Commit Graph

22167 Commits

Author SHA1 Message Date
Bruce Momjian a4a6ee4a5d Check for "msys" so it doesn't use 'con' by checking for an evironment
variable.
2006-03-05 05:33:12 +00:00
Bruce Momjian 5fde861375 Improve STRINGS_H macro test for MSVC extensions.
Add DLLIMPORT for V1 headers, in case Win32 don't export all symbols.
2006-03-05 04:43:57 +00:00
Tom Lane c1bb2877b2 Improve OS X shared-memory documentation: fix typos and provide a usable
example of /etc/sysctl.conf contents.
2006-03-05 03:50:44 +00:00
Tom Lane 5b8ac71042 Support include directives in postgresql.conf.
Patch by Joachim Wieland, somewhat reworked for clarity and portability.
2006-03-04 22:19:31 +00:00
Tom Lane 60d3c9fdf4 Declare the arguments of AllocateFile() as const char *, not char *.
This is consistent with the standard definition of fopen().
2006-03-04 21:32:47 +00:00
Tom Lane 2689abf078 Incorporate a couple of recent tuplesort.c improvements into tuplestore.c.
In particular, ensure that enlargement of the memtuples[] array doesn't
fall foul of MaxAllocSize when work_mem is very large, and don't bother
enlarging it if that would force an immediate switch into 'tape' mode anyway.
2006-03-04 19:30:12 +00:00
Tom Lane 20bdc71369 Prevent lazy_space_alloc from making requests that exceed MaxAllocSize,
per report from Stefan Kaltenbrunner.
2006-03-04 19:09:09 +00:00
Tom Lane 80cadb303c Prevent sorting from requesting a SortTuple array that exceeds MaxAllocSize;
we'll go over to disk-based sort if we reach that limit.
This fixes Stefan Kaltenbrunner's observation that sorting can suffer an
'invalid memory alloc request size' failure when sort_mem is set large
enough.  It's unfortunately not so easy to fix in 8.1 ...
2006-03-04 19:05:06 +00:00
Tatsuo Ishii b3d0442ab3 Tighten up SJIS byte sequence check. Now we reject invalid SJIS byte
sequence such as "0x95 0x27". Patches from Akio Ishida.
Also update copyright notice.
2006-03-04 10:57:35 +00:00
Bruce Momjian 18bc9ea561 Add:
> * Allow FSM page return free space based on table clustering, to assist
>   in maintaining clustering?
2006-03-04 05:02:03 +00:00
Bruce Momjian 3bce31f613 > gettimeofday.c:35: warning: integer constant is too large for "long"
> type

Wouldn't it be better to use the UINT64CONST macro?  I realize this
file is Windows-only, but we do worry about more than one compiler
on that platform.

Kris Jurka
2006-03-04 04:44:07 +00:00
Bruce Momjian f1d7f80c1a > It doesn't say that only the listed commands acquire ACCESS EXCLUSIVE,
> just that certain commands do.  TRUNCATE isn't shown.

Patch against HEAD to add TRUNCATE to the list of commands that aquire
ACCESS EXCLUSIVE.

Jim C. Nasby, Sr.
2006-03-04 04:41:36 +00:00
Bruce Momjian 64e7c8a951 Use DEVTTY as 'con' on Win32 as a replacement for /dev/tty. 2006-03-04 04:30:41 +00:00
Bruce Momjian 26d6054128 That was a typo in my comment before the code (the nutshell
descriptions after the code are correct). Only shmmax needs to be
multiples of the page size (at least, that's how I interpret the
Darwin code).

Chris Campbell
2006-03-04 03:47:29 +00:00
Bruce Momjian 295615a6ca This patch fixes this warning.
gettimeofday.c:35: warning: integer constant is too large for "long"
type

Kris Jurka
2006-03-03 23:59:14 +00:00
Bruce Momjian ef3f7c3f74 Avoid trying to open /dev/tty on Win32. Some Win32 systems have
/dev/tty, but it isn't a device file and doesn't work as expected.

This fixes a known bug where psql does not prompt for a password on some
Win32 systems.

Backpatch to 8.1.X.

Robert Kinberg
2006-03-03 23:49:12 +00:00
Tom Lane decdaf3592 Improve pg_dump and psql to use libpq's newer COPY support routines,
instead of the old deprecated ones.
Volkan Yazici, with some editorializing by moi.
2006-03-03 23:38:30 +00:00
Bruce Momjian 0b1b010c12 Fixes for Win32-client only compiles.
Hiroshi Saito
2006-03-03 23:11:48 +00:00
Bruce Momjian 43e9bab94a Rename Online Backup to Continuous Archiving. 2006-03-03 22:02:08 +00:00
Bruce Momjian bf8337b8af Update ipcclean to use try 'id' first for root check. 2006-03-03 21:52:37 +00:00
Bruce Momjian 28edbdb7be Add workaround so MSVC doesn't try to load strings.h, which it doesn't
have.  This happens when MSVC uses pg_config.h generated by MinGW.

Per report from Charles F. I. Savage
2006-03-03 21:35:46 +00:00
Tom Lane 523adeb111 Teach PQcmdTuples() that a COPY command tag might contain a row count,
and tighten up its sanity checking of the tag as a safety measure.
Volkan Yazici.
2006-03-03 20:57:32 +00:00
Bruce Momjian 502e9aefdc Clarify macro layout for win32 IMPORT. 2006-03-03 20:52:36 +00:00
Bruce Momjian f4a4755734 Done:
> * -Allow TRUNCATE ... CASCADE/RESTRICT
2006-03-03 20:43:00 +00:00
Bruce Momjian c776fba1e1 Done:
< 	o %Have COPY return the number of rows loaded/unloaded?
> 	o -Have COPY return the number of rows loaded/unloaded?
2006-03-03 20:37:29 +00:00
Tom Lane 023570f5e3 Make the COPY command return a command tag that includes the number of
rows copied.  Backend side of Volkan Yazici's recent patch, with
corrections and documentation.
2006-03-03 19:54:10 +00:00
Tom Lane 4e086f7cb5 Dept. of second thoughts: rejigger the TRUNCATE ... CASCADE patch so that
relations are still checked for permissions etc as soon as they are
opened.  The original form of the patch could hold exclusive lock for a
long time on relations that the user doesn't even have permissions to
access, let alone truncate.
2006-03-03 18:25:14 +00:00
Bruce Momjian a6add72ac3 In ipcclean, check LOGNAME only if USER is not set.
Fixes problem with 'su' on some platforms.
2006-03-03 16:49:21 +00:00
Neil Conway 587bc81887 Fix a typo. 2006-03-03 04:31:07 +00:00
Tom Lane 984a6ced3e Add CASCADE option to TRUNCATE. Joachim Wieland 2006-03-03 03:30:54 +00:00
Bruce Momjian 2a0ba3f8dd Add:
> * Allow FSM to return free space toward the beginning of the heap file,
>   in hopes that empty pages at the end can be truncated by VACUUM
2006-03-03 03:13:04 +00:00
Bruce Momjian b35440eae8 Appended is a small documentation patch that adds a note to the CREATE
ROLE page, based on what Tom Lane told me here:

	http://archives.postgresql.org/pgsql-general/2005-11/msg00998.php


Joachim Wieland
2006-03-03 03:06:05 +00:00
Bruce Momjian 66eccb466e Add:
> * Add missing parameter handling in to_char()
>
> 	http://archives.postgresql.org/pgsql-hackers/2005-12/msg00948.php
>
2006-03-03 02:38:57 +00:00
Bruce Momjian d65b4e30f0 Add:
> * Allow to_date() and to_timestamp() accept localized month names
2006-03-03 02:18:09 +00:00
Bruce Momjian 6839bc95d4 Add comment about localized month names for to_date and to_timestamp. 2006-03-03 02:17:30 +00:00
Tom Lane 9a506a6257 Arrange to call AbsorbFsyncRequests every so often while performing a
checkpoint in the bgwriter.  This forestalls overflow of the fsync request
queue, which is not fatal but causes considerable performance degradation
when it occurs (because backends then have to do their own fsyncs).  Per
patch from Itagaki Takahiro, modified a little bit by me.
2006-03-03 00:02:02 +00:00
Tom Lane f0bfc02001 Remove unnecessary lo_lseek call in lo_open. Apparently there was once
a need for it back in the neolithic era, but it's certainly dead code in
any PG release we would recognize as such.  Since it forces an additional
network round trip to the backend, getting rid of it should provide some
small performance improvement for large-object-using clients.
2006-03-02 21:56:14 +00:00
Tom Lane 56aa84a69c Fix ancient error in large objects usage example: overwrite() subroutine
was opening with INV_READ flag and then writing.  Prior to 8.1 the backend
did not reject this, but now it does.
2006-03-02 21:49:09 +00:00
Tom Lane 9356877bba Repair oidvectorrecv and int2vectorrecv, which I broke while changing
them to use array_recv :-(.  Per report from Tim Kordas.
2006-03-02 21:13:04 +00:00
Bruce Momjian 487b7f5de3 Update OS X shared memory documentation for 10.3.9 and later to use
/etc/sysctl.conf.

Chris Campbell
2006-03-02 20:30:21 +00:00
Bruce Momjian 33a84bb306 Mark tsearch2 item as Tom's, not Teodor's. 2006-03-02 19:34:43 +00:00
Bruce Momjian 989d94a171 Update TODO.detail/qsort. 2006-03-02 19:21:05 +00:00
Bruce Momjian 8da308036d Update TODO.detail/qsort. 2006-03-02 19:20:44 +00:00
Teodor Sigaev 38c4fe87ac Significantly improve ranking:
1) rank_cd now use weight of lexemes
2) rank_cd and rank can use any combination of normalization methods:
        no normalization
        normalization by log(length of document)
        -----/------- by length of document
        -----/------- by number of unique word in document
        -----/------- by log(number of unique word in document)
        -----/------- by number of covers (only rank_cd)

Improve cover's search.

TODO: changes in documentation
2006-03-02 19:07:19 +00:00
Bruce Momjian 85fa9f516c Update TODO.detail/qsort. 2006-03-02 18:20:25 +00:00
Bruce Momjian 1fa33539ab Add to TODO.detail for qsort. 2006-03-02 18:18:13 +00:00
Bruce Momjian f79df7fcc9 Add:
> * Improve port/qsort() to handle sorts with 50% unique and 50% duplicate
>   value [qsort]
>
>   This involves choosing better pivot points for the quicksort.
2006-03-02 18:18:00 +00:00
Tom Lane fa7f6ff0db Fix possible crash at transaction end when a plpgsql function is used and
then modified within the same transaction.  The code was using a linked list
of active PLpgSQL_expr structs, which was OK when it was written because
plpgsql never released any parse data structures for the life of the backend.
But since Neil fixed plpgsql's memory management, elements of the linked list
could be freed, leading to crash when the list is chased.  Per report and test
case from Kris Jurka.
2006-03-02 05:34:12 +00:00
Tom Lane 181f55e5fb Fix up pg_dump to emit shell-type definitions at the proper time, to
make use of the recently added ability to create a shell type explicitly.

I also put in place some infrastructure to allow dump/no dump decisions
to be made separately for each database object, rather than the former
hardwired 'dump if in a dumpable schema' policy.  This was needed anyway
for shell types so now seemed a convenient time to do it.  The flexibility
isn't exposed to the user yet, but is ready for future extensions.
2006-03-02 01:18:26 +00:00
Neil Conway 305fcb7aa4 Woops: also update the alternative "expected" files for contrib/cube's
regression tests to account for the new error message wording. It seems
today is not my day...
2006-03-01 23:00:56 +00:00