OSDN Git Service
Tom Lane [Fri, 15 May 2009 15:56:39 +0000 (15:56 +0000)]
Fix all the server-side SIGQUIT handlers (grumble ... why so many identical
copies?) to ensure they really don't run proc_exit/shmem_exit callbacks,
as was intended. I broke this behavior recently by installing atexit
callbacks without thinking about the one case where we truly don't want
to run those callback functions. Noted in an example from Dave Page.
Marc G. Fournier [Fri, 15 May 2009 02:18:27 +0000 (02:18 +0000)]
commit for BETA2
Tom Lane [Thu, 14 May 2009 22:22:01 +0000 (22:22 +0000)]
Include recovery_end_command in recovery.conf.sample.
Per suggestion of Jaime Casanova.
Tom Lane [Thu, 14 May 2009 22:17:28 +0000 (22:17 +0000)]
Update release notes to today.
Tom Lane [Thu, 14 May 2009 21:59:22 +0000 (21:59 +0000)]
Clean up overly hasty docs patch for pg_standby.
Alvaro Herrera [Thu, 14 May 2009 21:41:53 +0000 (21:41 +0000)]
Translation updates
Tom Lane [Thu, 14 May 2009 21:28:35 +0000 (21:28 +0000)]
Improve a couple of comments.
Heikki Linnakangas [Thu, 14 May 2009 21:01:49 +0000 (21:01 +0000)]
Add a note to release notes about the smart failover mode in pg_standby.
Heikki Linnakangas [Thu, 14 May 2009 20:31:09 +0000 (20:31 +0000)]
Add recovery_end_command option to recovery.conf. recovery_end_command
is run at the end of archive recovery, providing a chance to do external
cleanup. Modify pg_standby so that it no longer removes the trigger file,
that is to be done using the recovery_end_command now.
Provide a "smart" failover mode in pg_standby, where we don't fail over
immediately, but only after recovering all unapplied WAL from the archive.
That gives you zero data loss assuming all WAL was archived before
failover, which is what most users of pg_standby actually want.
recovery_end_command by Simon Riggs, pg_standby changes by Fujii Masao and
myself.
Tom Lane [Wed, 13 May 2009 22:32:55 +0000 (22:32 +0000)]
Add checks to DefineQueryRewrite() to prohibit attaching rules to relations
that aren't RELKIND_RELATION or RELKIND_VIEW, and to disallow attaching rules
to system relations unless allowSystemTableMods is on. This is to make the
behavior of CREATE RULE more like CREATE TRIGGER, which disallows the
comparable cases. Per discussion of bug #4808.
Tom Lane [Wed, 13 May 2009 21:53:41 +0000 (21:53 +0000)]
Remove a useless backslash from a pattern-match example. Michael Toews
Tom Lane [Wed, 13 May 2009 20:27:17 +0000 (20:27 +0000)]
Rewrite xml.c's memory management (yet again). Give up on the idea of
redirecting libxml's allocations into a Postgres context. Instead, just let
it use malloc directly, and add PG_TRY blocks as needed to be sure we release
libxml data structures in error recovery code paths. This is ugly but seems
much more likely to play nicely with third-party uses of libxml, as seen in
recent trouble reports about using Perl XML facilities in pl/perl and bug
#4774 about contrib/xml2.
I left the code for allocation redirection in place, but it's only
built/used if you #define USE_LIBXMLCONTEXT. This is because I found it
useful to corral libxml's allocations in a palloc context when hunting
for libxml memory leaks, and we're surely going to have more of those
in the future with this type of approach. But we don't want it turned on
in a normal build because it breaks exactly what we need to fix.
I have not re-indented most of the code sections that are now wrapped
by PG_TRY(); that's for ease of review. pg_indent will fix it.
This is a pre-existing bug in 8.3, but I don't dare back-patch this change
until it's gotten a reasonable amount of field testing.
Tom Lane [Tue, 12 May 2009 23:43:50 +0000 (23:43 +0000)]
Update release notes for changes through 2009-05-11. Also some minor
copy-editing and reordering of items.
Tom Lane [Tue, 12 May 2009 20:17:40 +0000 (20:17 +0000)]
Fix intratransaction memory leaks in xml_recv, xmlconcat, xmlroot, and
xml_parse, all arising from the same sloppy usage of parse_xml_decl.
The original coding had that function returning its output string
parameters in the libxml context, which is long-lived, and all but one
of its callers neglected to free the strings afterwards. The easiest
and most bulletproof fix is to return the strings in the local palloc
context instead, since that's short-lived. This was only costing a
dozen or two bytes per function call, but that adds up fast if the
function is called repeatedly ...
Noted while poking at the more general problem of what to do with our
libxml memory allocation hooks. Back-patch to 8.3, which has the
identical coding.
Tom Lane [Tue, 12 May 2009 16:43:32 +0000 (16:43 +0000)]
Fix LOCK TABLE to eliminate the race condition that could make it give weird
errors when tables are concurrently dropped. To do this we must take lock
on each relation before we check its privileges. The old code was trying
to do that the other way around, which is a bit pointless when there are lots
of other commands that lock relations before checking privileges. I did keep
it checking each relation's privilege before locking the next relation, which
is a detail that ALTER TABLE isn't too picky about.
Tom Lane [Tue, 12 May 2009 03:11:02 +0000 (03:11 +0000)]
Modify find_inheritance_children() and find_all_inheritors() to add the
ability to lock relations as they scan pg_inherits, and to ignore any
relations that have disappeared by the time we get lock on them. This
makes uses of these functions safe against concurrent DROP operations
on child tables: we will effectively ignore any just-dropped child,
rather than possibly throwing an error as in recent bug report from
Thomas Johansson (and similar past complaints). The behavior should
not change otherwise, since the code was acquiring those same locks
anyway, just a little bit later.
An exception is LockTableCommand(), which is still behaving unsafely;
but that seems to require some more discussion before we change it.
Tom Lane [Tue, 12 May 2009 00:56:05 +0000 (00:56 +0000)]
Do some minor code refactoring in preparation for changing the APIs of
find_inheritance_children() and find_all_inheritors(). I got annoyed that
these are buried inside the planner but mostly used elsewhere. So, create
a new file catalog/pg_inherits.c and put them there, along with a couple
of other functions that search pg_inherits.
The code that modifies pg_inherits is (still) in tablecmds.c --- it's
kind of entangled with unrelated code that modifies pg_depend and other
stuff, so pulling it out seemed like a bigger change than I wanted to make
right now. But this file provides a natural home for it if anyone ever
gets around to that.
This commit just moves code around; it doesn't change anything, except
I succumbed to the temptation to make a couple of trivial optimizations
in typeInheritsFrom().
Tom Lane [Mon, 11 May 2009 17:56:08 +0000 (17:56 +0000)]
Partially revert my patch of 2008-11-12 that installed a limit on the number
of AND/OR clause branches that predtest.c would attempt to deal with. As
noted in bug #4721, that change disabled proof attempts for sizes of problems
that people are actually expecting it to work for. The original complaint
it was trying to solve was O(N^2) behavior for long IN-lists, so let's try
applying the limit to just ScalarArrayOpExprs rather than everything.
Another case of "foolish consistency" I fear.
Back-patch to 8.2, same as the previous patch was.
Magnus Hagander [Mon, 11 May 2009 09:11:41 +0000 (09:11 +0000)]
Move crypt auth comment to proper section.
Add some details about the name=value format of auth options.
Magnus Hagander [Mon, 11 May 2009 09:00:10 +0000 (09:00 +0000)]
Edit the SSL and Kerberos parts of the release notes a bit, and add
a note about the certificates chains patch just applied.
Magnus Hagander [Mon, 11 May 2009 08:06:21 +0000 (08:06 +0000)]
Support SSL certificate chains in the server certificate file.
Andrew Gierth
Tom Lane [Sun, 10 May 2009 22:45:28 +0000 (22:45 +0000)]
Make a marginal performance improvement in predicate_implied_by and
predicate_refuted_by: if either top-level input is a single-element list,
reduce it to its lone member before proceeding. This avoids
a useless level of AND-recursion within the recursive proof routines.
It's worth doing because, for example, if the clause is a 100-element
list and the predicate is a 1-element list then we'd otherwise strip
the predicate's list structure 100 times as we iterate through the clause.
It's only needed at top level because there won't be any trivial ANDs below
that --- this situation is an artifact of the decision to represent even
single-item conditions as Lists in the "implicit AND" format, and that format
is only used at the top level of any predicate or restriction condition.
Tom Lane [Sun, 10 May 2009 02:51:44 +0000 (02:51 +0000)]
Adjust pg_dumpall so that it emits ENCODING, LC_COLLATE, and LC_CTYPE options
in its CREATE DATABASE commands only for databases that have settings
different from the installation defaults. This is a low-tech method of
avoiding unnecessary platform dependencies in dump files. Eventually we ought
to have a platform-independent way of specifying LC_COLLATE and LC_CTYPE, but
that's not going to happen for 8.4, and this patch at least avoids the issue
for people who aren't setting up per-database locales. ENCODING doesn't have
the platform dependency problem, but it seems consistent to make it act the
same as the locale settings.
Tom Lane [Sat, 9 May 2009 22:51:41 +0000 (22:51 +0000)]
Fix cost_nestloop and cost_hashjoin to model the behavior of semi and anti
joins a bit better, ie, understand the differing cost functions for matched
and unmatched outer tuples. There is more that could be done in cost_hashjoin
but this already helps a great deal. Per discussions with Robert Haas.
Heikki Linnakangas [Fri, 8 May 2009 14:48:06 +0000 (14:48 +0000)]
Add alternative expected output files for cs_CZ locale for btree_gist and
tsearch2 tests. This should make 'comet_moth' buildfarm member pass
contrib check. Zdenek Kotala.
Bruce Momjian [Fri, 8 May 2009 14:06:27 +0000 (14:06 +0000)]
Add missing third argument to open().
Bruce Momjian [Fri, 8 May 2009 03:21:35 +0000 (03:21 +0000)]
'PGDLLIMPORT' ShmemVariableCache, needed for pg_migrator.so function
linkage on Win32.
Tested by Hiroshi Saito
Tom Lane [Thu, 7 May 2009 22:58:28 +0000 (22:58 +0000)]
Add an option to AlterTableCreateToastTable() to allow its caller to force
a toast table to be built, even if the sum-of-column-widths calculation
indicates one isn't needed. This is needed by pg_migrator because if the
old table has a toast table, we have to migrate over the toast table since
it might contain some live data, even though subsequent column drops could
mean that no recently-added rows could require toasting.
Tom Lane [Thu, 7 May 2009 22:01:18 +0000 (22:01 +0000)]
Change pgbench to use the table names pgbench_accounts, pgbench_branches,
pgbench_history, and pgbench_tellers, rather than just accounts, branches,
history, and tellers. This is to prevent accidental conflicts with real
application tables, as has been reported to happen at least once. Also
remove the automatic "SET search_path = public" that it did at startup,
as this seems to restrict testing flexibility without actually buying much.
Per proposal by Joshua Drake and ensuing discussion.
Joshua Drake and Tom Lane
Tom Lane [Thu, 7 May 2009 20:13:09 +0000 (20:13 +0000)]
Ooops ... make_outerjoininfo wasn't actually enforcing the join order
restrictions specified for semijoins in optimizer/README, to wit that
you can't reassociate outer joins into or out of the RHS of a semijoin.
Per report from Heikki.
Heikki Linnakangas [Thu, 7 May 2009 11:25:25 +0000 (11:25 +0000)]
Request XLOG switch before writing checkpoint in pg_start_backup(). Otherwise
you can end up with an unrecoverable backup if you start a new base backup
right after finishing archive recovery. In that scenario, the redo pointer of
the checkpoint that pg_start_backup() writes points to the XLOG segment where
the timeline-changing end-of-archive-recovery checkpoint is. The beginning
of that segment contains pages with the old timeline ID, and we don't accept
that in recovery unless we find a history file covering the old timeline ID.
If you omit pg_xlog from the base backup and clear the archive directory
before starting the backup, there will be no such history file available.
The bug is present in all versions since PITR was introduced in 8.0, but I'm
back-patching only back to 8.2. Earlier versions didn't have XLOG switch
records, making this fix unfeasible. Given the lack of reports until now,
it doesn't seem worthwhile to spend more effort to fix 8.0 and 8.1.
Per report and suggestion by Mikael Krantz
Tom Lane [Wed, 6 May 2009 20:31:18 +0000 (20:31 +0000)]
Tweak distribute_qual_to_rels so that when we decide a pseudoconstant qual
can be pushed to the top of the join tree, we update both the relids and
qualscope variables to keep them in sync. This prevents a possible later
failure of an Assert clause, and affects nothing else since qualscope isn't
used later except for that Assert. At the moment the Assert shouldn't be
reachable when we've pushed the qual up; but this is cheap insurance, and
it's more sensible anyway in terms of the overall logic of the routine.
Per analysis of a bug report from Stefan Huehner.
I'm not back-patching this since it's just future-proofing; but if anyone
gets tempted to change check_outerjoin_delay again in the back branches,
this might be needed.
Tom Lane [Wed, 6 May 2009 16:15:21 +0000 (16:15 +0000)]
Modify CREATE DATABASE to enforce that the source database's encoding setting
must be used for the new database, except when copying from template0.
This is the same rule that we now enforce for locale settings, and it has
the same motivation: databases other than template0 might contain data that
would be invalid according to a different setting. This represents another
step in a continuing process of locking down ways in which encoding violations
could occur inside the backend. Per discussion of a few days ago.
In passing, fix pre-existing breakage of mbregress.sh, and fix up a couple
of ereport() calls in dbcommands.c that failed to specify sqlstate codes.
Tom Lane [Wed, 6 May 2009 03:22:15 +0000 (03:22 +0000)]
Fix ecpg tests for change that disallowed Unicode literals unless
standard_conforming_strings is on.
Tom Lane [Tue, 5 May 2009 23:39:55 +0000 (23:39 +0000)]
Tweak a comment to agree a bit better with the new dispensation that
locales are database-wide, not server-wide.
Tom Lane [Tue, 5 May 2009 21:51:46 +0000 (21:51 +0000)]
Minor improvement: avoid assuming that GetLastError value cannot be
affected by CloseHandle() or Sleep().
Tom Lane [Tue, 5 May 2009 21:09:23 +0000 (21:09 +0000)]
Make new complaint about unsafe Unicode literals include an error location.
Every other ereport in scan.l has one, this should too.
Tom Lane [Tue, 5 May 2009 20:06:07 +0000 (20:06 +0000)]
Install an atexit(2) callback that ensures that proc_exit's cleanup processing
will still be performed if something in a backend process calls exit()
directly, instead of going through proc_exit() as we prefer. This is a second
response to the issue that we might load third-party code that doesn't know it
should not call exit(). Such a call will now cause a reasonably graceful
backend shutdown, if possible. (Of course, if the reason for the exit() call
is out-of-memory or some such, we might not be able to recover, but at least
we will try.)
Tom Lane [Tue, 5 May 2009 19:59:00 +0000 (19:59 +0000)]
Install a "dead man switch" to allow the postmaster to detect cases where
a backend has done exit(0) or exit(1) without having disengaged itself
from shared memory. We are at risk for this whenever third-party code is
loaded into a backend, since such code might not know it's supposed to go
through proc_exit() instead. Also, it is reported that under Windows
there are ways to externally kill a process that cause the status code
returned to the postmaster to be indistinguishable from a voluntary exit
(thank you, Microsoft). If this does happen then the system is probably
hosed --- for instance, the dead session might still be holding locks.
So the best recovery method is to treat this like a backend crash.
The dead man switch is armed for a particular child process when it
acquires a regular PGPROC, and disarmed when the PGPROC is released;
these should be the first and last touches of shared memory resources
in a backend, or close enough anyway. This choice means there is no
coverage for auxiliary processes, but I doubt we need that, since they
shouldn't be executing any user-provided code anyway.
This patch also improves the management of the EXEC_BACKEND
ShmemBackendArray array a bit, by reducing search costs.
Although this problem is of long standing, the lack of field complaints
seems to mean it's not critical enough to risk back-patching; at least
not till we get some more testing of this mechanism.
Tom Lane [Tue, 5 May 2009 19:36:32 +0000 (19:36 +0000)]
Insert CHECK_FOR_INTERRUPTS() calls into btree and hash index scans at the
points where we step right or left to the next page. This should ensure
reasonable response time to a query cancel request during an unsuccessful
index scan, as seen in recent gripe from Marc Cousin. It's a bit trickier
than it might seem at first glance, because CHECK_FOR_INTERRUPTS() is a no-op
if executed while holding a buffer lock. So we have to do it just at the
point where we've dropped one page lock and not yet acquired the next.
Remove CHECK_FOR_INTERRUPTS calls at the top level of btgetbitmap and
hashgetbitmap, since they're pointless given the added checks.
I think that GIST is okay already --- at least, there's a CHECK_FOR_INTERRUPTS
at a plausible-looking place in gistnext(). I don't claim to know GIN well
enough to try to poke it for this, if indeed it has a problem at all.
This is a pre-existing issue, but in view of the lack of prior complaints
I'm not going to risk back-patching.
Tom Lane [Tue, 5 May 2009 19:02:22 +0000 (19:02 +0000)]
Update comment for _bt_relandgetbuf.
Peter Eisentraut [Tue, 5 May 2009 18:32:17 +0000 (18:32 +0000)]
Disable the use of Unicode escapes in string constants (U&'') when
standard_conforming_strings is not on, for security reasons.
Tom Lane [Tue, 5 May 2009 18:02:11 +0000 (18:02 +0000)]
Avoid integer overflow in the loop that extracts histogram entries from
ANALYZE's total sample. The original coding is at risk of overflow for
statistics targets exceeding about 2675; this was not a problem before
8.4 but it is now. Per bug #4793 from Dennis Noordsij.
Magnus Hagander [Tue, 5 May 2009 09:48:51 +0000 (09:48 +0000)]
Make the win32 shared memory code try 10 times instead of one if
it fails because the shared memory segment already exists. This
means it can take up to 10 seconds before it reports the error
if it *does* exist, but hopefully it will make the system capable
of restarting even when the server is under high load.
Tom Lane [Tue, 5 May 2009 02:29:06 +0000 (02:29 +0000)]
Code review for \df rewrite: fix assorted bugs, make type and
volatility columns localizable.
Heikki Linnakangas [Mon, 4 May 2009 17:31:35 +0000 (17:31 +0000)]
Fix the query used for \d against 8.2 and 8.3 servers.
Magnus Hagander [Mon, 4 May 2009 09:30:06 +0000 (09:30 +0000)]
Update no longer current comments in header.
Magnus Hagander [Mon, 4 May 2009 08:36:40 +0000 (08:36 +0000)]
Call SetLastError(0) before calling the file mapping functions
to make sure that the error code is reset, as a precaution in
case the API doesn't properly reset it on success. This could
be necessary, since we check the error value even if the function
doesn't fail for specific success cases.
Peter Eisentraut [Mon, 4 May 2009 08:08:47 +0000 (08:08 +0000)]
Replace a couple of references to files that no longer exist in the source
tree with references to the appropriate URLs.
Robert Haas
Tom Lane [Mon, 4 May 2009 02:46:36 +0000 (02:46 +0000)]
Fix missed usage of DLNewElem()
Alvaro Herrera [Mon, 4 May 2009 02:24:17 +0000 (02:24 +0000)]
Avoid a memory allocation in the backend startup code, to avoid having to check
whether it failed. Modelled after catcache.c's usage of DlList, per suggestion
from Tom.
Tom Lane [Sun, 3 May 2009 23:13:37 +0000 (23:13 +0000)]
Fix pg_resetxlog to remove archive status files along with WAL segment files.
Fujii Masao
Tom Lane [Sun, 3 May 2009 21:10:27 +0000 (21:10 +0000)]
Fix mis-description of XML Schema functions, per discussion.
Tom Lane [Sun, 3 May 2009 20:45:43 +0000 (20:45 +0000)]
Fix unintelligible description created by removing only part of
a parenthetical remark.
Tom Lane [Sun, 3 May 2009 20:09:54 +0000 (20:09 +0000)]
Fix assign_pgstat_temp_directory() to ensure the directory path is
canonicalized. Avoid the need to elog(FATAL) on out-of-memory.
Tom Lane [Sun, 3 May 2009 17:16:58 +0000 (17:16 +0000)]
Fix already-obsolete hint message ... sslverify parameter is no more.
Tatsuo Ishii [Sun, 3 May 2009 01:17:41 +0000 (01:17 +0000)]
Update UTF-8 <--> EUC_KR, JOHAB, UHC mappings.
Patch contributed by Chuck McDevitt
Tom Lane [Sat, 2 May 2009 22:02:37 +0000 (22:02 +0000)]
Install some simple defenses in postmaster startup to help ensure a useful
error message if the installation directory layout is messed up (or at least,
something more useful than the behavior exhibited in bug #4787). During
postmaster startup, check that get_pkglib_path resolves as a readable
directory; and if ParseTzFile() fails to open the expected timezone
abbreviation file, check the possibility that the directory is missing rather
than just the specified file. In case of either failure, issue a hint
suggesting that the installation is broken. These two checks cover the lib/
and share/ trees of a full installation, which should take care of most
scenarios where a sysadmin decides to get cute.
Tom Lane [Sat, 2 May 2009 20:28:17 +0000 (20:28 +0000)]
We don't need major_release_split any more.
Tom Lane [Sat, 2 May 2009 20:17:19 +0000 (20:17 +0000)]
Split the release notes into a separate file for each (active) major branch,
as per my recent proposal. release.sgml itself is now just a stub that should
change rarely; ideally, only once per major release to add a new include line.
Most editing work will occur in the release-N.N.sgml files. To update a back
branch for a minor release, just copy the appropriate release-N.N.sgml
file(s) into the back branch.
This commit doesn't change the end-product documentation at all, only the
source layout. However, it makes it easy to start omitting ancient information
from newer branches' documentation, should we ever decide to do that.
Tom Lane [Sat, 2 May 2009 17:27:57 +0000 (17:27 +0000)]
Fix plpgsql's EXIT so that an EXIT without a label only matches a loop,
never a BEGIN block. This is required for Oracle compatibility and is
also plainly stated to be the behavior by our original documentation
(up until 8.1, in which the docs were adjusted to match the code's behavior;
but actually the old docs said the correct thing and the code was wrong).
Not back-patched because this introduces an incompatibility that could
break working applications. Requires release note.
Tom Lane [Fri, 1 May 2009 23:57:34 +0000 (23:57 +0000)]
Fix a couple of cases where the plpgsql grammar looked for T_WORD and
failed to consider the possibility that it would get T_SCALAR, T_RECORD,
or T_ROW instead because the word happens to match a plpgsql variable name.
In particular, give "duplicate declaration" rather than generic "syntax error"
if the same identifier is declared twice in the same block, as per my recent
complaint. Also behave more sanely when decl_aliasitem or proc_condition or
opt_lblname is coincidentally not T_WORD. Refactor the related productions a
bit to reduce duplication.
This is a longstanding bug, but it doesn't seem critical enough to
back-patch.
Tom Lane [Fri, 1 May 2009 19:29:07 +0000 (19:29 +0000)]
When checking for datetime field overflow, we should allow a fractional-second
part that rounds up to exactly 1.0 second. The previous coding rejected input
like "00:12:57.
9999999999999999999999999999", with the exact number of nines
needed to cause failure varying depending on float-timestamp option and
possibly on platform. Obviously this should round up to the next integral
second, if we don't have enough precision to distinguish the value from that.
Per bug #4789 from Robert Kruus.
In passing, fix a missed check for fractional seconds in one copy of the
"is it greater than 24:00:00" code.
Broken all the way back, so patch all the way back.
Heikki Linnakangas [Thu, 30 Apr 2009 10:26:35 +0000 (10:26 +0000)]
Clean up check_keywords.pl script, making it 'strict' and removing a few
leftover unused variables.
Laurent Laborde
Heikki Linnakangas [Wed, 29 Apr 2009 05:05:57 +0000 (05:05 +0000)]
Add check_keyword.pl script to perform some basic sanity checks to the
keyword lists in gram.y and kwlist.h. It checks that all lists are in
alphabetical order, and that all keywords present in gram.y are listed
in kwlist.h in the right category, and that all keywords in kwlist.h are
also in gram.y. What's still missing is to check that all keywords
defined with "%token <keyword>" in gram.y are present in one of the
keyword lists in gram.y.
Tom Lane [Tue, 28 Apr 2009 21:31:16 +0000 (21:31 +0000)]
Improve pull_up_subqueries logic so that it doesn't insert unnecessary
PlaceHolderVar nodes in join quals appearing in or below the lowest
outer join that could null the subquery being pulled up. This improves
the planner's ability to recognize constant join quals, and probably
helps with detection of common sort keys (equivalence classes) as well.
Bruce Momjian [Tue, 28 Apr 2009 17:07:50 +0000 (17:07 +0000)]
Blank line Makefile cleanups.
Heikki Linnakangas [Tue, 28 Apr 2009 09:09:41 +0000 (09:09 +0000)]
Move SERVER to the right place in the alphabetically sorted keyword list.
Bruce Momjian [Tue, 28 Apr 2009 02:37:09 +0000 (02:37 +0000)]
In VACUUM, FREEZE must be before ANALYZE; fix this in vacuumdb. Docs
are already correct.
Bruce Momjian [Mon, 27 Apr 2009 16:27:36 +0000 (16:27 +0000)]
Proofreading adjustments for first two parts of documentation (Tutorial
and SQL).
Tom Lane [Sun, 26 Apr 2009 20:08:30 +0000 (20:08 +0000)]
Some more work on the 8.4 release notes. Document a few changes that Bruce
apparently found uninteresting, and do minor wordsmithing on a number of
the existing entries.
Tom Lane [Sun, 26 Apr 2009 15:31:50 +0000 (15:31 +0000)]
Remove Windows-specific definition of S_ISDIR(). This should not be here;
if there are any Windows configurations where port/win32.h fails to
provide the macro, it should be fixed in the latter file not here.
Tom Lane [Sun, 26 Apr 2009 02:56:52 +0000 (02:56 +0000)]
Typo fix.
Tom Lane [Sun, 26 Apr 2009 02:47:05 +0000 (02:47 +0000)]
Update citext expected output to exactly match the real output, rather
than having some whitespace discrepancy. Although whitespace is supposed
to be ignored in our regression tests, for some reason buildfarm member
spoonbill doesn't like it.
Tom Lane [Sat, 25 Apr 2009 16:44:56 +0000 (16:44 +0000)]
Fix the handling of sub-SELECTs appearing in the arguments of an outer-level
aggregate function. By definition, such a sub-SELECT cannot reference any
variables of query levels between itself and the aggregate's semantic level
(else the aggregate would've been assigned to that lower level instead).
So the correct, most efficient implementation is to treat the sub-SELECT as
being a sub-select of that outer query level, not the level the aggregate
syntactically appears in. Not doing so also confuses the heck out of our
parameter-passing logic, as illustrated in bug report from Daniel Grace.
Fortunately, we were already copying the whole Aggref expression up to the
outer query level, so all that's needed is to delay SS_process_sublinks
processing of the sub-SELECT until control returns to the outer level.
This has been broken since we introduced spec-compliant treatment of
outer aggregates in 7.4; so patch all the way back.
Bruce Momjian [Fri, 24 Apr 2009 20:46:16 +0000 (20:46 +0000)]
Improve server spoofing wording, per request from Magnus.
Tom Lane [Fri, 24 Apr 2009 16:09:50 +0000 (16:09 +0000)]
Fix some more 'variable may be used uninitialized' warnings from gcc 4.4.
Magnus Hagander [Fri, 24 Apr 2009 14:10:41 +0000 (14:10 +0000)]
More clearly document what the different sslmode options mean, both the new
and the old ones.
Consistently talk about certificate verification, and not validation.
Magnus Hagander [Fri, 24 Apr 2009 13:01:30 +0000 (13:01 +0000)]
Remove sslverify parameter from release notes, mention that sslmode is
used instead.
Magnus Hagander [Fri, 24 Apr 2009 09:43:10 +0000 (09:43 +0000)]
Remove sslverify parameter again, replacing it with two new sslmode values:
"verify-ca" and "verify-full".
Since "prefer" remains the default, this will make certificate validation
off by default, which should lead to less upgrade issues.
Magnus Hagander [Fri, 24 Apr 2009 08:43:51 +0000 (08:43 +0000)]
Move gettext encoding names into encnames.c, so we only have one place to update.
Per discussion.
Tom Lane [Thu, 23 Apr 2009 23:25:13 +0000 (23:25 +0000)]
Suppress some 'variable may be used uninitialized' warnings from gcc 4.4.
Tom Lane [Thu, 23 Apr 2009 17:39:21 +0000 (17:39 +0000)]
Don't use the result of strcmp as if it were a boolean.
A service of your local coding style police.
Heikki Linnakangas [Thu, 23 Apr 2009 10:20:27 +0000 (10:20 +0000)]
Mention that tables have a visibility map fork alongside the main fork
and FSM.
Heikki Linnakangas [Thu, 23 Apr 2009 10:09:11 +0000 (10:09 +0000)]
Improve the documentation on 8.4 visibility map related VACUUM changes.
Explain how vacuum_freeze_table_age should be tuned, and how it relates
to the other settings. Mention that vacuum_freeze_table_age also affects
when autovacuum scans the whole table.
Heikki Linnakangas [Thu, 23 Apr 2009 07:19:09 +0000 (07:19 +0000)]
varstr_cmp and any comparison function that piggybacks on it can return
any negative or positive number, not just -1 or 1. Fix comment on
varstr_cmp and citext test case accordingly.
As pointed out by Zdenek Kotala, and buildfarm member gothic moth.
Tom Lane [Thu, 23 Apr 2009 00:23:46 +0000 (00:23 +0000)]
Change the default value of max_prepared_transactions to zero, and add
documentation warnings against setting it nonzero unless active use of
prepared transactions is intended and a suitable transaction manager has been
installed. This should help to prevent the type of scenario we've seen
several times now where a prepared transaction is forgotten and eventually
causes severe maintenance problems (or even anti-wraparound shutdown).
The only real reason we had the default be nonzero in the first place was to
support regression testing of the feature. To still be able to do that,
tweak pg_regress to force a nonzero value during "make check". Since we
cannot force a nonzero value in "make installcheck", add a variant regression
test "expected" file that shows the results that will be obtained when
max_prepared_transactions is zero.
Also, extend the HINT messages for transaction wraparound warnings to mention
the possibility that old prepared transactions are causing the problem.
All per today's discussion.
Heikki Linnakangas [Wed, 22 Apr 2009 19:51:12 +0000 (19:51 +0000)]
After archive recovery, mark the last WAL segment from the parent timeline
ready for archival. It was marked at the next checkpoint anyway, but
waiting for the next checkpoint is an unnecessary delay.
Fujii Masao
Bruce Momjian [Wed, 22 Apr 2009 14:58:48 +0000 (14:58 +0000)]
Use brackets in psql \df \? help.
Tom Lane [Tue, 21 Apr 2009 21:15:50 +0000 (21:15 +0000)]
Remove the long-obsolete homebrew dl*() functions for AIX, in favor of just
using the system functions all the time. (These files are now just copies
of the osf.* files.) The homebrew functions were not getting used anyway
on AIX versions that have dlopen(), that is 4.3 and up, so they are not
needed on any AIX that is even remotely supported by the vendor anymore.
We'd have probably left them here anyway, except some questions were
raised about the copyright.
Tom Lane [Tue, 21 Apr 2009 21:05:25 +0000 (21:05 +0000)]
Fix obsolete cross-reference (this file isn't called alpha.c anymore)
Bruce Momjian [Tue, 21 Apr 2009 17:28:01 +0000 (17:28 +0000)]
Improve psql \df error handling.
David Fetter
Bruce Momjian [Tue, 21 Apr 2009 17:23:05 +0000 (17:23 +0000)]
Merge psql \df options into a single \? line, and update docs.
Bruce Momjian [Tue, 21 Apr 2009 15:49:06 +0000 (15:49 +0000)]
Add agg/normal/trigger/window flags for psql \df and in \df output.
David Fetter
Magnus Hagander [Mon, 20 Apr 2009 08:38:00 +0000 (08:38 +0000)]
Read nls.nk files to determine the proper name of NLS catalogs to install.
Hiroshi Inoue, with minor modifications by me
Tom Lane [Sun, 19 Apr 2009 22:37:13 +0000 (22:37 +0000)]
Assorted portability fixes for Borland C, from Pavel Golub.
Tom Lane [Sun, 19 Apr 2009 21:50:09 +0000 (21:50 +0000)]
Rethink the idea of having plpgsql depend on parser/gram.h. Aside from the
fact that this is breaking the MSVC build, it's probably not really a good
idea to expand the dependencies of gram.h any further than the core parser;
for instance the value of SCONST might depend on which bison version you'd
built with. Better to expose an additional call point in parser.c, so
move what I had put into pl_funcs.c into parser.c. Also PGDLLIMPORT'ify
the reference to standard_conforming_strings, per buildfarm results.
Tom Lane [Sun, 19 Apr 2009 21:08:54 +0000 (21:08 +0000)]
Fix de-escaping checks so that we will reject \000 as well as other invalidly
encoded sequences. Per discussion of a couple of days ago.
Tom Lane [Sun, 19 Apr 2009 20:36:06 +0000 (20:36 +0000)]
Fix textsearch documentation examples to not recommend concatenating separate
fields without putting a space between. Per gripe from Rick Schumeyer.
Tom Lane [Sun, 19 Apr 2009 19:46:33 +0000 (19:46 +0000)]
Fix estimate_num_groups() to not fail on PlaceHolderVars, per report from
Stefan Kaltenbrunner. The most reasonable behavior (at least for the near
term) seems to be to ignore the PlaceHolderVar and examine its argument
instead. In support of this, change the API of pull_var_clause() to allow
callers to request recursion into PlaceHolderVars. Currently
estimate_num_groups() is the only customer for that behavior, but where
there's one there may be others.