OSDN Git Service

pg-rex/syncrep.git
24 years agoUpdate sco faq.
Bruce Momjian [Wed, 3 May 2000 03:45:55 +0000 (03:45 +0000)]
Update sco faq.

24 years agoUpdate FAQ.
Bruce Momjian [Tue, 2 May 2000 21:57:59 +0000 (21:57 +0000)]
Update FAQ.

24 years agoRemove derived files from CVS. Now kept in
Thomas G. Lockhart [Tue, 2 May 2000 21:08:36 +0000 (21:08 +0000)]
Remove derived files from CVS. Now kept in
 hub.org:/home/projects/pgsql/ftp/www/html/users-lounge/7.0/docs/

24 years agoImprove section heading for date/time history.
Thomas G. Lockhart [Tue, 2 May 2000 20:36:21 +0000 (20:36 +0000)]
Improve section heading for date/time history.

24 years agoFixups in content and markup for 7.0 release.
Thomas G. Lockhart [Tue, 2 May 2000 20:02:03 +0000 (20:02 +0000)]
Fixups in content and markup for 7.0 release.

24 years agoUpdate HISTORY/sgml for new pgaccess version.
Bruce Momjian [Tue, 2 May 2000 17:06:10 +0000 (17:06 +0000)]
Update HISTORY/sgml for new pgaccess version.

24 years agoUpdate SCO FAQ. Billy G. Allie
Bruce Momjian [Tue, 2 May 2000 10:57:11 +0000 (10:57 +0000)]
Update SCO FAQ. Billy G. Allie

24 years agoModify getdatabaseencoding(), pg_encoding_to_char()
Tatsuo Ishii [Tue, 2 May 2000 08:13:08 +0000 (08:13 +0000)]
Modify getdatabaseencoding(), pg_encoding_to_char()
pg_char_to_encoding() in multibyte disbaled case so that it does not
throw an error, rather return HARD CODED default value (currently SQL_ASCII).
This would solve the "non-mb backend vs. mb-enabled frontend" problem.

24 years agoReset CurrentMemoryContext to TopMemoryContext at the beginning of error
Tom Lane [Sun, 30 Apr 2000 21:29:23 +0000 (21:29 +0000)]
Reset CurrentMemoryContext to TopMemoryContext at the beginning of error
cleanup, ie, as soon as we have caught the longjmp.  This ensures that
current context will be a valid context throughout error cleanup.  Before
it was possible that current context was pointing at a context that would
get deleted during cleanup, leaving any subsequent pallocs in deep
trouble.  I was able to provoke an Assert failure when compiled with
asserts + -DCLOBBER_FREED_MEMORY, if I did something that would cause
an error to be reported by the backend large-object code, because indeed
that code operates in a context that gets deleted partway through xact
abort --- and CurrentMemoryContext was still pointing at it!  Boo hiss.

24 years agoRemove bogus 'xid loop detected' check, which actually wasn't detecting
Tom Lane [Sun, 30 Apr 2000 21:23:31 +0000 (21:23 +0000)]
Remove bogus 'xid loop detected' check, which actually wasn't detecting
loops, but just arbitrarily failing at 1000 locks.

24 years agoClean up ecpg test files.
Bruce Momjian [Sat, 29 Apr 2000 03:25:24 +0000 (03:25 +0000)]
Clean up ecpg test files.

24 years agoUpdate TODO list.
Bruce Momjian [Sat, 29 Apr 2000 02:27:21 +0000 (02:27 +0000)]
Update TODO list.

24 years agoUpdate pgeasy examples
Bruce Momjian [Fri, 28 Apr 2000 22:49:25 +0000 (22:49 +0000)]
Update pgeasy examples

24 years agoUpdate TODO list.
Bruce Momjian [Fri, 28 Apr 2000 19:40:00 +0000 (19:40 +0000)]
Update TODO list.

24 years agoUpdate TODO list.
Bruce Momjian [Fri, 28 Apr 2000 19:15:49 +0000 (19:15 +0000)]
Update TODO list.

24 years agoUpdate libpgeasy define.
Bruce Momjian [Fri, 28 Apr 2000 17:19:42 +0000 (17:19 +0000)]
Update libpgeasy define.

24 years agoMore libpgeasy update of connectdb() parameter ordering
Bruce Momjian [Fri, 28 Apr 2000 15:03:39 +0000 (15:03 +0000)]
More libpgeasy update of connectdb() parameter ordering

24 years agoChange libpgeasy to take dbname at end like all other interfaces.
Bruce Momjian [Fri, 28 Apr 2000 14:58:51 +0000 (14:58 +0000)]
Change libpgeasy to take dbname at end like all other interfaces.

24 years agoUpdate TODO list.
Bruce Momjian [Fri, 28 Apr 2000 14:44:47 +0000 (14:44 +0000)]
Update TODO list.

24 years agoSetting statistic options from SET PG_OPTIONS caused a backend crash
Tom Lane [Fri, 28 Apr 2000 05:07:34 +0000 (05:07 +0000)]
Setting statistic options from SET PG_OPTIONS caused a backend crash
because StatFp never got set in that case.  Set it immediately before
use to eliminate such problems.

24 years agoplpgsql RAISE statement was careless about the possibility of a NULL
Tom Lane [Fri, 28 Apr 2000 00:12:44 +0000 (00:12 +0000)]
plpgsql RAISE statement was careless about the possibility of a NULL
field value being displayed; produced coredump instead of the expected
<NULL> display.

24 years agoExcept_Intersect_Rewrite() failed to ignore resjunk targetlist entries,
Tom Lane [Thu, 27 Apr 2000 20:32:41 +0000 (20:32 +0000)]
Except_Intersect_Rewrite() failed to ignore resjunk targetlist entries,
thus causing failure if one sub-select had resjunk entries that the other
did not (cf. bug report from Espinosa 4/27/00).

24 years agoUpdate libpgeasy e-mail address
Bruce Momjian [Thu, 27 Apr 2000 18:45:43 +0000 (18:45 +0000)]
Update libpgeasy e-mail address

24 years agoRepair problem noted by Elphick: make_rels_by_joins failed to handle
Tom Lane [Thu, 27 Apr 2000 18:35:04 +0000 (18:35 +0000)]
Repair problem noted by Elphick: make_rels_by_joins failed to handle
cases where joinclauses were present but some joins have to be made
by cartesian-product join anyway.  An example is
SELECT * FROM a,b,c WHERE (a.f1 + b.f2 + c.f3) = 0;
Even though all the rels have joinclauses, we must join two of them
in cartesian style before we can use the join clause...

24 years ago-D switch to postmaster should override any PGDATA environment variable
Tom Lane [Thu, 27 Apr 2000 02:59:17 +0000 (02:59 +0000)]
-D switch to postmaster should override any PGDATA environment variable
that might be hanging about.  Now it does ... amazing nobody noticed
this before ...

24 years agoTweak outUnique to include uniqColIdx[] field in the printout. This does
Tom Lane [Wed, 26 Apr 2000 23:39:10 +0000 (23:39 +0000)]
Tweak outUnique to include uniqColIdx[] field in the printout.  This does
not cause any compatibility problems because stored rules don't contain
plan nodes --- in fact, we don't even have a readfunc for Unique nodes.

24 years agoOn HPUX, shl_load should be called with options BIND_IMMEDIATE rather
Tom Lane [Wed, 26 Apr 2000 23:35:34 +0000 (23:35 +0000)]
On HPUX, shl_load should be called with options BIND_IMMEDIATE rather
than BIND_DEFERRED.  That way, if the loaded library has unresolved
references, shl_load fails cleanly.  As we had it, shl_load would
succeed and then the dynlinker would call abort() when we try to call
into the loaded library.  abort()ing a backend is uncool.

24 years agoFix include "" to <>
Bruce Momjian [Wed, 26 Apr 2000 22:51:14 +0000 (22:51 +0000)]
Fix include "" to <>

24 years agoRemove DriverClass.java. It is generated by the compile.
Bruce Momjian [Wed, 26 Apr 2000 17:54:28 +0000 (17:54 +0000)]
Remove DriverClass.java.  It is generated by the compile.

24 years agoAdd res clear to example
Bruce Momjian [Wed, 26 Apr 2000 17:51:03 +0000 (17:51 +0000)]
Add res clear to example

24 years agoInstall Peter's Makefile.
Bruce Momjian [Wed, 26 Apr 2000 14:19:29 +0000 (14:19 +0000)]
Install Peter's Makefile.

24 years agoReinstalled revision 1.36 (looks Peter Mount installed
Jan Wieck [Wed, 26 Apr 2000 09:03:10 +0000 (09:03 +0000)]
Reinstalled revision 1.36 (looks Peter Mount installed
a new JDBC Makefile here by accident)

Jan

24 years agothird attempt
Peter Mount [Wed, 26 Apr 2000 05:50:18 +0000 (05:50 +0000)]
third attempt

24 years agoAttempt III
Peter Mount [Wed, 26 Apr 2000 05:39:32 +0000 (05:39 +0000)]
Attempt III

24 years agoAnother attempt
Peter Mount [Wed, 26 Apr 2000 05:32:01 +0000 (05:32 +0000)]
Another attempt

24 years agoUpdate libpgeasy readme.
Bruce Momjian [Wed, 26 Apr 2000 00:53:11 +0000 (00:53 +0000)]
Update libpgeasy readme.

24 years agoMake c++ examples return 0 from main().
Bruce Momjian [Tue, 25 Apr 2000 21:00:29 +0000 (21:00 +0000)]
Make c++ examples return 0 from main().

24 years agoUpdate readme for 7.0.
Bruce Momjian [Tue, 25 Apr 2000 18:43:14 +0000 (18:43 +0000)]
Update readme for 7.0.

24 years agoFix libpq example return values
Bruce Momjian [Tue, 25 Apr 2000 16:39:07 +0000 (16:39 +0000)]
Fix libpq example return values

24 years agochange reindex ERROR/NOTICE message
Hiroshi Inoue [Tue, 25 Apr 2000 10:38:38 +0000 (10:38 +0000)]
change reindex ERROR/NOTICE message

24 years agoinitdb didn't always remove temp file
Peter Eisentraut [Tue, 25 Apr 2000 08:29:02 +0000 (08:29 +0000)]
initdb didn't always remove temp file

24 years agoCheck that user-specified opclass in CREATE INDEX corresponds to operators
Tom Lane [Tue, 25 Apr 2000 02:45:54 +0000 (02:45 +0000)]
Check that user-specified opclass in CREATE INDEX corresponds to operators
that will actually work on the column datatype.

24 years agoUpdate pg_ctl so that it does not redirect outputs from postmaster
Tatsuo Ishii [Tue, 25 Apr 2000 01:07:23 +0000 (01:07 +0000)]
Update pg_ctl so that it does not redirect outputs from postmaster
to a temp file.

24 years agoUpdate SCO FAQ>
Bruce Momjian [Mon, 24 Apr 2000 23:33:07 +0000 (23:33 +0000)]
Update SCO FAQ>

24 years agoFix \h to not go past array bounds
Bruce Momjian [Mon, 24 Apr 2000 22:41:43 +0000 (22:41 +0000)]
Fix \h to not go past array bounds

24 years agoAdd mention of int in variable.
Bruce Momjian [Sun, 23 Apr 2000 11:42:51 +0000 (11:42 +0000)]
Add mention of int in variable.

24 years agoAdd mention of int for variable in examples
Bruce Momjian [Sun, 23 Apr 2000 11:39:08 +0000 (11:39 +0000)]
Add mention of int for variable in examples

24 years agoOur test to see if we had permission to install into Perl5 install area
Tom Lane [Sun, 23 Apr 2000 04:26:32 +0000 (04:26 +0000)]
Our test to see if we had permission to install into Perl5 install area
always failed if Perl makefile's INSTALLSITELIB variable was specified
in terms of another variable.  Fix by adding an echo-installdir target
to the Perl makefile, which the upper-level Makefile can invoke.

24 years agoUpdate obsolete info in CREATE INDEX ref page. (I had fixed the text
Tom Lane [Sun, 23 Apr 2000 02:08:33 +0000 (02:08 +0000)]
Update obsolete info in CREATE INDEX ref page.  (I had fixed the text
description in indices.sgml, but missed the near-duplicate prose in
the reference page...)

24 years agoProduce an appropriate error message when opclass is not supported by
Tom Lane [Sun, 23 Apr 2000 01:44:55 +0000 (01:44 +0000)]
Produce an appropriate error message when opclass is not supported by
specified index access method.  Clean up wording of some existing error
messages, too.

24 years agoList of available pg_options items was a tad out-of-date.
Tom Lane [Sun, 23 Apr 2000 00:25:06 +0000 (00:25 +0000)]
List of available pg_options items was a tad out-of-date.

24 years ago7.0 buffer manager can support different backends running with different
Tom Lane [Sun, 23 Apr 2000 00:13:16 +0000 (00:13 +0000)]
7.0 buffer manager can support different backends running with different
fsync settings, so the -F option no longer needs to be treated as secure.

24 years agoRemove broken tracing code (which would be dangerous if it did work...)
Tom Lane [Sat, 22 Apr 2000 22:39:15 +0000 (22:39 +0000)]
Remove broken tracing code (which would be dangerous if it did work...)
libpq++.h contained copies of the class declarations in the other libpq++
include files, which was bogus enough, but the declarations were not
completely in step with the real declarations.  Remove these in favor
of including the headers with #include.  Make PgConnection destructor
virtual (not absolutely necessary, but seems like a real good idea
considering the number of subclasses derived from it).  Give all classes
declared private copy constructors and assignment operators, to prevent
compiler from thinking it can copy these objects safely.

24 years agoUpdate example: PgLargeObject constructor now takes a conninfo string,
Tom Lane [Sat, 22 Apr 2000 22:15:48 +0000 (22:15 +0000)]
Update example: PgLargeObject constructor now takes a conninfo string,
not a bare database name.

24 years agoUpdate TODO list.
Bruce Momjian [Sat, 22 Apr 2000 20:33:54 +0000 (20:33 +0000)]
Update TODO list.

24 years agopltcl didn't work well at all when Tcl had been built with a different
Tom Lane [Fri, 21 Apr 2000 03:28:17 +0000 (03:28 +0000)]
pltcl didn't work well at all when Tcl had been built with a different
compiler than the one selected to build Postgres with.  It was trying
to feed Postgres-compiler switches to Tcl's compiler.  (Seen this before
with the perl5 interface...) Fix to use only CFLAGS taken from Tcl's
configure information, plus -I which is pretty universal.

24 years agoFinally figured out that HP's cpp won't do ANSI preprocessing constructs
Tom Lane [Fri, 21 Apr 2000 03:07:51 +0000 (03:07 +0000)]
Finally figured out that HP's cpp won't do ANSI preprocessing constructs
unless you feed it -Aa or -Ae switch.  Autoconf does not know about this,
but we can fix it in the hpux_cc template file.  I knew templates were
good for something ;-)

24 years agoFix still more static-declaration-vs-nonstatic-definition glitches.
Tom Lane [Fri, 21 Apr 2000 03:01:54 +0000 (03:01 +0000)]
Fix still more static-declaration-vs-nonstatic-definition glitches.
gcc doesn't think these are a problem, but somewhere out there is a
compiler that will spit up.

24 years agoClean up const-vs-not-const compiler warning in MULTIBYTE code.
Tom Lane [Thu, 20 Apr 2000 22:40:18 +0000 (22:40 +0000)]
Clean up const-vs-not-const compiler warning in MULTIBYTE code.
'Twas my fault, I think.

24 years agoAllow libpq++ compile failure to stop entire compile.
Bruce Momjian [Thu, 20 Apr 2000 17:21:46 +0000 (17:21 +0000)]
Allow libpq++ compile failure to stop entire compile.

24 years agoAdd a regress test case for SELECT count(*) FROM view, so that we'll
Tom Lane [Thu, 20 Apr 2000 00:32:57 +0000 (00:32 +0000)]
Add a regress test case for SELECT count(*) FROM view, so that we'll
know if that case ever breaks again...

24 years agoCorrect error in rewriter that caused SELECT count(*) FROM view
Tom Lane [Thu, 20 Apr 2000 00:31:49 +0000 (00:31 +0000)]
Correct error in rewriter that caused SELECT count(*) FROM view
to give wrong results: it should be looking at inJoinSet not inFromCl.
Also, make 'modified' flag be local to ApplyRetrieveRule: we should
append a rule's quals to the query iff that particular rule applies,
not if we have fired any previously-considered rule for the query!

24 years agolibpq++.sgml fixes from Tom Vijlbrief
Bruce Momjian [Wed, 19 Apr 2000 21:21:38 +0000 (21:21 +0000)]
libpq++.sgml fixes from  Tom Vijlbrief

24 years agoUpdate TODO list.
Bruce Momjian [Wed, 19 Apr 2000 20:50:34 +0000 (20:50 +0000)]
Update TODO list.

24 years agoUpdate TODO list.
Bruce Momjian [Wed, 19 Apr 2000 00:03:44 +0000 (00:03 +0000)]
Update TODO list.

24 years agoMore cleanup
Bruce Momjian [Tue, 18 Apr 2000 23:44:58 +0000 (23:44 +0000)]
More cleanup

24 years agoRename binary object to binary format.
Bruce Momjian [Tue, 18 Apr 2000 23:43:24 +0000 (23:43 +0000)]
Rename binary object to binary format.

24 years agoUpdate COPY manual page for \N.
Bruce Momjian [Tue, 18 Apr 2000 23:21:04 +0000 (23:21 +0000)]
Update COPY manual page for \N.

24 years agoRemove old TODO items. Add some to release.sgml.
Bruce Momjian [Tue, 18 Apr 2000 21:18:43 +0000 (21:18 +0000)]
Remove old TODO items.  Add some to release.sgml.

24 years agoFix \N mention in manual
Bruce Momjian [Tue, 18 Apr 2000 18:41:15 +0000 (18:41 +0000)]
Fix \N mention in manual

24 years agoClean up description of SET CONSTRAINTS.
Thomas G. Lockhart [Tue, 18 Apr 2000 15:23:34 +0000 (15:23 +0000)]
Clean up description of SET CONSTRAINTS.

24 years agoUpdate list of supported platforms to include SCO info.
Thomas G. Lockhart [Tue, 18 Apr 2000 15:23:10 +0000 (15:23 +0000)]
Update list of supported platforms to include SCO info.
Move several platforms to the unsupported list :(

24 years agoSecurity fix for plperl.
Bruce Momjian [Tue, 18 Apr 2000 15:04:02 +0000 (15:04 +0000)]
Security fix for plperl.

24 years agoUpdate release.sgml
Bruce Momjian [Tue, 18 Apr 2000 14:55:27 +0000 (14:55 +0000)]
Update release.sgml

24 years agolinux_alpha pattern should allow for more-detailed machine type such
Tom Lane [Tue, 18 Apr 2000 05:57:45 +0000 (05:57 +0000)]
linux_alpha pattern should allow for more-detailed machine type such
as 'alphaev5', cf report from Stepanov 13-Apr-00.

24 years agoRepair coredump seen when a view refers to an inheritance group
Tom Lane [Tue, 18 Apr 2000 05:52:35 +0000 (05:52 +0000)]
Repair coredump seen when a view refers to an inheritance group
(SELECT FROM table*).  Cause was reference to 'eref' field of an RTE,
which is null in an RTE loaded from a stored rule parsetree.  There
wasn't any good reason to be touching the refname anyway...

24 years agoCorrect oversight in hashjoin cost estimation: nodeHash sizes its hash
Tom Lane [Tue, 18 Apr 2000 05:43:02 +0000 (05:43 +0000)]
Correct oversight in hashjoin cost estimation: nodeHash sizes its hash
table for an average of NTUP_PER_BUCKET tuples/bucket, but cost_hashjoin
was assuming a target load of one tuple/bucket.  This was causing a
noticeable underestimate of hashjoin costs.

24 years agoInclude information for armv4l from Mark Knox <segfault@hardline.org>.
Thomas G. Lockhart [Tue, 18 Apr 2000 05:35:38 +0000 (05:35 +0000)]
Include information for armv4l from Mark Knox <segfault@hardline.org>.

24 years agoRevert no-longer-needed relaxation of compiler errors.
Tom Lane [Tue, 18 Apr 2000 00:27:34 +0000 (00:27 +0000)]
Revert no-longer-needed relaxation of compiler errors.

24 years agoMake ECPGraise's str parameter const to suppress warnings from gcc
Tom Lane [Tue, 18 Apr 2000 00:24:30 +0000 (00:24 +0000)]
Make ECPGraise's str parameter const to suppress warnings from gcc
and errors from pickier compilers.

24 years agoAdd comment for -qhalt=w removal so it can be re-added later
Bruce Momjian [Mon, 17 Apr 2000 22:21:07 +0000 (22:21 +0000)]
Add comment for -qhalt=w removal so it can be re-added later

24 years agoAnother attempt at 7.0
Peter Mount [Mon, 17 Apr 2000 20:07:56 +0000 (20:07 +0000)]
Another attempt at 7.0

24 years agoAix additions
Bruce Momjian [Mon, 17 Apr 2000 18:11:43 +0000 (18:11 +0000)]
Aix additions

24 years agoAix patches from Zeugswetter Andrea
Bruce Momjian [Mon, 17 Apr 2000 18:05:00 +0000 (18:05 +0000)]
Aix patches from Zeugswetter Andrea

24 years agoUpdate TODO list.
Bruce Momjian [Mon, 17 Apr 2000 16:35:23 +0000 (16:35 +0000)]
Update TODO list.

24 years agoRemove -qhalt=w flag from aix42.
Bruce Momjian [Mon, 17 Apr 2000 15:50:18 +0000 (15:50 +0000)]
Remove -qhalt=w flag from aix42.

24 years agoAdd code to test for non-shared libperl and fail gracefully.
Tom Lane [Mon, 17 Apr 2000 00:49:58 +0000 (00:49 +0000)]
Add code to test for non-shared libperl and fail gracefully.
Not a big step forward, but it's something...

24 years agosquished \dS+ bug pointed out by Mike Mascari
Peter Eisentraut [Sun, 16 Apr 2000 20:04:51 +0000 (20:04 +0000)]
squished \dS+ bug pointed out by Mike Mascari

24 years agoAdd scripts to enable/disable use of new LIKE/regexp selectivity
Tom Lane [Sun, 16 Apr 2000 18:41:30 +0000 (18:41 +0000)]
Add scripts to enable/disable use of new LIKE/regexp selectivity
estimation operators.  See contrib/likeplanning/README for info.

24 years agoTweak create_help.pl so it will work under either perl 4.* or perl 5.*.
Tom Lane [Sun, 16 Apr 2000 18:07:22 +0000 (18:07 +0000)]
Tweak create_help.pl so it will work under either perl 4.* or perl 5.*.
Remove knowledge of path to documentation source directory from perl
script, instead have Makefile pass it to script.

24 years agomore psql bug squashing:
Peter Eisentraut [Sun, 16 Apr 2000 15:46:40 +0000 (15:46 +0000)]
more psql bug squashing:
\copy without arguments failed
commands with too many arguments were too silent

24 years agoAdd new selectivity estimation functions for pattern-matching operators
Tom Lane [Sun, 16 Apr 2000 04:41:03 +0000 (04:41 +0000)]
Add new selectivity estimation functions for pattern-matching operators
(LIKE and regexp matches).  These are not yet referenced in pg_operator,
so by default the system will continue to use eqsel/neqsel.
Also, tweak convert_to_scalar() logic so that common prefixes of strings
are stripped off, allowing better accuracy when all strings in a table
share a common prefix.

24 years agoAllow COPY WITH OIDS to system OID values --- rely on unique indexes to
Tom Lane [Sun, 16 Apr 2000 04:27:52 +0000 (04:27 +0000)]
Allow COPY WITH OIDS to system OID values --- rely on unique indexes to
prevent duplicate OIDs from being added.  Clean up redundant error
messages.

24 years agoShow failing OID in 'cache lookup failed' messages.
Tom Lane [Sun, 16 Apr 2000 04:25:42 +0000 (04:25 +0000)]
Show failing OID in 'cache lookup failed' messages.

24 years agoShow failing OID in 'cache lookup failed' messages;
Tom Lane [Sun, 16 Apr 2000 04:19:41 +0000 (04:19 +0000)]
Show failing OID in 'cache lookup failed' messages;
print OIDs as %u not %d.

24 years agoShow failing OID in 'cache lookup failed' messages.
Tom Lane [Sun, 16 Apr 2000 04:16:55 +0000 (04:16 +0000)]
Show failing OID in 'cache lookup failed' messages.

24 years agoPrint OIDs as %u not %d.
Tom Lane [Sun, 16 Apr 2000 04:14:49 +0000 (04:14 +0000)]
Print OIDs as %u not %d.

24 years agoChange postmaster.opts.default.sample. remove -S option.
Tatsuo Ishii [Sun, 16 Apr 2000 03:50:00 +0000 (03:50 +0000)]
Change postopts.default.sample. remove -S option.
With the default option, pg_ctl can show messages from
postmaster upon its failure.

24 years agoget_relattval() should treat a NULL constant as a non-constant expression,
Tom Lane [Sun, 16 Apr 2000 01:55:45 +0000 (01:55 +0000)]
get_relattval() should treat a NULL constant as a non-constant expression,
since it has no way to indicate to its caller that the constant is
actually NULL.  This prevents coredump in cases like
  WHERE textfield < null::text;

24 years agoSYstem user not implemented
Bruce Momjian [Sat, 15 Apr 2000 23:29:58 +0000 (23:29 +0000)]
SYstem user not implemented