a more logical ordering.

> 	o Add SET SCHEMA

formatting (\timing) works correctly.  Change "Total time" to "Time"
since there is nothing that "total" refers to.  Remove non-multibyte
code.

except that it's more robust, reconnects less often, and is NLS'ed.

(using an elog(LOG) ), as well as includes a few more comment fixes.

Neil Conway

> * Allow free space map to be auto-sized or warn when it is too small

> * Allow REINDEX to rebuild all indexes, remove /contrib/reindex
> 	o Allow CLUSTER to cluster all tables, remove clusterdb

from Tom Lane.

> 	o -Cluster all tables at once using pg_index.indisclustered set during

anonymous return type SRF code. It gets rid of the superflous
'pg_locks_result' that Bruce/Tom had commented on. Otherwise, no
changes in functionality.

Neil Conway

the 'override CPPFLAGS' to include the source directory during compile,
and makes the install target look in the proper place for the man page.

Changes are only required when building outside the source directory.

J. R. Nield

copying into a fixed-size buffer (in this case, a buffer of
NAMEDATALEN bytes). AFAICT nothing to worry about here, but worth
fixing anyway...

Neil Conway

bit on the indexes.

I also attach clusterdb and clusterdb.sgml; both of them are blatant
rips of vacuumdb and vacuumdb.sgml, but get the job done.  Please review
them, as I'm probably making a lot of mistakes with SGML and I can't
compile it here.

vacuumdb itself is not very comfortable to use when the databases have
passwords, because it has to connect once for each table (I can probably
make it connect only once for each database; should I?).  Because of
this I added a mention of PGPASSWORDFILE in the documentation, but I
don't know if that is the correct place for that.

Alvaro Herrera

>   partial page writes [wal]

> * Add CORRESPONDING BY to UNION/INTERSECT/EXCEPT

< 	o Add plsh server-side shell language (Peter E)