Bruce Momjian [Mon, 22 Jan 2001 17:20:12 +0000 (17:20 +0000)]
Update FAQ.
Michael Meskes [Mon, 22 Jan 2001 17:05:50 +0000 (17:05 +0000)]
Synced preproc.y with gram.y and added missing include file to pgc.l.
Bruce Momjian [Mon, 22 Jan 2001 16:35:35 +0000 (16:35 +0000)]
Update FAQ.
Tom Lane [Mon, 22 Jan 2001 16:11:17 +0000 (16:11 +0000)]
Minor fixes.
Bruce Momjian [Mon, 22 Jan 2001 06:27:41 +0000 (06:27 +0000)]
Update FAQ.
Bruce Momjian [Mon, 22 Jan 2001 05:56:26 +0000 (05:56 +0000)]
Update FAQ.
Bruce Momjian [Mon, 22 Jan 2001 03:47:22 +0000 (03:47 +0000)]
Add.
Bruce Momjian [Mon, 22 Jan 2001 03:36:58 +0000 (03:36 +0000)]
Update.
Bruce Momjian [Mon, 22 Jan 2001 03:33:55 +0000 (03:33 +0000)]
Make Kevin Lo's MS FAQ the default.
Tom Lane [Mon, 22 Jan 2001 00:50:07 +0000 (00:50 +0000)]
Clean up per-tuple memory leaks in trigger firing and plpgsql
expression evaluation.
Tom Lane [Mon, 22 Jan 2001 00:18:13 +0000 (00:18 +0000)]
All the global memory contexts should be DLLIMPORT, if any are.
Bruce Momjian [Sun, 21 Jan 2001 22:16:56 +0000 (22:16 +0000)]
Update FAQ.
Peter Eisentraut [Sun, 21 Jan 2001 22:02:01 +0000 (22:02 +0000)]
Description of table expressions, including join syntax, from Robert B.
Easter <reaster@comptechnews.com>, heavily massaged by me. Also cleaned up
value expressions a bit.
Bruce Momjian [Sun, 21 Jan 2001 05:16:45 +0000 (05:16 +0000)]
Put quotes around environment variables.
Bruce Momjian [Sun, 21 Jan 2001 05:01:57 +0000 (05:01 +0000)]
Update FAQ.
Bruce Momjian [Sun, 21 Jan 2001 03:50:25 +0000 (03:50 +0000)]
Back out patch for BLOB operations until approval.
Bruce Momjian [Sun, 21 Jan 2001 03:49:14 +0000 (03:49 +0000)]
Hello,
here is the patch attached which do check in each BLOB operation, if we are
in transaction, and raise an error otherwise. This will prevent such mistakes.
--
Sincerely Yours,
Denis Perchine
Tom Lane [Sun, 21 Jan 2001 00:59:26 +0000 (00:59 +0000)]
Deal with C++ incompatibility of sys_nerr declaration by taking it out
of c.h altogether, and putting it into the only places that use it
(elog.c and exc.c), instead. Modify these routines to check for a
NULL or empty-string return from strerror, too, since some platforms
define strerror to return empty string for unknown errors (what a useless
definition that is ...). Clean up some cruft in ExcPrint while at it.
Tom Lane [Sat, 20 Jan 2001 23:07:27 +0000 (23:07 +0000)]
Get rid of sunos4-only strerror() macro, and arrange to use the
implementation in backend/port/strerror.c if configure finds no strerror
in libc, same as we do for snprintf and inet_aton.
Tom Lane [Sat, 20 Jan 2001 22:56:33 +0000 (22:56 +0000)]
Remove no-longer-used STRERROR2 config symbol.
Tom Lane [Sat, 20 Jan 2001 22:09:24 +0000 (22:09 +0000)]
Get rid of initdb -t bugs by the simple expedient of getting rid of
initdb -t. This option is obsoleted by 7.1's ability to drop and
recreate template1 during normal operation.
Peter Eisentraut [Sat, 20 Jan 2001 20:59:29 +0000 (20:59 +0000)]
Replace some oldish, non-SQL'ish elements with more standard forms. (cast
syntax, type names, function names, etc.)
Bruce Momjian [Sat, 20 Jan 2001 18:55:35 +0000 (18:55 +0000)]
Add.
Tom Lane [Sat, 20 Jan 2001 17:37:52 +0000 (17:37 +0000)]
Give a good error message for what's likely to be a common syntax error,
namely omitting the alias clause for a sub-SELECT in FROM.
Bruce Momjian [Sat, 20 Jan 2001 05:00:03 +0000 (05:00 +0000)]
Add emails about transaction rollover.
Bruce Momjian [Sat, 20 Jan 2001 04:59:36 +0000 (04:59 +0000)]
Update TODO list.
Bruce Momjian [Sat, 20 Jan 2001 04:57:11 +0000 (04:57 +0000)]
Update TODO list.
Bruce Momjian [Sat, 20 Jan 2001 04:46:46 +0000 (04:46 +0000)]
Add.
Bruce Momjian [Sat, 20 Jan 2001 04:46:04 +0000 (04:46 +0000)]
Update TODO list.
Bruce Momjian [Sat, 20 Jan 2001 04:16:55 +0000 (04:16 +0000)]
Move anoncvs to top of docs, then put cvs tree. Hope that is OK. Seems
more logical.
Peter Eisentraut [Sat, 20 Jan 2001 00:15:59 +0000 (00:15 +0000)]
Add missing piece of BitString support to node output functions. Expand
and remove IsA_Value macro.
Bruce Momjian [Sat, 20 Jan 2001 00:05:54 +0000 (00:05 +0000)]
In the Programmer's Guide, the sample code for accessing large objects
from libpq has two functions with memory leaks.
The functions pickout() and overwrite() malloc space for buf which is
never freed.
See
http://www.postgresql.org/users-lounge/docs/7.0/programmer/largeobjects3207.htm
This problem is also in the 6.5 docs at
http://www.postgresql.org/users-lounge/docs/6.5/programmer/x3184.htm
Nishad Prakash
Tom Lane [Sat, 20 Jan 2001 00:03:55 +0000 (00:03 +0000)]
Still further tweaking of s_lock assembler: do not assume that leading
whitespace is unimportant in assembly code. Also, move VAX definition
of typedef slock_t to port header files to be like all the other ports.
Note that netbsd.h and openbsd.h are now identical, and I rather think
that freebsd.h is broken in the places where it doesn't agree --- but
I'll leave it to the freebsders to look at that.
Peter Eisentraut [Fri, 19 Jan 2001 23:43:36 +0000 (23:43 +0000)]
From Jason Tishler <jt@dothill.com>
* doc/FAQ_MSWIN: Update to be consistent with software -- mainly change
comment from lack of Cygwin UNIX domain socket support and to list of
current Cygwin UNIX domain socket issues.
* src/include/config.h.in: Enable UNIX domain sockets for Cygwin.
* src/include/port/win.h: Disable UNIX domain sockets for Cygwin b20.1.
* src/test/regress/pg_regress.sh: Use UNIX domain sockets for Cygwin
instead of TCP/IP.
Bruce Momjian [Fri, 19 Jan 2001 22:34:45 +0000 (22:34 +0000)]
Add.
Tom Lane [Fri, 19 Jan 2001 22:08:47 +0000 (22:08 +0000)]
Make critical sections (elog->crash) and interrupt holdoff sections
into distinct concepts, per recent discussion on pghackers.
Bruce Momjian [Fri, 19 Jan 2001 21:09:57 +0000 (21:09 +0000)]
cleanup.
Bruce Momjian [Fri, 19 Jan 2001 20:39:16 +0000 (20:39 +0000)]
Remove ; and add \n to ASM code.
Bruce Momjian [Fri, 19 Jan 2001 19:50:04 +0000 (19:50 +0000)]
update
Peter Eisentraut [Fri, 19 Jan 2001 19:43:27 +0000 (19:43 +0000)]
Update to reality.
Tom Lane [Fri, 19 Jan 2001 19:39:23 +0000 (19:39 +0000)]
Make pqexpbuffer a little more robust, per bug report from Heinz Ekker.
Peter Eisentraut [Fri, 19 Jan 2001 19:15:33 +0000 (19:15 +0000)]
Repair circular dependencies and broken clean commands.
Peter Eisentraut [Fri, 19 Jan 2001 19:06:48 +0000 (19:06 +0000)]
Change
#! /usr/local/bin/perl -w
to
#! /usr/bin/perl
The path is probably more portable, and the -w was kind of silly for a six
line script that produces two warnings as it stands.
Peter Eisentraut [Fri, 19 Jan 2001 18:35:40 +0000 (18:35 +0000)]
Change
#! /usr/local/bin/python
to
#! /usr/bin/env python
which is the recommended way.
Bruce Momjian [Fri, 19 Jan 2001 18:19:34 +0000 (18:19 +0000)]
Add email.
Bruce Momjian [Fri, 19 Jan 2001 18:00:22 +0000 (18:00 +0000)]
Add more about LRU-2.
Bruce Momjian [Fri, 19 Jan 2001 16:54:31 +0000 (16:54 +0000)]
Add to optimizer file.
Bruce Momjian [Fri, 19 Jan 2001 16:54:01 +0000 (16:54 +0000)]
Update TODO list.
Bruce Momjian [Fri, 19 Jan 2001 16:34:14 +0000 (16:34 +0000)]
Update TODO list.
Bruce Momjian [Fri, 19 Jan 2001 16:14:36 +0000 (16:14 +0000)]
> > I have attached a simple change to src/pl/plperl/plperl.c to
> > enable the :bash_math opcodes. Currently plperl.c only
> > enables the :default opcodes. This leave out about five of six
> > math functions including sqrt().
Travis Bauer
Peter Mount [Fri, 19 Jan 2001 08:49:06 +0000 (08:49 +0000)]
Fri Jan 19 08:47:00 GMT 2001 peter@retep.org.uk
- Applied patch submitted by John Schutz <schutz@austin.rr.com> that
fixed a bug with ANT's SQL functions (not needed for building but nice
to have fixed).
Bruce Momjian [Fri, 19 Jan 2001 07:03:53 +0000 (07:03 +0000)]
Fix alignment
Bruce Momjian [Fri, 19 Jan 2001 06:59:59 +0000 (06:59 +0000)]
Fix univel asm alignment
Tom Lane [Fri, 19 Jan 2001 06:57:25 +0000 (06:57 +0000)]
Suppress compiler warning in MULTIBYTE code.
Tom Lane [Fri, 19 Jan 2001 06:54:57 +0000 (06:54 +0000)]
Suppress compiler warning in MULTIBYTE case.
Tom Lane [Fri, 19 Jan 2001 06:50:23 +0000 (06:50 +0000)]
Suppress unused-variable warning in non-Assert compilations.
Bruce Momjian [Fri, 19 Jan 2001 05:09:47 +0000 (05:09 +0000)]
New emails.
Bruce Momjian [Fri, 19 Jan 2001 04:57:43 +0000 (04:57 +0000)]
Add to type conversion TODO emails.
Tom Lane [Fri, 19 Jan 2001 04:47:50 +0000 (04:47 +0000)]
Update docs to explain that 7.1 locks down LC_COLLATE and LC_CTYPE at
initdb time. A few copy-editing cleanups, too.
Bruce Momjian [Fri, 19 Jan 2001 04:07:58 +0000 (04:07 +0000)]
Update TODO list.
Bruce Momjian [Fri, 19 Jan 2001 03:58:35 +0000 (03:58 +0000)]
Add __volatile__ to all __asm__ and make consistent indenting
Bruce Momjian [Fri, 19 Jan 2001 02:58:59 +0000 (02:58 +0000)]
New ASM format:
/*
* Standard __asm__ format:
*
* __asm__(
* "command;"
* "command;"
* "command;"
* : "=r"(_res) return value, in register
* : "r"(lock) argument, 'lock pointer', in register
* : "r0"); inline code uses this register
*/
Bruce Momjian [Fri, 19 Jan 2001 02:53:52 +0000 (02:53 +0000)]
Update TODO list.
Bruce Momjian [Fri, 19 Jan 2001 02:05:51 +0000 (02:05 +0000)]
Update TODO list.
Bruce Momjian [Thu, 18 Jan 2001 23:40:26 +0000 (23:40 +0000)]
Fix VAX ASM '1 f' -> '1f'.
Bruce Momjian [Thu, 18 Jan 2001 19:12:36 +0000 (19:12 +0000)]
Oops. Remove extra semicolon in comment.
Bruce Momjian [Thu, 18 Jan 2001 18:52:20 +0000 (18:52 +0000)]
Update TODO list.
Vadim B. Mikheev [Thu, 18 Jan 2001 18:33:45 +0000 (18:33 +0000)]
Comment out xlrec in xact_redo - no support for file unlinking on
commit yet.
Peter Mount [Thu, 18 Jan 2001 17:38:50 +0000 (17:38 +0000)]
Forgot to cvs add UpdateableResultSet.java ;-)
Peter Mount [Thu, 18 Jan 2001 17:37:15 +0000 (17:37 +0000)]
Thu Jan 18 17:37:00 GMT 2001 peter@retep.org.uk
- Added new error message into errors.properties "postgresql.notsensitive"
This is used by jdbc2.ResultSet when a method is called that should
fetch the current value of a row from the database refreshRow() for
example.
- These methods no longer throw the not implemented but the new noupdate
error. This is in preparation for the Updateable ResultSet support
which will overide these methods by extending the existing class to
implement that functionality, but needed to show something other than
notimplemented:
moveToCurrentRow()
moveToInsertRow()
rowDeleted()
rowInserted()
all update*() methods, except those that took the column as a String
as they were already implemented to convert the String to an int.
- getFetchDirection() and setFetchDirection() now throws
"postgresql.notimp" as we only support one direction.
The CursorResultSet will overide this when its implemented.
- Created a new class under jdbc2 UpdateableResultSet which extends
ResultSet and overides the relevent update methods.
This allows us to implement them easily at a later date.
- In jdbc2.Connection, the following methods are now implemented:
createStatement(type,concurrency);
getTypeMap();
setTypeMap(Map);
- The JDBC2 type mapping scheme almost complete, just needs SQLInput &
SQLOutput to be implemented.
- Removed some Statement methods that somehow appeared in Connection.
- In jdbc2.Statement()
getResultSetConcurrency()
getResultSetType()
setResultSetConcurrency()
setResultSetType()
- Finally removed the old 6.5.x driver.
Peter Mount [Thu, 18 Jan 2001 14:50:15 +0000 (14:50 +0000)]
Thu Jan 18 12:24:00 GMT 2001 peter@retep.org.uk
- These methods in org.postgresql.jdbc2.ResultSet are now implemented:
getBigDecimal(int) ie: without a scale (why did this get missed?)
getBlob(int)
getCharacterStream(int)
getConcurrency()
getDate(int,Calendar)
getFetchDirection()
getFetchSize()
getTime(int,Calendar)
getTimestamp(int,Calendar)
getType()
NB: Where int represents the column name, the associated version
taking a String were already implemented by calling the int
version.
- These methods no longer throw the not implemented but the new noupdate
error. This is in preparation for the Updateable ResultSet support
which will overide these methods by extending the existing class to
implement that functionality, but needed to show something other than
notimplemented:
cancelRowUpdates()
deleteRow()
- Added new error message into errors.properties "postgresql.noupdate"
This is used by jdbc2.ResultSet when an update method is called and
the ResultSet is not updateable. A new method notUpdateable() has been
added to that class to throw this exception, keeping the binary size
down.
- Added new error message into errors.properties "postgresql.psqlnotimp"
This is used instead of unimplemented when it's a feature in the
backend that is preventing this method from being implemented.
- Removed getKeysetSize() as its not part of the ResultSet API
Thu Jan 18 09:46:00 GMT 2001 peter@retep.org.uk
- Applied modified patch from Richard Bullington-McGuire
<rbulling@microstate.com>. I had to modify it as some of the code
patched now exists in different classes, and some of it actually
patched obsolete code.
Wed Jan 17 10:19:00 GMT 2001 peter@retep.org.uk
- Updated Implementation to include both ANT & JBuilder
- Updated README to reflect the changes since 7.0
- Created jdbc.jpr file which allows JBuilder to be used to edit the
source. JBuilder _CAN_NOT_ be used to compile. You must use ANT for
that. It's only to allow JBuilders syntax checking to improve the
drivers source. Refer to Implementation for more details
Hiroshi Inoue [Thu, 18 Jan 2001 07:29:04 +0000 (07:29 +0000)]
Fix some leaks(was my fault).
Thomas G. Lockhart [Thu, 18 Jan 2001 07:22:43 +0000 (07:22 +0000)]
Fix up "Postgres-style" time interval representation when fields have
mixed-signs. Previous effort left way too many minus signs, and was at
least as broken as the one before that :(
Clean up "ISO-style" time interval representation to omit zero fields if
there is at least one non-zero field. Supress some leading plus signs
when not necessary for clarity.
Replace every #ifdef __CYGWIN__ block with a cleaner TIMEZONE_GLOBAL macro
defined in datetime.h.
Thomas G. Lockhart [Thu, 18 Jan 2001 07:18:39 +0000 (07:18 +0000)]
Fix incorrect placeholder name in example.
Thomas G. Lockhart [Thu, 18 Jan 2001 07:16:56 +0000 (07:16 +0000)]
Add "--nodata" option to allow schema conversion only.
Tom Lane [Thu, 18 Jan 2001 07:12:37 +0000 (07:12 +0000)]
Fix performance issue with qualifications on VIEWs: outer query should
try to push restrictions on the view down into the view subquery,
so that they can become indexscan quals or what-have-you rather than
being applied at the top level of the subquery. 7.0 and before were
able to do this, though in a much klugier way, and I'd hate to have
anyone complaining that 7.1 is stupider than 7.0 ...
Thomas G. Lockhart [Thu, 18 Jan 2001 07:11:36 +0000 (07:11 +0000)]
Replace contraction with long form. Cosmetic only.
Bruce Momjian [Thu, 18 Jan 2001 04:12:47 +0000 (04:12 +0000)]
Add "OK, OK, Hiroshi's right" thread.
Hiroshi Inoue [Thu, 18 Jan 2001 04:01:42 +0000 (04:01 +0000)]
Change LockClassinfoForUpdate() to retry mark4update() in case
the tuple is already uodated. (If LockClassinfoForUpdate() is
thought to be useful).
Bruce Momjian [Thu, 18 Jan 2001 03:15:55 +0000 (03:15 +0000)]
Update TODO list.
Bruce Momjian [Thu, 18 Jan 2001 03:15:26 +0000 (03:15 +0000)]
Update TODO list.
Bruce Momjian [Wed, 17 Jan 2001 22:13:33 +0000 (22:13 +0000)]
Functions -> Function's
Bruce Momjian [Wed, 17 Jan 2001 22:11:19 +0000 (22:11 +0000)]
attached is a patch that makes SysV semaphore emulation
using POSIX semaphores more robust on Darwin 1.2/Mac OS X
Public Beta. this is for the version of 7.1 available
via anon cvs as of Jan 14 2001 14:00 PST.
since the semaphores and shared memory created by this
emulator are shared with the backends via fork(), their
persistent names are not necessary. removing their
names with shm_unlink() and sem_unlink() after creation
obviates the need for any "ipcclean" function. further,
without these changes, the shared memory (and, therefore,
the semaphores) will not be re-initialized/re-created after
the first execution of the postmaster, until reboot
or until some (non-existent) ipcclean function is executed.
this patch does the following:
1) if the shared memory segment "SysV_Sem_Info" already
existed, it is cleaned up. it shouldn't be there anyways.
2) the real indicator for whether the shared memory/semaphore
emulator has been initialized is if "SemInfo" has been
initialized. the shared memory and semaphores must be
initialized regardless of whether there was a garbage shared
memory segment lying around.
3) the shared memory segment "SysV_Sem_Info" is created with "O_EXCL"
to catch the case where two postmasters might be starting
simultaneously, so they don't both end up with the same shared
memory (one will fail). note that this can't be done with the
semaphores because Darwin 1.2 has a bug where attempting to
open an existing semaphore with "O_EXCL" set will ruin the
semaphore until the next reboot.
4) the shared memory segment "SysV_Sem_Info" is unlinked after
it is created. it will then exist without a name until the
postmaster and all backend children exit.
5) all semaphores are unlinked after they are created. they'll
then exist without names until the postmaster and all backend
children exit.
-michael thornburgh, zenomt@armory.com
Bruce Momjian [Wed, 17 Jan 2001 17:26:45 +0000 (17:26 +0000)]
Change lcons(x, NIL) to makeList(x) where appropriate.
Thomas G. Lockhart [Wed, 17 Jan 2001 16:46:56 +0000 (16:46 +0000)]
Change comparisons of tm->tm_isdst from "nonzero" to "greater than zero".
Not sure why some were this way, and others were already correct, but it
seems to have been like this for several years.
This caused problems on a few damaged platforms like AIX and IRIX which do
not support DST calculations for years before 1970.
Thanks to Andreas Zeugswetter <ZeugswetterA@wien.spardat.at> for finding
the problem.
Bruce Momjian [Wed, 17 Jan 2001 16:34:34 +0000 (16:34 +0000)]
There are misprints in postgres doc., in :
Chapter 10. PL/pgSQL - SQL Procedural Language (
c40914117.htm)
Statements
...
(resulting in a PL/pgSQL internal SELECT).
But there are cases where someone isn't interested int
-----------------------------------------(have to be)-->
But there are cases where someone isn't interested in
the functions result.
RAISE level format''
--(have to be)-->
RAISE level 'format'
Bruce Momjian [Wed, 17 Jan 2001 06:41:31 +0000 (06:41 +0000)]
Move structure comments from the top block down to the line entries for
this file to match all the other files, and to be clearer.
Tom Lane [Tue, 16 Jan 2001 22:48:34 +0000 (22:48 +0000)]
Remove bogus backslashes in sed command.
Tom Lane [Tue, 16 Jan 2001 20:59:34 +0000 (20:59 +0000)]
Oops, I had managed to break query-cancel-while-waiting-for-lock.
Tom Lane [Tue, 16 Jan 2001 06:11:34 +0000 (06:11 +0000)]
Rename fields of lock and lockholder structures to something a tad less
confusing, and clean up documentation.
Tom Lane [Mon, 15 Jan 2001 22:20:00 +0000 (22:20 +0000)]
Show intarray contrib module in contrib/README and Makefile.
Peter Eisentraut [Mon, 15 Jan 2001 21:17:27 +0000 (21:17 +0000)]
Adjust file names.
Tom Lane [Mon, 15 Jan 2001 20:36:36 +0000 (20:36 +0000)]
Fix problems with parentheses around sub-SELECT --- for the last time,
I hope. I finally realized that we were going at it backwards: when
there are excess parentheses, they need to be treated as part of the
sub-SELECT, not as part of the surrounding expression. Although either
choice yields an unambiguous grammar, only this way produces a grammar
that is LALR(1). With the old approach we were guaranteed to fail on
either 'SELECT (((SELECT 2)) + 3)' or
'SELECT (((SELECT 2)) UNION SELECT 2)' depending on which way we
resolve the initial shift/reduce conflict. With the new way, the same
reduction track can be followed in both cases until we have advanced
far enough to know whether we are done with the sub-SELECT or not.
Tom Lane [Mon, 15 Jan 2001 05:29:19 +0000 (05:29 +0000)]
Tweak heap_update/delete so that we do not hold the buffer context lock
on the old tuple's page while we are doing TOAST pushups.
Tom Lane [Sun, 14 Jan 2001 22:21:54 +0000 (22:21 +0000)]
Minor coding cleanups.
Tom Lane [Sun, 14 Jan 2001 22:21:05 +0000 (22:21 +0000)]
Another go-round on making GetRawDatabaseInfo behave as well as it can,
given the fundamental restriction of not looking at transaction commit
data in pg_log. Use code that is actually based on tqual.c rather than
ad-hoc tests. Also write the tuple fetch loop using standard access
macros rather than ad-hoc code.
Tom Lane [Sun, 14 Jan 2001 22:17:22 +0000 (22:17 +0000)]
pg_database's datpath column must not be marked toastable, because
GetRawDatabaseInfo() won't cope with a compressed path spec (much less
a moved-off one). I'm not going to force an initdb for this change,
because it's noncritical --- we're not actually using datpath at all
right now. But it seems a good idea to apply the fix while I'm thinking
about it.
Tom Lane [Sun, 14 Jan 2001 22:14:10 +0000 (22:14 +0000)]
Need to do BufferSync at end of DROP DATABASE as well as CREATE DATABASE.
Otherwise, newly connecting backends will still think the deleted DB is
valid, and will generate unexpected error messages.
Tom Lane [Sun, 14 Jan 2001 19:23:27 +0000 (19:23 +0000)]
Make aclcontains() do something that's at least vaguely reasonable:
it now returns true if the aclitem argument exactly matches any one of
the elements of the aclitem[] argument. Per complaint from Wolff 1/10/01.
Tom Lane [Sun, 14 Jan 2001 05:08:17 +0000 (05:08 +0000)]
Restructure backend SIGINT/SIGTERM handling so that 'die' interrupts
are treated more like 'cancel' interrupts: the signal handler sets a
flag that is examined at well-defined spots, rather than trying to cope
with an interrupt that might happen anywhere. See pghackers discussion
of 1/12/01.