OSDN Git Service

pg-rex/syncrep.git
15 years agoAdd URL to:
Bruce Momjian [Thu, 31 Jul 2008 20:47:30 +0000 (20:47 +0000)]
Add URL to:

* Consider decreasing the I/O caused by updating tuple hint bits

>   http://archives.postgresql.org/pgsql-patches/2008-07/msg00199.php

15 years agoRequire superuser privilege to create base types (but not composites, enums,
Tom Lane [Thu, 31 Jul 2008 16:27:16 +0000 (16:27 +0000)]
Require superuser privilege to create base types (but not composites, enums,
or domains).  This was already effectively required because you had to own
the I/O functions, and the I/O functions pretty much have to be written in
C since we don't let PL functions take or return cstring.  But given the
possible security consequences of a malicious type definition, it seems
prudent to enforce superuser requirement directly.  Per recent discussion.

15 years agoAllow I/O conversion casts to be applied to or from any type that is a member
Tom Lane [Wed, 30 Jul 2008 21:23:17 +0000 (21:23 +0000)]
Allow I/O conversion casts to be applied to or from any type that is a member
of the STRING type category, thereby opening up the mechanism for user-defined
types.  This is mainly for the benefit of citext, though; there aren't likely
to be a lot of types that are all general-purpose character strings.
Per discussion with David Wheeler.

15 years agoFlip the default typispreferred setting from true to false. This affects
Tom Lane [Wed, 30 Jul 2008 19:35:13 +0000 (19:35 +0000)]
Flip the default typispreferred setting from true to false.  This affects
only type categories in which the previous coding made *every* type
preferred; so there is no change in effective behavior, because the function
resolution rules only do something different when faced with a choice
between preferred and non-preferred types in the same category.  It just
seems safer and less surprising to have CREATE TYPE default to non-preferred
status ...

15 years agoAdjust citext to make use of the new ability to declare its type category:
Tom Lane [Wed, 30 Jul 2008 17:08:52 +0000 (17:08 +0000)]
Adjust citext to make use of the new ability to declare its type category:
by putting it into the standard string category, we cause casts from citext
to text to be recognized as "preferred" casts.  This eliminates the need
for creation of alias functions and operators that only serve to prevent
ambiguous-function errors; get rid of the ones that were in the original
commit.

15 years agoReplace the hard-wired type knowledge in TypeCategory() and IsPreferredType()
Tom Lane [Wed, 30 Jul 2008 17:05:05 +0000 (17:05 +0000)]
Replace the hard-wired type knowledge in TypeCategory() and IsPreferredType()
with system catalog lookups, as was foreseen to be necessary almost since
their creation.  Instead put the information into two new pg_type columns,
typcategory and typispreferred.  Add support for setting these when
creating a user-defined base type.

The category column is just a "char" (i.e. a poor man's enum), allowing
a crude form of user extensibility of the category list: just use an
otherwise-unused character.  This seems sufficient for foreseen uses,
but we could upgrade to having an actual category catalog someday, if
there proves to be a huge demand for custom type categories.

In this patch I have attempted to hew exactly to the behavior of the
previous hardwired logic, except for introducing new type categories for
arrays, composites, and enums.  In particular the default preferred state
for user-defined types remains TRUE.  That seems worth revisiting, but it
should be done as a separate patch from introducing the infrastructure.
Likewise, any adjustment of the standard set of categories should be done
separately.

15 years agoAdd a new, improved version of citext as a contrib module.
Tom Lane [Tue, 29 Jul 2008 18:31:20 +0000 (18:31 +0000)]
Add a new, improved version of citext as a contrib module.

David E. Wheeler

15 years agoCleanup reference to config.pl so it makes sense not only in
Magnus Hagander [Tue, 29 Jul 2008 15:23:44 +0000 (15:23 +0000)]
Cleanup reference to config.pl so it makes sense not only in
SGML source but in the actual web/pdf viewer...

15 years agoDocument which versions of ActivePerl and ActiveTcl are required
Magnus Hagander [Tue, 29 Jul 2008 15:17:27 +0000 (15:17 +0000)]
Document which versions of ActivePerl and ActiveTcl are required
for building on MSVC, and that the free distribution is enough
(no need for the enterprise version).

Per gripe from Martin Zaun.

16 years agoAs noted by Andrew Gierth, there's really no need any more to force a junk
Tom Lane [Sat, 26 Jul 2008 19:15:35 +0000 (19:15 +0000)]
As noted by Andrew Gierth, there's really no need any more to force a junk
filter to be used when INSERT or SELECT INTO has a plan that returns raw
disk tuples.  The virtual-tuple-slot optimizations that were put in place
awhile ago mean that ExecInsert has to do ExecMaterializeSlot, and that
already copies the tuple if it's raw (and does so more efficiently than
a junk filter, too).  So get rid of that logic.  This in turn means that
we can throw away ExecMayReturnRawTuples, which wasn't used for any other
purpose, and was always a kluge anyway.

In passing, move a couple of SELECT-INTO-specific fields out of EState
and into the private state of the SELECT INTO DestReceiver, as was foreseen
in an old comment there.  Also make intorel_receive use ExecMaterializeSlot
not ExecCopySlotTuple, for consistency with ExecInsert and to possibly save
a tuple copy step in some cases.

16 years agoFix parsing of LDAP URLs so it doesn't reject spaces in the "suffix" part.
Tom Lane [Thu, 24 Jul 2008 17:51:55 +0000 (17:51 +0000)]
Fix parsing of LDAP URLs so it doesn't reject spaces in the "suffix" part.
Per report from César Miguel Oliveira Alves.

16 years agoRemove some redundant tests and improve comments in next_token().
Tom Lane [Thu, 24 Jul 2008 17:43:45 +0000 (17:43 +0000)]
Remove some redundant tests and improve comments in next_token().
Cosmetic, but it might make this a bit less confusing to the next reader.

16 years agoRatchet up patch to improve autovacuum wraparound messages.
Alvaro Herrera [Wed, 23 Jul 2008 20:20:10 +0000 (20:20 +0000)]
Ratchet up patch to improve autovacuum wraparound messages.

Simon Riggs

16 years agoUse guc.c's parse_int() instead of pg_atoi() to parse fillfactor in
Tom Lane [Wed, 23 Jul 2008 17:29:53 +0000 (17:29 +0000)]
Use guc.c's parse_int() instead of pg_atoi() to parse fillfactor in
default_reloptions().  The previous coding was really a bug because pg_atoi()
will always throw elog on bad input data, whereas default_reloptions is not
supposed to complain about bad input unless its validate parameter is true.
Right now you could only expose the problem by hand-modifying
pg_class.reloptions into an invalid state, so it doesn't seem worth
back-patching; but we should get it right in HEAD because there might be other
situations in future.  Noted while studying GIN fast-update patch.

16 years agoAbort if Tcl support was configured and no tcl shell was found.
Peter Eisentraut [Wed, 23 Jul 2008 17:07:50 +0000 (17:07 +0000)]
Abort if Tcl support was configured and no tcl shell was found.
This is required because the value is substituted into the pltcl_*mod
scripts.

16 years agoUpdate link to Oleg and Teodor's GIN page.
Tom Lane [Tue, 22 Jul 2008 22:05:24 +0000 (22:05 +0000)]
Update link to Oleg and Teodor's GIN page.
(Extracted from fast-insert patch, since it ought to be back-patched)

16 years agoPublish more openly the fact that autovacuum is working for wraparound
Alvaro Herrera [Mon, 21 Jul 2008 15:27:02 +0000 (15:27 +0000)]
Publish more openly the fact that autovacuum is working for wraparound
protection.

Simon Riggs

16 years agoAdd comment about the two different query strings that ExecuteQuery()
Tom Lane [Mon, 21 Jul 2008 15:26:55 +0000 (15:26 +0000)]
Add comment about the two different query strings that ExecuteQuery()
has to deal with.

16 years agoCode review for array_fill patch: fix inadequate check for array size overflow
Tom Lane [Mon, 21 Jul 2008 04:47:00 +0000 (04:47 +0000)]
Code review for array_fill patch: fix inadequate check for array size overflow
and bogus documentation (dimension arrays are int[] not anyarray).  Also the
errhint() messages seem to be really errdetail(), since there is nothing
heuristic about them.  Some other trivial cosmetic improvements.

16 years agoAdd a pg_dump option --lock-wait-timeout to allow failing the dump if unable
Tom Lane [Sun, 20 Jul 2008 18:43:30 +0000 (18:43 +0000)]
Add a pg_dump option --lock-wait-timeout to allow failing the dump if unable
to acquire shared table locks within a specified amount of time.

David Gould

16 years agoRevert patch so .psqlrc can suppress startup banner:
Bruce Momjian [Sun, 20 Jul 2008 06:08:38 +0000 (06:08 +0000)]
Revert patch so .psqlrc can suppress startup banner:

  In psql, run .psqlrc _after_ printing warnings and banner.

16 years agoAvoid substituting NAMEDATALEN, FLOAT4PASSBYVAL, and FLOAT8PASSBYVAL into
Tom Lane [Sat, 19 Jul 2008 04:01:29 +0000 (04:01 +0000)]
Avoid substituting NAMEDATALEN, FLOAT4PASSBYVAL, and FLOAT8PASSBYVAL into
the postgres.bki file during build, because we want that file to be entirely
platform- and configuration-independent; else it can't safely be put into
/usr/share on multiarch machines.  We can do the substitution during initdb,
instead.  FLOAT4PASSBYVAL and FLOAT8PASSBYVAL are new breakage as of 8.4,
while the NAMEDATALEN hazard has been there all along but I guess no one
tripped over it.  Noticed while trying to build "universal" OS X binaries.

16 years agoAdjust things so that the query_string of a cached plan and the sourceText of
Tom Lane [Fri, 18 Jul 2008 20:26:06 +0000 (20:26 +0000)]
Adjust things so that the query_string of a cached plan and the sourceText of
a portal are never NULL, but reliably provide the source text of the query.
It turns out that there was only one place that was really taking a short-cut,
which was the 'EXECUTE' utility statement.  That doesn't seem like a
sufficiently critical performance hotspot to justify not offering a guarantee
of validity of the portal source text.  Fix it to copy the source text over
from the cached plan.  Add Asserts in the places that set up cached plans and
portals to reject null source strings, and simplify a bunch of places that
formerly needed to guard against nulls.

There may be a few places that cons up statements for execution without
having any source text at all; I found one such in ConvertTriggerToFK().
It seems sufficient to inject a phony source string in such a case,
for instance
        ProcessUtility((Node *) atstmt,
                       "(generated ALTER TABLE ADD FOREIGN KEY command)",
                       NULL, false, None_Receiver, NULL);

We should take a second look at the usage of debug_query_string,
particularly the recently added current_query() SQL function.

ITAGAKI Takahiro and Tom Lane

16 years agoProvide a function hook to let plug-ins get control around ExecutorRun.
Tom Lane [Fri, 18 Jul 2008 18:23:47 +0000 (18:23 +0000)]
Provide a function hook to let plug-ins get control around ExecutorRun.

ITAGAKI Takahiro

16 years agoProperly document archive/restore command examples on Windows.
Bruce Momjian [Fri, 18 Jul 2008 17:33:17 +0000 (17:33 +0000)]
Properly document archive/restore command examples on Windows.

ITAGAKI Takahiro

16 years agoIn psql, run .psqlrc _after_ printing warnings and banner.
Bruce Momjian [Fri, 18 Jul 2008 17:19:55 +0000 (17:19 +0000)]
In psql, run .psqlrc _after_ printing warnings and banner.

16 years agoFix a race condition that I introduced into sinvaladt.c during the recent
Tom Lane [Fri, 18 Jul 2008 14:45:48 +0000 (14:45 +0000)]
Fix a race condition that I introduced into sinvaladt.c during the recent
rewrite.  When called from SIInsertDataEntries, SICleanupQueue releases
the write lock if it has to issue a kill() to signal some laggard backend.
That still seems like a good idea --- but it's possible that by the time
we get the lock back, there are no longer enough free message slots to
satisfy SIInsertDataEntries' requirement.  Must recheck, and repeat the
whole SICleanupQueue process if not.  Noted while reading code.

16 years agoSuppress compiler warning, and not incidentally make the code more
Tom Lane [Fri, 18 Jul 2008 04:20:24 +0000 (04:20 +0000)]
Suppress compiler warning, and not incidentally make the code more
robust.  The previous coding was quite risky because it was testing
conditions different from 'is the array really allocated?'.

16 years agoImplement SQL-spec RETURNS TABLE syntax for functions.
Tom Lane [Fri, 18 Jul 2008 03:32:53 +0000 (03:32 +0000)]
Implement SQL-spec RETURNS TABLE syntax for functions.

(Unlike the original submission, this patch treats TABLE output parameters
as being entirely equivalent to OUT parameters -- tgl)

Pavel Stehule

16 years agoAdd URL for:
Bruce Momjian [Fri, 18 Jul 2008 02:03:28 +0000 (02:03 +0000)]
Add URL for:

* Implement SQL:2003 window functions

>   http://archives.postgresql.org/pgsql-hackers/2008-07/msg00232.php

16 years agoAdd MSVC++ debug libraries to .cvsignore.
Alvaro Herrera [Thu, 17 Jul 2008 21:16:23 +0000 (21:16 +0000)]
Add MSVC++ debug libraries to .cvsignore.

16 years agoAvoid crashing when a table is deleted while we're on the process of checking
Alvaro Herrera [Thu, 17 Jul 2008 21:02:31 +0000 (21:02 +0000)]
Avoid crashing when a table is deleted while we're on the process of checking
it.

Per report from Tom Lane based on buildfarm evidence.

16 years agoAdd dump support for SortBy nodes. Needed this while debugging a reported
Tom Lane [Thu, 17 Jul 2008 16:02:12 +0000 (16:02 +0000)]
Add dump support for SortBy nodes.  Needed this while debugging a reported
problem with DISTINCT, so might as well commit it.

16 years agoAdd to TODO:
Bruce Momjian [Thu, 17 Jul 2008 01:34:33 +0000 (01:34 +0000)]
Add to TODO:
>
> * Reduce locking requirements for creating a trigger
>
>   http://archives.postgresql.org/pgsql-hackers/2008-06/msg00635.php
>

16 years agoAdd URL for:
Bruce Momjian [Wed, 16 Jul 2008 21:03:45 +0000 (21:03 +0000)]
Add URL for:

* Implement SQL:2003 window functions
>
>   http://archives.postgresql.org/pgsql-hackers/2008-06/msg00380.php
>

16 years agoFix previous patch so that it actually works --- consider TRUNCATE foo, public.foo
Tom Lane [Wed, 16 Jul 2008 19:33:25 +0000 (19:33 +0000)]
Fix previous patch so that it actually works --- consider TRUNCATE foo, public.foo

16 years agoAdd a "provariadic" column to pg_proc to eliminate the remarkably expensive
Tom Lane [Wed, 16 Jul 2008 16:55:24 +0000 (16:55 +0000)]
Add a "provariadic" column to pg_proc to eliminate the remarkably expensive
need to deconstruct proargmodes for each pg_proc entry inspected by
FuncnameGetCandidates().  Fixes function lookup performance regression
caused by yesterday's variadic-functions patch.

In passing, make pg_proc.probin be NULL, rather than a dummy value '-',
in cases where it is not actually used for the particular type of function.
This should buy back some of the space cost of the extra column.

16 years agoAllow TRUNCATE foo, foo to succeed, per report from Nikhils.
Bruce Momjian [Wed, 16 Jul 2008 16:54:08 +0000 (16:54 +0000)]
Allow TRUNCATE foo, foo to succeed, per report from Nikhils.

16 years agoAdd to TODO:
Bruce Momjian [Wed, 16 Jul 2008 15:30:19 +0000 (15:30 +0000)]
Add to TODO:

>
>  o Add external tool to auto-tune some postgresql.conf parameters
>
>    http://archives.postgresql.org/pgsql-hackers/2008-06/msg00000.php
>

16 years agoSupport "variadic" functions, which can accept a variable number of arguments
Tom Lane [Wed, 16 Jul 2008 01:30:23 +0000 (01:30 +0000)]
Support "variadic" functions, which can accept a variable number of arguments
so long as all the trailing arguments are of the same (non-array) type.
The function receives them as a single array argument (which is why they
have to all be the same type).

It might be useful to extend this facility to aggregates, but this patch
doesn't do that.

This patch imposes a noticeable slowdown on function lookup --- a follow-on
patch will fix that by adding a redundant column to pg_proc.

Pavel Stehule

16 years agoAdd array_fill() to create arrays initialized with a value.
Bruce Momjian [Wed, 16 Jul 2008 00:48:54 +0000 (00:48 +0000)]
Add array_fill() to create arrays initialized with a value.

Pavel Stehule

16 years agoFix alignment of SGML array docs.
Bruce Momjian [Tue, 15 Jul 2008 18:24:59 +0000 (18:24 +0000)]
Fix alignment of SGML array docs.

16 years agoAdd Swedish_Sweden.1252 Windows locale example to docs.
Bruce Momjian [Tue, 15 Jul 2008 17:45:03 +0000 (17:45 +0000)]
Add Swedish_Sweden.1252 Windows locale example to docs.

16 years agoAddendum: psql sequence value display patch was originally written by
Bruce Momjian [Tue, 15 Jul 2008 16:06:06 +0000 (16:06 +0000)]
Addendum:  psql sequence value display patch was originally written by
Euler Taveira de Oliveira.

16 years agoAdd to TODO:
Bruce Momjian [Tue, 15 Jul 2008 15:03:41 +0000 (15:03 +0000)]
Add to TODO:

>
>  o Reduce PITR WAL file size by removing full page writes and
>    by removing trailing bytes to improve compression

16 years agoAdd TODO:
Bruce Momjian [Tue, 15 Jul 2008 13:55:20 +0000 (13:55 +0000)]
Add TODO:

>
> * Consider decreasing the I/O caused by updating tuple hint bits
>
>   http://archives.postgresql.org/pgsql-hackers/2008-05/msg00847.php
>

16 years agoDone:
Bruce Momjian [Tue, 15 Jul 2008 03:17:51 +0000 (03:17 +0000)]
Done:

>  o -Have psql show current values for a sequence

16 years agoHave psql \d show the value of sequence columns.
Bruce Momjian [Tue, 15 Jul 2008 03:16:03 +0000 (03:16 +0000)]
Have psql \d show the value of sequence columns.

Dickson S. Guedes

16 years agoClarify that locale names on Windows are more verbose.
Bruce Momjian [Tue, 15 Jul 2008 01:35:23 +0000 (01:35 +0000)]
Clarify that locale names on Windows are more verbose.

Report from Martin Saschek

16 years agoAdd comment about literal strings in our syntax not being translated in
Bruce Momjian [Mon, 14 Jul 2008 23:13:04 +0000 (23:13 +0000)]
Add comment about literal strings in our syntax not being translated in
psql.

16 years agoAdd to TODO:
Bruce Momjian [Mon, 14 Jul 2008 22:57:21 +0000 (22:57 +0000)]
Add to TODO:

> * Improve ability to modify views via ALTER TABLE
>
>   http://archives.postgresql.org/pgsql-hackers/2008-05/msg00691.php

16 years agoAdd column storage type to psql \d+ display.
Bruce Momjian [Mon, 14 Jul 2008 22:51:48 +0000 (22:51 +0000)]
Add column storage type to psql \d+ display.

Gregory Stark

16 years agoIn psql, rename trans_* variables to translate_*, for clarity.
Bruce Momjian [Mon, 14 Jul 2008 22:00:04 +0000 (22:00 +0000)]
In psql, rename trans_* variables to translate_*, for clarity.

16 years agoDone:
Bruce Momjian [Mon, 14 Jul 2008 20:54:18 +0000 (20:54 +0000)]
Done:

< * Add temporal versions of generate_series()
> * -Add temporal versions of generate_series()

16 years agoAdd URL for:
Bruce Momjian [Mon, 14 Jul 2008 17:54:54 +0000 (17:54 +0000)]
Add URL for:

* Consider allowing control of upper/lower case folding of unquoted
  identifiers

>   http://archives.postgresql.org/pgsql-hackers/2008-07/msg00415.php

16 years agoDone, per Simon:
Bruce Momjian [Mon, 14 Jul 2008 17:16:13 +0000 (17:16 +0000)]
Done, per Simon:

<  o Fix server restart problem when the server was shutdown during
>  o -Fix server restart problem when the server was shutdown during

16 years agoClean up buildfarm failures arising from the seemingly straightforward page
Tom Lane [Mon, 14 Jul 2008 03:22:32 +0000 (03:22 +0000)]
Clean up buildfarm failures arising from the seemingly straightforward page
macros patch :-(.  Results from both baiji and mastodon imply that MSVC
fails to perceive offsetof(PageHeaderData, pd_linp[0]) as a constant
expression in some contexts where offsetof(PageHeaderData, pd_linp) works
fine.  Sloth, thy name is Micro.

16 years agoCreate a type-specific typanalyze routine for tsvector, which collects stats
Tom Lane [Mon, 14 Jul 2008 00:51:46 +0000 (00:51 +0000)]
Create a type-specific typanalyze routine for tsvector, which collects stats
on the most common individual lexemes in place of the mostly-useless default
behavior of counting duplicate tsvectors.  Future work: create selectivity
estimation functions that actually do something with these stats.

(Some other things we ought to look at doing: using the Lossy Counting
algorithm in compute_minimal_stats, and using the element-counting idea for
stats on regular arrays.)

Jan Urbanski

16 years agoChange the PageGetContents() macro to guarantee its result is maxalign'd,
Tom Lane [Sun, 13 Jul 2008 21:50:04 +0000 (21:50 +0000)]
Change the PageGetContents() macro to guarantee its result is maxalign'd,
thereby forestalling any problems with alignment of the data structure placed
there.  Since SizeOfPageHeaderData is maxalign'd anyway in 8.3 and HEAD, this
does not actually change anything right now, but it is foreseeable that the
header size will change again someday.  I had to fix a couple of places that
were assuming that the content offset is just SizeOfPageHeaderData rather than
MAXALIGN(SizeOfPageHeaderData).  Per discussion of Zdenek's page-macros patch.

16 years agoClean up the use of some page-header-access macros: principally, use
Tom Lane [Sun, 13 Jul 2008 20:45:47 +0000 (20:45 +0000)]
Clean up the use of some page-header-access macros: principally, use
SizeOfPageHeaderData instead of sizeof(PageHeaderData) in places where that
makes the code clearer, and avoid casting between Page and PageHeader where
possible.  Zdenek Kotala, with some additional cleanup by Heikki Linnakangas.

I did not apply the parts of the proposed patch that would have resulted in
slightly changing the on-disk format of hash indexes; it seems to me that's
not a win as long as there's any chance of having in-place upgrade for 8.4.

16 years agoFix a couple of stray misuses of "binary compatible".
Tom Lane [Sat, 12 Jul 2008 16:20:06 +0000 (16:20 +0000)]
Fix a couple of stray misuses of "binary compatible".

16 years agoMore replacements of binary compatible to binary coercible.
Peter Eisentraut [Sat, 12 Jul 2008 10:44:56 +0000 (10:44 +0000)]
More replacements of binary compatible to binary coercible.

16 years agoDon't make --enable-cassert turn on RANDOMIZE_ALLOCATED_MEMORY automatically;
Tom Lane [Sat, 12 Jul 2008 02:28:43 +0000 (02:28 +0000)]
Don't make --enable-cassert turn on RANDOMIZE_ALLOCATED_MEMORY automatically;
it's just too dang expensive.  Per recent discussion, but I just got my
nose rubbed in it again while doing some performance checking.

16 years agoConst-ify the arguments of str_tolower() and friends to suppress compile
Tom Lane [Sat, 12 Jul 2008 00:44:38 +0000 (00:44 +0000)]
Const-ify the arguments of str_tolower() and friends to suppress compile
warnings.  Clean up various unneeded cruft that was left behind after
creating those routines.  Introduce some convenience functions str_tolower_z
etc to eliminate tedious and error-prone double arguments in formatting.c.
(Currently there seems no need to export the latter, but maybe reconsider
this later.)

16 years agoMulti-column GIN indexes. Teodor Sigaev
Tom Lane [Fri, 11 Jul 2008 21:06:29 +0000 (21:06 +0000)]
Multi-column GIN indexes.  Teodor Sigaev

16 years agoAdd caching of query to GIN/GiST consistent function.
Teodor Sigaev [Fri, 11 Jul 2008 11:56:48 +0000 (11:56 +0000)]
Add caching of query to GIN/GiST consistent function.
Per performance gripe from nomao.com

16 years agoAllow binary-coercible types for cast function arguments and return types.
Peter Eisentraut [Fri, 11 Jul 2008 07:02:43 +0000 (07:02 +0000)]
Allow binary-coercible types for cast function arguments and return types.
Document return type of cast functions.

Also change documentation to prefer the term "binary coercible" in its
present sense instead of the previous term "binary compatible".

16 years agoMake sure we only try to free snapshots that have been passed through
Alvaro Herrera [Fri, 11 Jul 2008 02:10:14 +0000 (02:10 +0000)]
Make sure we only try to free snapshots that have been passed through
CopySnapshot, per Neil Conway.  Also add a comment about the assumption in
GetSnapshotData that the argument is statically allocated.

Also, fix some more typos in comments in snapmgr.c.

16 years agoFix a few typos in comments in snapmgr.c, and sort header inclusions
Neil Conway [Fri, 11 Jul 2008 00:00:29 +0000 (00:00 +0000)]
Fix a few typos in comments in snapmgr.c, and sort header inclusions
alphabetically.

16 years agoAdd unchangeable GUC "variables" segment_size, wal_block_size, and
Tom Lane [Thu, 10 Jul 2008 22:08:17 +0000 (22:08 +0000)]
Add unchangeable GUC "variables" segment_size, wal_block_size, and
wal_segment_size to make those configuration parameters available to clients,
in the same way that block_size was previously exposed.  Bernd Helmle, with
comments from Abhijit Menon-Sen and some further tweaking by me.

16 years agoTighten up SS_finalize_plan's computation of valid_params to exclude Params of
Tom Lane [Thu, 10 Jul 2008 02:14:03 +0000 (02:14 +0000)]
Tighten up SS_finalize_plan's computation of valid_params to exclude Params of
the current query level that aren't in fact output parameters of the current
initPlans.  (This means, for example, output parameters of regular subplans.)
To make this work correctly for output parameters coming from sibling
initplans requires rejiggering the API of SS_finalize_plan just a bit:
we need the siblings to be visible to it, rather than hidden as
SS_make_initplan_from_plan had been doing.  This is really part of my response
to bug #4290, but I concluded this part probably shouldn't be back-patched,
since all that it's doing is to make a debugging cross-check tighter.

16 years agoFix mis-calculation of extParam/allParam sets for plan nodes, as seen in
Tom Lane [Thu, 10 Jul 2008 01:17:29 +0000 (01:17 +0000)]
Fix mis-calculation of extParam/allParam sets for plan nodes, as seen in
bug #4290.  The fundamental bug is that masking extParam by outer_params,
as finalize_plan had been doing, caused us to lose the information that
an initPlan depended on the output of a sibling initPlan.  On reflection
the best thing to do seemed to be not to try to adjust outer_params for
this case but get rid of it entirely.  The only thing it was really doing
for us was to filter out param IDs associated with SubPlan nodes, and that
can be done (with greater accuracy) while processing individual SubPlan
nodes in finalize_primnode.  This approach was vindicated by the discovery
that the masking method was hiding a second bug: SS_finalize_plan failed to
remove extParam bits for initPlan output params that were referenced in the
main plan tree (it only got rid of those referenced by other initPlans).
It's not clear that this caused any real problems, given the limited use
of extParam by the executor, but it's certainly not what was intended.

I originally thought that there was also a problem with needing to include
indirect dependencies on external params in initPlans' param sets, but it
turns out that the executor handles this correctly so long as the depended-on
initPlan is earlier in the initPlans list than the one using its output.
That seems a bit of a fragile assumption, but it is true at the moment,
so I just documented it in some code comments rather than making what would
be rather invasive changes to remove the assumption.

Back-patch to 8.1.  Previous versions don't have the case of initPlans
referring to other initPlans' outputs, so while the existing logic is still
questionable for them, there are not any known bugs to be fixed.  So I'll
refrain from changing them for now.

16 years agoIncrease PG_SYSLOG_LIMIT (the max line length sent to syslog()) from 128 to
Tom Lane [Wed, 9 Jul 2008 15:56:49 +0000 (15:56 +0000)]
Increase PG_SYSLOG_LIMIT (the max line length sent to syslog()) from 128 to
1024 to improve performance when sending large elog messages.  Also add a
comment about why we use that number.

Since this represents an externally visible behavior change, and might
possibly result in portability issues, it seems best not to back-patch it.

16 years agoFix performance bug in write_syslog(): the code to preferentially break the
Tom Lane [Tue, 8 Jul 2008 22:17:41 +0000 (22:17 +0000)]
Fix performance bug in write_syslog(): the code to preferentially break the
log message at newlines cost O(N^2) for very long messages with few or no
newlines.  For messages in the megabyte range this became the dominant cost.
Per gripe from Achilleas Mantzios.

Patch all the way back, since this is a safe change with no portability
risks.  I am also thinking of increasing PG_SYSLOG_LIMIT, but that should
be done separately.

16 years agoFix WAL file cutoff point calculation in pg_standby.
Heikki Linnakangas [Tue, 8 Jul 2008 15:11:58 +0000 (15:11 +0000)]
Fix WAL file cutoff point calculation in pg_standby.

Patch by Simon Riggs, per bug report from Ferenc Felhoffer

16 years agoMinor improvements to the Gin internal documentation.
Neil Conway [Tue, 8 Jul 2008 03:25:42 +0000 (03:25 +0000)]
Minor improvements to the Gin internal documentation.

16 years agoAdd comment for deadlock_timeout:
Bruce Momjian [Tue, 8 Jul 2008 02:07:29 +0000 (02:07 +0000)]
Add comment for deadlock_timeout:

        /* This is PGC_SIGHUP so all backends have the same value. */

16 years agoFix estimate_num_groups() to assume that GROUP BY expressions yielding boolean
Tom Lane [Mon, 7 Jul 2008 20:24:55 +0000 (20:24 +0000)]
Fix estimate_num_groups() to assume that GROUP BY expressions yielding boolean
results always contribute two groups, regardless of the expression contents.
This is very substantially more accurate than the regular heuristic for
certain boolean tests like "col IS NULL".  Per gripe from Sam Mason.

Back-patch to all supported releases, since the behavior of
estimate_num_groups() hasn't changed all that much since 7.4.

16 years agoFix AT TIME ZONE (in all three variants) so that we first try to interpret
Tom Lane [Mon, 7 Jul 2008 18:09:46 +0000 (18:09 +0000)]
Fix AT TIME ZONE (in all three variants) so that we first try to interpret
the timezone argument as a timezone abbreviation, and only try it as a full
timezone name if that fails.  The zic database has four zones (CET, EET, MET,
WET) that are full daylight-savings zones and yet have names that are the
same as their abbreviations for standard time, resulting in ambiguity.
In the timestamp input functions we resolve the ambiguity by preferring the
abbreviation, and AT TIME ZONE should work the same way.  (No functionality
is lost because the zic database also has other names for these zones, eg
Europe/Zurich.)  Per gripe from Jaromir Talir.

Backpatch to 8.1.  Older releases did not have the issue because AT TIME ZONE
only accepted abbreviations not zone names.  (Thus, this patch also arguably
fixes a compatibility botch introduced at 8.1: in ambiguous cases we now
behave the same as 8.0 did.)

16 years agoAdded documentation for function xmlagg.
Peter Eisentraut [Mon, 7 Jul 2008 16:11:35 +0000 (16:11 +0000)]
Added documentation for function xmlagg.

16 years agoPrevent integer overflows during units conversion when displaying a GUC
Tom Lane [Sun, 6 Jul 2008 19:48:45 +0000 (19:48 +0000)]
Prevent integer overflows during units conversion when displaying a GUC
variable that has units.  Per report from Stefan Kaltenbrunner.

Backport to 8.2.  I also backported my patch of 2007-06-21 that prevented
comparable overflows on the input side, since that now seems to have enough
field track record to be back-patched safely.  That patch included addition
of hints listing the available unit names, which I did not bother to strip
out of it --- this will make a little more work for the translators, but
they can copy the translation from 8.3, and anyway an untranslated hint
is better than no hint.

16 years agoFix initialization of GinScanEntryData.partialMatch
Teodor Sigaev [Fri, 4 Jul 2008 13:21:18 +0000 (13:21 +0000)]
Fix initialization of GinScanEntryData.partialMatch

16 years agoFix a couple of bugs in win32 shmem name generation:
Magnus Hagander [Fri, 4 Jul 2008 10:50:18 +0000 (10:50 +0000)]
Fix a couple of bugs in win32 shmem name generation:
* Don't cut off the prefix. With this fix, it's again readable.
* Properly store it in the Global namespace as intended.

16 years agoAdd a function pg_get_keywords() to let clients find out the set of keywords
Tom Lane [Thu, 3 Jul 2008 20:58:47 +0000 (20:58 +0000)]
Add a function pg_get_keywords() to let clients find out the set of keywords
known to the SQL parser.  Dave Page

16 years agoDon't refer to the database name "regression" inside the regression test
Peter Eisentraut [Thu, 3 Jul 2008 16:01:10 +0000 (16:01 +0000)]
Don't refer to the database name "regression" inside the regression test
scripts, to allow running the test successfully with another database name.

16 years agoDon't print the name of the database in psql \z.
Peter Eisentraut [Thu, 3 Jul 2008 15:59:55 +0000 (15:59 +0000)]
Don't print the name of the database in psql \z.

16 years agoClean up weird whitespace. Separate patch to simplifiy the next change.
Peter Eisentraut [Thu, 3 Jul 2008 15:36:20 +0000 (15:36 +0000)]
Clean up weird whitespace.  Separate patch to simplifiy the next change.

16 years agoWhen an ERROR happens on a dblink remote connection, take
Joe Conway [Thu, 3 Jul 2008 03:56:57 +0000 (03:56 +0000)]
When an ERROR happens on a dblink remote connection, take
pains to pass the ERROR message components locally, including
using the passed SQLSTATE. Also wrap the passed info in an
appropriate CONTEXT message. Addresses complaint by Henry
Combrinck. Joe Conway, with much good advice from Tom Lane.

16 years agoFix psql's \d and allied commands to work with all server versions back to 7.4.
Tom Lane [Thu, 3 Jul 2008 03:37:17 +0000 (03:37 +0000)]
Fix psql's \d and allied commands to work with all server versions back to 7.4.
Guillaume Lelarge, with some additional fixes by me.

16 years agoUpdate source code comment about when to use gettext_noop().
Bruce Momjian [Thu, 3 Jul 2008 02:49:54 +0000 (02:49 +0000)]
Update source code comment about when to use gettext_noop().

16 years agoFix transaction-lifespan memory leak in xpath(). Report by Matt Magoffin,
Tom Lane [Thu, 3 Jul 2008 00:04:24 +0000 (00:04 +0000)]
Fix transaction-lifespan memory leak in xpath().  Report by Matt Magoffin,
fix by Kris Jurka.

16 years agoDocumentation patch by Kevin L. McBride explaining GUC lock variables,
Bruce Momjian [Tue, 1 Jul 2008 21:49:04 +0000 (21:49 +0000)]
Documentation patch by Kevin L. McBride explaining GUC lock variables,
which are available if LOCK_DEBUG is defined.

16 years agoRemove GUC extra_desc strings that are redundant with the enum value lists.
Tom Lane [Tue, 1 Jul 2008 21:07:33 +0000 (21:07 +0000)]
Remove GUC extra_desc strings that are redundant with the enum value lists.

16 years agoAdd URL for:
Bruce Momjian [Tue, 1 Jul 2008 18:26:30 +0000 (18:26 +0000)]
Add URL for:

* Add database and transaction-level triggers

16 years agoAdd to TODO:
Bruce Momjian [Tue, 1 Jul 2008 18:09:09 +0000 (18:09 +0000)]
Add to TODO:

> * Fix TRUNCATE ... RESTART IDENTITY so its affect on sequences is rolled
>   back on transaction abort

16 years agoMove volatility, language, etc. modifiers before function body in the pg_dump
Heikki Linnakangas [Tue, 1 Jul 2008 11:46:48 +0000 (11:46 +0000)]
Move volatility, language, etc. modifiers before function body in the pg_dump
output for CREATE FUNCTION. This makes it easier to read especially if the
function body is long.

Original idea and patch by Greg Sabino Mullane, though this is a stripped
down version of that.

16 years agoExtend VacAttrStats to allow typanalyze functions to store statistic values
Heikki Linnakangas [Tue, 1 Jul 2008 10:33:09 +0000 (10:33 +0000)]
Extend VacAttrStats to allow typanalyze functions to store statistic values
of different types than the underlying column. The capability isn't yet
used for anything, but will be required by upcoming patch to analyze
tsvector columns.

Jan Urbanski

16 years ago"debug" level was supposed to be hidden, since it's just an alias for debug2.
Magnus Hagander [Tue, 1 Jul 2008 06:36:11 +0000 (06:36 +0000)]
"debug" level was supposed to be hidden, since it's just an alias for debug2.

16 years agoSplit apart message_level_options into one set for server-side settings and
Magnus Hagander [Tue, 1 Jul 2008 06:08:31 +0000 (06:08 +0000)]
Split apart message_level_options into one set for server-side settings and
one for client-side, restoring the previous behaviour with different
sort order for the 'log' level. Also, remove redundant list of available
options, since the enum code will output it automatically.

16 years agoFix identify_system_timezone() so that it tests the behavior of the system
Tom Lane [Tue, 1 Jul 2008 03:40:55 +0000 (03:40 +0000)]
Fix identify_system_timezone() so that it tests the behavior of the system
timezone setting in the current year and for 100 years back, rather than
always examining years 1904-2004.  The original coding would have problems
distinguishing zones whose behavior diverged only after 2004; which is a
situation we will surely face sometime, if it's not out there already.

In passing, also prevent selection of the dummy "Factory" timezone, even
if that's exactly what the system is using.  Reporting time as GMT seems
better than that.