clustering options.

node of a SubLink or SubPlan testexpr field.  Bug resulted from replacing
the old lefthand/exprs list fields with a simple expression field, and not
remembering that expression_tree_walker is coded to save a few cycles by
recursing directly to self on list fields (on the assumption the walker
isn't interested in List nodes per se).  On non-list fields it must of
course call the walker.  Possibly that hack isn't worth the risk of more
such bugs, but I'll leave it be for now.  Per bug report from James Robinson.

outer joins.  Originally it was only looking for overlap of the righthand
side of a left join, but we have to do it on the lefthand side too.
Per example from Jean-Pierre Pelletier.

CopyFileEx.  This avoids a warning about the function not being
present on older Windows versions.  Magnus Hagander

* Add estimated_count(*) to return an estimate of COUNT(*)
>   http://archives.postgresql.org/pgsql-hackers/2005-11/msg00943.php

* Speed up COUNT(*)
>   http://archives.postgresql.org/pgsql-hackers/2005-11/msg00943.php

instead of the SQL92 output-column-ID case.

than just showing the incomprehensible formulas.

that are almost never used.  Put documentation-related variables into the
documentation chapter.

This was required back when RESUME_INTERRUPTS could actually
execute ProcessInterrupts, but that hasn't been true since 2001...

newlines.

because the function didn't exist in 7.4.X.

Marko Kreen

>   http://archives.postgresql.org/pgsql-hackers/2006-10/msg00820.php

the politically correct view that backslash escapes are deprecated.

scenarios. With multiple clinets, only the first client got the right
scaling factor and this gave a illusion of better performance in case
of the scaling factor greater than 1.

and avoid scribbling on its result (might be safe but why risk it)

Going to wait for buildfarm results before backpatching, this time.

Remove obsolete note about logging of statements generated by plpgsql
function execution ... that doesn't happen anymore.

a fastpath function call.

provide a macro so code can still just say qsort).  Avoids linker warnings
on pickier platforms such as Darwin, and outright failure on MSVC.

instead of inserting an MSVC dependency.

< * Improve port/qsort() to handle sorts with 50% unique and 50% duplicate
<   value [qsort]
<
<   This involves choosing better pivot points for the quicksort.
<