Features Download
From: David Fetter <david <at> fetter.org>
Subject: == PostgreSQL Weekly News - August 23 2009 ==
Newsgroups: gmane.comp.db.postgresql.announce
Date: Monday 24th August 2009 02:42:58 UTC (over 7 years ago)
== PostgreSQL Weekly News - August 23 2009 ==

The first commitfest of the 8.5 cycle has now closed, with 35 patches
done.  Kudos to Robert Haas, commitfest mom, and all the submitters
and reviewers.

New Survey: Which of these DBA mistakes has caused you the most
down time?

== PostgreSQL Product News ==

check_postgres 2.11.0, a Nagios plugin for monitoring PostgreSQL, released.

== PostgreSQL Jobs for August ==


== PostgreSQL Local ==

PyCon Argentina will be in Beunos Aires on September 4-5 2009.

Andreas (ads) Scherbaum will be teaching a "PostgreSQL for Corporate
Use" course at the adult education center in Magdeburg, Germany on
September 7-11, 2009.  Details below:

There will be a conference in Athens, Georgia, USA on September 19,
2009.  The CfP is open.

There will be a conference in Seattle, Washington, USA October 16-18,
2009.  The CfP is open.

PGCon Brazil will be take place October 23-24 2009 at Unicamp in
Campinas, Sao Paulo state.  The CfP is open!

PGDay.EU 2009 will be at Telecom ParisTech in Paris, France on
November 6-7, 2009.  The CfP is out.  Submit!

OpenSQL Camp in Portland is looking for sponsors.  Make your travel plans
now! :)

JPUG 10th Anniversary Conference is November 20-21, 2009 in Tokyo, Japan.

FOSDEM 2010 will be in Brussels on February 6-7, 2010.

== PostgreSQL in the News ==

Planet PostgreSQL: http://planet.postgresql.org/

PostgreSQL Weekly News is brought to you this week by David Fetter
and Josh Berkus.

Submit news and announcements by Sunday at 3:00pm Pacific time.
Please send English language ones to [email protected], German language
to [email protected], Italian language to [email protected]

== Applied Patches ==

Tom Lane committed:

- In pgsql/src/backend/access/common/tupconvert.c, department of
  marginal improvements: teach tupconvert.c to avoid doing a physical
  conversion when there are dropped columns in the same places in the
  input and output tupdescs.  This avoids possible performance loss
  from the recent patch to improve dropped-column handling, in some
  cases where the old code would have worked.

- In pgsql/contrib/unaccent/unaccent.c, suppress compiler warnings
  about uninitialized variables.

- In pgsql/contrib/unaccent/Makefile, fix copy-and-pasteo that might
  explain some of the buildfarm's indigestion about this module.

- In pgsql/contrib/unaccent/unaccent.c, fix some *other* compiler
  warnings from a different gcc version.

- Print the actual DB encoding in the unaccent regression test.  This
  is to help make it more obvious what the problem is, if the encoding
  isn't what the test expects.

- Fix overflow for INTERVAL 'x ms' where x is more than a couple
  million, and integer datetimes are in use.  Per bug report from
  Hubert Depesz Lubaczewski.  Alex Hunsaker.

- Allow mixing of traditional and SQL:2008 LIMIT/OFFSET syntax.  Being
  rigid about it doesn't simplify the grammar at all, and it does
  invite confusion among those who only read the SELECT syntax summary
  and not the full details.  Per gripe from Jaime Casanova.

- In pgsql/src/tools/version_stamp.pl, fix version_stamp.pl to allow
  'alphaN' version stamps.  Not sure how Peter Eisentraut did that
  without fixing this ...

- In pgsql/src/backend/commands/explain.c, include resjunk columns in
  EXPLAIN VERBOSE output lists.  Per discussion.

- Tweak ExecIndexEvalRuntimeKeys to forcibly detoast any toasted
  comparison values before they get passed to the index access method.
  This avoids repeated detoastings that will otherwise ensue as the
  comparison value is examined by various index support functions.  We
  have seen a couple of reports of cases where repeated detoastings
  result in an order-of-magnitude slowdown, so it seems worth adding a
  bit of extra logic to prevent this.  I had previously proposed
  trying to avoid duplicate detoastings in general, but this fix takes
  care of what seems the most important case in practice with very
  little effort or risk.  Back-patch to 8.4 so that the PostGIS folk
  won't have to wait a year to have this fix in a production release.
  (The issue exists further back, of course, but the code's diverged
  enough to make backpatching further a higher-risk action.  Also it
  appears that the possible gains may be limited in prior releases
  because of different handling of lossy operators.)

- Make TRUNCATE do truncate-in-place when processing a relation that
  was created or previously truncated in the current (sub)transaction.
  This is safe since if the (sub)transaction later rolls back, we'd
  just discard the rel's current physical file anyway.  This avoids
  unreasonable growth in the number of transient files when a relation
  is repeatedly truncated.  Per a performance gripe a couple weeks ago
  from Todd Cook.

- Fix a violation of WAL coding rules in the recent patch to include
  an "all tuples visible" flag in heap page headers.  The flag update
  *must* be applied before calling XLogInsert, but heap_update and the
  tuple moving routines in VACUUM FULL were ignoring this rule.  A
  crash and replay could therefore leave the flag incorrectly set,
  causing rows to appear visible in seqscans when they should not be.
  This might explain recent reports of data corruption from Jeff Ross
  and others.  In passing, do a bit of editorialization on comments in

Peter Eisentraut committed:

- In pgsql/doc/src/sgml/Makefile, make version.sgml depend on
  configure instead of Makefile.global.  This cheats a bit, but it
  avoids unsatisfied dependencies in distribution tarballs.  (found by
  make distcheck)

- Add release notes for 8.5alpha1

- In pgsql/doc/src/sgml/release-8.5.sgml, remove duplicate from
  release notes, and reorganize slightly.

- In pgsql/doc/src/sgml/release-8.5.sgml, update alpha release notes
  to current.

- In pgsql/doc/src/sgml/release-8.5.sgml, alter release note item
  about PL/pgSQL functions and dropped columns.  per Pavel Stehule

- Version stamp 8.5alpha1

Teodor Sigaev committed:

- Introduce filtering dictionary support to tsearch.  Propagate
  --nolocale option to CREATE DATABASE command in pg_regress to allow
  correct checking of locale-sensitive contrib modules.

- Add unaccent dictionary to text search.

Andrew Dunstan committed:

- In pgsql/src/tools/msvc/vcregress.pl, pick up REGRESS_OPTS from
  contrib makefiles.  Along the way, fix ordering of makefile tests to
  mimic gmake.

Bruce Momjian committed:

- In pgsql/doc/src/sgml/unaccent.sgml, remove tabs from SGML.

== Rejected Patches (for now) ==

No one was disappointed this week :-)

== Pending Patches ==

ITAGAKI Takahiro sent in a WIP patch which adds buffer usage
information to EXPLAIN and contrib/pg_stat_statements.

Zoltan Boszormenyi sent in another revision of his ECPG patch.

Andrew Dunstan sent in a patch which sets the "explain" root element
for auto-explain XML/JSON output.

Jan Otto sent in a patch to work around a bug in OS/X Snow Leopard's

Tom Lane and Josh Berkus sent in revisions of the alpha1 release

Tom Lane sent in a WIP patch to make a case common in PostGIS work
faster by not repeatedly de-TOASTing under certain conditions.

Stef Walter sent in another revision of the patch to add samehost and
samenet directives to pg_hba.conf.

ITAGAKI Takahiro sent in another WIP patch to base dblink on SQL/MED
Foreign Data Wrappers.

ITAGAKI Takahiro sent in a patch intended to fix a bug in handling
default values in reloptions.

ITAGAKI Takahiro sent in a patch to fix the cache lookup failure in
auto_explain log_verbose

ITAGAKI Takahiro sent in a patch intended to help add 2PC semantics to
foreign data wrappers.

Robert Haas sent in an updated version of the hot standby patch.

Kevin Grittner sent in a patch to change the supplied module for Linux
startup scripts.

Heikki Linnakangas sent in two patches to return data from b-tree
indexes and use it to satisfy qualifiers.  The first patch changes the
API for indexam by splitting indexam_getnext into two pieces.  The
second does the actual work.

Robert Haas sent in another update to the hot standby patch.

James Pye sent in another revision of his PL/Python3 patch.

Roger Leigh sent in a flock of patches intended to allow psql to use
Unicode box-drawing characters for borders.

Paul Matthews sent in a patch to add a hypot() function.

---------------------------(end of broadcast)---------------------------
-To unsubscribe from this list, send an email to:

               [email protected]
CD: 3ms