<html><body>
<br /><div align="center"><em>--- forwarded message ---</em></div><br />Mon 25 Dec 2006 11:13:27 PM EST from David Fetter to PostgreSQL Announce <br />Subject: [ANNOUNCE] == PostgreSQL Weekly News - December 25 2006 ==<br /><br />== PostgreSQL Weekly News - December 25 2006 ==<br /><br />PostgreSQL Weekly News extends a warm welcome to PostreSQL's<br />Indonesian community.<br />a xhref=http://tech.groups.yahoo.com/group/postgresql_indo/ TARGET=webcit01http://tech.groups.yahoo.com/group/postgresql_indo//A<br /><br />Peter Eisentraut committed preliminary support for the XML data type<br />per SQL:2003's SQL/XML standard, and lots of people have been testing<br />it.<br />a xhref=http://developer.postgresql.org/pgdocs/postgres/datatype-xml.html TARGET=webcit01http://developer.postgresql.org/pgdocs/postgres/datatype-xml.html/A<br /><br />Oleg Bartunov an Teodor Sigaev submitted a patch to tsearch2 which<br />adds lots of new features, sponsored by EnterpriseDB<br />a xhref=http://mira.sai.msu.su/~megera/pgsql/ftsdoc/ TARGET=webcit01http://mira.sai.msu.su/~megera/pgsql/ftsdoc//A<br /><br />Sebastien Lardiere's new book,<br />PostgreSQL 8.1 Administration et Exploitation d'une Base de Donnees,<br />is out.<br />a xhref=http://www.postgresqlfr.org/?q=node/1031 TARGET=webcit01http://www.postgresqlfr.org/?q=node/1031/A<br /><br />== PostgreSQL Product News ==<br /><br />phpPgAdmin is now in Fedora Extras.<br />http://people.planetpostgresql.org/devrim/index.php?/archives/73-Pushed-phpPgAdmin-into-Fedora-Core-Extras.html<br /><br />Update Scout is now tracking PostgreSQL<br />a xhref=http://www.update-scout.com TARGET=webcit01http://www.update-scout.com/A<br /><br />SuSE PostgreSQL RPMs released.<br />a xhref=ftp://ftp.suse.com/pub/projects/postgresql/ TARGET=webcit01ftp://ftp.suse.com/pub/projects/postgresql//A<br /><br />Grifinor, an open source 3D virtual globe platform, uses PostgreSQL.<br />a xhref=http://www.grifinor.net/ TARGET=webcit01http://www.grifinor.net//A<br /><br />== PostgreSQL Jobs for December ==<br /><br />a xhref=http://archives.postgresql.org/pgsql-jobs/2006-12/threads.php TARGET=webcit01http://archives.postgresql.org/pgsql-jobs/2006-12/threads.php/A<br /><br />== PostgreSQL Local ==<br /><br />The Italian PostgreSQL community will be holding a PostgreSQL day this<br />summer. Bookmark the link below to participate.<br />a xhref=http://www.pgday.it TARGET=webcit01http://www.pgday.it/A<br /><br />PostgreSQL Weekly News extends a warm welcome to PostreSQL's<br />Indonesian community.<br />a xhref=http://tech.groups.yahoo.com/group/postgresql_indo/ TARGET=webcit01http://tech.groups.yahoo.com/group/postgresql_indo//A<br /><br />Gavin Sherry is running a PostgreSQL miniconf at Linux.Conf.Au in<br />Sydney on Tuesday the 16th of January 2007.<br />a xhref=http://lca2007.linux.org.au/Miniconfs/PostgreSQL TARGET=webcit01http://lca2007.linux.org.au/Miniconfs/PostgreSQL/A If you would like to<br />attending, email gavin AT alcove . com . au<br /><br />== PostgreSQL in the News ==<br /><br />Planet PostgreSQL: a xhref=http://www.planetpostgresql.org/ TARGET=webcit01http://www.planetpostgresql.org//A<br /><br />General Bits, Archives and occasional new articles:<br />a xhref=http://www.varlena.com/GeneralBits/ TARGET=webcit01http://www.varlena.com/GeneralBits//A<br /><br />PostgreSQL Weekly News is brought to you this week by David Fetter,<br />Pavel Stehule, Robert Treat<br /><br />== Applied Patches ==<br /><br />Andrew Dunstan committed:<br /><br />- A Tom Lane/Andrew Dunstan patch to interpret a dbName param to<br /> PQsetdbLogin as a conninfo string if it contains an = sign.<br /><br />- Fix thinko in placement of TimeValStruct typedef in Windows case, as<br /> reported by Magnus Hagander.<br /><br />- In psql, enable \timing output for \copy commands<br /><br />Tom Lane committed:<br /><br />- Make HISTCONTROL=ignoredups work again (broken by misordering of<br /> operations during recent code refactoring). Per bug #2840 from Ned<br /> Crigler.<br /><br />- Bring some order and sanity to error handling in the xml patch. Use<br /> a TRY block instead of (inadequate) ad-hoc coding to ensure that<br /> libxml is cleaned up after a failure. Report the intended SQLCODE<br /> instead of defaulting to XX000. Avoid risking use of a dangling<br /> pointer by keeping the persistent error buffer in TopMemoryContext.<br /> Be less trusting that error messages don't contain %.<br /><br />- Fix machine-dependent crash in sqlchar_to_unicode(). Get rid of<br /> bletcherous and unsafe manipulation of global encoding setting.<br /> Clean up libxml reporting mechanism a bit (it still looks like a<br /> dangling-pointer crash waiting to happen, though, not to mention<br /> being far less than sane from a localization standpoint).<br /><br />- Code review for XML patch. Instill a bit of sanity in the location<br /> of the XmlExpr code in various lists, use a representation that has<br /> some hope of reverse-listing correctly (though it's still a<br /> de-escaping function shy of correctness), generally try to make it<br /> look more like Postgres coding conventions.<br /><br />- Suppress various compiler warnings in new xml code.<br /><br />- Restructure operator classes to allow improved handling of<br /> cross-data-type cases. Operator classes now exist within "operator<br /> families". While most families are equivalent to a single class,<br /> related classes can be grouped into one family to represent the fact<br /> that they are semantically compatible. Cross-type operators are now<br /> naturally adjunct parts of a family, without having to wedge them<br /> into a particular opclass as we had done originally. This commit<br /> restructures the catalogs and cleans up enough of the fallout so<br /> that everything still works at least as well as before, but most of<br /> the work needed to actually improve the planner's behavior will come<br /> later. Also, there are not yet CREATE/DROP/ALTER OPERATOR FAMILY<br /> commands; the only way to create a new family right now is to allow<br /> CREATE OPERATOR CLASS to make one by default. I owe some more<br /> documentation work, too. But that can all be done in smaller pieces<br /> once this infrastructure is in place.<br /><br />- Set pg_am.amstrategies to zero for index AMs that don't have fixed<br /> operator strategy numbers, ie, GiST and GIN. This is almost<br /> cosmetic enough to not need a catversion bump, but since the<br /> opr_sanity regression test has to change in sync with the catalog<br /> entry, I figured I'd better do one.<br /><br />Bruce Momjian committed:<br /><br />- Remove unnecessary parentheses in if() statements in<br /> pgsql/src/backend/utils/adt/float.c<br /><br />- Change a VACUUM manual page word from 'deleted' to 'expired', so<br /> DELETE and UPDATE are clearly covered by the term.<br /><br />- Change documentation to state that it is a bad idea to CREATE TYPE<br /> names that begin with an underscore. Doing so is not actually<br /> disallowed.<br /><br />- For GUC values, check for partial string matches on 'on' and 'off',<br /> but require at least two characters for uniqueness. This now<br /> matches the behavior of other boolean strings we support, per report<br /> from Gurjeet Singh.<br /><br />- Add a link to the developer's FAQ for my article about how companies<br /> can work effectively with open source communities.<br /><br />- Slight adjustments to xml documentation to reflect functionality<br /> just added.<br /><br />- 8.3 release schedule is year 2007, not 2006.<br /><br />- Add timeline for next release to developer's FAQ.<br /><br />- Change TODO to reflect done-ness of "Improve xid wraparound<br /> detection by recording per-table rather than per-database.<br /><br />Peter Eisentraut committed:<br /><br />- Initial SQL/XML support: xml data type and initial set of functions,<br /> complete with version bump and expected output fixes.<br /><br />Teodor Sigaev committed:<br /><br />- In tsearch2, fix conversion for 'PFX flag N num'.<br /><br />== Rejected Patches (for now) ==<br /><br />No one was disappointed this week :-)<br /><br />== Pending Patches ==<br /><br />Simon Riggs sent in a patch which implements explain_analyze_timer per<br />previous discussions.<br /><br />Tom Dunstan sent in a new version of his enums patch, now current with<br />CVS HEAD and with docs, bounds checks and error codes.<br /><br />ITAGAKI Takahiro sent in a patch for load distributed checkpoint per<br />his proposal below.<br />a xhref=http://archives.postgresql.org/pgsql-hackers/2006-12/msg00337.php TARGET=webcit01http://archives.postgresql.org/pgsql-hackers/2006-12/msg00337.php/A<br /><br />Glen Parker submitted a patch which adds two C functions which expose<br />n_live_tuples and n_dead_tuples, SQL functions to expose them to SQL<br />land, and corresponding fields added to pg_stat_all_tables.<br /><br />Guillaume LeLarge sent in a patch which changes the capitalization of<br />DateStyle to datestyle, bringing it into line with other similar<br />names.<br /><br />Jeremy Drake sent in a patch which makes psql's \lo_* commands respect<br />the -q flag (or other methods of setting quiet mode) as well as HTML<br />output mode.<br /><br />Bruce Momjian sent in a patch per Roman Kononov's but #2846 to handle<br />underflow and NaN better.<br /><br /><br /><br />---------------------------(end of broadcast)---------------------------<br />-To unsubscribe from this list, send an email to:<br /><br /> pgsql-announce-unsubscribe@postgresql.org<br /><br />
</body></html>