:12 +00:00
Tom Lane f6474586ea Use English-style quotes in error messages, per Serguei Mokhov. 2004-11-14 23:36:53 +00:00
Tom Lane ea23ec82c2 Remove GUC USERLIMIT variable category, making the affected variables
plain SUSET instead.  Also delay processing of options received in
client connection request until after we know if the user is a superuser,
so that SUSET values can be set that way by legitimate superusers.
Per recent discussion.
2004-11-14 19:35:35 +00:00
Neil Conway a236dd9536 There is no need for ReadBuffer() call sites to check that the returned
buffer is valid, as ReadBuffer() will elog on error. Most of the call
sites of ReadBuffer() got this right, but this patch fixes those call
sites that did not.
2004-11-14 02:04:14 +00:00
Bruce Momjian 1c5125adf3 Not entirely sure when this broke, but libpq's VC++ build is currently
lacking pqsignal which is now required. This was found and fixed for
VC++ by Shachar Shemesh, I simply duplicated the fix for the Borland
makefile (untested, as I don't have that compiler).

Dave Page
2004-11-13 03:18:50 +00:00
Tom Lane c7866f6645 Fix obsolete comments. 2004-11-12 20:08:40 +00:00
Bruce Momjian 03b12ef987 > I think in addition the system global name "sharemem.1" should be made more
> pg specific, like "PostgreSQL.1". I have not done this since a new compile
> would not detect a running old beta. But now would be the time (or never).

Zeugswetter Andreas
2004-11-12 17:59:42 +00:00
Tom Lane 5666485aa8 Remember to close the file on failure (pretty much redundant, really,
since this path will lead to postmaster exit anyway...)
2004-11-12 00:08:23 +00:00
Tom Lane 664f93ce18 Un-break custom_variable_classes kluge ... mea culpa. 2004-11-11 23:45:13 +00:00
Bruce Momjian fc402fdf2d pg_restore --help improvements, from Tom Lane:
!       printf(_("  -d, --dbname=NAME        connect to database name\n"));

!       printf(_("  -C, --create             create the target database\n"));
2004-11-11 17:06:46 +00:00
Neil Conway 4d0f669f3c Remove obsolete comment from btbuild() and hashbuild(): we no longer use
a global variable to control building indexes.
2004-11-11 00:32:50 +00:00
Michael Meskes 67e1ae9b03 List VALUE as a keyword. 2004-11-10 13:48:10 +00:00
Tom Lane 9a633cbb6c Allow planner to fold "stable" functions to constants when forming
selectivity estimates, per recent discussion.
2004-11-09 21:42:53 +00:00
Tom Lane 8f6278d907 Put in place some defenses against being fooled by accidental match of
shared memory segment ID.  If we can't access the existing shmem segment,
it must not be relevant to our data directory.  If we can access it,
then attach to it and check for an actual match to the data directory.
This should avoid some cases of failure-to-restart-after-boot without
introducing any significant risk of failing to detect a still-running
old backend.
2004-11-09 21:30:18 +00:00
Peter Eisentraut e9c05281b5 Get rid of perror(), substitute some better phrased error messages.
malloc() doesn't set errno, so most uses were buggy anyway.
2004-11-09 15:57:57 +00:00
Peter Eisentraut 1813d94664 Translation updates 2004-11-09 14:46:37 +00:00
Peter Eisentraut 7d8578a0c2 Clarify some strings 2004-11-09 14:39:44 +00:00
Peter Eisentraut ea182a9ef4 Translation update 2004-11-09 14:38:48 +00:00
Peter Eisentraut 3c093ff151 Clarify some error messages 2004-11-09 13:01:27 +00:00
Peter Eisentraut 5c398e6e70 Translation update 2004-11-09 13:00:05 +00:00
Neil Conway f5ef59cf9c Delete the `remove-old-headers' target, as it is no longer useful. Patch
from Alvaro Herrera.
2004-11-09 06:23:50 +00:00
Tom Lane 547bb4a7f2 Use a hopefully-more-reliable method of detecting default selectivity
estimates when combining the estimates for a range query.  As pointed out
by Miquel van Smoorenburg, the existing check for an impossible combined
result would quite possibly fail to detect one default and one non-default
input.  It seems better to use the default range query estimate in such
cases.  To do so, add a check for an estimate of exactly DEFAULT_INEQ_SEL.
This is a bit ugly because it introduces additional coupling between
clauselist_selectivity and scalarltsel/scalargtsel, but it's not like
there wasn't plenty already...
2004-11-09 00:34:46 +00:00
Bruce Momjian 9d95cdcb07 Update workding on why we use is_absolute_path() as a macro. 2004-11-08 16:34:23 +00:00
Bruce Momjian 0d3b8e9a50 Allow win32/cygwin link against the first matching library symbol rather
than erroring out.  This is the Unix behavior.
2004-11-08 05:23:26 +00:00
Tom Lane 3a372d61d0 Kris Jurka pointed out that the qualified_name production wasn't
working as intended --- for some reason, FROM a.b.c was getting
parsed as if it were a function name and not a qualified name.
I think there must be a bug in bison, because it should have
complained that the grammar was ambiguous.  Anyway, fix it along
the same lines previously used for func_name vs columnref, and get
rid of the right-recursion in attrs that seems to have confused
bison.
2004-11-08 04:02:20 +00:00
Tom Lane 0d1ca2a474 Fix unportable code in SockAddr_cidr_mask: you can't assume that
shifting left by full word width gives zero.  Per bug report from
Tyson Thomson.
2004-11-08 01:54:40 +00:00
Peter Eisentraut c22b4fa386 New translation 2004-11-07 23:38:19 +00:00
Peter Eisentraut 22aef89cbf Translation update 2004-11-07 23:29:35 +00:00
Peter Eisentraut 58b33f2e85 Translation updates 2004-11-07 23:28:20 +00:00
Bruce Momjian 33be9d359a Removes duplicate slashes from the path in canonicalize_path(). It
preserve double leading slashes on Win32.

e.g.    ////a////b => /a/b
2004-11-07 02:12:17 +00:00
Tom Lane 336969e490 Add code to find_my_exec() to resolve a symbolic link down to the
actual executable location.  This allows people to continue to use
setups where, eg, postmaster is symlinked from a convenient place.
Per gripe from Josh Berkus.
2004-11-06 23:06:29 +00:00
Tom Lane bffe45a72c Simplify relative-path logic, on the assumption that compiled-in paths
won't have trivial differences in their common prefix.
2004-11-06 21:39:45 +00:00
Tom Lane 04baa0ebf9 Update pg_dump to use SET DEFAULT_TABLESPACE instead of explicit
tablespace clauses; this should improve compatibility of dump files.
Philip Warner, some rework by Tom Lane.
2004-11-06 19:36:02 +00:00
Bruce Momjian 1d13bed735 Remove directory from cmd.exe default shell. 2004-11-06 17:56:40 +00:00
Tom Lane f245c4eb1a When implementing a coercion to a domain type with a combined
type-and-length coercion function, make sure that the coercion function
is told the correct typmod.  Fixes Kris Jurka's example of a domain
over bit(N).
2004-11-06 17:46:38 +00:00
Bruce Momjian fa00650d8a Now that we advertize only CIDR address in pg_hba.conf, remove
duplicates sample entries from comments.
2004-11-06 05:32:05 +00:00
Bruce Momjian 86100de856 Variable initializer no longer necessary. 2004-11-06 05:20:41 +00:00
Bruce Momjian 7ba377431e Tests for COMSPEC in the right place and supply a reasonable shell
default on Win32.
2004-11-06 04:29:40 +00:00
Tom Lane d2f6c3e56c Fix sizeof(EXE) arithmetic, per Andrew Dunstan. 2004-11-06 04:24:14 +00:00
Tom Lane 4867afef7a Code cleanup in path.c and exec.c. Handle Windows drive and network specs
everywhere not just some places, get rid of . and .. when joining path
sections together.  This should eliminate most of the ugly paths like
/foo/bar/./baz that we've been generating.
2004-11-06 01:16:22 +00:00
Tom Lane 3d6e538edf pred_test() logic was being too narrow-minded about where it might find
RestrictInfo nodes in the query expression.  Per example from James Robinson.
2004-11-05 20:45:10 +00:00
Tom Lane 98e8b48053 Create 'default_tablespace' GUC variable that supplies a TABLESPACE
clause implicitly whenever one is not given explicitly.  Remove concept
of a schema having an associated tablespace, and simplify the rules for
selecting a default tablespace for a table or index.  It's now just
(a) explicit TABLESPACE clause; (b) default_tablespace if that's not an
empty string; (c) database's default.  This will allow pg_dump to use
SET commands instead of tablespace clauses to determine object locations
(but I didn't actually make it do so).  All per recent discussions.
2004-11-05 19:17:13 +00:00
Peter Eisentraut 0ed3c7665e Small message clarifications 2004-11-05 17:11:34 +00:00
Peter Eisentraut a23db90ce0 Translation update 2004-11-05 17:08:11 +00:00
Bruce Momjian 54691d414d Fix psql \e and \! for Win32. 2004-11-04 22:25:14 +00:00
Tom Lane e48b9b5570 Minor documentation updates from Simon Riggs. 2004-11-04 19:08:42 +00:00
Neil Conway f4127feacf Don't create $DESTDIR/include/internal/lib, as it is no longer used. 2004-11-03 07:38:08 +00:00
Peter Eisentraut d0066b9fa4 Tab completion for CREATE TABLESPACE arguments. Patch by Joachim Wieland. 2004-11-02 16:10:05 +00:00
Peter Eisentraut bb2d1ce8f1 Translation updates 2004-11-02 09:28:41 +00:00
Peter Eisentraut 005b1b5fc8 Translation updates 2004-11-02 09:20:22 +00:00
Bruce Momjian 0c3663b47d Add comment to postmaster.c that get_progname() will call exit if it
can't strdup().
2004-11-02 03:34:50 +00:00
Bruce Momjian b0830e39c3 Add comment about memory failure of get_progname exiting the postmaster:
exit(1);    /* This could exit the postmaster */
2004-11-02 03:09:06 +00:00
Tom Lane c4acbb843b timestamptz_trunc() should only recalculate the timezone when truncating
to DAY precision or coarser; leave the timezone alone when precision is
HOUR or less.  This avoids surprises for inputs near a DST transition
time, as per example from Matthew Gabeler-Lee.  (The only reason we
recalculate at all is so that outputs that are supposed to represent
days will come out as local midnight, and that's not relevant for sub-day
precision.)
2004-11-01 22:00:30 +00:00
Tom Lane 5ba04cd9f1 Invent pg_next_dst_boundary() and rewrite DetermineLocalTimeZone() to
use it, as per my proposal of yesterday.  This gives us a means of
determining the zone offset to impute to an unlabeled timestamp that
is both efficient and reliable, unlike all our previous tries involving
mktime() and localtime().  The behavior for invalid or ambiguous times
at a DST transition is fixed to be really and truly "assume standard
time", fixing a bug that has come and gone repeatedly but was back
again in 7.4.  (There is some ongoing discussion about whether we should
raise an error instead, but for the moment I'll make it do what it was
previously intended to do.)
2004-11-01 21:34:44 +00:00
Bruce Momjian 7fad5fff0d That's just the lovely way windows handles a "segfault". Gotta be really
interesting for MS to catch all those dumps...

Anyway. Oops. Seems I ran my regression tests with the old psql, and
just managed to update the backend, when I tested that patch. Turns out
there are codepaths where we'd access the Critical Section before it was
initialized. Attached patch breaks the initializeation off to a separate
part and adds that one to a much earlier position in the program.

Magnus Hagander
2004-11-01 19:21:50 +00:00
Bruce Momjian 9c3d654a16 Update comment to point to proper file. 2004-11-01 14:33:10 +00:00
Dave Cramer 7914deecb8 fixed bug in segfault REVOKE statement 2004-11-01 13:17:12 +00:00
Bruce Momjian 584bfe3d70 Adjust use of sizeof(EXE). 2004-11-01 04:46:56 +00:00
Bruce Momjian b2b585fb2d Clarify coding of .exe patch 2004-11-01 04:25:18 +00:00
Tom Lane e837e4be8f Use psql -q option instead of 2>/dev/null during first invocation
of psql; this should make it easier to diagnose client-side problems,
such as library version mismatch.  Also, consistently use -X option
to avoid problems from weird .psqlrc settings.
2004-10-31 19:14:16 +00:00
Tom Lane 44e8a968e3 Invent a new, more thread-safe version of PQrequestCancel, called PQcancel.
Use this new function in psql.  Implement query cancellation in psql for
Windows.  Code by Magnus Hagander, documentation and minor editorialization
by Tom Lane.
2004-10-30 23:11:27 +00:00
Tom Lane 80559fa9e9 I found a corner case in which it is possible for RI_FKey_check's call
of HeapTupleSatisfiesItself() to trigger a hint-bit update on the tuple:
if the row was updated or deleted by a subtransaction of my own transaction
that was later rolled back.  This cannot occur in pre-8.0 of course, so
the hint-bit patch applied a couple weeks ago is OK for existing releases.
But for 8.0 it seems we had better fix things so that RI_FKey_check can
pass the correct buffer number to HeapTupleSatisfiesItself.  Accordingly,
add fields to the TriggerData struct to carry the buffer ID(s) for the
old and new tuple(s).  There are other possible solutions but this one
seems cleanest; it will allow other AFTER-trigger functions to safely
do tqual.c calls if they want to.  Put new fields at end of struct so
that there is no API breakage.
2004-10-30 20:53:06 +00:00
Tom Lane 88868d4fbc Change COMMIT back to the old behavior of emitting command tag COMMIT,
not ROLLBACK, for the case of COMMIT outside a transaction block.
Alvaro Herrera
2004-10-30 20:44:43 +00:00
Peter Eisentraut e5ac8db035 Translation update 2004-10-30 08:23:32 +00:00
Tom Lane 23f264d125 Rearrange order of pre-commit operations: must close cursors before doing
ON COMMIT actions.  Per bug report from Michael Guerin.
2004-10-29 22:19:53 +00:00
Bruce Momjian 4eb1165ff3 No need to set errorMessage length --- already set in goto target. 2004-10-29 19:30:02 +00:00
Tom Lane f05cfd2c73 Fix failure to think clearly about encoding conversion errors in COPY.
We can't regurgitate the unconverted string as I first thought, because
the elog.c mechanisms will assume the error message data is in the server
encoding and attempt a reverse conversion.  Eventually it might be worth
providing a short-circuit path to support this, but for now the simplest
solution is to abandon trying to report back the line contents after a
conversion failure.  Per bug report from Sil Lee, 27-Oct-2004.
2004-10-29 19:18:22 +00:00
Bruce Momjian 08510856a4 Set errorMessage length in PQrequestCancel() in all places. 2004-10-29 17:53:13 +00:00
Tom Lane ee69be44d5 Add DEBUG1-level logging of checkpoint start and end. Also, reduce the
'recycled log files' and 'removed log files' messages from DEBUG1 to
DEBUG2, replacing them with a count of files added/removed/recycled in
the checkpoint end message, as per suggestion from Simon Riggs.
2004-10-29 00:16:08 +00:00
Tom Lane 5a5f34e022 Code cleanup in dirmod.c. Andrew Dunstan, some further mods by moi. 2004-10-28 22:09:31 +00:00
Tom Lane 319902dc8c Fix to_number for the case of a trailing S.
Karel Zak
2004-10-28 18:55:08 +00:00
Peter Eisentraut 5fef3c6ef0 Translation update 2004-10-28 09:38:27 +00:00
Peter Eisentraut 94c502f1db Translation updates 2004-10-28 09:01:06 +00:00
Peter Eisentraut fc4cfc366d Translation updates 2004-10-28 08:54:09 +00:00
Neil Conway 6f1b3cf19c Use AllocateFile(), FreeFile() and palloc() rather than fopen(), fclose()
and malloc() in pgstat.c, respectively. This simplifies error recovery,
as well as being more consistent with the rest of the backend.
2004-10-28 01:38:41 +00:00
Tom Lane e6f9bf9b7f On Windows, force a checkpoint just before dropping a database's physical
files and directories.  This ensures that the bgwriter will close any open
file references it is holding for files therein, which is needed for the
rmdir() to succeed.  Andrew Dunstan and Tom Lane.
2004-10-28 00:39:59 +00:00
Bruce Momjian a7049b52e3 Have pg_ctl status always output to stdout, even stopped status, not stderr. 2004-10-27 19:44:14 +00:00
Bruce Momjian 43dc050525 Fix Cygwin compile for timezone. 2004-10-27 19:14:07 +00:00
Tom Lane b2b0673e4b When displaying a Var that is a reference to a column of an unnamed join,
try to display it as a reference to the underlying column instead.  This
is a legitimate substitution (it wouldn't be for a named join) and it
fixes some cases where the display would otherwise be ambiguous.  Per
example from Sim Zacks.
2004-10-27 18:09:41 +00:00
Bruce Momjian 3fe704209a Canonicalize Win32 path coming in from pg_ctl -D, idea from Magnus. 2004-10-27 17:17:09 +00:00
Peter Eisentraut 118bd91809 Translation update 2004-10-27 11:52:28 +00:00
Neil Conway 3bc89ac9eb Add a missing dependency: the "install" target requires "all" to have
been built before it. Per report from Kris Jurka.
2004-10-27 11:51:40 +00:00
Peter Eisentraut 2514e904da New translation 2004-10-27 11:44:09 +00:00
Peter Eisentraut ad6b3975c9 Translation updates 2004-10-27 11:22:18 +00:00
Peter Eisentraut 2db0bde40e Translation update 2004-10-27 11:13:29 +00:00
Peter Eisentraut cd92d85d11 Translation updates 2004-10-27 11:09:33 +00:00
Peter Eisentraut 979b0be4c3 Translation updates 2004-10-27 10:22:47 +00:00
Peter Eisentraut eb7fc3eb1f Translation updates 2004-10-27 10:13:49 +00:00
Neil Conway 28d8ff7948 Trivial fix: clarify a comment. 2004-10-27 07:26:56 +00:00
Bruce Momjian b69170481e Add reminder to update copyright in sgml file. 2004-10-26 23:42:49 +00:00
Tom Lane 83cd2d8b0f Make heap_fetch API more consistent by having the buffer remain pinned
in all cases when keep_buf = true.  This allows ANALYZE's inner loop to
use heap_release_fetch, which saves multiple buffer lookups for the same
page and avoids overestimation of cost by the vacuum cost mechanism.
2004-10-26 16:05:03 +00:00
Tom Lane 2c66dcf684 In the new dispensation where REINDEX doesn't take exclusive lock on
the parent table, it's essential that all index accesses take some kind
of lock on the index.  I had missed vacuumlazy.c :-( ...
2004-10-25 15:42:02 +00:00
Neil Conway 1d6b096941 Update copyright information in documentation. Also, add a note
to RELEASE_CHANGES so that this is not forgotten in the future.
2004-10-25 06:40:39 +00:00
Neil Conway ea7f2f6a58 Remove three unnecessary casts from a pointer type to char * when calling
pfree().
2004-10-25 06:27:21 +00:00
Neil Conway 86bcfc788b Fix tyop in comment. 2004-10-25 03:23:02 +00:00
Tom Lane 02ca529dd8 Make error message more verbose, in hopes of avoiding misunderstandings
such as bug #1293.
2004-10-25 03:08:29 +00:00
Neil Conway 8ec05b28b7 Modify hash_create() to elog(ERROR) if an error occurs, rather than
returning a NULL pointer (some callers remembered to check the return
value, but some did not -- it is safer to just bail out).

Also, cleanup pgstat.c to use elog(ERROR) rather than elog(LOG) followed
by exit().
2004-10-25 00:46:43 +00:00
Tom Lane 549a4bd6fe Adjust message to not be misleading on Windows.
Andrew Dunstan
2004-10-24 22:09:33 +00:00
Tom Lane 39fccf0277 On Windows, cause get_progname to strip any .EXE suffix.
Andrew Dunstan
2004-10-24 22:08:19 +00:00
Tom Lane 85b2facdd8 Write config files as text not binary, per Magnus Hagander. 2004-10-24 15:55:29 +00:00
Tom Lane 15db03181a Sync timezone data with latest zic database (dated Oct 11 2004). 2004-10-24 15:09:57 +00:00
Tom Lane 261f184f0c Update RELEASE_CHANGES to mention updating the timezone database as
a routine part of release prep.
2004-10-24 15:01:54 +00:00
Dennis Bjorklund 1b9c10275a Translation update 2004-10-24 14:55:04 +00:00
Dennis Bjorklund 8b9aaea888 Translation update 2004-10-23 14:48:03 +00:00
Dennis Bjorklund fea9cf6231 Translation updates 2004-10-23 07:29:27 +00:00
Dennis Bjorklund 3b15cc9129 Translation update 2004-10-23 06:48:59 +00:00
Dennis Bjorklund f4eaed6bd9 Translation updates 2004-10-23 06:40:24 +00:00
Dennis Bjorklund 0b7705fff8 Translation updates 2004-10-23 06:32:53 +00:00
Tom Lane 529db99c6e Avoid overflow in cost_sort when work_mem exceeds 1Gb. 2004-10-23 00:05:27 +00:00
Tom Lane 6e3cc2029d Fix tsearch build problems.
Magnus Hagander
2004-10-22 22:33:58 +00:00
Tom Lane 521408a56f Don't use LC_MESSAGES value on WIN32, since it doesn't work.
Magnus Hagander
2004-10-22 22:30:57 +00:00
Tom Lane 9e83d73b6c Add a GUC_SUPERUSER_ONLY flag to mark GUC variables that should not be
examinable by non-superusers, and use it to protect the recently-added
GUC variables for data directory and config files.  For now I have only
flagged those variables that could be used to deduce something about
the server's filesystem layout, but possibly we should also mark vars
related to logging settings and other admin-only information?
2004-10-22 19:48:19 +00:00