]> granicus.if.org Git - postgresql/log
postgresql
25 years agoUndo collateral damage from recent patch: Makefile had
Tom Lane [Sun, 16 May 1999 14:56:17 +0000 (14:56 +0000)]
Undo collateral damage from recent patch: Makefile had
lost most of target list and thus 'make clean' didn't clean up very well.

25 years agoChange iostream to iostream.h, strstream to strstream.h
Tom Lane [Sun, 16 May 1999 14:34:59 +0000 (14:34 +0000)]
Change iostream to iostream.h, strstream to strstream.h
for compatibility with older C++ libraries.

25 years agoI made it so it rolled over files at 1MB. My table ended up with 120
Bruce Momjian [Sat, 15 May 1999 22:31:07 +0000 (22:31 +0000)]
I made it so it rolled over files at 1MB.  My table ended up with 120
segments, and my indexes had 3(Yes, it DOES work!).
DROP TABLE removed ALL segments from the table, but only the main index
segment.

So it looks like removing the table itself is using mdunlink in md.c,
while removing indexes uses FileNameUnlink() which only unlinks 1 file.
As far as I can tell, calling FileNameUnlink() and mdunlink() is basically
the same, except mdunlink() deletes any extra segments.

I've done some testing and it seems to work.  It also passes regression
tests(except float8, geometry and rules, but that's normal).

If this patch is right, this fixes all known multi-segment problems on
Linux.

Ole Gjerde

25 years agoI've got 2 pretty small patches.
Bruce Momjian [Sat, 15 May 1999 22:18:51 +0000 (22:18 +0000)]
I've got 2 pretty small patches.

configtype.patch simply fixes a typo in config.h.in
pg_dump.c.patch Updates a bunch of error messages to include a reason
from
                the backend, and also removes a couple of unnecessary
if's

Ole Gjerde

25 years ago*** empty log message ***
Michael Meskes [Fri, 14 May 1999 06:56:18 +0000 (06:56 +0000)]
*** empty log message ***

25 years agoHi, Bruce!
Bruce Momjian [Thu, 13 May 1999 15:55:45 +0000 (15:55 +0000)]
Hi, Bruce!

These are my last changes to lmgr fixing deadlock handling.
Please apply them to cvs...

Vadim

25 years agoAdd double quotes around the sequence name generated to support the
Thomas G. Lockhart [Thu, 13 May 1999 15:01:32 +0000 (15:01 +0000)]
Add double quotes around the sequence name generated to support the
 SERIAL data type DEFAULT clause.
This fixes a problem finding the sequence name when mixed case table names
 are involved.

25 years agoSurround a variable declaration with ENABLE_OUTER_JOINS to suppress
Thomas G. Lockhart [Thu, 13 May 1999 14:59:05 +0000 (14:59 +0000)]
Surround a variable declaration with ENABLE_OUTER_JOINS to suppress
 compiler warnings about an unused variable.

25 years agoset client_encoding to <nothing> crashes backend.
Tatsuo Ishii [Thu, 13 May 1999 10:28:26 +0000 (10:28 +0000)]
set client_encoding to <nothing> crashes backend.

25 years agoRip out QueryTreeList structure, root and branch. Querytree
Tom Lane [Thu, 13 May 1999 07:29:22 +0000 (07:29 +0000)]
Rip out QueryTreeList structure, root and branch.  Querytree
lists are now plain old garden-variety Lists, allocated with palloc,
rather than specialized expansible-array data allocated with malloc.
This substantially simplifies their handling and eliminates several
sources of memory leakage.
Several basic types of erroneous queries (syntax error, attempt to
insert a duplicate key into a unique index) now demonstrably leak
zero bytes per query.

25 years agoUpdate driver to 6-40-0006
Byron Nikolaidis [Thu, 13 May 1999 03:33:00 +0000 (03:33 +0000)]
Update driver to 6-40-0006

25 years agoHere's a small patch to cause pg_dump to emit the
Bruce Momjian [Thu, 13 May 1999 02:35:44 +0000 (02:35 +0000)]
Here's a small patch to cause pg_dump to emit the
scale and precision for NUMERIC type column defs.

Keith Parks

25 years agoRelease allocated memory during AtAbort_Memory.
Tom Lane [Thu, 13 May 1999 00:34:57 +0000 (00:34 +0000)]
Release allocated memory during AtAbort_Memory.

25 years agoDo not refer to stdin in static variable initializer ...
Tom Lane [Wed, 12 May 1999 23:26:03 +0000 (23:26 +0000)]
Do not refer to stdin in static variable initializer ...
apparently some systems choke on that :-(.

25 years agoFixed small bug in ruleutils and added output of pg_views and
Jan Wieck [Wed, 12 May 1999 17:59:32 +0000 (17:59 +0000)]
Fixed small bug in ruleutils and added output of pg_views and
pg_rules to rules regression test.

Jan

25 years agoFixed wrong hasAggs when aggregate columns of view aren't
Jan Wieck [Wed, 12 May 1999 17:04:47 +0000 (17:04 +0000)]
Fixed wrong hasAggs when aggregate columns of view aren't
selected.

Disabled ability of defining DISTINCT or ORDER BY on views.

Jan

25 years agoReplaced targetlist entry in GroupClause by reference number
Jan Wieck [Wed, 12 May 1999 15:02:39 +0000 (15:02 +0000)]
Replaced targetlist entry in GroupClause by reference number
in Resdom and GroupClause so changing of resno's doesn't confuse
the grouping any more.

Jan

25 years agoI am sorry, I misinterpreted the still failing trigger regression test.
Bruce Momjian [Wed, 12 May 1999 12:47:24 +0000 (12:47 +0000)]
I am sorry, I misinterpreted the still failing trigger regression test.
The
offending code
has been removed, the action is now always dependent :-)

I suggest the following patch, to finally make trigger regression happy
again:

 <<refint1.patch>>
After that you can remove the following from TODO:
Remove ERROR:  check_primary_key: even number of arguments should be
specified
Trigger regression test fails

Andreas

25 years agoAdded installation of created procedural languages to initdb
Jan Wieck [Wed, 12 May 1999 10:35:44 +0000 (10:35 +0000)]
Added installation of created procedural languages to initdb

Jan

25 years agoMinor wording change.
Thomas G. Lockhart [Wed, 12 May 1999 07:34:02 +0000 (07:34 +0000)]
Minor wording change.

25 years agoInclude mention of CASE, COALESCE, and IFNULL.
Thomas G. Lockhart [Wed, 12 May 1999 07:32:47 +0000 (07:32 +0000)]
Include mention of CASE, COALESCE, and IFNULL.
Add date/time parsing procedure (perhaps should be in appendix).
Add time zone information (ditto).
Update keyword list.

25 years agoAdjust elog NOTICE messages to surround table and column names with single
Thomas G. Lockhart [Wed, 12 May 1999 07:24:01 +0000 (07:24 +0000)]
Adjust elog NOTICE messages to surround table and column names with single
 quotes.

25 years agoAdd keywords to implement Vadim's transaction isolation
Thomas G. Lockhart [Wed, 12 May 1999 07:22:52 +0000 (07:22 +0000)]
Add keywords to implement Vadim's transaction isolation
 and lock syntax as fully parsed tokens.
Two keywords for isolation are non-reserved SQL92
 (COMMITTED, SERIALIZABLE).
All other new keywords are non-reserved Postgres (not SQL92)
 (ACCESS, EXCLUSIVE, MODE, SHARE).
Add syntax to allow CREATE [GLOBAL|LOCAL] TEMPORARY TABLE, throwing an
 error if GLOBAL is specified.

25 years agoFix problem with multiple indices defined if using column- and table-
Thomas G. Lockhart [Wed, 12 May 1999 07:17:18 +0000 (07:17 +0000)]
Fix problem with multiple indices defined if using column- and table-
 constraints. Reported by Tom Lane.
Now, check for duplicate indices and retain the one which is a primary-key.
Adjust elog NOTICE messages to surround table and column names with single
 quotes.

25 years agoHandle conversion of floating point constants to internal strings.
Thomas G. Lockhart [Wed, 12 May 1999 07:14:24 +0000 (07:14 +0000)]
Handle conversion of floating point constants to internal strings.

25 years agoKeep long non-quoted numeric strings *as* untyped strings if they fail
Thomas G. Lockhart [Wed, 12 May 1999 07:12:51 +0000 (07:12 +0000)]
Keep long non-quoted numeric strings *as* untyped strings if they fail
 the obvious conversion.
Define a new pattern "decimal" which is non-exponential floating point
 for use with numeric() and decimal() types.

25 years agoFix bogus assumption that MAXALIGN is at least sizeof(pointer).
Tom Lane [Wed, 12 May 1999 04:38:24 +0000 (04:38 +0000)]
Fix bogus assumption that MAXALIGN is at least sizeof(pointer).

25 years agoFix configure to generate correct expansion of library directory
Tom Lane [Tue, 11 May 1999 22:57:50 +0000 (22:57 +0000)]
Fix configure to generate correct expansion of library directory
in pl/plpgsql/src/mklang.sql.

25 years agoJust a couple more files to untar in src/test/regress
Bruce Momjian [Tue, 11 May 1999 16:29:45 +0000 (16:29 +0000)]
Just a couple more files to untar in src/test/regress
(int2,int4,geometry).

Cheers,

Patrick Welche

25 years agoChanged debug options:
Jan Wieck [Tue, 11 May 1999 09:06:35 +0000 (09:06 +0000)]
Changed debug options:

-d4 now prints compressed trees from nodeToString()
-d5 prints pretty trees via nodeDisplay()

new pg_options: pretty_plan, pretty_parse, pretty_rewritten

Jan

25 years agoput back mklang.sql.in file.
Bruce Momjian [Tue, 11 May 1999 03:30:41 +0000 (03:30 +0000)]
put back mklang.sql.in file.

25 years agoclean up comments
Bruce Momjian [Tue, 11 May 1999 03:28:43 +0000 (03:28 +0000)]
clean up comments

25 years agoUpdate man page for select
Bruce Momjian [Mon, 10 May 1999 18:22:48 +0000 (18:22 +0000)]
Update man page for select

25 years agoCheck for NUMERIC overflow a second time after rounding
Jan Wieck [Mon, 10 May 1999 18:17:44 +0000 (18:17 +0000)]
Check for NUMERIC overflow a second time after rounding

Jan

25 years agoAllow perl install as non-root, from Geoff Keating
Bruce Momjian [Mon, 10 May 1999 17:01:25 +0000 (17:01 +0000)]
Allow perl install as non-root, from Geoff Keating

25 years agoUpdate to PyGreSQL 2.3.
Bruce Momjian [Mon, 10 May 1999 16:10:51 +0000 (16:10 +0000)]
Update to PyGreSQL 2.3.

25 years ago libpq++ uses fe_setauthsvc which is deprecated and results in an error
Bruce Momjian [Mon, 10 May 1999 15:27:19 +0000 (15:27 +0000)]
  libpq++ uses fe_setauthsvc which is deprecated and results in an error
on connection. This patch changes it to use PQconnectdb rather than
{fe_setauthsvc,PQsetdb}. This still isn't the complete solution, as
there
is no provision for user,password in class PgEnv, but it does get rid of
the error message. Tested with gcc version egcs-2.91.60 19981201
(egcs-1.1.1 release) under NetBSD-1.3K/i386.

Cheers,

Patrick Welche

25 years agoRename MAP_FILE to USERMAP_FILE for Digital Unix.
Bruce Momjian [Mon, 10 May 1999 15:17:17 +0000 (15:17 +0000)]
Rename MAP_FILE to USERMAP_FILE for Digital Unix.

25 years agoI send you a attach of my modified refint.c that
Bruce Momjian [Mon, 10 May 1999 15:12:29 +0000 (15:12 +0000)]
I send you a attach of my modified refint.c that
works with a new policy  in cascade mode .

Please Read README.MAX .
I do not know if you are the author of refint.c ,
but if not please tell me who is .

Thank you ( excuse me for my bad english) .
Massimo Lambertini massimo.lambertini@everex.it

25 years agoThis patch is to add more comments
Bruce Momjian [Mon, 10 May 1999 04:57:07 +0000 (04:57 +0000)]
This patch is to add more comments
to postgres.init.sh , clairify the options
available, and to add easy support
for installation of postgres into the
runlevel system.

"sh postgres.init.sh install"

Will now install "postgres" in the
/etc/rc.d/init.d directory and execute
/sbin/chkconfig to hook up the symbolic
links.  An uninstall option is also added.

Enclosed is the patch and the patched file

Clark

25 years agoadd mention of large object orphaning
Bruce Momjian [Mon, 10 May 1999 04:19:30 +0000 (04:19 +0000)]
add mention of large object orphaning

25 years agoFix initdb problem introduced by recent patch.
Bruce Momjian [Mon, 10 May 1999 04:02:07 +0000 (04:02 +0000)]
Fix initdb problem introduced by recent patch.

25 years agoChange error messages to oids come out as %u and not %d. Change has no
Bruce Momjian [Mon, 10 May 1999 00:46:32 +0000 (00:46 +0000)]
Change error messages to oids come out as %u and not %d.  Change has no
real affect now.

25 years agoCorrect type_sanity test so it doesn't spit up on new
Tom Lane [Sun, 9 May 1999 23:43:25 +0000 (23:43 +0000)]
Correct type_sanity test so it doesn't spit up on new
definition of numeric_in.

25 years agoRearrange top-level rewrite operations so that EXPLAIN works
Tom Lane [Sun, 9 May 1999 23:31:47 +0000 (23:31 +0000)]
Rearrange top-level rewrite operations so that EXPLAIN works
on queries involving UNION, EXCEPT, INTERSECT.

25 years agoRunning lo_read/lo_write under different memory context
Tatsuo Ishii [Sun, 9 May 1999 15:00:18 +0000 (15:00 +0000)]
Running lo_read/lo_write under different memory context
cause troubles. See
Message-Id: <199905090312.MAA00466@ext16.sra.co.jp>
for more details.

25 years agoMistyping by me.
Vadim B. Mikheev [Sun, 9 May 1999 14:00:29 +0000 (14:00 +0000)]
Mistyping by me.
Fixed by Hiroshi.

25 years agoRepair incorrect pg_proc entry for numeric_in.
Tom Lane [Sun, 9 May 1999 02:22:16 +0000 (02:22 +0000)]
Repair incorrect pg_proc entry for numeric_in.

25 years agoFix some miscellaneous places that were using raw open() or
Tom Lane [Sun, 9 May 1999 00:54:30 +0000 (00:54 +0000)]
Fix some miscellaneous places that were using raw open() or
fopen(), instead of going through fd.c ... naughty naughty.

25 years agoUpdate hash and join routines to use fd.c's new temp-file
Tom Lane [Sun, 9 May 1999 00:53:22 +0000 (00:53 +0000)]
Update hash and join routines to use fd.c's new temp-file
code, instead of not-very-bulletproof stuff they had before.

25 years agoAdd 'temporary file' facility to fd.c, and arrange for temp
Tom Lane [Sun, 9 May 1999 00:52:08 +0000 (00:52 +0000)]
Add 'temporary file' facility to fd.c, and arrange for temp
files to be closed automatically at transaction abort or commit, should
they still be open.  Also close any still-open stdio files allocated with
AllocateFile at abort/commit.  This should eliminate problems with leakage
of file descriptors after an error.  Also, put in some primitive buffered-IO
support so that psort.c can use virtual files without severe performance
penalties.

25 years agoAdd new netbsd file.
Bruce Momjian [Sat, 8 May 1999 19:28:20 +0000 (19:28 +0000)]
Add new netbsd file.

25 years ago*** empty log message ***
Michael Meskes [Fri, 7 May 1999 18:03:37 +0000 (18:03 +0000)]
*** empty log message ***

25 years agoRemove unused Netbsd files.
Bruce Momjian [Fri, 7 May 1999 02:46:24 +0000 (02:46 +0000)]
Remove unused Netbsd files.

25 years agoPlease apply the following patch for regress.sh to do something useful with
Bruce Momjian [Fri, 7 May 1999 02:37:08 +0000 (02:37 +0000)]
Please apply the following patch for regress.sh to do something useful with
"SYSTEM", and unpack the files in the uuencoded .tar.gz file at the end in
src/test/regress so that the int2, int4 and geometry tests pass on NetBSD/i386.
They just fail on different wording of error messages and eg printing "0"
rather than "-0". At a guess the same will be true for the other NetBSD ports,
but I can't test them.

Cheers,

Patrick

25 years agoPlease apply the following patch for regress.sh to do something useful
Bruce Momjian [Fri, 7 May 1999 02:31:43 +0000 (02:31 +0000)]
Please apply the following patch for regress.sh to do something useful
with
"SYSTEM", Patrick Welche

25 years agoFix LMGR for MVCC.
Vadim B. Mikheev [Fri, 7 May 1999 01:23:11 +0000 (01:23 +0000)]
Fix LMGR for MVCC.
Get rid of Extend lock mode.

25 years agoUpdate regress test for CASE to enable tests involving joins.
Tom Lane [Thu, 6 May 1999 23:09:30 +0000 (23:09 +0000)]
Update regress test for CASE to enable tests involving joins.

25 years agoFix oversights in flatten_tlistentry and replace_clause_joinvar_refs
Tom Lane [Thu, 6 May 1999 23:07:33 +0000 (23:07 +0000)]
Fix oversights in flatten_tlistentry and replace_clause_joinvar_refs
that led to CASE expressions not working very well in joined queries.

25 years agofix_indxqual_references didn't cope with ArrayRef nodes,
Tom Lane [Thu, 6 May 1999 01:30:58 +0000 (01:30 +0000)]
fix_indxqual_references didn't cope with ArrayRef nodes,
meaning that this failed:
select proname,typname,prosrc from pg_proc,pg_type
where proname = 'float8' and pg_proc.proargtypes[0] = pg_type.oid;

25 years agoFix some nasty coredump bugs in hashjoin. This code was just
Tom Lane [Thu, 6 May 1999 00:30:47 +0000 (00:30 +0000)]
Fix some nasty coredump bugs in hashjoin.  This code was just
about certain to fail anytime it decided the relation to be hashed was
too big to fit in memory --- the code for 'batching' a series of hashjoins
had multiple errors.  I've fixed the easier problems.  A remaining big
problem is that you can get 'hashtable out of memory' if the code's
guesstimate about how much overflow space it will need turns out wrong.
That will require much more extensive revisions to fix, so I'm committing
these fixes now before I start on that problem.

25 years agoI have two patches for 6.5.0:
Bruce Momjian [Wed, 5 May 1999 21:38:40 +0000 (21:38 +0000)]
I have two patches for 6.5.0:

arrayfuncs.patch        fixes a small bug in my previous patches for
arrays

array-regress.patch     adds _bpchar and _varchar to regression tests

--
Massimo Dal Zotto

25 years agosmall error message improvement from Dmitry Samersoff
Bruce Momjian [Wed, 5 May 1999 13:51:38 +0000 (13:51 +0000)]
small error message improvement from Dmitry Samersoff

25 years agoRepair incorrectly-figured snprintf length restriction.
Tom Lane [Tue, 4 May 1999 23:39:20 +0000 (23:39 +0000)]
Repair incorrectly-figured snprintf length restriction.

25 years agoUse sprintf() to convert float8 to a string during conversion to numeric.
Thomas G. Lockhart [Tue, 4 May 1999 15:50:24 +0000 (15:50 +0000)]
Use sprintf() to convert float8 to a string during conversion to numeric.
Original code used float8out(), but the resulting exponential notation
 was not handled (e.g. '3E9' was decoded as '3').

25 years agoAllow -t tablename to preserve case if specified as "tablename".
Thomas G. Lockhart [Tue, 4 May 1999 15:47:35 +0000 (15:47 +0000)]
Allow -t tablename to preserve case if specified as "tablename".

25 years agoFix up examples of sh/ksh PATH environment variable definition.
Thomas G. Lockhart [Tue, 4 May 1999 02:57:13 +0000 (02:57 +0000)]
Fix up examples of sh/ksh PATH environment variable definition.

25 years agoClean up markup for first useful version.
Thomas G. Lockhart [Tue, 4 May 1999 02:43:55 +0000 (02:43 +0000)]
Clean up markup for first useful version.

25 years agoInclude new introductory chapter on SQL from Stefan S.
Thomas G. Lockhart [Tue, 4 May 1999 02:26:06 +0000 (02:26 +0000)]
Include new introductory chapter on SQL from Stefan S.
Should this be in the tutorial instead?

25 years agoIndent markup using emacs.
Thomas G. Lockhart [Tue, 4 May 1999 02:24:26 +0000 (02:24 +0000)]
Indent markup using emacs.

25 years agoInclude specific info on available timezones.
Thomas G. Lockhart [Tue, 4 May 1999 02:22:13 +0000 (02:22 +0000)]
Include specific info on available timezones.
Document date/time input parsing procedure.

25 years agoInclude chapters on security and an intro to SQL.
Thomas G. Lockhart [Tue, 4 May 1999 02:19:20 +0000 (02:19 +0000)]
Include chapters on security and an intro to SQL.

25 years agoFirst version of markup which can run through jade for html.
Thomas G. Lockhart [Tue, 4 May 1999 02:18:34 +0000 (02:18 +0000)]
First version of markup which can run through jade for html.
Probably not acceptable yet. Need to check some math symbols and markup.

25 years agoInclude chapter on security.
Thomas G. Lockhart [Tue, 4 May 1999 02:16:57 +0000 (02:16 +0000)]
Include chapter on security.

25 years agoMake sure targetlist generated for subplan does not share
Tom Lane [Tue, 4 May 1999 00:00:20 +0000 (00:00 +0000)]
Make sure targetlist generated for subplan does not share
nodes with HAVING qualifier of upper plan.  Have not seen any failures,
just being a little bit paranoid...

25 years agoCorrect declaration of array_map() so that it doesn't make
Tom Lane [Mon, 3 May 1999 23:48:26 +0000 (23:48 +0000)]
Correct declaration of array_map() so that it doesn't make
gcc quite so unhappy.

25 years agohere are some patches for 6.5.0 which I already submitted but have never
Bruce Momjian [Mon, 3 May 1999 19:10:48 +0000 (19:10 +0000)]
here are some patches for 6.5.0 which I already submitted but have never
been applied. The patches are in the .tar.gz attachment at the end:

varchar-array.patch     this patch adds support for arrays of bpchar() and
                        varchar(), which where always missing from postgres.

                        These datatypes can be used to replace the _char4,
                        _char8, etc., which were dropped some time ago.

block-size.patch        this patch fixes many errors in the parser and other
                        program which happen with very large query statements
                        (> 8K) when using a page size larger than 8192.

                        This patch is needed if you want to submit queries
                        larger than 8K. Postgres supports tuples up to 32K
                        but you can't insert them because you can't submit
                        queries larger than 8K. My patch fixes this problem.

                        The patch also replaces all the occurrences of `8192'
                        and `1<<13' in the sources with the proper constants
                        defined in include files. You should now never find
                        8192 hardwired in C code, just to make code clearer.

--
Massimo Dal Zotto

25 years agoRevise union_planner and associated routines to clean up breakage
Tom Lane [Mon, 3 May 1999 00:38:44 +0000 (00:38 +0000)]
Revise union_planner and associated routines to clean up breakage
from EXCEPT/HAVING patch.  Cases involving nontrivial GROUP BY expressions
now work again.  Also, the code is at least somewhat better documented...

25 years agoClean up cost_sort some more: most callers were double-counting
Tom Lane [Sat, 1 May 1999 19:47:42 +0000 (19:47 +0000)]
Clean up cost_sort some more: most callers were double-counting
the cost of reading the source data.

25 years agoArrange for VACUUM to delete the init file that relcache.c uses
Tom Lane [Sat, 1 May 1999 19:09:46 +0000 (19:09 +0000)]
Arrange for VACUUM to delete the init file that relcache.c uses
to save a little bit of backend startup time.  This way, the first
backend started after a VACUUM will rebuild the init file with up-to-date
statistics for the critical system indexes.

25 years ago-T was omitted from getopt() call.
Tom Lane [Sat, 1 May 1999 17:16:25 +0000 (17:16 +0000)]
-T was omitted from getopt() call.

25 years agoFix compile failures in dt.c --- line broken in middle of
Tom Lane [Sat, 1 May 1999 17:14:56 +0000 (17:14 +0000)]
Fix compile failures in dt.c --- line broken in middle of
an identifier :-(.  Sloppy transmission of a patch, likely.

25 years agoPatch from "Hiroshi Inoue" <Inoue@tpf.co.jp> for
Vadim B. Mikheev [Sat, 1 May 1999 16:09:45 +0000 (16:09 +0000)]
Patch from "Hiroshi Inoue" <Inoue@tpf.co.jp> for
FATAL 1:btree: BTP_CHAIN flag was expected

25 years agoUse page-level ExtendLock lock instead of table-level -
Vadim B. Mikheev [Sat, 1 May 1999 15:04:46 +0000 (15:04 +0000)]
Use page-level ExtendLock lock instead of table-level -
should be faster.

25 years agocleanup
Bruce Momjian [Fri, 30 Apr 1999 17:03:04 +0000 (17:03 +0000)]
cleanup

25 years agoClean up typo causing compile failure.
Bruce Momjian [Fri, 30 Apr 1999 16:22:46 +0000 (16:22 +0000)]
Clean up typo causing compile failure.

25 years agoFill in reasonable-looking cost estimates in inserted nodes.
Tom Lane [Fri, 30 Apr 1999 04:04:27 +0000 (04:04 +0000)]
Fill in reasonable-looking cost estimates in inserted nodes.
This makes no difference to the optimizer, which has already decided what
it's gonna do, but it makes the output of EXPLAIN much more plausible.

25 years agoClean up some bogosities in path cost estimation, like
Tom Lane [Fri, 30 Apr 1999 04:01:44 +0000 (04:01 +0000)]
Clean up some bogosities in path cost estimation, like
sometimes estimating an index scan of a table to be cheaper than a
sequential scan of the same tuples...

25 years agoFix nasty little typo that prevented get_cheapest_path_for_joinkeys
Tom Lane [Fri, 30 Apr 1999 03:59:06 +0000 (03:59 +0000)]
Fix nasty little typo that prevented get_cheapest_path_for_joinkeys
from ever returning a path.  This put a bit of a crimp in the system's
ability to generate intelligent merge-join plans...

25 years agoPrevent priority inversion in locking by checking for existing locks
Bruce Momjian [Fri, 30 Apr 1999 02:04:51 +0000 (02:04 +0000)]
Prevent priority inversion in locking by checking for existing locks
before going into queue behind person with higher piority.

25 years agoBugfix (bug by me in 1.4) in backparsing INSERT ... SELECT
Jan Wieck [Thu, 29 Apr 1999 15:52:01 +0000 (15:52 +0000)]
Bugfix (bug by me in 1.4) in backparsing INSERT ... SELECT

Jan

25 years agoDefend against 'update oid'. Someday we might want to support
Tom Lane [Thu, 29 Apr 1999 03:01:50 +0000 (03:01 +0000)]
Defend against 'update oid'.  Someday we might want to support
that, but it'd be a New Feature, wouldn't it ... in the meantime,
avoiding a backend crash seems worthwhile.

25 years agoAggregate functions didn't work on subscripted array references.
Tom Lane [Thu, 29 Apr 1999 01:13:13 +0000 (01:13 +0000)]
Aggregate functions didn't work on subscripted array references.
Things are better now.

25 years agoSeveral routines in setrefs.c would crash on array refs
Tom Lane [Thu, 29 Apr 1999 00:20:27 +0000 (00:20 +0000)]
Several routines in setrefs.c would crash on array refs
due to lack of check for recursing into a null subexpression.

25 years agoFix stupid typo that broke empty-query response... odd that
Tom Lane [Wed, 28 Apr 1999 22:17:58 +0000 (22:17 +0000)]
Fix stupid typo that broke empty-query response... odd that
this is not revealed by any of our regression tests...

25 years agoFixed DECIMAL data type to handle specified precision in atttypmod
Jan Wieck [Tue, 27 Apr 1999 13:33:43 +0000 (13:33 +0000)]
Fixed DECIMAL data type to handle specified precision in atttypmod

Jan

25 years agoFix for _copyUnique() suggested by Hiroshi Inoue
Tatsuo Ishii [Tue, 27 Apr 1999 09:49:36 +0000 (09:49 +0000)]
Fix for _copyUnique() suggested by Hiroshi Inoue

25 years agoThese are no longer used ...
Marc G. Fournier [Mon, 26 Apr 1999 13:37:14 +0000 (13:37 +0000)]
These are no longer used ...

25 years agoprovides string to be used for expected/* files for system specific output...
Marc G. Fournier [Mon, 26 Apr 1999 13:33:19 +0000 (13:33 +0000)]
provides string to be used for expected/* files for system specific output...

25 years ago*** empty log message ***
Michael Meskes [Mon, 26 Apr 1999 05:28:48 +0000 (05:28 +0000)]
*** empty log message ***