== PostgreSQL Weekly News - August 31 2008 ==

== PostgreSQL Weekly News - August 31 2008 ==

September's Commitfest starts tomorrow.

A PostgreSQL developer room is available for the afternoon of October
16th, the day before PGDay 2008 starts.  Let's make great use of that!

September 10 is the last day to sign for PGDay 2008's "conference
bags" advertising partnership.

For a free entry PGDay 2008 edition we need a few more partners,
especially from Silver level up.  Please visit the partnership page
for more information: http://www.pgday.org/en/sponsors/campaign

PostgreSQL RPM Building Project pushed new updates to 8.3 and 8.2
branches.
http://yum.pgsqlrpms.org

Victor Vislobokov has published a short features overview in Russian.
http://postgresql.ru.net/docs/overview.html

== PostgreSQL Product News ==

Benetl 2.3 released.
http://www.benetl.net

DBD::Pg 2.10.3 released.
http://search.cpan.org/dist/DBD-Pg/

== PostgreSQL Jobs for August ==

http://archives.postgresql.org/pgsql-jobs/2008-08/threads.php

== PostgreSQL Local ==

PostgreSQL has a table at LinuxLive, Olympia, London, UK on 23-25
October, 2008.  Write to Dave Page to participate.
dpage AT pgamin DOT org

PGCon Brasil 2008 registration is open.
http://pgcon.postgresql.org.br/inscricoes.en.html

Stefan Kaltenbrunner is giving a talk about monitoring strategies for
postgresql.org at the Nagios conference 2008 September 11-12 in
Nuremberg, Germany.
http://www.netways.de/nagios_konferenz/y2008/programm/v/postgresql_monitoring/

The Prato Linux User Group will be having PostgreSQL talks in
September.  The schedule in Italian is:
http://www.prato.linux.it/serate_a_tema_2008

PGCon Brazil 2008 will be on September 26-27 at Unicamp in Campinas.
http://pgcon.postgresql.org.br/index.en.html

PgDay.fr will be October 4 in Toulouse.  The Call for Papers is open:
http://www.postgresqlfr.org/?q=node/1686
Registration:
http://www.pgday.fr/doku.php/inscription

The Open Source Day in Magdeburg, Germany will be on October 11 2008.
http://www.open-source-tag.de/
Send in your interesting talks for the PostgreSQL track!
The call for papers is open until August 31 2008.
http://www.open-source-tag.de/cfp/index.html (in german)

PostgreSQL Conference West 2008 will be October 10-12 at Portland
State University in Portland, Oregon.
http://www.postgresqlconference.org/
Talk submission at:
http://www.postgresqlconference.org/west08/talk_submission/

PGDay.(IT|EU) 2008 will be October 17 and 18 in Prato.  Registration
will open soon.
http://www.pgday.org/en/

== PostgreSQL in the News ==

Planet PostgreSQL: http://www.planetpostgresql.org/

PostgreSQL Weekly News is brought to you this week by David Fetter
Gabriele Bartolini

Submit news and announcements by Sunday at 3:00pm Pacific time.
Please send English language ones to david@fetter.org, German language
to pwn@pgug.de, Italian language to pwn@itpug.org.

== Applied Patches ==

Magnus Hagander committed:

- Convert remaining builtin set-returning functions to use OUT
  parameters, making it possible to call them without specifying a
  column list.  Jaime Casanova.

- Make stats_temp_directory PGC_SIGHUP, and document how it may cause
  a temporary "outage" of the statistics views.  This requires making
  the stats collector respond to SIGHUP, like the other utility
  processes already did.

- Unconditionally write the statsfile when SIGHUP is received, to
  minimize the window during which backends have no statistics file to
  read.

Alvaro Herrera committed:

- In pgsql/doc/src/sgml/Makefile, have man page generation work on
  VPATH builds too.

- In pgsql/src/backend/utils/hash/pg_crc.c, update URL to Ross
  William's paper.  Devrim GUNDUZ.

- In pgsql/src/bin/psql/help.c, document that \t and \x are now
  settable.

- Fixup pg_dumpall adding --lock-wait-timeout, to match pg_dump.
  David Gould.

Tom Lane committed:

- In pgsql/doc/src/sgml/config.sgml, add standard boilerplate sentence
  about when stats_temp_directory can be changed.

- Get rid of the last remaining uses of var_is_rel(), to wit some
  debugging checks in ExecIndexBuildScanKeys() that were inadequate
  anyway: it's better to verify the correct varno on an expected index
  key, not just reject OUTER and INNER.  This makes the entire current
  contents of nodeFuncs.c dead code.  I'll be replacing it with some
  other stuff later, as per recent proposal.

- Move exprType(), exprTypmod(), expression_tree_walker(), and related
  routines into nodes/nodeFuncs, so as to reduce wanton
  cross-subsystem #includes inside the backend.  There's probably more
  that should be done along this line, but this is a start anyway.

- Add %option noinput to contrib's flex scanners, to suppress gcc 4.3
  warnings.  Peter did this for core awhile ago but evidently missed
  contrib.

- In pgsql/doc/src/sgml/ref/pg_dump.sgml, fix pg_dump docs to
  acknowledge that you can use -Z with plain text output.  Pointed out
  by Daniel Migowski.

- In pgsql/src/backend/optimizer/util/clauses.c, teach
  eval_const_expressions() to simplify an ArrayCoerceExpr to a
  constant when its input is constant and the element coercion
  function is immutable (or nonexistent, ie, binary-coercible case).
  This is an oversight in the 8.3 implementation of ArrayCoerceExpr,
  and its result is that certain cases involving IN or NOT IN with
  constants don't get optimized as they should be.  Per
  experimentation with an example from Ow Mun Heng.

- Extend the parser location infrastructure to include a location
  field in most node types used in expression trees (both before and
  after parse analysis).  This allows us to place an error cursor in
  many situations where we formerly could not, because the information
  wasn't available beyond the very first level of parse analysis.
  There's a fair amount of work still to be done to persuade
  individual ereport() calls to actually include an error location,
  but this gets the initdb-forcing part of the work out of the way;
  and the situation is already markedly better than before for
  complaints about unimplementable implicit casts, such as CASE and
  UNION constructs with incompatible alternative data types.  Per my
  proposal of a few days ago.

- In pgsql/src/backend/utils/mb/conversion_procs/euc_jis_2004_and_shift_jis_2004/euc_jis_2004_and_shift_jis_2004.c,
  suppress gcc warning about possibly-uninitialized variable.  It's
  not clear to me why I'd not seen this message before --- on F-9 it
  seems to only happen if Asserts are disabled, which ought to be
  irrelevant.  Maybe that affects a decision whether to inline
  get_ten(), which would be needed to expose the warning condition to
  the compiler?  Anyway, the fix is clear.

- In pgsql/src/backend/parser/parse_expr.c, fix bug in original
  implementation of xmlserialize(): if user specifies a target type
  that isn't acceptable, the code failed to raise the proper error.
  The result instead was to return a NULL expression tree, which in a
  quick test led to a 'cache lookup failed for type 0' error later.
  Patch 8.3 only --- I fixed this in HEAD as part of recent locations
  patch.

- In GCC-based builds, use a better newNode() macro that relies on
  GCC-specific syntax to avoid a useless store into a global variable.
  Per experimentation, this works better than my original thought of
  trying to push the code into an out-of-line subroutine.

- Fix the raw-parsetree representation of star (as in SELECT * FROM or
  SELECT foo.*) so that it cannot be confused with a quoted identifier
  "*".  Instead create a separate node type A_Star to represent this
  notation.  Per pgsql-hackers discussion of 2007-Sep-27.

- Fix (hopefully) some oversights in recent Bison cleanup patch.  Per
  buildfarm results.

Peter Eisentraut committed:

- Remove all traces that suggest that a non-Bison yacc might be
  supported, and change build system to use only Bison.  Simplify
  build rules, make file names uniform.  Don't build the token table
  header file where it is not needed.

== Rejected Patches (for now) ==

No one was disappointed this week :-)

== Pending Patches ==

Laszlo Benedek sent in two revisions of a patch to let people set a
role to a user before dumping the database.

ITAGAKI Takahiro sent in an auto-explain patch.

Michelle Caisse sent in a patch to generate code coverage statistics.

Martin Pihlak sent in another revision of his plan invalidation patch
for stored procedures.

ITAGAKI Takahiro sent in a contrib version of his auto-explain patch.

Alvaro Herrera sent in a patch to add source file and line numbers to
GUC variables in the pg_settings view.

Heikki Linnakangas sent in another revision of his FSM patch.

Hitoshi Harada sent in another revision of his patch implementing
windowing functions.

Karl Schnaitter sent in a patch intended to fix a but in
backend/utils/time/combocid.c, changing the assumption that the raw
command id corresponds to cmin to allow for the case when it may be a
combo id.

Brendan Jurd sent in a patch to improve error handling in to_date()
and to_timestamp() per previous discussions.

KaiGai Kohei sent in another revision of his SE-PostgreSQL patches.

Simon Riggs sent in a patch to add resource manager hooks along with a
contrib module that uses same.

Simon Riggs sent in a WIP patch to remove JOINs under some
circumstances.

Simon Riggs sent in a patch to change the recovery infrastructure
which decouples the death of the startup process to the change of
state in the postmaster following recovery.

Ryan Bradetich sent in a pair of patches implementing unsigned integer
types and doing regression tests on same.

Radek Strnad sent in a WIP patch for collation support.

Zdenek Kotala sent in a prototype of his in-place upgrade patch.

David Rowley sent in two revisions of a patch intended to implement
Boyer-Moore string searching.

Marko Kreen sent in a patch to make gram.y use palloc and pfree for
memory management.

Marko Kreen sent in a patch intended to change from case-sensitive
units ("kB" vs. "KB") in GUC variables to case-insensitive ones, and
to clean up the way GUCs are handled on INT64_IS_BUSTED platforms.

Marko Kreen sent in a patch to clarify the error for mismatched magic
blocks in the module and server for SPI.

Gregory Stark sent in a patch to improve CLUSTER performance by having
it use a sort.