Remove halt.c, improve comments, rename manual page file.

In Danish collations, there are letter combinations which sort
higher than 'Z'.  A test for values > 'WA' was picking up rows
where the value started with 'AA', causing the test to fail.

Backpatch to 9.2, where the failing test was added.

Per report from Svenne Krap and analysis by Jeff Janes

ALTER TABLE .. VALIDATE CONSTRAINT previously
gave incorrect details about lock levels and
therefore incomplete reasons to use the option.

Initial bug report and fix from Marko Tiikkaja
Reworded by me to include comments by Kevin Grittner

MarkBufferDirtyHint() writes WAL, and should know if it's got a
standard buffer or not. Currently, the only callers where buffer_std
is false are related to the FSM.

In passing, rename XLOG_HINT to XLOG_FPI, which is more descriptive.

Back-patch to 9.3.

This avoids platform-dependent behavior wherein pg_sleep() might fail to be
interrupted by statement timeout, query cancel, SIGTERM, etc.  Also, since
there's no reason to wake up once a second any more, we can reduce the
power consumption of a sleeping backend a tad.

Back-patch to 9.3, since use of SA_RESTART for SIGALRM makes this a bigger
issue than it used to be.

Back-patch to 9.1 where the directory archive was introduced.

The exclusion of SIGALRM dates back to Berkeley days, when Postgres used
SIGALRM in only one very short stretch of code.  Nowadays, allowing it to
interrupt kernel calls doesn't seem like a very good idea, since its use
for statement_timeout means SIGALRM could occur anyplace in the code, and
there are far too many call sites where we aren't prepared to deal with
EINTR failures.  When third-party code is taken into consideration, it
seems impossible that we ever could be fully EINTR-proof, so better to
use SA_RESTART always and deal with the implications of that.  One such
implication is that we should not assume pg_usleep() will be terminated
early by a signal.  Therefore, long sleeps should probably be replaced
by WaitLatch operations where practical.

Back-patch to 9.3 so we can get some beta testing on this change.

This is just neatnik-ism, since all the tests in the code are #ifdefs,
but we shouldn't specify symbols as "Define to 1 ..." and then not
actually define them that way.

This makes it easier to write custom scripts that have different logic for
each client.

Gurjeet Singh, with some changes by me.

Let the hacking begin ...

SP-GiST's original scheme for avoiding deadlocks during concurrent index
insertions doesn't work, as per report from Hailong Li, and there isn't any
evident way to make it work completely.  We could possibly lock individual
inner tuples instead of their whole pages, but preliminary experimentation
suggests that the performance penalty would be huge.  Instead, if we fail
to get a buffer lock while descending the tree, just restart the tree
descent altogether.  We keep the old tuple positioning rules, though, in
hopes of reducing the number of cases where this can happen.

Teodor Sigaev, somewhat edited by Tom Lane

elog.c has historically treated LOG messages as low-priority during
bootstrap and standalone operation.  This has led to confusion and even
masked a bug, because the normal expectation of code authors is that
elog(LOG) will put something into the postmaster log, and that wasn't
happening during initdb.  So get rid of the special-case rule and make
the priority order the same as it is in normal operation.  To keep from
cluttering initdb's output and the behavior of a standalone backend,
tweak the severity level of three messages routinely issued by xlog.c
during startup and shutdown so that they won't appear in these cases.
Per my proposal back in December.

pg_filedump and other external utility programs are likely to want to be
able to check Postgres page checksums.  To avoid messy duplication of code,
move the checksumming functionality into an exported header file, much as
we did awhile back for the CRC code.

In passing, get rid of an unportable assumption that a static char[] array
will be word-aligned, and do some other minor code beautification.

Memory was allocated based on the sizeof a type that was not the type of
the pointer that the result was being assigned to.  The types happen to
be of the same size, but it's still wrong.

In most scenarios a portal without a ResourceOwner is dead and not subject
to any further execution, but a portal for a cursor WITH HOLD remains in
existence with no ResourceOwner after the creating transaction is over.
In this situation, if we attempt to "execute" the portal directly to fetch
data from it, we were setting CurrentResourceOwner to NULL, leading to a
segfault if the datatype output code did anything that required a resource
owner (such as trying to fetch system catalog entries that weren't already
cached).  The case appears to be impossible to provoke with stock libpq,
but psqlODBC at least is able to cause it when working with held cursors.

Simplest fix is to just skip the assignment to CurrentResourceOwner, so
that any resources used by the data output operations will be managed by
the transaction-level resource owner instead.  For consistency I changed
all the places that install a portal's resowner as current, even though
some of them are probably not reachable with a held cursor's portal.

Per report from Joshua Berry (with thanks to Hiroshi Inoue for developing
a self-contained test case).  Back-patch to all supported versions.

Several loops in the JSON parser examined a byte in memory just before
checking whether its address was in-bounds, so they could read one byte
beyond the datum's allocation.  A SIGSEGV is possible.  New in 9.3, so
no back-patch.

We would wrongly overwrite the prior stack byte if it happened to
contain '\n' or '\r'.  New in 9.3, so no back-patch.

Since the structure ends with a flexible array, doing so truncates any
vector having more than one element.  New in 9.3, so no back-patch.

Since commit f21bb9cf, this function
ignores the caller-provided length and loops until it finds a
terminator, which GetVirtualXIDsDelayingChkpt() never adds.  Restore the
previous loop control logic.  In passing, revert the addition of an
unused variable by the same commit, presumably a debugging relic.

Consumers are entitled to read the full 64 bytes pertaining to a Name;
using a shorter NULL-terminated string leads to reading beyond the end
its allocation; a SIGSEGV is possible.  Use the frequent idiom of
copying to a NameData on the stack.  New in 9.3, so no back-patch.

Extend the FDW API (which we already changed for 9.3) so that an FDW can
report whether specific foreign tables are insertable/updatable/deletable.
The default assumption continues to be that they're updatable if the
relevant executor callback function is supplied by the FDW, but finer
granularity is now possible.  As a test case, add an "updatable" option to
contrib/postgres_fdw.

This patch also fixes the information_schema views, which previously did
not think that foreign tables were ever updatable, and fixes
view_is_auto_updatable() so that a view on a foreign table can be
auto-updatable.

initdb forced due to changes in information_schema views and the functions
they rely on.  This is a bit unfortunate to do post-beta1, but if we don't
change this now then we'll have another API break for FDWs when we do
change it.

Dean Rasheed, somewhat editorialized on by Tom Lane

Per discussion  on -hackers. We treat Unicode escapes when unescaping
them similarly to the way we treat them in PostgreSQL string literals.
Escapes in the ASCII range are always accepted, no matter what the
database encoding. Escapes for higher code points are only processed in
UTF8 databases, and attempts to process them in other databases will
result in an error. \u0000 is never unescaped, since it would result in
an impermissible null byte.

Patch by me, reviewed by Tatsuo Ishii.

We need to increment the refcount on the composite type's cached tuple
descriptor while we do lookups of its column types.  Otherwise a cache
flush could occur and release the tuple descriptor before we're done with
it.  This fails reliably with -DCLOBBER_CACHE_ALWAYS, but the odds of a
failure in a production build seem rather low (since the pfree'd descriptor
typically wouldn't get scribbled on immediately).  That may explain the
lack of any previous reports.  Buildfarm issue noted by Christian Ullrich.

Back-patch to 9.1 where the bogus code was added.

lo_read()/lo_write() in libpq to avoid confusion.

pg_isready displays the host name and the port number that it uses to connect
to the server. So far, pg_isready didn't use the conninfo specified in -d option
for calculating those host name and port number. This can lead to wrong display
to a user. This commit changes pg_isready so that it uses the conninfo for that
calculation.

Original patch by Phil Sorber, modified by me.

getSchemaData() must identify extension member objects and mark them
as not to be dumped. This must happen after reading all objects that can be
direct members of extensions, but before we begin to process table subsidiary
objects. Both rules and event triggers were wrong in this regard.

Backport rules portion of patch to 9.1 -- event triggers do not exist prior to 9.3.
Suggested fix by Tom Lane, initial complaint and patch by me.

Autovacuum occurring while the test runs could allow some of the inserts to
go into recycled space, thus changing the output ordering of later queries.
While we could complicate those queries to force sorting of their output
rows, it doesn't seem like that would make the test better in any
meaningful way, and conceivably it could hide unexpected diffs.  Instead,
tweak the affected queries so that the inserted rows aren't updated by the
following UPDATE.  Per buildfarm.

When the existing code here was written, it made sense to special-case
RowExprs because that was the only way that we could handle row comparisons
at all.  Now that we have record_eq() and arrays of composites, the generic
logic for "scalar" types will in fact work on RowExprs too, so there's no
reason to throw error for combinations of RowExprs and other ways of
forming composite values, nor to ignore the possibility of using a
ScalarArrayOpExpr.  But keep using the old logic when comparing two
RowExprs, for consistency with the main transformAExprOp() logic.  (This
allows some cases with not-quite-identical rowtypes to succeed, so we might
get push-back if we removed it.)  Per bug #8198 from Rafal Rzepecki.

Back-patch to all supported branches, since this works fine as far back as
8.4.

Rafal Rzepecki and Tom Lane

Per discussion, this restriction isn't needed for any real security reason,
and it seems to confuse people more often than it helps them.  It could
also result in some database states being unrestorable.  So just drop it.

Back-patch to 9.0, where ALTER DEFAULT PRIVILEGES was introduced.

AllocateFile(), AllocateDir(), and some sister routines share a small array
for remembering requests, so that the files can be closed on transaction
failure.  Previously that array had a fixed size, MAX_ALLOCATED_DESCS (32).
While historically that had seemed sufficient, Steve Toutant pointed out
that this meant you couldn't scan more than 32 file_fdw foreign tables in
one query, because file_fdw depends on the COPY code which uses
AllocateFile().  There are probably other cases, or will be in the future,
where this nonconfigurable limit impedes users.

We can't completely remove any such limit, at least not without a lot of
work, since each such request requires a kernel file descriptor and most
platforms limit the number we can have.  (In principle we could
"virtualize" these descriptors, as fd.c already does for the main VFD pool,
but not without an additional layer of overhead and a lot of notational
impact on the calling code.)  But we can at least let the array size be
configurable.  Hence, change the code to allow up to max_safe_fds/2
allocated file requests.  On modern platforms this should allow several
hundred concurrent file_fdw scans, or more if one increases the value of
max_files_per_process.  To go much further than that, we'd need to do some
more work on the data structure, since the current code for closing
requests has potentially O(N^2) runtime; but it should still be all right
for request counts in this range.

Back-patch to 9.1 where contrib/file_fdw was introduced.

Long-standing code has called tolower() on identifier character bytes
with the high bit set. This is clearly an error and produces junk output
when the encoding is multi-byte. This patch therefore restricts this
activity to cases where there is a character with the high bit set AND
the encoding is single-byte.

There have been numerous gripes about this, most recently from Martin
Schäfer.

Backpatch to all live releases.

In 9.2, Unicode escape sequences are not analysed at all other than
to make sure that they are in the form \uXXXX. But in 9.3 many of the
new operators and functions try to turn JSON text values into text in
the server encoding, and this includes de-escaping Unicode escape
sequences. This processing had not taken into account the possibility
that this might contain a surrogate pair to designate a character
outside the BMP. That is now handled correctly.

This also enforces correct use of surrogate pairs, something that is not
done by the type's input routines. This fact is noted in the docs.

Although the DTD technically allows this, the resulting HTML is invalid
because it puts block elements inside inline elements.  DocBook 5.0 also
doesn't allow it anymore, so it's fair to assume that this was never
really intended to work.  Replace <synopsis> with <literal>, which is
the markup used elsewhere in the documentation in similar cases.

It claimed the value was always zero; it is really always -1.

Per report from Hari Babu

Swap the order of a couple of phrases to clarify what the adjective
"subsequent" applies to.

Joshua Tolley