OSDN Git Service

pg-rex/syncrep.git
25 years agoAllow insert statements to have every column
Thomas G. Lockhart [Thu, 3 Sep 1998 14:21:06 +0000 (14:21 +0000)]
Allow insert statements to have every column
 supplied by a DEFAULT clause.  Enables INSERT INTO TABLE DEFAULT VALUES...

25 years agoUpdate to match newest gram.c since both are generated from gram.y.
Thomas G. Lockhart [Thu, 3 Sep 1998 12:26:20 +0000 (12:26 +0000)]
Update to match newest gram.c since both are generated from gram.y.
 Should be done every time gram.c is refreshed.

25 years agoOops missing file.
Bruce Momjian [Thu, 3 Sep 1998 05:09:27 +0000 (05:09 +0000)]
Oops missing file.

25 years agopatch cleanups.
Bruce Momjian [Thu, 3 Sep 1998 05:08:28 +0000 (05:08 +0000)]
patch cleanups.

25 years ago[Part #1: Type: text/plain, Encoding: 7bit, Size: 59]
Bruce Momjian [Thu, 3 Sep 1998 02:34:35 +0000 (02:34 +0000)]
[Part #1: Type: text/plain, Encoding: 7bit, Size: 59]

I will be cleaning this up more before the Oct 1 deadline.

David Hartwig.  AND/OR fix.

25 years agoFix ifdef.
Bruce Momjian [Thu, 3 Sep 1998 02:32:41 +0000 (02:32 +0000)]
Fix ifdef.

25 years agoNew stuff from Peter Mount for jdbc.
Bruce Momjian [Thu, 3 Sep 1998 02:31:38 +0000 (02:31 +0000)]
New stuff from Peter Mount for jdbc.

25 years agoNew stuff for 6.4 jdbc.
Bruce Momjian [Thu, 3 Sep 1998 02:29:45 +0000 (02:29 +0000)]
New stuff for 6.4 jdbc.

Peter mount

25 years ago> Yikes, that is certainly not standard C. I have never seen that
Bruce Momjian [Thu, 3 Sep 1998 02:20:30 +0000 (02:20 +0000)]
> Yikes, that is certainly not standard C.  I have never seen that
before.
> Looks like a GNU-ism.  I nice one, but still a GNU-ism.

Sorry, I didn't know it is a GNU extension. I have written this patch
which should fix the problem. Let me know if you still have problems.

Massimo Dal Zotto

25 years agoRemove \r\n in windows makefile.
Bruce Momjian [Thu, 3 Sep 1998 02:16:27 +0000 (02:16 +0000)]
Remove \r\n in windows makefile.

25 years agoO2. Seems tas() for PPC (storage/buffer/s_lock.c) never works if
Bruce Momjian [Thu, 3 Sep 1998 02:14:39 +0000 (02:14 +0000)]
O2. Seems tas() for PPC (storage/buffer/s_lock.c) never works if
compiled with -O0. Included are patches that should fix the problem
(of course I have confirmed -O2 works with this patch).

BTW, here is a platforms/regression test failure(serious one--backend
death) matrix.

Tatsuo Ishii

25 years ago Oops. I'll fix this.
Bruce Momjian [Thu, 3 Sep 1998 02:12:14 +0000 (02:12 +0000)]
 Oops. I'll fix this.

Here are patches for bin/createdb/createdb.sh.
--
Tatsuo Ishii
t-ishii@sra.co.jp

25 years agoAttached is a patch to remove the definitions of libpq's internal
Bruce Momjian [Thu, 3 Sep 1998 02:10:56 +0000 (02:10 +0000)]
Attached is a patch to remove the definitions of libpq's internal
structs from libpq-fe.h, as we previously discussed.

There turned out to be sloppy coding practices in more places than
I had realized :-(, but all in all I think it was a well-worth-while
exercise.

I ended up adding several routines to libpq's API in order to respond
to application requirements that were exposed by this work.  I owe the
docs crew updates for libpq.sgml to describe these changes.  I'm way too
tired to work on the docs tonight, however.

This is the last major change I intend to submit for 6.4.  I do want
to see if I can make libpgtcl work with Tcl 8.0 before we go final,
but hopefully that will be a minor bug fix.

25 years ago> David Hartwig wrote:
Bruce Momjian [Wed, 2 Sep 1998 23:37:13 +0000 (23:37 +0000)]
> David Hartwig wrote:
>
> Please apply this HAVING regression patch.
> > My bad.   It is caused by a known bug having to do with GROUP BY.
It ain't$
> > nothing to do with HAVING.  For some reason the bug went away for a
while, $
> > script.  It must have, because that is how I created the expected
file.   :(
> >
> > A patch to the regression will be forthcoming.
>

25 years agoFix for indexing problems.
Bruce Momjian [Wed, 2 Sep 1998 23:05:37 +0000 (23:05 +0000)]
Fix for indexing problems.

25 years agoUpdate to support the CREATE TABLE DEFAULT VALUES statement in gram.y.
Thomas G. Lockhart [Wed, 2 Sep 1998 15:48:28 +0000 (15:48 +0000)]
Update to support the CREATE TABLE DEFAULT VALUES statement in gram.y.

25 years agoSupport CREATE TABLE DEFAULT VALUES statement.
Thomas G. Lockhart [Wed, 2 Sep 1998 15:47:30 +0000 (15:47 +0000)]
Support CREATE TABLE DEFAULT VALUES statement.

25 years agoAdd more new pages.
Thomas G. Lockhart [Wed, 2 Sep 1998 03:01:02 +0000 (03:01 +0000)]
Add more new pages.

25 years agoautoconf
Bruce Momjian [Wed, 2 Sep 1998 02:38:29 +0000 (02:38 +0000)]
autoconf

25 years agoFix for install path
Bruce Momjian [Wed, 2 Sep 1998 02:34:00 +0000 (02:34 +0000)]
Fix for install path

25 years agoupdate
Bruce Momjian [Wed, 2 Sep 1998 02:08:02 +0000 (02:08 +0000)]
update

25 years agoinstall file search fix
Bruce Momjian [Wed, 2 Sep 1998 02:04:39 +0000 (02:04 +0000)]
install file search fix

25 years agoautoconf
Bruce Momjian [Wed, 2 Sep 1998 02:00:07 +0000 (02:00 +0000)]
autoconf

25 years agoConfigure fix for install script search.
Bruce Momjian [Wed, 2 Sep 1998 01:54:18 +0000 (01:54 +0000)]
Configure fix for install script search.

25 years agore-add bsdi makefile
Bruce Momjian [Tue, 1 Sep 1998 21:43:16 +0000 (21:43 +0000)]
re-add bsdi makefile

25 years agobsdi cleanup
Bruce Momjian [Tue, 1 Sep 1998 16:31:19 +0000 (16:31 +0000)]
bsdi cleanup

25 years agocleanup
Bruce Momjian [Tue, 1 Sep 1998 16:21:47 +0000 (16:21 +0000)]
cleanup

25 years agoFix message delimiters for substitution by sed script during installation.
Thomas G. Lockhart [Tue, 1 Sep 1998 15:57:00 +0000 (15:57 +0000)]
Fix message delimiters for substitution by sed script during installation.
 They were missing a "PG_" prefix so were not substituted properly.

25 years agoAdd new files from Oliver and Thomas.
Thomas G. Lockhart [Tue, 1 Sep 1998 15:53:09 +0000 (15:53 +0000)]
Add new files from Oliver and Thomas.

25 years agoRemove mention of the migration directory since this info has already
Thomas G. Lockhart [Tue, 1 Sep 1998 15:50:53 +0000 (15:50 +0000)]
Remove mention of the migration directory since this info has already
 been integrated into here.

25 years agoPut in explicit reference to the ref/ subdirectory to help emacs.
Thomas G. Lockhart [Tue, 1 Sep 1998 15:49:22 +0000 (15:49 +0000)]
Put in explicit reference to the ref/ subdirectory to help emacs.

25 years agoInitial information for Linux/Unix installation.
Thomas G. Lockhart [Tue, 1 Sep 1998 15:48:00 +0000 (15:48 +0000)]
Initial information for Linux/Unix installation.

25 years agocleanups of indexing.c
Bruce Momjian [Tue, 1 Sep 1998 06:51:35 +0000 (06:51 +0000)]
cleanups of indexing.c

25 years agocleanup
Bruce Momjian [Tue, 1 Sep 1998 06:22:46 +0000 (06:22 +0000)]
cleanup

25 years agocleanup
Bruce Momjian [Tue, 1 Sep 1998 05:34:16 +0000 (05:34 +0000)]
cleanup

25 years agoOK, folks, here is the pgindent output.
Bruce Momjian [Tue, 1 Sep 1998 04:40:42 +0000 (04:40 +0000)]
OK, folks, here is the pgindent output.

25 years agoRenaming cleanup, no pgindent yet.
Bruce Momjian [Tue, 1 Sep 1998 03:29:17 +0000 (03:29 +0000)]
Renaming cleanup, no pgindent yet.

25 years agoindexing.c cleanups.
Bruce Momjian [Mon, 31 Aug 1998 17:49:18 +0000 (17:49 +0000)]
indexing.c cleanups.

25 years agoFix for regproc
Bruce Momjian [Mon, 31 Aug 1998 07:55:50 +0000 (07:55 +0000)]
Fix for regproc

25 years agoFix for regprocout, use underscore, not parens.
Bruce Momjian [Mon, 31 Aug 1998 07:35:44 +0000 (07:35 +0000)]
Fix for regprocout, use underscore, not parens.

25 years agoOR clause index fix
Bruce Momjian [Mon, 31 Aug 1998 07:19:56 +0000 (07:19 +0000)]
OR clause index fix

25 years agoTwo spaces after a colon.
Bruce Momjian [Mon, 31 Aug 1998 04:48:36 +0000 (04:48 +0000)]
Two spaces after a colon.

25 years agoPreserver pg_log and pg_variable in upgrade. Cleanups.
Bruce Momjian [Mon, 31 Aug 1998 04:32:18 +0000 (04:32 +0000)]
Preserver pg_log and pg_variable in upgrade.  Cleanups.

25 years agoFix for possible releasebuffer bug.
Bruce Momjian [Sun, 30 Aug 1998 23:25:56 +0000 (23:25 +0000)]
Fix for possible releasebuffer bug.

25 years agoFrom: Massimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Sun, 30 Aug 1998 21:05:27 +0000 (21:05 +0000)]
From: Massimo Dal Zotto <dz@cs.unitn.it>

25 years agoTypes removed from Postgres: oidint2, oidint4, oidname.
Thomas G. Lockhart [Sun, 30 Aug 1998 19:54:04 +0000 (19:54 +0000)]
Types removed from Postgres: oidint2, oidint4, oidname.
Fix example columns in alter_table.out to use datetime and timespan
 as a substitute for oidint4 and oidname.

25 years agoFrom: Tom Lane <tgl@sss.pgh.pa.us>
Marc G. Fournier [Sun, 30 Aug 1998 19:41:50 +0000 (19:41 +0000)]
From: Tom Lane <tgl@sss.pgh.pa.us>

We're carrying around a copy of install-sh in case the local system
has no install script.  It's wasted baggage, because configure doesn't
know it's there :-(.  (Apparently everyone who's used postgres lately
already had an install script somewhere in their path.  I happened to
try to run configure with a minimal PATH tonight, and it promptly
gave up for lack of an install program.)  Here's the patch.

25 years agosequence contrib directory removed, as already integrated into the main
Marc G. Fournier [Sun, 30 Aug 1998 19:39:01 +0000 (19:39 +0000)]
sequence contrib directory removed, as already integrated into the main
backend...

25 years agoFrom: Massimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Sun, 30 Aug 1998 19:37:51 +0000 (19:37 +0000)]
From: Massimo Dal Zotto <dz@cs.unitn.it>

Here is a tar file the new directories, which substitute the old ones
in contrib. Please remove the old directories array, datetime, miscutil,
string and userlock before unpacking the tar file in contrib.

Note that as the modules are now installed in lib/modules I install all
my sql code in lib/sql. In my opinion also the other contributors should
follow these rules.

25 years agoFrom: Massimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Sun, 30 Aug 1998 19:30:38 +0000 (19:30 +0000)]
From: Massimo Dal Zotto <dz@cs.unitn.it>

After some playing with gdb I found that in printtup() there is a non null
attribute with typeinfo->attrs[i]->atttypid = 0 (invalid oid). Unfortunately
attibutes with invalid type are neither printed nor marked as null, and this
explains why psql doesn't get all the expected data.

So I made this patch to printtup():

25 years agoMake old db directory configurable.
Bruce Momjian [Sun, 30 Aug 1998 13:14:00 +0000 (13:14 +0000)]
Make old db directory configurable.

25 years agoUpdate man pages.
Bruce Momjian [Sun, 30 Aug 1998 05:25:47 +0000 (05:25 +0000)]
Update man pages.

25 years agoUpdate INSTALL to mention pg_upgrade.
Bruce Momjian [Sun, 30 Aug 1998 05:09:03 +0000 (05:09 +0000)]
Update INSTALL to mention pg_upgrade.

25 years agoNew pg_upgrade command.
Bruce Momjian [Sun, 30 Aug 1998 05:06:54 +0000 (05:06 +0000)]
New pg_upgrade command.

25 years agoUpdate FAQ.
Bruce Momjian [Sun, 30 Aug 1998 04:01:31 +0000 (04:01 +0000)]
Update FAQ.

25 years agoImprove distclean.
Bruce Momjian [Sun, 30 Aug 1998 03:34:05 +0000 (03:34 +0000)]
Improve distclean.

25 years agoAdd missing pgaccess file.
Bruce Momjian [Sun, 30 Aug 1998 03:32:47 +0000 (03:32 +0000)]
Add missing pgaccess file.

25 years agoautoconf
Bruce Momjian [Sun, 30 Aug 1998 03:25:22 +0000 (03:25 +0000)]
autoconf

25 years ago--without fix from Brook
Bruce Momjian [Sun, 30 Aug 1998 03:19:47 +0000 (03:19 +0000)]
--without fix from Brook

25 years agoAdd missing file.
Bruce Momjian [Sun, 30 Aug 1998 01:43:20 +0000 (01:43 +0000)]
Add missing file.

25 years agoUpdate INSTALL, etc. for release 6.4. Update pgaccess to 0.88.
Bruce Momjian [Sun, 30 Aug 1998 01:40:52 +0000 (01:40 +0000)]
Update INSTALL, etc. for release 6.4.  Update pgaccess to 0.88.

25 years agoUpdate of changes
Bruce Momjian [Sun, 30 Aug 1998 00:17:42 +0000 (00:17 +0000)]
Update of changes

25 years agoPrevent file descriptor leak from failed COPY.
Bruce Momjian [Sat, 29 Aug 1998 18:19:59 +0000 (18:19 +0000)]
Prevent file descriptor leak from failed COPY.

25 years agoOk. BTW Mr. Kataoka who is maintaing Japanese version of PostgreSQL
Bruce Momjian [Sat, 29 Aug 1998 18:06:57 +0000 (18:06 +0000)]
Ok. BTW Mr. Kataoka who is maintaing Japanese version of PostgreSQL
ODBC driver have found a bug in 6.3.2 pg_dump and have made patches.
I confirmed that the same bug still exists in the current source
tree. So I made up patches based on Kataoka's. Here are some
explanations.

o fmtId() returns pointer to a static memory in it. In the meantime
there is a line where is fmtId() called twice without saving the
first value returned by fmtId(). So second call to fmtId() will
break the first one.

o findTableByName() looks up a table by its name. if a table name
contanins upper letters or non ascii chars, fmtId() will returns
a name quoted in double quotes, which will not what findTableByName()
wants. The result is SEG fault.  -- Tatsuo Ishii t-ishii@sra.co.jp

25 years agoFix scanner name length trimming.
Bruce Momjian [Sat, 29 Aug 1998 05:27:17 +0000 (05:27 +0000)]
Fix scanner name length trimming.

25 years agoBetter error message if table exists.
Bruce Momjian [Sat, 29 Aug 1998 04:19:08 +0000 (04:19 +0000)]
Better error message if table exists.

25 years agoThis patch resolves some regression test failures caused by platform
Bruce Momjian [Sat, 29 Aug 1998 04:10:40 +0000 (04:10 +0000)]
This patch resolves some regression test failures caused by platform
dependencies.

David Hartwig

25 years agoThis is the first (of hopefully few) AIX port patches. This patch
Bruce Momjian [Sat, 29 Aug 1998 04:09:29 +0000 (04:09 +0000)]
This is the first (of hopefully few) AIX port patches.   This patch
was tested with Linux/GCC.  I still have some issues with with the
snprintf() function.

David Hartwig

25 years agoupdate libpq.rc version.
Bruce Momjian [Sat, 29 Aug 1998 04:07:08 +0000 (04:07 +0000)]
update libpq.rc version.

25 years agoHello!
Bruce Momjian [Sat, 29 Aug 1998 04:05:46 +0000 (04:05 +0000)]
Hello!

Here is a new patch for libpq, to make it work on Win32 again (since
the latest modifications broke it a little).

Please also add the file "libpq.rc" to the interfaces/libpq directory.
This will allow version-stamping of the generated DLL file, so that
automatic install programs (and interested users) can determine
the version of the file.  The file is currently set as "prerelease".
Before the release, somebody should change the line "FILEFLAGS
VS_FF_PRERELEASE" to "FILEFLAGS 0".  That information should probably
go into toos\RELEASE_CHANGES.

The patch is against the cvs as of ~ 1998-08-26 14:30 CEST.

//Magnus

25 years agoTruncate identifiers at NAMEDATALEN length.
Bruce Momjian [Sat, 29 Aug 1998 02:36:18 +0000 (02:36 +0000)]
Truncate identifiers at NAMEDATALEN length.

25 years agolibpq's pqGetnchar() should not return a null-terminated value
Bruce Momjian [Sat, 29 Aug 1998 02:09:27 +0000 (02:09 +0000)]
libpq's pqGetnchar() should not return a null-terminated value
anymore.  Fix for large objects.

25 years agoFrom: Michael Meskes <meskes@online-club.de>
Marc G. Fournier [Fri, 28 Aug 1998 17:47:53 +0000 (17:47 +0000)]
From: Michael Meskes <meskes@online-club.de>

This one is against the current archive (so it contains the one I send the
other day). It should fix the AIX problems. Andreas, could you please try
it? Thanks.

+ Wed Aug 26 16:17:39 CEST 1998
+
+       - Sync preproc.y with gram.y
+
+ Thu Aug 27 15:32:23 CEST 1998
+
+       - Fix some minor glitches that the AIX compiler complains about
+       - Added patchlevel to library
+
+ Fri Aug 28 15:36:58 CEST 1998
+
+       - Removed one line of code that AIX complains about since it was not
+         needed anyway
+       - Set library version to 2.6.1

25 years agoIgnore spaces when reporting regression failures.
Bruce Momjian [Fri, 28 Aug 1998 14:17:57 +0000 (14:17 +0000)]
Ignore spaces when reporting regression failures.

25 years agoFrom: Massimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Fri, 28 Aug 1998 12:08:03 +0000 (12:08 +0000)]
From: Massimo Dal Zotto <dz@cs.unitn.it>

I have this small patch which fixes a typo in my previous lock patches.
Fortunately the bug affects only user locks.

-

25 years agovacuum fix.
Bruce Momjian [Fri, 28 Aug 1998 04:57:21 +0000 (04:57 +0000)]
vacuum fix.

25 years agoHi all,
Bruce Momjian [Fri, 28 Aug 1998 03:36:31 +0000 (03:36 +0000)]
Hi all,

    I  don't know if this is really related to the initdb problem
    discussion (haven't followed it enough). But seems so because
    it  fixes  a  damn  problem  during  index tuple insertion on
    CREATE TABLE into pg_attribute_relid_attnum_index.

    Anyway - this bug was really hard to find. During startup the
    relcache  reads  in  some  prepared  information  about index
    strategies from a file and then  reinitializes  the  function
    pointers  inside  the  scanKey data.  But for sake it assumed
    single attribute index tuples (hasn't that changed recently).
    Thus not all the strategies scanKey entries where initialized
    properly,  resulting  in  invalid  addresses  for  the  btree
    comparision functions.

    With  the  patch  at  the  end  the  regression  tests passed
    excellent except for the sanity_check that crashed at  vacuum
    and the misc test where the select unique1 from onek2 outputs
    the two rows in different order.

Jan

25 years agoFrom: Magnus Hagander <mha@edu.sollentuna.se>
Marc G. Fournier [Thu, 27 Aug 1998 13:25:40 +0000 (13:25 +0000)]
From: Magnus Hagander <mha@edu.sollentuna.se>

Ok. Here is a patch to make psql work on Win32 (as a console mode
application, of course).
It requires getopt.c to be in src/utils - works fine with the FreeBSD
version of it.

Also, the file win32.mak should go into src/bin/psql.

25 years agoMake 'name' type int aligned, like char/varchar.
Bruce Momjian [Thu, 27 Aug 1998 05:06:57 +0000 (05:06 +0000)]
Make 'name' type int aligned, like char/varchar.

25 years agoFix 'd ' should be 'd'.
Bruce Momjian [Thu, 27 Aug 1998 01:04:23 +0000 (01:04 +0000)]
Fix 'd ' should be 'd'.

25 years agoFix problem where attbyval was wrong for xid types.
Bruce Momjian [Wed, 26 Aug 1998 17:12:12 +0000 (17:12 +0000)]
Fix problem where attbyval was wrong for xid types.

25 years agoFix atttypmod alignment again, and re-enable ecpg.
Bruce Momjian [Wed, 26 Aug 1998 16:43:54 +0000 (16:43 +0000)]
Fix atttypmod alignment again, and re-enable ecpg.

25 years agoFix atttypmod to be int32 in system catalogs. Oops.
Bruce Momjian [Wed, 26 Aug 1998 05:51:24 +0000 (05:51 +0000)]
Fix atttypmod to be int32 in system catalogs.  Oops.

25 years agoMake attalign match type alignment.
Bruce Momjian [Wed, 26 Aug 1998 05:22:58 +0000 (05:22 +0000)]
Make attalign match type alignment.

25 years agoFix up crashing symptoms for new serial type by making sure constraint
Thomas G. Lockhart [Wed, 26 Aug 1998 04:20:27 +0000 (04:20 +0000)]
Fix up crashing symptoms for new serial type by making sure constraint
 and index name fields are pstrdup'd (copied) rather than reused.

25 years agoFix for select bug.
Bruce Momjian [Wed, 26 Aug 1998 03:17:32 +0000 (03:17 +0000)]
Fix for select bug.

25 years agocvs add'd two files for the tprintf() patch...
Marc G. Fournier [Tue, 25 Aug 1998 21:43:47 +0000 (21:43 +0000)]
cvs add'd two files for the tprintf() patch...

25 years agoFrom: Massimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Tue, 25 Aug 1998 21:37:08 +0000 (21:37 +0000)]
From: Massimo Dal Zotto <dz@cs.unitn.it>

>       these patches define the UNLISTEN sql command. The code already
>       existed but it was unknown to the parser. Now it can be used
>       like the listen command.
>       You must make clean and delete gram.c and parser.h before make.

25 years agoFrom: Massimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Tue, 25 Aug 1998 21:34:10 +0000 (21:34 +0000)]
From: Massimo Dal Zotto <dz@cs.unitn.it>

> tprintf.patch
>
>       tprintf.patch
>
>       adds functions and macros which implement a conditional trace package
>       with the ability to change flags and numeric options of running
>       backends at runtime.
>       Options/flags can be specified in the command line and/or read from
>       the file pg_options in the data directory.

25 years agoMassimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Tue, 25 Aug 1998 21:32:10 +0000 (21:32 +0000)]
Massimo Dal Zotto <dz@cs.unitn.it>

> socket-flock.patch
>
>       use advisory locks to check if the unix socket can be deleted.
>       A running postmaster keeps a lock on that file. A starting
>       postmaster exits if the file exists and is locked, otherwise
>       it deletes the sockets and proceeds.
>       This avoid the need to remove manually the file after a postmaster
>       or system crash.
>       I don't know if flock is available on any system. If not we could
>       define a HAVE_FLOCK set by configure.

25 years agoFrom: Massimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Tue, 25 Aug 1998 21:31:20 +0000 (21:31 +0000)]
From: Massimo Dal Zotto <dz@cs.unitn.it>

> sinval.patch
>
>       fixes a problem in SI cache which causes table overflow if some
>       backend is idle for a long time while other backends keep adding
>       entries.
>       It uses the new signal handling implemented in tprintf.patch.
>       I have also increacasesed the max number of backends from 32 to 64
>       and the table size from 1000 to 5000.
>       I don't know if anybody is working on SI, but until another
>       solution is found this patch fixes the problem. I have received
>       messages from other people reporting the same problem which I
>       fixed many months ago.

25 years agomissed adding a new include file
Marc G. Fournier [Tue, 25 Aug 1998 21:29:16 +0000 (21:29 +0000)]
missed adding a new include file

25 years agoFrom: Massimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Tue, 25 Aug 1998 21:25:46 +0000 (21:25 +0000)]
From: Massimo Dal Zotto <dz@cs.unitn.it>

> sequence.patch
>
>       adds the missing setval command to sequences. Owner of sequences
>       can now set the last value to any value between min and max
>       without recreating the sequence. This is useful after loading
>       data from external files.

25 years agoFrom: Massimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Tue, 25 Aug 1998 21:24:12 +0000 (21:24 +0000)]
From: Massimo Dal Zotto <dz@cs.unitn.it>

> ps-status.patch
>
>       macros for ps status, used by postgres.c and utility.c.
>       Unfortunately ps status is system dependent and the current
>       code doesn't work on linux. The use of macros confines system
>       dependency to into one file (ps-status.h). Users of other
>       operating systems should check this code and submit new macros.

25 years agoFrom: Massimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Tue, 25 Aug 1998 21:22:27 +0000 (21:22 +0000)]
From: Massimo Dal Zotto <dz@cs.unitn.it>

> pqpacket.patch
>
>       fixed indentation.

25 years agoFrom: Massimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Tue, 25 Aug 1998 21:20:32 +0000 (21:20 +0000)]
From: Massimo Dal Zotto <dz@cs.unitn.it>

lock.patch

        I have rewritten lock.c cleaning up the code and adding better
        assert checking I have also added some fields to the lock and
        xid tags for better support of user locks. There is also a new
        function which returns an array of pids owning a lock.
        I'm using this code from over six months and it works fine.

25 years agoFrom: Massimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Tue, 25 Aug 1998 21:17:11 +0000 (21:17 +0000)]
From: Massimo Dal Zotto <dz@cs.unitn.it>

config.patch

        remove old defines no more used in the code.

25 years agoFrom: Massimo Dal Zotto <dz@cs.unitn.it>
Marc G. Fournier [Tue, 25 Aug 1998 21:04:41 +0000 (21:04 +0000)]
From: Massimo Dal Zotto <dz@cs.unitn.it>

assert.patch

        adds a switch to turn on/off the assert checking if enabled at compile
        time. You can now compile postgres with assert checking and disable it
        at runtime in a production environment.

25 years agoCan someone please apply this portability patch to genbki.sh ?
Bruce Momjian [Tue, 25 Aug 1998 17:36:18 +0000 (17:36 +0000)]
Can someone please apply this portability patch to genbki.sh ?
(Mark or Bruce?) It fixes a problem when cpp gives a warning when
precompiling /dev/null like:  "/dev/null", line 1: 1506-229 (W)
File is empty.  This leads to a hangup when doing the description
load during initdb, since stderr also ends up in the global1.description
and local1_template1.description

stderr has to be redirected to /dev/null:

Andreas Zeugswetter

25 years agoAdd is_sequence flag to ColumnDef structure. Used to implement SERIAL type.
Thomas G. Lockhart [Tue, 25 Aug 1998 15:09:31 +0000 (15:09 +0000)]
Add is_sequence flag to ColumnDef structure. Used to implement SERIAL type.