'Twas my fault, I think.

know if that case ever breaks again...

to give wrong results: it should be looking at inJoinSet not inFromCl.
Also, make 'modified' flag be local to ApplyRetrieveRule: we should
append a rule's quals to the query iff that particular rule applies,
not if we have fired any previously-considered rule for the query!

Move several platforms to the unsupported list :(

as 'alphaev5', cf report from Stepanov 13-Apr-00.

(SELECT FROM table*).  Cause was reference to 'eref' field of an RTE,
which is null in an RTE loaded from a stored rule parsetree.  There
wasn't any good reason to be touching the refname anyway...

table for an average of NTUP_PER_BUCKET tuples/bucket, but cost_hashjoin
was assuming a target load of one tuple/bucket.  This was causing a
noticeable underestimate of hashjoin costs.

and errors from pickier compilers.

Not a big step forward, but it's something...

estimation operators.  See contrib/likeplanning/README for info.

Remove knowledge of path to documentation source directory from perl
script, instead have Makefile pass it to script.

\copy without arguments failed
commands with too many arguments were too silent

(LIKE and regexp matches).  These are not yet referenced in pg_operator,
so by default the system will continue to use eqsel/neqsel.
Also, tweak convert_to_scalar() logic so that common prefixes of strings
are stripped off, allowing better accuracy when all strings in a table
share a common prefix.

prevent duplicate OIDs from being added.  Clean up redundant error
messages.

print OIDs as %u not %d.

With the default option, pg_ctl can show messages from
postmaster upon its failure.