OSDN Git Service
Tom Lane [Tue, 4 Apr 2000 21:44:40 +0000 (21:44 +0000)]
Fix bug noted by Bruce: FETCH in an already-aborted transaction block
would crash, due to premature invocation of SetQuerySnapshot(). Clean
up problems with handling of multiple queries by splitting
pg_parse_and_plan into two routines. The old code would not, for
example, do the right thing with END; SELECT... submitted in one query
string when it had been in transaction abort state, because it'd decide
to skip planning the SELECT before it had executed the END. New
arrangement is simpler and doesn't force caller to plan if only
parse+rewrite is needed.
Bruce Momjian [Tue, 4 Apr 2000 19:42:36 +0000 (19:42 +0000)]
Remove FETCH syntax heading from MOVE manual page.
Bruce Momjian [Tue, 4 Apr 2000 19:29:18 +0000 (19:29 +0000)]
Remove from FETCH manual page:
Once all rows are fetched, every other fetch access
returns no rows.
Bruce Momjian [Tue, 4 Apr 2000 13:51:38 +0000 (13:51 +0000)]
No ID file needed.
Bruce Momjian [Tue, 4 Apr 2000 13:49:30 +0000 (13:49 +0000)]
Update QNX FAQ
Tom Lane [Tue, 4 Apr 2000 05:22:46 +0000 (05:22 +0000)]
Add a check to pg_dump to see whether backend is same version as pg_dump.
If not, abort by default. Abort can be prevented by using -i or
--ignore-version switch.
Bruce Momjian [Tue, 4 Apr 2000 04:07:53 +0000 (04:07 +0000)]
Update regression for bsdi.
Tom Lane [Tue, 4 Apr 2000 03:00:16 +0000 (03:00 +0000)]
Correct off-by-one error in strncat() usage.
Tom Lane [Tue, 4 Apr 2000 02:30:52 +0000 (02:30 +0000)]
When rewriting an aggregate introduced into WHERE, allow agg argument to
be an expression not just a simple Var, so long as only one table is
referenced (so that code isn't really any more difficult than before).
This whole thing is still fundamentally bogus, but at least we can accept
a few more cases than before.
Tom Lane [Tue, 4 Apr 2000 01:47:59 +0000 (01:47 +0000)]
For no good reason at all, libplpgsql was getting linked with libpq.
Since plpgsql is a backend extension, it doesn't need libpq...
Tom Lane [Tue, 4 Apr 2000 01:21:48 +0000 (01:21 +0000)]
Fix extremely nasty little bug observed when a sub-SELECT appears in
WHERE in a place where it can be part of a nestloop inner indexqual.
As the code stood, it put the same physical sub-Plan node into both
indxqual and indxqualorig of the IndexScan plan node. That confused
later processing in the optimizer (which expected that tracing the
subPlan list would visit each subplan node exactly once), and would
probably have blown up in the executor if the planner hadn't choked first.
Fix by making the 'fixed' indexqual be a complete deep copy of the
original indexqual, rather than trying to share nodes below the topmost
operator node. This had further ramifications though, because we were
making the aforesaid list of sub-Plan nodes during SS_process_sublinks
which is run before construction of the 'fixed' indexqual, meaning that
the copy of the sub-Plan didn't show up in that list. Fix by rearranging
logic so that the sub-Plan list is built by the final set_plan_references
pass, not in SS_process_sublinks. This may sound like a mess, but it's
actually a good deal cleaner now than it was before, because we are no
longer dependent on the assumption that planning will never make a copy
of a sub-Plan node.
Thomas G. Lockhart [Tue, 4 Apr 2000 01:07:54 +0000 (01:07 +0000)]
Augment the function call map logic with code from Tom Lane.
Should be more robust to overflows.
Pass through an unmapped function unchanged, rather than rejecting it.
Add a few more functions, but comment out those which can go through as-is.
Can be used with contrib/odbc/ package, though that isn't committed yet.
Bruce Momjian [Mon, 3 Apr 2000 20:56:40 +0000 (20:56 +0000)]
Hi,
here is an updated version of the bit type with a bugfix and all the necessa
ry
SQL functions defined. This should replace what is currently in contrib. I'd
appreciate any comments on what is there.
Kind regards,
Adriaan
Michael Meskes [Mon, 3 Apr 2000 19:34:26 +0000 (19:34 +0000)]
*** empty log message ***
Tom Lane [Mon, 3 Apr 2000 00:33:54 +0000 (00:33 +0000)]
Looks like Peter forgot to commit configure with configure.in.
Tom Lane [Sun, 2 Apr 2000 22:59:40 +0000 (22:59 +0000)]
Copy-editing, mostly.
Tom Lane [Sun, 2 Apr 2000 22:28:54 +0000 (22:28 +0000)]
Mention DST boundaries as a possible cause of regress test failure.
Michael Meskes [Sun, 2 Apr 2000 08:51:44 +0000 (08:51 +0000)]
*** empty log message ***
Peter Eisentraut [Fri, 31 Mar 2000 22:03:52 +0000 (22:03 +0000)]
Adjusted for new (again!) config.guess output under QNX
Tom Lane [Fri, 31 Mar 2000 19:39:22 +0000 (19:39 +0000)]
Fix low-probability bug in relcache startup: write_irels wrote the
pg_internal.init file in-place, which meant that if another backend
started at about the same time, it might read the incomplete file.
init_irels tries to guard against that, but I have now seen a crash
due to reading bad data from a partly-written file. (This may indicate
a kernel bug on my platform? Not sure.) Anyway, clearly the safest
course is to write the new pg_internal.init file under a unique temporary
filename, and rename it into place only after it's all written.
Tom Lane [Fri, 31 Mar 2000 17:45:32 +0000 (17:45 +0000)]
Correct typo.
Tom Lane [Fri, 31 Mar 2000 17:45:00 +0000 (17:45 +0000)]
Improve wording a little bit.
Tom Lane [Fri, 31 Mar 2000 17:18:26 +0000 (17:18 +0000)]
Update index cost estimation docs to final 7.0 scheme.
Tom Lane [Fri, 31 Mar 2000 15:00:14 +0000 (15:00 +0000)]
Minor copy-editing.
Tom Lane [Fri, 31 Mar 2000 14:57:05 +0000 (14:57 +0000)]
CREATE AGGREGATE ref page claimed that a noise-word AS was allowed in
the syntax ... but gram.y doesn't think so.
Bruce Momjian [Fri, 31 Mar 2000 14:14:36 +0000 (14:14 +0000)]
I have tested the beta3 on WinNT and here are the results:
- I was unable to compile ecpg due to the ":=" instead of "=" in defining
LIBPQDIR and some other variables in Makefile.global.in
- pg_id (and also pg_encoding) executable was not removed during "make
clean" - there was no $(X) appended to the executable name for rm
- I have added result for int2, int4, float8 and geometry regression tests
- int2, int2 - yet another message for too large numbers ;-)
- float8 - it is problably a bug in the newlib C library - it has no
error message for numbers with exponent -400
- geometry - differences in precision of float numbers
- I have added appropriate lines into resultmap file
- I have modified the script regress.sh to use "case" statement when testing
the hostname. For cygwin the script is called with "i686-pc-cygwin" (on my
machine) as a parameter and this was not catched with the "if" statement.
The check was done for PORTNAME (win) and not HOSTNAME (i.86-pc-cygwin*).
The patch for described modifications is included.
All this modifications can be applied to "current" tree too.
The compilation was done on CygwinB20.1 with gcc 2.95, cygipc library 1.05.
The binaries were able to run also on the newest development snapshot
(2000-03-25).
Dan
Bruce Momjian [Fri, 31 Mar 2000 14:05:07 +0000 (14:05 +0000)]
Fix pgaccess __wish__ thing.
Tatsuo Ishii [Fri, 31 Mar 2000 13:07:03 +0000 (13:07 +0000)]
Regenerate configure (for configure.in version 1.305)
Tatsuo Ishii [Fri, 31 Mar 2000 13:04:31 +0000 (13:04 +0000)]
Fix configure.in. Remove SJIS, BIG5, WIN1250 from the selection list of
multibyte encodings. These cannot be used as a backend/database encoding
(OK to use as a client encoding).
Bruce Momjian [Fri, 31 Mar 2000 11:30:10 +0000 (11:30 +0000)]
Add new pgaccess languages.
Bruce Momjian [Fri, 31 Mar 2000 11:22:36 +0000 (11:22 +0000)]
Update pgaccess to 0.98.5
Hiroshi Inoue [Fri, 31 Mar 2000 08:16:33 +0000 (08:16 +0000)]
*** empty log message ***
Thomas G. Lockhart [Fri, 31 Mar 2000 06:17:52 +0000 (06:17 +0000)]
Include all information from the old man pages.
Thomas G. Lockhart [Fri, 31 Mar 2000 06:17:21 +0000 (06:17 +0000)]
Use a uniform chapter title for all of the procedural languages.
Tom Lane [Fri, 31 Mar 2000 05:36:54 +0000 (05:36 +0000)]
Update from Andreas Kardos.
Tom Lane [Fri, 31 Mar 2000 05:30:31 +0000 (05:30 +0000)]
Add expected output for netbsd, per report from Patrick Welche.
Tom Lane [Fri, 31 Mar 2000 05:13:39 +0000 (05:13 +0000)]
Add resultmap entries for datetime-related tests on alpha-dec-osf.
Tom Lane [Fri, 31 Mar 2000 05:11:13 +0000 (05:11 +0000)]
Update horology-solaris-1947.out per results from SL Baur. These are
actually from an alpha-dec-osf machine, but as far as can be told the
two platforms have the same ideas about past DST rules.
Tom Lane [Fri, 31 Mar 2000 05:03:39 +0000 (05:03 +0000)]
Update alpha templates as suggested by SL Baur and Adriaan Joubert.
Make similar changes to hpux templates. Might want to do the same for
other foo_cc and foo_gcc pairs, but will desist until I hear from
someone who uses those platforms.
Tom Lane [Fri, 31 Mar 2000 05:00:36 +0000 (05:00 +0000)]
Use the CXXFLAGS exported by configure (amazingly, we weren't before!)
and do not arbitrarily pull in CFLAGS instead. This caters to platforms
where the C++ compiler does not like all the same switches the C compiler
wants.
Thomas G. Lockhart [Fri, 31 Mar 2000 03:27:42 +0000 (03:27 +0000)]
Minor markup changes. Refer to (".../catalog") in the emacs hints.
Thomas G. Lockhart [Fri, 31 Mar 2000 03:26:21 +0000 (03:26 +0000)]
Add Bruce's pgeasy doc from the man page.
Thomas G. Lockhart [Fri, 31 Mar 2000 03:20:17 +0000 (03:20 +0000)]
Document recent changes in syntax, including examples.
Tom Lane [Fri, 31 Mar 2000 02:43:31 +0000 (02:43 +0000)]
Get rid of SetBufferWriteMode(), which was an accident waiting to happen.
In the event of an elog() while the mode was set to immediate write,
there was no way for it to be set back to the normal delayed write.
The mechanism was a waste of space and cycles anyway, since the only user
was varsup.c, which could perfectly well call FlushBuffer directly.
Now it does just that, and the notion of a write mode is gone.
Tom Lane [Fri, 31 Mar 2000 02:11:03 +0000 (02:11 +0000)]
New coding for SET provoked a 'var might be used uninitialized' warning
from gcc. Which wasn't actually a code bug, but I don't like warnings.
Bruce Momjian [Fri, 31 Mar 2000 01:41:27 +0000 (01:41 +0000)]
Update make_mkid for mkid version 4.0.
Tom Lane [Thu, 30 Mar 2000 23:42:34 +0000 (23:42 +0000)]
Make discussion of names clearer and more accurate.
Thomas G. Lockhart [Thu, 30 Mar 2000 22:34:29 +0000 (22:34 +0000)]
Remove PL/perl language chapter, since it shows up in the User's Guide.
Thomas G. Lockhart [Thu, 30 Mar 2000 22:22:41 +0000 (22:22 +0000)]
Accumulated fixups.
Add some chapters on new topics.
Change to referencing OASIS/Docbook v3.1 rather than Davenport/Docbook v3.0
Grepped for and fixed apparent tag mangling from emacs
"Normalize" operation. Should be the last of those.
Thomas G. Lockhart [Thu, 30 Mar 2000 22:19:47 +0000 (22:19 +0000)]
Renamed from bug-reports.sgml. Not all problems are bugs ;)
Thomas G. Lockhart [Thu, 30 Mar 2000 22:18:54 +0000 (22:18 +0000)]
Writeup from Tom Lane on how costs are estimated.
Thomas G. Lockhart [Thu, 30 Mar 2000 22:17:50 +0000 (22:17 +0000)]
Hints on how to fill a database.
Thomas G. Lockhart [Thu, 30 Mar 2000 22:17:16 +0000 (22:17 +0000)]
Add reference to the original Postgres papers at a UCB web site.
Thomas G. Lockhart [Thu, 30 Mar 2000 22:14:47 +0000 (22:14 +0000)]
Small emacs fixup for catalogs.
Thomas G. Lockhart [Thu, 30 Mar 2000 22:13:30 +0000 (22:13 +0000)]
Split off language-specific docs to their own chapters.
Add PL/perl docs from Mark Hollomon.
Put language-specific docs into the User's Guide, and move the rest
to the Programmer's Guide.
Michael Meskes [Thu, 30 Mar 2000 11:41:46 +0000 (11:41 +0000)]
*** empty log message ***
Tom Lane [Thu, 30 Mar 2000 07:49:39 +0000 (07:49 +0000)]
alpha-dec-osf host pattern must become alpha.*-dec-osf to match
newer Alpha platforms.
Tom Lane [Thu, 30 Mar 2000 07:46:00 +0000 (07:46 +0000)]
Update alternate float8 output files to match current float8.sql.
Tom Lane [Thu, 30 Mar 2000 07:13:20 +0000 (07:13 +0000)]
Update numeric_big regress test for 7.0. This has apparently been
broken almost since the word go ... I guess no one ever ran it ...
Thomas G. Lockhart [Thu, 30 Mar 2000 06:02:36 +0000 (06:02 +0000)]
Enable more flexible syntax for the SET command. Now allows single floats,
single integers, and lists of names, without surrounding them with quotes.
Remove all tokens which are defined as operators from ColID and ColLabel
to avoid precedence confusion. Thanks to Tom Lane for catching this.
Tom Lane [Thu, 30 Mar 2000 05:57:30 +0000 (05:57 +0000)]
Makefile should not pre-empt user's decision about whether
to use -g ... especially not when this can cause real problems on some
platforms ...
Tom Lane [Thu, 30 Mar 2000 05:49:47 +0000 (05:49 +0000)]
Unixware fix for new config.guess output,
from Bill.Allie@mug.org.
Tom Lane [Thu, 30 Mar 2000 05:30:42 +0000 (05:30 +0000)]
Make use of configure symbols for unportable constructs. Make
inclusions of system headers more consistent.
Tom Lane [Thu, 30 Mar 2000 05:29:44 +0000 (05:29 +0000)]
Autoconf.
Tom Lane [Thu, 30 Mar 2000 05:29:21 +0000 (05:29 +0000)]
Add configure checks to see if 'using namespace std' and
'#include <string>' work in the local C++ compiler.
Thomas G. Lockhart [Thu, 30 Mar 2000 05:07:48 +0000 (05:07 +0000)]
Change cross reference to CREATE AGGREGATE to a citation, since the
Programmer's Guide does not have access to the reference page.
Normalize markup.
Tom Lane [Thu, 30 Mar 2000 02:59:14 +0000 (02:59 +0000)]
Remove extraneous semicolon.
Tom Lane [Thu, 30 Mar 2000 02:51:12 +0000 (02:51 +0000)]
Replace refs to v6.6 with v7.0.
Tom Lane [Thu, 30 Mar 2000 00:53:30 +0000 (00:53 +0000)]
Tweak indexscan cost estimation: round estimated # of tuples visited up
to next integer. Previously, if selectivity was small, we could compute
very tiny scan cost on the basis of estimating that only 0.001 tuple
would be fetched, which is silly. This naturally led to some rather
silly plans...
Peter Eisentraut [Wed, 29 Mar 2000 16:12:31 +0000 (16:12 +0000)]
Added latest config.guess and config.sub
Thomas G. Lockhart [Wed, 29 Mar 2000 03:57:18 +0000 (03:57 +0000)]
Change rules for interpreting date/time input to disallow 1 and 3 character
years. Rejects dates like '0.085', which were accepted previously.
Thomas G. Lockhart [Wed, 29 Mar 2000 03:57:11 +0000 (03:57 +0000)]
Change rules for interpreting date/time input to disallow 1 and 3 character
years. Rejects dates like '0.085', which were accepted previously.
Minor mods of ODBC markup.
Thomas G. Lockhart [Tue, 28 Mar 2000 14:35:27 +0000 (14:35 +0000)]
Fix markup for terminology.
Thomas G. Lockhart [Tue, 28 Mar 2000 14:35:13 +0000 (14:35 +0000)]
Add mention of join syntax to release notes, and reformat a few lines.
Update porting info. Still need a lot of platforms tested.
Fix small errors in markup.
Thomas G. Lockhart [Tue, 28 Mar 2000 14:16:07 +0000 (14:16 +0000)]
Update SGML catalog references to DocBook 3.1 on FreeBSD.
Matches postgresql.org/hub.org environment.
Tom Lane [Tue, 28 Mar 2000 03:39:50 +0000 (03:39 +0000)]
Small improvements to user's guide description of arrays.
Tom Lane [Tue, 28 Mar 2000 02:59:55 +0000 (02:59 +0000)]
Oops, missed some < and > that need to be < and > ...
Tom Lane [Tue, 28 Mar 2000 02:53:02 +0000 (02:53 +0000)]
Clean up major bogosity in description of creating a new index operator
class. This chapter was apparently sewed together from several
inconsistent examples... and perhaps from old docs that no longer
apply at all.
Tom Lane [Tue, 28 Mar 2000 02:51:09 +0000 (02:51 +0000)]
Add a little to index operator class discussion.
Tom Lane [Tue, 28 Mar 2000 02:49:19 +0000 (02:49 +0000)]
Fix some bogosity in the tutorial examples.
Bruce Momjian [Mon, 27 Mar 2000 22:39:13 +0000 (22:39 +0000)]
Allow compile to finish even if plperl fails, which it does now.
Peter Eisentraut [Mon, 27 Mar 2000 21:11:37 +0000 (21:11 +0000)]
Fixed bug with repeated \e in psql (failed to clear buffers correctly)
Thomas G. Lockhart [Mon, 27 Mar 2000 17:14:43 +0000 (17:14 +0000)]
Start updating for the v7.0 release.
Use "generic functions" for math and other routines.
Use SQL92 "type 'literal'" syntax rather than Postgres "'literal'::type".
Thomas G. Lockhart [Mon, 27 Mar 2000 17:12:06 +0000 (17:12 +0000)]
Allow full type names in CREATE FUNCTION arguments and return type.
Move CREATE FUNCTION/WITH clause to end of statement to get around
shift/reduce conflicts with type names containing "WITH".
Add lots of tokens as allowed ColId's and/or ColLabel's,
so this should be a complete set for the v7.0 release.
Thomas G. Lockhart [Mon, 27 Mar 2000 17:07:48 +0000 (17:07 +0000)]
Fix up comments where had been uglified by the automated reformatter.
Thomas G. Lockhart [Mon, 27 Mar 2000 17:04:47 +0000 (17:04 +0000)]
Change form of query which used "DISTINCT ON" to help support primary keys.
We still have an internal limit in the ODBC code of 8 columns per key,
but this should lay the groundwork for resolving that.
Includes reformulated query from Tom Lane.
Bruce Momjian [Mon, 27 Mar 2000 16:07:59 +0000 (16:07 +0000)]
Add html FAQ and FAQ_DEV sources
Tatsuo Ishii [Mon, 27 Mar 2000 02:12:03 +0000 (02:12 +0000)]
Enhance pg_ctl so that it prints error messages from postmaster
if it fails to start up it (this is only vaild if -w is given).
Tom Lane [Sun, 26 Mar 2000 19:47:17 +0000 (19:47 +0000)]
Update obsolete statement that indexes can have only 7 columns.
Reorganize description of index features for more clarity.
Tom Lane [Sun, 26 Mar 2000 19:45:21 +0000 (19:45 +0000)]
Update/improve documentation about creating aggregate functions.
Tom Lane [Sun, 26 Mar 2000 19:43:58 +0000 (19:43 +0000)]
nodeAgg has always been willing to accept an aggregate with a finalFunc
and only one transition state, but the CREATE AGGREGATE code rejected
this combination.
Peter Eisentraut [Sun, 26 Mar 2000 18:32:30 +0000 (18:32 +0000)]
Updated user's guide to match new psql's output format
Fixed bug in createdb/alternative location
Tom Lane [Sun, 26 Mar 2000 07:04:54 +0000 (07:04 +0000)]
Improve descriptions of postmaster switches.
Tom Lane [Sun, 26 Mar 2000 07:03:38 +0000 (07:03 +0000)]
Rearrange steps in recommended install procedure to something more
reasonable, ie configure and build first, then optionally run regress
tests using new parallel (non-installed) test method, and only then
backup and kill old installation.
Tom Lane [Sun, 26 Mar 2000 07:01:19 +0000 (07:01 +0000)]
Add description of new parallel regression testing method; update
procedure to cover running either test script; other minor improvements.
Tom Lane [Sun, 26 Mar 2000 06:59:31 +0000 (06:59 +0000)]
Update descriptions of configure's options, and document how to specify
multiple directories for --with-includes or --with-libraries.
Tom Lane [Sun, 26 Mar 2000 06:58:17 +0000 (06:58 +0000)]
Fix erroneous claim that 'postmaster -S' leaves the postmaster running
in the foreground --- in fact, it auto-detaches.
Bruce Momjian [Sun, 26 Mar 2000 03:40:37 +0000 (03:40 +0000)]
Update initdb flags.
Tatsuo Ishii [Sun, 26 Mar 2000 02:35:01 +0000 (02:35 +0000)]
Change resultmap:
powerpc-unknown-linux-gnu --> powerpc-unknown-linux-gnulibc1
this is because for new config.guess.
Tatsuo Ishii [Sun, 26 Mar 2000 02:28:07 +0000 (02:28 +0000)]
New platform specific geometry regression expected file for
powerpc-linux-gnulibc1.