Tom Lane [Mon, 7 Aug 2006 17:41:42 +0000 (17:41 +0000)]
On some platforms, pg_usleep isn't interruptible by signals; fix
archiver to behave per original coder's expectation on these machines.
We already know this everywhere else AFAICT.
Tom Lane [Mon, 7 Aug 2006 16:57:57 +0000 (16:57 +0000)]
Make recovery from WAL be restartable, by executing a checkpoint-like
operation every so often. This improves the usefulness of PITR log
shipping for hot standby: formerly, if the standby server crashed, it
was necessary to restart it from the last base backup and replay all
the WAL since then. Now it will only need to reread about the same
amount of WAL as the master server would. The behavior might also
come in handy during a long PITR replay sequence. Simon Riggs,
with some editorialization by Tom Lane.
Tom Lane [Sun, 6 Aug 2006 04:35:21 +0000 (04:35 +0000)]
Tweak sanity_check regression test to display more tables (viz, those
without indexes) but not to display temp tables. It's a bit hard to
credit that sanity_check could get through a database-wide VACUUM
while the preceding create_index test is still trying to clean up
its temp tables ... but I see no other explanation for the current
failure report from buildfarm member sponge.
Tom Lane [Sun, 6 Aug 2006 03:53:44 +0000 (03:53 +0000)]
Add support for forcing a switch to a new xlog file; cause such a switch
to happen automatically during pg_stop_backup(). Add some functions for
interrogating the current xlog insertion point and for easily extracting
WAL filenames from the hex WAL locations displayed by pg_stop_backup
and friends. Simon Riggs with some editorialization by Tom Lane.
Tom Lane [Sat, 5 Aug 2006 17:21:52 +0000 (17:21 +0000)]
Fix inheritance_planner() to delete dummy subplans from its Append plan
list, when some of the child rels have been excluded by constraint
exclusion. This doesn't save a huge amount of time but it'll save some,
and it makes the EXPLAIN output look saner. We already did the
equivalent thing in set_append_rel_pathlist(), but not here.
Tom Lane [Sat, 5 Aug 2006 00:22:49 +0000 (00:22 +0000)]
Extend relation_excluded_by_constraints() to check for mutually
contradictory WHERE-clauses applied to a relation. This makes the
GUC variable constraint_exclusion rather inappropriately named,
but I've refrained for the moment from renaming it.
Per example from Martin Lesser.
Tom Lane [Sat, 5 Aug 2006 00:21:14 +0000 (00:21 +0000)]
Teach predicate_refuted_by() how to do proofs involving NOT-clauses.
This doesn't matter too much for ordinary NOTs, since prepqual.c does
its best to get rid of those, but it helps with IS NOT TRUE clauses
which the rule rewriter likes to insert. Per example from Martin Lesser.
Tom Lane [Fri, 4 Aug 2006 21:33:36 +0000 (21:33 +0000)]
Fix domain_in() bug exhibited by Darcy Buskermolen. The idea of an EState
that's shorter-lived than the expression state being evaluated in it really
doesn't work :-( --- we end up with fn_extra caches getting deleted while
still in use. Rather than abandon the notion of caching expression state
across domain_in calls altogether, I chose to make domain_in a bit cozier
with ExprContext. All we really need for evaluating variable-free
expressions is an ExprContext, not an EState, so I invented the notion of a
"standalone" ExprContext. domain_in can prevent resource leakages by doing
a ReScanExprContext on this rather than having to free it entirely; so we
can make the ExprContext have the same lifespan (and particularly the same
per_query memory context) as the expression state structs.
Tom Lane [Fri, 4 Aug 2006 14:09:51 +0000 (14:09 +0000)]
Teach eval_const_expressions to simplify BooleanTest nodes that have
constant input. Seems worth doing because rule rewriter inserts
IS NOT TRUE tests into WHERE clauses.
Tom Lane [Thu, 3 Aug 2006 20:57:06 +0000 (20:57 +0000)]
Cause ALTER TABLE to perform ALTER COLUMN DROP DEFAULT operations during
the DROP pass rather than the ADD_CONSTR pass. On examining the code I
think this was just an oversight rather than intentional, and it seems
to satisfy the principle of least surprise better than the alternative
solution that was discussed. Add an example to the ref page showing how
to do ALTER TYPE and update the default in one command. Per gripe from
Markus Bertheau that that wasn't possible.
Tom Lane [Thu, 3 Aug 2006 17:04:00 +0000 (17:04 +0000)]
Remove use of a regex character class in resultmap (for freebsd version
check). This isn't supported by pg_regress since the recent rewrite
into C. While we could add char classes to pg_regress.c's code, it's
not really needed at the moment: thanks to Andrew's patch to make
pg_regress always accept the 'standard' comparison file, we can just
drop the version check.
Tom Lane [Thu, 3 Aug 2006 16:04:41 +0000 (16:04 +0000)]
Fix AfterTriggerExecute() to pass tg_trigtuple and tg_newtuple as NULLs
rather than pointers to garbage, when calling AFTER STATEMENT triggers.
Michael Fuhr
Bruce Momjian [Wed, 2 Aug 2006 21:48:43 +0000 (21:48 +0000)]
Move 'ddl' right after 'none' in postgresql.conf. It used to be between
'mod' and 'all', which is not the right order. SGML documentation order
was correct. Report from Chander Ganesan.
Tom Lane [Wed, 2 Aug 2006 18:58:21 +0000 (18:58 +0000)]
Arrange for ValuesScan to keep per-sublist expression eval state in a
temporary context that can be reset when advancing to the next sublist.
This is faster and more thorough at recovering space than the previous
method; moreover it will do the right thing if something in the sublist
tries to register an expression context callback.
Tom Lane [Wed, 2 Aug 2006 16:29:49 +0000 (16:29 +0000)]
Fix documentation error: GRANT/REVOKE for roles only accept role names
as grantees, not PUBLIC ... and you can't say GROUP either. Noted by
Brian Hurt.
Tom Lane [Wed, 2 Aug 2006 13:58:52 +0000 (13:58 +0000)]
Add a HINT per suggestion from Michael Glaesemann. Also, tweak OLD/NEW
test to avoid expensive contain_vars_of_level() scan in the normal case
where we're not inside a rule.
Joe Conway [Wed, 2 Aug 2006 01:59:48 +0000 (01:59 +0000)]
Add support for multi-row VALUES clauses as part of INSERT statements
(e.g. "INSERT ... VALUES (...), (...), ...") and elsewhere as allowed
by the spec. (e.g. similar to a FROM clause subselect). initdb required.
Joe Conway and Tom Lane.
Bruce Momjian [Tue, 1 Aug 2006 18:07:36 +0000 (18:07 +0000)]
Done:
< o %Allow pg_dump to use multiple -t and -n switches [pg_dump]
<
< http://archives.postgresql.org/pgsql-patches/2006-01/msg00190.php
<
> o -Allow pg_dump to use multiple -t and -n switches, exclusion
> ability, and regular expression object matching
Bruce Momjian [Tue, 1 Aug 2006 18:05:04 +0000 (18:05 +0000)]
Allow multiple -n (schema) and -t (table) pg_dump options, and add -T
and -N options to exclude objects. Also support regular expressions for
option object names.
Tom Lane [Mon, 31 Jul 2006 20:09:10 +0000 (20:09 +0000)]
Change the relation_open protocol so that we obtain lock on a relation
(table or index) before trying to open its relcache entry. This fixes
race conditions in which someone else commits a change to the relation's
catalog entries while we are in process of doing relcache load. Problems
of that ilk have been reported sporadically for years, but it was not
really practical to fix until recently --- for instance, the recent
addition of WAL-log support for in-place updates helped.
Along the way, remove pg_am.amconcurrent: all AMs are now expected to support
concurrent update.
Tom Lane [Mon, 31 Jul 2006 01:16:38 +0000 (01:16 +0000)]
Change the bootstrap sequence so that toast tables for system catalogs are
created in the bootstrap phase proper, rather than added after-the-fact
by initdb. This is cleaner than before because it allows us to retire the
undocumented ALTER TABLE ... CREATE TOAST TABLE command, but the real reason
I'm doing it is so that toast tables of shared catalogs will now have
predetermined OIDs. This will allow a reasonably clean solution to the
problem of locking tables before we load their relcache entries, to appear
in a forthcoming patch.
Modify snapshot definition so that lazy vacuums are ignored by other
vacuums. This allows a OLTP-like system with big tables to continue
regular vacuuming on small-but-frequently-updated tables while the
big tables are being vacuumed.
Original patch from Hannu Krossing, rewritten by Tom Lane and updated
by me.
Bruce Momjian [Sun, 30 Jul 2006 01:45:21 +0000 (01:45 +0000)]
Fix WIN32 wait() return value macros to be accurate, particularly
because they are used for testing the return value from system().
(WIN32 doesn't overlay the return code with other failure conditions
like Unix does, so they are just simple macros.)
Fix regression checks to properly handle diff failures on Win32 using
the new macros.
Andrew Dunstan [Sat, 29 Jul 2006 19:55:18 +0000 (19:55 +0000)]
prevent multiplexing Windows kernel event objects we listen for across various sockets - should fix the occasional stats test regression failures we see.
Tom Lane [Sat, 29 Jul 2006 03:02:56 +0000 (03:02 +0000)]
Adjust initialization sequence for timezone_abbreviations so that
it's handled just about like timezone; in particular, don't try
to read anything during InitializeGUCOptions. Should solve current
startup failure on Windows, and avoid wasted cycles if a nondefault
setting is specified in postgresql.conf too. Possibly we need to
think about a more general solution for handling 'expensive to set'
GUC options.
Change the delta val from 0 and 10000 to -5000 and 5000 per recent
discussion in hackers list. Also enhance predefined benchmark
scenarios to reflect the scaling factor parameter flexibly.
Tom Lane [Fri, 28 Jul 2006 18:33:04 +0000 (18:33 +0000)]
SQL2003-standard statistical aggregates, by Sergey Koposov. I've added only
the float8 versions of the aggregates, which is all that the standard requires.
Sergey's original patch also provided versions using numeric arithmetic,
but given the size and slowness of the code, I doubt we ought to include
those in core.
Tom Lane [Thu, 27 Jul 2006 19:52:07 +0000 (19:52 +0000)]
Aggregate functions now support multiple input arguments. I also took
the opportunity to treat COUNT(*) as a zero-argument aggregate instead
of the old hack that equated it to COUNT(1); this is materially cleaner
(no more weird ANYOID cases) and ought to be at least a tiny bit faster.
Original patch by Sergey Koposov; review, documentation, simple regression
tests, pg_dump and psql support by moi.
Tom Lane [Thu, 27 Jul 2006 15:37:19 +0000 (15:37 +0000)]
Original coding of pg_regress.c made the results and log directories
with restrictive permissions, which was not the behavior of the shell
script and doesn't seem very desirable. Use the umask setting instead.