reviewed by Robert Haas.

In an inherited UPDATE/DELETE, each target table has its own subplan,
because it might have a column set different from other targets.  This
means that the resjunk columns we add to support EvalPlanQual might be
at different physical column numbers in each subplan.  The EvalPlanQual
rewrite I did for 9.0 failed to account for this, resulting in possible
misbehavior or even crashes during concurrent updates to the same row,
as seen in a recent report from Gordon Shannon.  Revise the data structure
so that we track resjunk column numbers separately for each subplan.

I also chose to move responsibility for identifying the physical column
numbers back to executor startup, instead of assuming that numbers derived
during preprocess_targetlist would stay valid throughout subsequent
massaging of the plan.  That's a bit slower, so we might want to consider
undoing it someday; but it would complicate the patch considerably and
didn't seem justifiable in a bug fix that has to be back-patched to 9.0.

This reverts commit a8a88679, committed
by me earlier today (2011-01-12).  This isn't safe inside an aborted
transaction.

Noted by Tom Lane.

Stephen Frost, with some editorialization by me.

This was lost during the recent recursive make change.

Some versions of gcc complain about "variable `tablespaces' might be
clobbered by `longjmp' or `vfork'" with the original coding.  Fix by
moving the PG_TRY block into a separate subroutine.

Per my note of a couple days ago, create_index_paths would refuse to
consider any path at all for GIN indexes if the selectivity estimate came
out as 1.0; not even if you tried to force it with enable_seqscan.  While
this isn't really a bad outcome in practice, it could be annoying for
testing purposes.  Adjust the test for "is this path only useful for
sorting" so that it doesn't fire on paths with nil pathkeys, which will
include all GIN paths.

Josh Kupershmidt

When in streaming mode we can never get out, so it will never
be required, but after a base backup (or other operations)
we can get back to the loop, so the title needs to be cleared.

remove them.  Also other renaming.

Add BASE_BACKUP command to walsender, allowing it to stream a
base backup to the client (in tar format). The syntax is still
far from ideal, that will be fixed in the switch to use a proper
grammar for walsender.

No client included yet, will come as a separate commit.

Magnus Hagander and Heikki Linnakangas

No actual change in functionality ... just get rid of uselessly complex
code to pass the number of keys via extra_data.

In particular, make hstore @> '' succeed for all hstores, likewise
hstore ?& '{}'.  Previously the results were inconsistent and could
depend on whether you were using a GiST index, GIN index, or seqscan.

Move the actual functionality into a separate function that's
easier to call internally, and change the SQL-callable function
to be a wrapper calling this.

Also create a pg_abort_backup() function, only callable internally,
that does only the most vital parts of pg_stop_backup(), making it
safe(r) to call from error handlers.

This bug was exercised by contrib/intarray/bench, as noted by Tom Lane.

No need for the empty-prefix-match kluge to force a full scan anymore.

This applies the fix for bug #5784 to remaining places where we wish
to reject nulls in user-supplied arrays.  In all these places, there's
no reason not to allow a null bitmap to be present, so long as none of
the current elements are actually null.

I did not change some other places where we are looking at system catalog
entries or aggregate transition values, as the presence of a null bitmap
in such an array would be suspicious.

Result of bad testing of my last commit.

This file is now needed by pgAdmin builds, which started
failing since it was missing in the installer builds.

Add support for reading back information about the symbolic
links we've created with pgsymlink(), which are actually
Junction Points. Just like pgsymlink() can only create directory
symlinks, pgreadlink() can only read directory symlinks.

The array containment operators now behave per mathematical expectation
for empty arrays (ie, an empty array is contained in anything).
Both these operators and the query_int operators now work as expected in
GiST and GIN index searches, rather than having corner cases where the
index searches gave different answers.

Also, fix unexpected failures where the operators would claim that an array
contained nulls, when in fact there was no longer any null present (similar
to bug #5784).  The restriction to not have nulls is still there, as
removing it would take a lot of added code complexity and probably slow
things down significantly.

Also, remove the arbitrary restriction to 1-D arrays; unlike the other
restriction, this was buying us nothing performance-wise.

Assorted cosmetic improvements and marginal performance improvements, too.

This will support fixing contrib/intarray (and probably other places)
so that they don't have to fail on arrays that contain a null bitmap
but no live null entries.

The only reason this wasn't crashing while testing the core anyarray
operators was that it was disabled for those cases because of passing the
wrong type information to get_opfamily_proc :-(.  So fix that too, and make
it insist on finding the support proc --- in hindsight, silently doing
nothing is not as sane a coping mechanism as all that.

The only use we have had for amindexnulls is in determining whether an
index is safe to cluster on; but since the addition of the amclusterable
flag, that usage is pretty redundant.

In passing, clean up assorted sloppiness from the last patch that touched
pg_am.h: Natts_pg_am was wrong, and ambuildempty was not documented.

The original coding could combine duplicate entries only when they
originated from the same qual condition.  In particular it could not
combine cases where multiple qual conditions all give rise to full-index
scan requests, which is an expensive case well worth optimizing.  Refactor
so that duplicates are recognized across all the quals.

up relations, but rather order old/new relations and use the same array
index value for both.  This should speed up pg_upgrade for databases
with many relations.

that we restore by oid;  they can be handled like regular tables when
creating the file mapping structure.

Josh Kupershmidt

The underlying C code still needs work, but this at least gets its
current regression test passing again.

in pg_largeobject_metadata).