avoid repeated evaluations in cost_qual_eval().  This turns out to save
a useful fraction of planning time.  No change to external representation
of RestrictInfo --- although that node type doesn't appear in stored
rules anyway.

the first paragraph:

    As an example, say we wish to find all the records that
    are in the  temperature  range  of  other records. In
    effect, we need to compare the temp_lo and temp_hi
    attributes of each EMP instance to the temp_lo and
    temp_hi  attributes of all other EMP instances.

I believe that EMP should read WEATHER, as the example query that
follows joins WEATHER to itself.

EMP is often used in Oracle examples.

Regards,
Graham

Other RULE cleanups

ISO-compliant UNION/INTERSECT/EXCEPT.  Revise discussion of rule
rewriter to reflect new subselect-in-FROM implementation of views.
Miscellaneous other cleanups.

not match.

all forms of foreign keys be exposed to SQLForeignKeys.  This patch is in
addition to the ones I mailed yesterday (forget had I changed that as
well....)

Michael Fork - CCNA - MCP - A+
Network Support - Toledo Internet Access - Toledo Ohio

return foreign key information based on the pg_trigger system table.  I
have tested the patch with (what I believe) is all possible
primary/foreign key combinations -- however I may have missed some, so if
anyone feels like taking the patch for a test drive, here are some useful
links:

Michael Fork

line 1324:
"left-hand" should be "right-hand"

BTW: new document looks very good!
And the new configure/build process seems much better then before!

Thanks!

Laser

is the only diff not accounted for by fmgr rewrite...

Thanks Chih-Chang Hsieh <cch@cc.kmu.edu.tw> for finding the bug.

varlena type.  (I did not force initdb, but you won't see the fix
unless you do one.)  Also, make sure all index support operators and
functions are careful not to leak memory for toasted inputs; I had
missed some hash and rtree support ops on this point before.

mostly just on the WAL logfile nowadays.  But if people want to disable
fsync for performance, why should we say no?