git.postgresql.org Git - postgresql.git/log
Update TODO list.
Gen_fmgrtab.sh is strange: it is a platform dependent way (because it uses
CPP) to create platform independent files. Unfortunately, that means that
every config.status (or configure) run invariably causes a relink of the
postmaster and also that we can't put these files in the distribution
(usefully). So we make it a little smarter: when the output files already
exist and it notices that it would recreate them in identical form, it
doesn't touch them. In order to avoid re-running the make rule all the time
we update a timestamp file instead.
Update release_prep accordingly. Also make Gen_fmgrtab.sh use the awk that
is detected at configure time, not necessarily named `awk' and have it check
for exit statuses a little better.
In other news... Remove USE_LOCALE from the templates, it was set to `no'
everywhere anyway. Also remove YACC and YFLAGS from the templates, configure
is smart enough to find bison or yacc itself. Use AC_PROG_YACC for that
instead of the hand-crafted code. Do not set YFLAGS to `-d'. The make rules
that need this flag should explicitly invoke it. YFLAGS should be a user
variable. Update the makefiles to that effect.
Re-order pg_listener index so it can later be used in an index scan.
New system index, initdb everyone.
Add index on pg_index.indrelid for Tom Lane.
Moved configure script from src/ to the top level directory. Moved
configuration helper things into config/ dir. Adjusted some relative paths
in makefiles.
Tom Lane [Tue, 6 Jun 2000 17:44:25 +0000 (17:44 +0000)]
init_fcache was being careless about using SearchSysCacheTuple result
over multiple lookups --- it should use SearchSysCacheTupleCopy instead.
This accounts for rare failures like 'init_fcache: null probin for procedure 481'
when running concurrently with a VACUUM.
Tom Lane [Tue, 6 Jun 2000 17:02:38 +0000 (17:02 +0000)]
Improve comments for SearchSysCacheTuple and SearchSysCacheTupleCopy.
Tom Lane [Tue, 6 Jun 2000 16:50:37 +0000 (16:50 +0000)]
typeTypeName() must return a pstrdup'd copy of the type name, not a
direct pointer into the syscache entry for the type. In some cases
the syscache entry might get flushed before we are done using the
returned type name. This bug accounts for difficult-to-repeat
failures seen when INSERTs into columns of certain data types are
run in parallel with VACUUMs of system tables. There may be related
problems elsewhere --- we need to take a harder look at uses of
syscache data.
PGPORT envar was erroneously ignored by the backend
Update FAQ.
Peter Mount [Tue, 6 Jun 2000 11:06:09 +0000 (11:06 +0000)]
Added org/postgresql/DriverClass.java to the list of files removed by make clean (it's dynamically built)
Fixed Statement, so that the update count is valid when an SQL DELETE operation is done.
While fixing the update count, made it easier to get the OID of the last insert as well. Example is in example/basic.java
Peter Mount [Tue, 6 Jun 2000 07:45:07 +0000 (07:45 +0000)]
Added some missing org.'s that prevented the use of the geometric types
Peter Mount [Tue, 6 Jun 2000 07:24:06 +0000 (07:24 +0000)]
Removed hardwired 8k limit on queries
Update for 7.0.2.
Update for 7.0.2.
Tom Lane [Mon, 5 Jun 2000 07:29:25 +0000 (07:29 +0000)]
Latest round of fmgr updates. All functions with bool,char, or int2
inputs have been converted to newstyle. This should go a long way towards
fixing our portability problems with platforms where char and short
parameters are passed differently from int-width parameters. Still
more to do for the Alpha port however.
Tom Lane [Mon, 5 Jun 2000 07:16:13 +0000 (07:16 +0000)]
Simplify build/install process for bki and description files. There's
no reason for them to be copied into src/backend rather than being
installed straight from the catalog subdirectory. This also avoids
some peculiar behavior (bugs?) present in at least gmake 3.78.1: it
won't always update the bki files in backend/ even when the ones in
backend/catalog/ are newer.
Tom Lane [Sun, 4 Jun 2000 22:08:53 +0000 (22:08 +0000)]
If user re-DECLAREs an existing cursor name, close the old cursor of
that name and issue a NOTICE to the effect that we did. Previously,
code would try to assign the new cursor declaration to the old portal,
but this didn't work reliably since new parsetree is still sitting in
blank portal and is likely to get clobbered.
Tom Lane [Sun, 4 Jun 2000 22:04:32 +0000 (22:04 +0000)]
Disallow CLOSE of reserved system portal names.
Tom Lane [Sun, 4 Jun 2000 20:50:50 +0000 (20:50 +0000)]
Repair bug reported by ldm@apartia.com: Append nodes, which don't
actually use their targetlist, are given a targetlist that is just a
pointer to the first appended plan's targetlist. This is OK, but what
is not OK is that any sub-select expressions in said tlist were being
entered in the subPlan lists of both the Append and the first appended
plan. That led to two startup and two shutdown calls for the same
plan node at exec time, which led to crashes. Fix is to not generate
a list of subPlans for an Append node. Same problem and fix apply
to other node types that don't have a real, functioning targetlist:
Material, Sort, Unique, Hash.
Remove FROM warning by fixing query.
Removed ELOG_TIMESTAMPS #define in favor of two run-time
configuration options `Log_timestamp' and `Log_pid'.
New ps display code, works on more platforms.
Install a default configuration file.
Clean up some funny business in the config file code.
New warning code about auto-created range table entries.
More cleanup of c.h binary macros
Reverse PG_BINARY defines
Remove NT-specific file open defines by defining our own open macros for
"rb" and "wb".
heap' xlog records
Tom Lane [Fri, 2 Jun 2000 04:04:54 +0000 (04:04 +0000)]
If create/drop database are going to call closeAllVfds(), they ought
to do it at the last moment before calling system() ... not at some
randomly-chosen earlier point in the routine ...
Tom Lane [Fri, 2 Jun 2000 03:58:34 +0000 (03:58 +0000)]
Create an fd.c entry point that is just like plain open(2) except that
it will close VFDs if necessary to surmount ENFILE or EMFILE failures.
Make use of this in md.c, xlog.c, and user.c routines that were
formerly vulnerable to these failures. In particular, this should
handle failures of mdblindwrt() that have been observed under heavy
load conditions. (By golly, every other process on the system may
crash after Postgres eats up all the kernel FDs, but Postgres will
keep going!)
Update TODO detail files.
Update faq's.
Tom Lane [Fri, 2 Jun 2000 02:00:28 +0000 (02:00 +0000)]
Correct portability problem introduced by yours truly --- I used a
conditional expression x?y:z in an awk program. Seems old versions
of awk don't have that ...
Add fix to index length
Tom Lane [Fri, 2 Jun 2000 00:04:26 +0000 (00:04 +0000)]
Keep CVS quiet about derived file guc-file.c.
createlang fix
Tom Lane [Thu, 1 Jun 2000 22:21:05 +0000 (22:21 +0000)]
Lexer defended us against overlength plain identifiers, but not against
overlength quoted identifiers. Death and destruction ensue...
Add BSDI readme
Update replication todo.
Update TODO list.
Add lock todo items
The definition for "ProcessConfigFile()" in guc-file.l does not match
the prototype in guc.h. The following patch corrects that.
Billy G. Allie
Tom Lane [Thu, 1 Jun 2000 14:52:25 +0000 (14:52 +0000)]
Simplify overly-clever Make rule, which evidently confuses at least
some versions of gmake (mine didn't do the right thing, anyway).
Peter Mount [Thu, 1 Jun 2000 06:33:01 +0000 (06:33 +0000)]
Removed timezone in ResultSet.getTimestamp()
Update 7.0.1 for Tatsuo
Mention pg_upgrade
Update 7.0.1
Fixups for 7.0.1
More perl cleanup
Rename perl example eg directory to examples.
Update pgcvslog
all array stuff fixed.
Tom Lane [2000年5月31日 15:38:53 +0000 (15:38 +0000)]
Tweak recognition of range-clause pairs so that 'var > 1ドル AND var < 2ドル'
(ie, parameters instead of consts) will be treated as a range query.
We do not know the actual selectivities involved, but it seems like
a good idea to use a smaller estimate than we would use for two unrelated
inequalities.
The heralded `Grand Unified Configuration scheme' (GUC)
That means you can now set your options in either or all of $PGDATA/configuration,
some postmaster option (--enable-fsync=off), or set a SET command. The list of
options is in backend/utils/misc/guc.c, documentation will be written post haste.
pg_options is gone, so is that pq_geqo config file. Also removed were backend -K,
-Q, and -T options (no longer applicable, although -d0 does the same as -Q).
Added to configure an --enable-syslog option.
changed all callers from TPRINTF to elog(DEBUG)
Typo fix
Update pg_hba.conf with more examples
Update pg_hba.conf.sample with better examples and descriptions
Copy pg_ident.conf.sample into /lib directory. Rename USERAUTH to
AUTHTYPE in config file. Patch both branches.
Tom Lane [2000年5月30日 07:09:23 +0000 (07:09 +0000)]
Ooops, looks like I broke fastpath in last changes...
Added header files required for ALTER TABLE DROP COLUMN
trial implementation.
Tom Lane [2000年5月30日 04:25:00 +0000 (04:25 +0000)]
Third round of fmgr updates: eliminate calls using fmgr() and
fmgr_faddr() in favor of new-style calls. Lots of cleanup of
sloppy casts to use XXXGetDatum and DatumGetXXX ...
Remove unused include files. Do not touch /port or includes used by defines.
More <> cleanup
Cleanup of <> and ""
Tom Lane [2000年5月29日 21:02:32 +0000 (21:02 +0000)]
Convert array_map to use new fmgr interface.
Tom Lane [2000年5月29日 20:51:35 +0000 (20:51 +0000)]
Neglected to add fmgr.h to set of installed headers...
Tom Lane [2000年5月29日 20:30:11 +0000 (20:30 +0000)]
Update array_iterator to use new fmgr interface.
Tom Lane [2000年5月29日 20:18:30 +0000 (20:18 +0000)]
Tweak fmgrtab generation so that the F_XXX macros that give OIDs for
built-in procedures are named after the prosrc field of pg_proc (ie,
the actual C function name), not the proname field. This did not use
to make a difference back when the two were always the same, but in the
presence of overloaded proname values we'd best try to use the C name
instead. AFAICT this change affects no existing code, but it is
necessary to be able to get at some built-in functions that no macro
was being generated for before.
Tom Lane [2000年5月29日 19:16:57 +0000 (19:16 +0000)]
Repair problems with overrun of timezone name length. Increase MAXTZLEN
to 10, and be consistent about whether it counts the trailing null (it
does not). Also increase MAXDATELEN to be sure no buffer overflows are
caused by the longer MAXTZLEN.
Add analyze.c file for split.
Split vacuum and analyze into separate files
Update messages.
Make analyze do vacuum/analyze in one step.
More vacuum cleanup
more cleanup
Add analyze log messages for verbose mode.
cleanup
Allow vacuum to perform analyze with shared lock. Update cvs manual.
Tom Lane [2000年5月29日 05:45:56 +0000 (05:45 +0000)]
Generated header files parse.h and fmgroids.h are now copied into
the src/include tree, so that -I backend is no longer necessary anywhere.
Also, clean up some bit rot in contrib tree.
Tom Lane [2000年5月29日 01:59:17 +0000 (01:59 +0000)]
Second round of fmgr changes: triggers are now invoked in new style,
CurrentTriggerData is history.
More vacuum cleanups
More vacuum renaming.
Tom Lane [2000年5月28日 20:34:52 +0000 (20:34 +0000)]
Miscellaneous cleanups of places that needed to account for new
pg_language entries.
Tom Lane [2000年5月28日 20:33:28 +0000 (20:33 +0000)]
Constant-expression simplifier now knows how to simplify strict functions
that have at least one constant-NULL input, even if other inputs are
not constants.
Tom Lane [2000年5月28日 18:06:55 +0000 (18:06 +0000)]
Install fmgr rewrite doc as README file.
Need to update user docs still ...
Tom Lane [2000年5月28日 17:56:29 +0000 (17:56 +0000)]
First round of changes for new fmgr interface. fmgr itself and the
key call sites are changed, but most called functions are still oldstyle.
An exception is that the PL managers are updated (so, for example, NULL
handling now behaves as expected in plperl and plpgsql functions).
NOTE initdb is forced due to added column in pg_proc.
Tom Lane [2000年5月27日 18:18:31 +0000 (18:18 +0000)]
'CREATE FUNC TION' => 'CREATE FUNCTION'.
Tom Lane [2000年5月27日 05:13:38 +0000 (05:13 +0000)]
define YY_NEVER_INTERACTIVE for flex, to persuade flex that it's not
necessary to call isatty() for each and every received query. That's
one less kernel call per query cycle ...
Update kerberos patch
Update readme.kerberos
Back out kerberos changes. Causes compile problems.
Add README.kerbros
Patch for Kerberos V.
Most (nearly all) of the work was done by David Wragg <dpw@doc.ic.ac.uk>
He patched 6.5.3. I've updated it for 7.0RC5.
It works for MIT kerberos 1.1.1 (and previously for 1.0.6 as well).
I've got the patch against 6.5.3, plus kerberized RPMS.
Mike Wyer <mw@doc.ic.ac.uk> || "Woof?"
Remove // comments from ODBC.
Remove pgaccess dll's from the distribution.
Clean up pg_hba.conf
Improve pg_hba.conf examples
Tom Lane [2000年5月26日 17:19:15 +0000 (17:19 +0000)]
Reduce eqsel()'s fudge-factor for estimating the frequency of values
other than the most common value in a column. We had had 0.5, make it
0.1 to make it more likely that an indexscan will be chosen. Really
need better statistics instead, but this should stem the bleeding
meanwhile ...
Fix \h SELECT to not show SELECT INTO.
This is the main PostgreSQL git repository.
RSS
Atom