Skip to content

Wöchentlicher PostgreSQL Newsletter - 25. Juli 2010

Der Originalartikel befindet sich unter:

== Wöchentlicher PostgreSQL Newsletter - 25. Juli 2010 ==

== PostgreSQL Produkt Neuigkeiten ==

Another PostgreSQL Diff Tool 2.0 Beta 2, ein Werkzeug um Datenbankdiffs
zu erstellen, ist erschienen.

JASPA 0.1RC1, ein Java GIS welches PostgreSQL nutzen kann, ist

MyJSQLView 3.22, ein GUI Werkzeug welches mit PostgreSQL verwendet
werden kann, ist erschienen.

Simpycity 0.3.1, ein Python Mapping System für
PostgreSQL, ist erschienen.

tail_n_mail 1.15.0, ein Perl Programm zum Suchen in Logdateien und
Mailen der Ergebnisse, ist erschienen.

== PostgreSQL Jobs im Juli ==

== PostgreSQL Lokal ==

Reuven Lerner gibt einen 5-tägigen PostgreSQL Kurs vom 1. bis 5.
August am Hi-Tech College in Herzliya, Israel.

FrOSCon 2010 findet in St. Augustin, Deutschland am 21. und 22. August
2010 statt. Die Deutsche PostgreSQL Usergruppe hat ihren eigenen Devroom
und sucht noch Vorträge. Weitere Informationen unter:

Der Call for Papers für West ist bis zum 5. September 2010 offen.
Details unter:

== PostgreSQL in den News ==

Planet PostgreSQL:

Dieser wöchentliche PostgreSQL Newsletter wurde erstellt von David

Sende Neuigkeiten und Ankündigungen bis Sonntag, 15 Uhr Pazifischer
Zeit. Bitte sende englische Beiträge an, deutsche an, italienische an

== Reviews ==

KaiGai Kohei reviewed Robert Haas's patch to add get_whatever_oid

== Applied Patches ==

Peter Eisentraut committed:

- In pgsql/src/interfaces/libpq/fe-connect.c, portability fixes for
  Solaris for requirepeer feature patch per report from Dave Page

- Add more checks against altering typed tables.  1. Prohibit altering
  column type.  2. Prohibit changing inheritance.  3. Move checks from
  Exec to Prep phases in ALTER TABLE code.  Backpatched to 9.0

- In pgsql/doc/src/sgml/config.sgml, remove tab from SGML file

- In pgsql/doc/src/sgml/datatype.sgml, tidy up boolean data type page.
  Thom Brown.

- In pgsql/doc/src/sgml/datatype.sgml, fix typo.

- In pgsql/doc/src/sgml/xfunc.sgml, add INSERT statement to example so
  that it can be reproduced from John Gage

Robert Haas committed:

- Change the default value of standard_conforming_strings to on.  This
  change should be publicized to driver maintainers at once and
  release-noted as an incompatibility with previous releases.

- Add restart_after_crash GUC.  Normally, we automatically restart
  after a backend crash, but in some cases when PostgreSQL is invoked
  by clusterware it may be desirable to suppress this behavior, so we
  provide an option which does this.  Since no existing GUC group
  quite fits, create a new group called "error handling options" for
  this and the previously undocumented GUC exit_on_error, which is now
  documented.  Review by Fujii Masao.

- Make hstore regression tests independent of
  standard_conforming_strings.  Per buildfarm.

- Make ECPG regression tests independent of
  standard_conforming_strings.  Per buildfarm, again.

- Add \conninfo command to psql, to show current connection info.
  David Christensen. Reviewed by Steve Singer.  Some further changes
  by me.

- In pgsql/src/bin/psql/command.c, have \conninfo mention the port
  even for local sockets.  Per discussion with David Christensen,
  there can be multiple instances of PG accessible via local sockets,
  and you need the port to see which one you're actually connected to.
  David's original patch worked this way, but I inadvertently ripped
  it out during commit.

- Centralize DML permissions-checking logic.  Remove bespoke code in
  DoCopy and RI_Initial_Check, which now instead fabricate call
  ExecCheckRTPerms with a manufactured RangeTblEntry.  This is
  intended to make it feasible for an enhanced security provider to
  actually make use of ExecutorCheckPerms_hook, but also has the
  advantage that RI_Initial_Check can allow use of the fast-path when
  column-level but not table-level permissions are present.  KaiGai
  Kohei.  Reviewed (in an earlier version) by Stephen Frost, and by
  me.  Some further changes to the comments by me.

- Add options to force quoting of all identifiers.  I've added a
  quote_all_identifiers GUC which affects the behavior of the backend,
  and a --quote-all-identifiers argument to pg_dump and pg_dumpall
  which sets the GUC and also affects the quoting done internally by
  those applications.  Design by Tom Lane; review by Alex Hunsaker; in
  response to bug #5488 filed by Hartmut Goebel.

- In pgsql/src/bin/pg_dump/pg_dumpall.c, fix bogus server version in
  pg_dumpall --quote-all-identifiers.

- In pgsql/src/backend/replication/walsender.c, add missing function
  prototype.  Fujii Masao.

- In pgsql/src/backend/access/transam/xact.c, avoid deep recursion
  when assigning XIDs to multiple levels of subxacts.  Backpatch to
  8.0.  Andres Freund, with cleanup and adjustment for older branches
  by me.

- In pgsql/src/bin/psql/command.c, have psql avoid describing local
  sockets as host names.  We now use the phrase 'via local socket in'
  rather than 'on host' in both \c and \conninfo output, when
  applicable.  Fujii Masao, with some kibitzing by me.

- CREATE TABLE IF NOT EXISTS.  Reviewed by Bernd Helmle.

Bruce Momjian committed:

- Properly replay CREATE TABLESPACE during crash recovery by deleting
  directory/symlink before creation.  Report from Tom Lane.  Backpatch
  to 9.0.

- Prevent pg_upgrade from migrating databases that use reg* data types
  where the oid is not preserved by pg_upgrade (everything but
  pg_type).  Update documentation.  Per bug report from

- In contrib/pg_upgrade, adjust
  check_for_isn_and_int8_passing_mismatch() so it is called for all
  migrations.  Backpatch to 9.0.

- In pgsql/doc/src/sgml/ref/create_table.sgml, remove tabs from sgml

Magnus Hagander committed:

- Backpatch reservation of shared memory region during backend startup
  on Windows, so that memory allocated by starting third party DLLs
  doesn't end up conflicting.  The same functionality has been in 8.3
  and 8.4 for almost a year, and seems to have solved some of the more
  common shared memory errors on Windows.

== Abgelehnte Patches (bis jetzt) ==

No one was disappointed this week :-)

== Eingesandte Patches ==

KaiGai Kohei sent in another revision of the patch to rework DML
permisssion checks.

ITAGAKI Takahiro sent in another revision of the patch to implement
more of SQL/MED.

Pavel Stehule sent in two more revisions of the patch to implement
to_string() and to_array().

Etienne Dube sent in a patch to fix a "could not attach to shared
memory" issue on Windows in the 8.2 branch.

ITAGAKI Takahiro and Pavel Stehule traded revisions of the string
functions patch.

Marc Cousin sent in a patch to correct a cursor declaration in the
"migrating from Oracle" section of the PL/pgsql docs.

Daniele Varrazzo sent in a patch to make more links from the libpq
docs.  This came from actual usage in psycopg.

Satoshi Nagayasu sent in another revision of the patch to redo the
ECPG docs.

Alexander Korotkov sent in two more revisions of the patch to make
levenshtein functions multibyte-aware.

Leonardo Francalanci sent in another revision of the patch to improve
the CLUSTER implentation.

Pavel Stehule sent in two more revisions of the patch to enhance \ef
and add \sf (show function) in psql.

Fujii Masao sent in another revision of the patch to do synchronous

Teodor Sigaev sent in another flock of patches implementing
K-Nearest-Neighbor for GiST indexes.

Alex Hunsaker sent in two more revisions of the patch to add
functional depency inference to GROUP BY.

Kevin Grittner sent in another revision of the patch to make a more
correct implementation of hypotenuse.

Joseph Adams sent in a patch to implement the JSON data type.

Mike Fowler sent in another revision of the patch to add XMLEXISTS to
the grammar.

Joseph Adams sent in a patch to add a utf8_to_unicode() function,
similar to the corresponding unicode_to_utf8() function.  Among other
things, this smooths the way for the upcoming JSON data type.


No Trackbacks


Display comments as Linear | Threaded

No comments

Add Comment

Enclosing asterisks marks text as bold (*word*), underscore are made via _word_.
E-Mail addresses will not be displayed and will only be used for E-Mail notifications.
To leave a comment you must approve it via e-mail, which will be sent to your address after submission.
Form options