OSDN Git Service
Marc G. Fournier [Mon, 17 Aug 1998 03:53:37 +0000 (03:53 +0000)]
From: Tom Lane <tgl@sss.pgh.pa.us>To: pgsql-patches@postgreSQL.org
Sigh. That tweak needs a tweak --- I didn't realize that ".DEFAULT"
processing ignores dependencies, at least in the version of gmake I
have here (not sure if it's a bug or not). Apply this patch aftermy previous one...
Marc G. Fournier [Mon, 17 Aug 1998 03:52:36 +0000 (03:52 +0000)]
Just a couple of "after-commit" cleanups...
Marc G. Fournier [Mon, 17 Aug 1998 03:50:43 +0000 (03:50 +0000)]
Date: Sun, 16 Aug 1998 14:56:48 -0400
From: Tom Lane <tgl@sss.pgh.pa.us>
Attached is a patch for this weekend's work on libpq. I've dealt
with several issues:
<for details: see message, in pgsql-patches archive for above data>
Marc G. Fournier [Mon, 17 Aug 1998 03:35:05 +0000 (03:35 +0000)]
From: Garrett Wollman <wollman@khavrinen.lcs.mit.edu>
Here is some more contrib-fodder, based on TIH's IP address type,
for ISBN and ISSN identifiers (which I just happened to need to keep
track of the things in my library).
Thomas G. Lockhart [Sun, 16 Aug 1998 05:38:41 +0000 (05:38 +0000)]
Allow a null pointer to be returned from get_opname().
Previously, had thrown an error, but looking for alternate strategies
for table indices utilization would prefer to continue.
Thomas G. Lockhart [Sun, 16 Aug 1998 05:37:04 +0000 (05:37 +0000)]
Check for null pointer returned from get_opname().
Don't bother checking for alternate strategies if so
since it was more likely a function or some other non-operator anyway.
Thomas G. Lockhart [Sun, 16 Aug 1998 05:35:35 +0000 (05:35 +0000)]
Check for null pointer returned from get_opname().
Thomas G. Lockhart [Sun, 16 Aug 1998 04:06:55 +0000 (04:06 +0000)]
Disable not-ready-to-use support code for the line data type.
Bracket things with #ifdef ENABLE_LINE_TYPE.
The line data type has always been used internally to support other types,
but I/O routines have never been defined for it.
Thomas G. Lockhart [Sat, 15 Aug 1998 16:39:21 +0000 (16:39 +0000)]
This README only mentions nroff problems with page.5, which has been
removed from the man pages and put into the new sgml docs.
Thomas G. Lockhart [Sat, 15 Aug 1998 16:36:22 +0000 (16:36 +0000)]
Include a sentence in the top pointing to the new docs.
pgbuiltin.3 is obsolete for sure, and libpq.3 can become so since the
size and scope of this man page is not appropriate in a man page format.
Thomas G. Lockhart [Sat, 15 Aug 1998 13:01:26 +0000 (13:01 +0000)]
Converted to page.sgml. Not needed in a man page.
Thomas G. Lockhart [Sat, 15 Aug 1998 07:03:59 +0000 (07:03 +0000)]
Information moved to sgml source files.
The "Oracle compatibility" page should have always been in with functions
anyway. The BKI information is not really appropriate for a man page.
Thomas G. Lockhart [Sat, 15 Aug 1998 07:00:37 +0000 (07:00 +0000)]
Make minor formatting improvements to statement synopses.
Make sure all files end with a newline.
Thomas G. Lockhart [Sat, 15 Aug 1998 06:56:36 +0000 (06:56 +0000)]
Update for new information on PQsetNoticeProcessor().
Now back in sync with libpq.3 man page, which should be obsolete soon.
Thomas G. Lockhart [Sat, 15 Aug 1998 06:55:05 +0000 (06:55 +0000)]
Change Id field in chapter tag to change html output file name.
Thomas G. Lockhart [Sat, 15 Aug 1998 06:53:52 +0000 (06:53 +0000)]
Include working list of all documentation sources, with current status
and plans for some of them.
Thomas G. Lockhart [Sat, 15 Aug 1998 06:52:54 +0000 (06:52 +0000)]
Include new chapters.
Thomas G. Lockhart [Sat, 15 Aug 1998 06:52:03 +0000 (06:52 +0000)]
Nice exposition on indices and keys from Herouth Maoz which appeared
on the mailing lists a while ago. Maybe slightly changed to fit docs.
Will go into the User's Guide.
Thomas G. Lockhart [Sat, 15 Aug 1998 06:49:33 +0000 (06:49 +0000)]
Convert information from man pages. Not really suitable for man pages
anyway, so this will replace them as of now.
Thomas G. Lockhart [Sat, 15 Aug 1998 06:47:39 +0000 (06:47 +0000)]
Remove single-argument trim() function from table.
Never seen because the parser frontend converts all trim() calls to
btrim(), ltrim(), and rtime() calls before execution.
Thomas G. Lockhart [Sat, 15 Aug 1998 06:45:10 +0000 (06:45 +0000)]
Some old cleanup fixes for close_ps() from G. Thaker.
Thomas G. Lockhart [Fri, 14 Aug 1998 16:13:07 +0000 (16:13 +0000)]
Allow binary-compatible indices to be considered when checking for valid
indices for restriction clauses containing a constant.
Note that if an index does not match directly (usually because the types
on both side of the clause don't match), and if a binary-compatible index
is identified, then the operator function will be replaced by a new
one. Should not be a problem, but be sure that if types are listed as
being binary compatible (in parse_coerce.h) then the comparison functions
are also binary-compatible, giving equivalent results.
Thomas G. Lockhart [Fri, 14 Aug 1998 16:07:00 +0000 (16:07 +0000)]
Include OID as a built-in type.
Thomas G. Lockhart [Fri, 14 Aug 1998 16:05:51 +0000 (16:05 +0000)]
Check for bad result from pg_id. A bad result can come from shared library
trouble, and the name of the shared library has been changed recently.
Had to rerun ldconfig on my machine to get it working again.
Give an error message with a helpful hint if so...
Bruce Momjian [Tue, 11 Aug 1998 22:39:33 +0000 (22:39 +0000)]
fix typo.
Bruce Momjian [Tue, 11 Aug 1998 20:55:52 +0000 (20:55 +0000)]
Strategy cleanups.
Bruce Momjian [Tue, 11 Aug 1998 19:32:39 +0000 (19:32 +0000)]
index strategy cleanup
Bruce Momjian [Tue, 11 Aug 1998 18:38:07 +0000 (18:38 +0000)]
the following patch fixes a bug in the oracle compatibility
functions btrim() ltrim() and rtrim().
The error was that the character after the set was included
in the tests (ptr2 pointed to the character after the vardata
part of set if no match found, so comparing *ptr or *end
against *ptr2 MAY match -> strip).
Jan
--
#======================================================================#
# It's easier to get forgiveness for being wrong than for being
right. # # Let's break this rule - forgive me.
# #======================================== jwieck@debis.com (Jan
Wieck) #
Bruce Momjian [Tue, 11 Aug 1998 18:33:37 +0000 (18:33 +0000)]
This one cleans the cursor problems ecpg had so far. It is now able
to understand cursors with variables.
Michael
Bruce Momjian [Tue, 11 Aug 1998 18:28:49 +0000 (18:28 +0000)]
Change owner from oid to int4 type.
Bruce Momjian [Tue, 11 Aug 1998 14:32:03 +0000 (14:32 +0000)]
More op_class cleanup.
Bruce Momjian [Tue, 11 Aug 1998 05:32:46 +0000 (05:32 +0000)]
Remove NOBTREE defines, and make findoidlinks handle regproc.
Bruce Momjian [Tue, 11 Aug 1998 05:09:30 +0000 (05:09 +0000)]
cleanup.
Bruce Momjian [Tue, 11 Aug 1998 02:32:00 +0000 (02:32 +0000)]
New findoidjoins examines oid columns to find join relationships.
Bruce Momjian [Mon, 10 Aug 1998 20:31:42 +0000 (20:31 +0000)]
Add new \w write command to psql.
Bruce Momjian [Mon, 10 Aug 1998 14:39:45 +0000 (14:39 +0000)]
Update flowchart.
Bruce Momjian [Mon, 10 Aug 1998 14:32:46 +0000 (14:32 +0000)]
Update flowchart.
Bruce Momjian [Mon, 10 Aug 1998 04:49:39 +0000 (04:49 +0000)]
Cleanups of optimizer.
Bruce Momjian [Mon, 10 Aug 1998 02:26:40 +0000 (02:26 +0000)]
Cleanup optimizer function names and clarify code.
Bruce Momjian [Sun, 9 Aug 1998 17:57:31 +0000 (17:57 +0000)]
Update pgindent.
Bruce Momjian [Sun, 9 Aug 1998 04:59:10 +0000 (04:59 +0000)]
New pgindent.
Bruce Momjian [Sun, 9 Aug 1998 04:17:38 +0000 (04:17 +0000)]
cleanups.
Bruce Momjian [Sun, 9 Aug 1998 02:59:33 +0000 (02:59 +0000)]
The attached patch implements some changes that were discussed a
couple weeks ago on the hackers and interfaces lists:
1. When the backend sends a NOTICE message and closes the connection
(typically, because it was told to by the postmaster after
another backend coredumped), libpq will now print the notice
and close the connection cleanly. Formerly, the frontend app
would usually terminate ungracefully due to a SIGPIPE. (I am
not sure if 6.3.2 behaved that way, but the current cvs sources
do...)
2. libpq's various printouts to stderr are now fed through a single
"notice processor" routine, which can be overridden by the
application to direct notices someplace else. This should ease
porting libpq to Windows.
I also noticed and fixed a problem in PQprint: when sending output
to a pager subprocess, it would disable SIGPIPE in case the pager
terminates early (this is good) --- but afterwards it reset SIGPIPE
to SIG_DFL, rather than restoring the application's prior setting
(bad).
regards, tom lane
Bruce Momjian [Fri, 7 Aug 1998 05:02:32 +0000 (05:02 +0000)]
OPTIMIZER_DEBUG additions.
Bruce Momjian [Thu, 6 Aug 1998 05:13:14 +0000 (05:13 +0000)]
Make large objects their own relkind type. Fix dups in pg_class_mb
files. Fix sequence creation hack for relkind type.
Bruce Momjian [Wed, 5 Aug 1998 19:14:23 +0000 (19:14 +0000)]
flowchart update
Bruce Momjian [Wed, 5 Aug 1998 19:01:58 +0000 (19:01 +0000)]
flowchart update
Bruce Momjian [Wed, 5 Aug 1998 16:23:40 +0000 (16:23 +0000)]
Fix for \d index display.
Marc G. Fournier [Wed, 5 Aug 1998 04:49:19 +0000 (04:49 +0000)]
From: David Hartwig <daybee@bellatlantic.net>
I have attached a patch to allow GROUP BY and/or ORDER BY function or
expressions. Note worthy items:
1. The expression or function need not be in the target list.
Example:
SELECT name FROM foo GROUP BY lower(name);
2. Simplified the grammar to use expressions only.
3. Cleaned up earlier patch in this area to make use of existing
utility functions.
3. Reduced some of the members in the SortGroupBy parse node. The
original data members were redundant with the new expression node.
(MUST do a "make clean" now)
4. Added a new parse node "JoinUsing". The JOIN USING clause was
overloading this SortGroupBy structure. With the afore mentioned
reduction of members, the two clauses lost all their commonality.
5. A bug still exist where, if a function or expression is GROUPed BY,
and an aggregate function does not include a attribute from the
expression or function, the backend crashes. (or something like
that) The bug pre-dates this patch. Example:
SELECT lower(a) AS lowcase, count(b) FROM foo GROUP BY lowcase;
*** BOOM ***
--Also when not in target list
SELECT count(b) FROM foo GROUP BY lower(a);
*** BOOM AGAIN ***
Marc G. Fournier [Wed, 5 Aug 1998 04:47:54 +0000 (04:47 +0000)]
From: Dr. Michael Meskes <meskes@online-club.de>
So this should finally get cursors working. There was an ugly bug in it.
Bruce Momjian [Tue, 4 Aug 1998 18:42:39 +0000 (18:42 +0000)]
OR processing cleanup.
Bruce Momjian [Tue, 4 Aug 1998 18:29:41 +0000 (18:29 +0000)]
Add Indices display to \d command.
Bruce Momjian [Tue, 4 Aug 1998 17:37:48 +0000 (17:37 +0000)]
Fix encoding grammer problem.
Bruce Momjian [Tue, 4 Aug 1998 16:44:31 +0000 (16:44 +0000)]
MergeSort was sometimes called mergejoin and was confusing. Now
it is now only mergejoin.
Bruce Momjian [Tue, 4 Aug 1998 15:00:28 +0000 (15:00 +0000)]
EXPLAIN VERBOSE prints the plan, and now pretty-prints the plan to
the postmaster log file.
Bruce Momjian [Tue, 4 Aug 1998 04:50:15 +0000 (04:50 +0000)]
Make EXPLAIN show output more clearly.
Bruce Momjian [Tue, 4 Aug 1998 00:42:14 +0000 (00:42 +0000)]
Update flow chart.
Bruce Momjian [Mon, 3 Aug 1998 19:41:35 +0000 (19:41 +0000)]
Update mark/reset index code for multiple indexes, (OR code).
Thanks for Vadim for fixes.
Bruce Momjian [Mon, 3 Aug 1998 05:54:30 +0000 (05:54 +0000)]
Fix typo in man pages.
Bruce Momjian [Mon, 3 Aug 1998 05:49:24 +0000 (05:49 +0000)]
Cleanup of OR processing.
Thomas G. Lockhart [Sun, 2 Aug 1998 13:34:26 +0000 (13:34 +0000)]
Allows the following query to succeed: "SELECT NULL ORDER BY 1;"
There are three or four cases in transformSortClause() and I had fixed
only one case for UNION. A second case is now fixed, in the same way; I
assigned INT4OID to the column type for the "won't actually happen"
sort. Didn't want to skip the code entirely, since the backend needs to
_try_ a sort to get the NULLs right. I'm not certain under what
circumstances the other cases are invoked and these are not yet
fixed up, though perhaps they don't need to be...
Bruce Momjian [Sun, 2 Aug 1998 07:10:38 +0000 (07:10 +0000)]
Fix for OR handling with multiple indexes.
Bruce Momjian [Sun, 2 Aug 1998 00:34:05 +0000 (00:34 +0000)]
run autoconf
Bruce Momjian [Sun, 2 Aug 1998 00:21:39 +0000 (00:21 +0000)]
Fix MULTIBYTE typo.
Bruce Momjian [Sat, 1 Aug 1998 22:57:41 +0000 (22:57 +0000)]
Fix duplicate inside initdb.sh
Bruce Momjian [Sat, 1 Aug 1998 22:44:55 +0000 (22:44 +0000)]
Cleanup of OR fix.
Bruce Momjian [Sat, 1 Aug 1998 22:12:13 +0000 (22:12 +0000)]
Allow index use with OR clauses.
Marc G. Fournier [Sat, 1 Aug 1998 19:30:29 +0000 (19:30 +0000)]
Adrian Hall reported a problem to me that snprintf() doesn't exist in, at
least, Solaris 2.5.1. We use it in backend/utils/adt/int8.c.
Add a check to configure so that we see if it exists or not, and, if not,
compile in snprintf.c from backend/port, which was taken from, and falls under
the same Berkeley license as us, the FreeBSD libc/stdio ...
Vadim B. Mikheev [Sat, 1 Aug 1998 15:26:38 +0000 (15:26 +0000)]
Lmgr cleanup, new locking modes for LLL.
Vadim B. Mikheev [Fri, 31 Jul 1998 15:10:40 +0000 (15:10 +0000)]
Matching clauses to functional indices was broken, now fixed.
Vadim B. Mikheev [Thu, 30 Jul 1998 05:05:05 +0000 (05:05 +0000)]
Fix scan adjustment.
Thomas G. Lockhart [Wed, 29 Jul 1998 15:23:57 +0000 (15:23 +0000)]
Removed from distribution. Completely converted to SGML.
This was the Postgres95 documentation written by Jolly and Chu...
Thomas G. Lockhart [Wed, 29 Jul 1998 07:08:00 +0000 (07:08 +0000)]
Use libpq only once to avoid file naming conflicts in the html output.
Thomas G. Lockhart [Wed, 29 Jul 1998 07:01:58 +0000 (07:01 +0000)]
Somehow had substituted this file for the contents of allfiles.sgml.
This should be the right thing now...
Thomas G. Lockhart [Wed, 29 Jul 1998 06:50:04 +0000 (06:50 +0000)]
Add id field to chapter and book tags to allow output file names
to be meaningful.
Thomas G. Lockhart [Wed, 29 Jul 1998 06:35:19 +0000 (06:35 +0000)]
Add port information for NS32532.
Fix Id tag in chapter headings to allow meaningful file names in html.
Thomas G. Lockhart [Wed, 29 Jul 1998 06:33:37 +0000 (06:33 +0000)]
Include information on documentation products and sources.
Include tables of possible source files throughout tree.
Thomas G. Lockhart [Wed, 29 Jul 1998 06:32:47 +0000 (06:32 +0000)]
Add missing information.
Fix examples for regex operators.
Thomas G. Lockhart [Wed, 29 Jul 1998 06:29:04 +0000 (06:29 +0000)]
Use option to jade to produce html file names from id field in tag.
Thomas G. Lockhart [Wed, 29 Jul 1998 06:23:26 +0000 (06:23 +0000)]
Add new files from Oliver Elphick.
Thomas G. Lockhart [Wed, 29 Jul 1998 06:21:57 +0000 (06:21 +0000)]
Fix minor end-of-file formatting.
Thomas G. Lockhart [Wed, 29 Jul 1998 06:13:48 +0000 (06:13 +0000)]
Information moved to doc/src/sgml/release.sgml.
Thomas G. Lockhart [Wed, 29 Jul 1998 06:07:54 +0000 (06:07 +0000)]
Information moved to doc/src/sgml/geqo.sgml.
Vadim B. Mikheev [Mon, 27 Jul 1998 19:38:40 +0000 (19:38 +0000)]
Use Snapshot in heap access methods.
Marc G. Fournier [Mon, 27 Jul 1998 03:21:58 +0000 (03:21 +0000)]
Missed a few files in the last round of commits from Tatsuo, as well
as needed to run autoconf ...
Marc G. Fournier [Sun, 26 Jul 1998 04:31:41 +0000 (04:31 +0000)]
From: t-ishii@sra.co.jp
As Bruce mentioned, this is due to the conflict among changes we made.
Included patches should fix the problem(I changed all MB to
MULTIBYTE). Please let me know if you have further problem.
P.S. I did not include pathces to configure and gram.c to save the
file size(configure.in and gram.y modified).
Marc G. Fournier [Sun, 26 Jul 1998 04:22:42 +0000 (04:22 +0000)]
I missed addding initdb.sh ...
Bruce Momjian [Sun, 26 Jul 1998 02:17:55 +0000 (02:17 +0000)]
Update flowchart. Add arrow.
Bruce Momjian [Sun, 26 Jul 1998 01:18:09 +0000 (01:18 +0000)]
Fix compile error. Make transaction/work optional on all transaction
statements. More cleanups of psql help. Fix for shift/reduce on
UNION in subselect.
Bruce Momjian [Sat, 25 Jul 1998 00:17:30 +0000 (00:17 +0000)]
Update psql help syntax to remove <> and uppercaese keywords.
Bruce Momjian [Fri, 24 Jul 1998 16:44:04 +0000 (16:44 +0000)]
Add mention of actual SQL command to create*/destroy* commands.
Bruce Momjian [Fri, 24 Jul 1998 16:36:56 +0000 (16:36 +0000)]
Add mention of SQL commands to create/destroy man pages.
Bruce Momjian [Fri, 24 Jul 1998 16:10:19 +0000 (16:10 +0000)]
Add mention of psql and template1 to create*, destroy* command
manual pages.
Bruce Momjian [Fri, 24 Jul 1998 15:54:10 +0000 (15:54 +0000)]
Fix HAVING patch missing cast.
Bruce Momjian [Fri, 24 Jul 1998 04:03:10 +0000 (04:03 +0000)]
I'm sorry, but I think I introduced a little bug with my last patch.
Everyone using an [NOT] EXISTS subquery will have noticed that
already.
The bug is in "subselect.c" in the function "SS_process_sublinks()".
Here the whole function as it *SHOULD BE*:
Stephan
Marc G. Fournier [Fri, 24 Jul 1998 03:32:46 +0000 (03:32 +0000)]
I really hope that I haven't missed anything in this one...
From: t-ishii@sra.co.jp
Attached are patches to enhance the multi-byte support. (patches are
against 7/18 snapshot)
* determine encoding at initdb/createdb rather than compile time
Now initdb/createdb has an option to specify the encoding. Also, I
modified the syntax of CREATE DATABASE to accept encoding option. See
README.mb for more details.
For this purpose I have added new column "encoding" to pg_database.
Also pg_attribute and pg_class are changed to catch up the
modification to pg_database. Actually I haved added pg_database_mb.h,
pg_attribute_mb.h and pg_class_mb.h. These are used only when MB is
enabled. The reason having separate files is I couldn't find a way to
use ifdef or whatever in those files. I have to admit it looks
ugly. No way.
* support for PGCLIENTENCODING when issuing COPY command
commands/copy.c modified.
* support for SQL92 syntax "SET NAMES"
See gram.y.
* support for LATIN2-5
* add UNICODE regression test case
* new test suite for MB
New directory test/mb added.
* clean up source files
Basic idea is to have MB's own subdirectory for easier maintenance.
These are include/mb and backend/utils/mb.
Byron Nikolaidis [Wed, 22 Jul 1998 23:47:48 +0000 (23:47 +0000)]
Version 06-30-0248
Vadim B. Mikheev [Wed, 22 Jul 1998 05:48:59 +0000 (05:48 +0000)]
Fix _lo_commit: core if LO memmory context was not created (LO were not used?)
Vadim B. Mikheev [Tue, 21 Jul 1998 06:17:39 +0000 (06:17 +0000)]
VariableCache (next XID generator) is placed in shmem.
Bruce Momjian [Tue, 21 Jul 1998 04:48:31 +0000 (04:48 +0000)]
Update.