error messages look at least a little bit like the message style
guidelines say.

not-too-bright compilers.  Per buildfarm results.

three slightly different copies of this file?)

theoretically vary depending on what the compile-time locale setting is.
Hence, force it to see LC_CTYPE=C to ensure consistent build results.
(It's likely that this makes no difference in practice, since our
specification for "identifier" surely includes both ends of any possible
uppercase/lowercase pair anyway.  But it should silence warnings about
ambiguous character classes that are reported by some buildfarm members.)

sanely if the loop value overflows int32 on the way to the end value.
Avoid useless computation of "SELECT 1" when BY is omitted.  Avoid some
type-punning between Datum and int4 that dates from the original coding.

Andrew Dunstan.  Minor other improvements in documentation of integer
FOR loops.

of variable substitution and plan caching behavior in dedicated sections.
(A lot of this material existed already, but was scattered in various places
in the chapter.)  Reorganize material a little bit, mostly to try to avoid
diving into deep details in the first introductory sections.  Document some
fine points that had escaped treatment before, notably the ability to qualify
plpgsql variable names with block labels.  Some minor wordsmithing here and
there.

by Solaris 10 and possibly others.

Stefan Kaltenbrunner

from old versions of gcc.  It's not clear to me that this is really
necessary for correctness, but less warnings are always good.
Per buildfarm results and local testing.

columns, per my gripe earlier today.  Make it look a bit less like
someone's first effort at backend coding.

unsigned char).  Fortunately we still have buildfarm machines that
will flag this.  Seems to be new in CVS HEAD, so no backpatch.

on Cygwin (answer: we don't).  Also try to unwind the #ifdef spaghetti
a little bit.  Untested but hopefully I didn't break anything.

add dashes to the start/end of the comment block to try to prevent
this happening in the future.

may or may not be harmless.  Report from Stefan, patch from Heikki.

define pg_dlsym() as returning a PGFunction pointer, not just any
pointer-to-function.  But many are not.  Suppress compiler warnings
on platforms that aren't careful by inserting explicit casts at the
two call sites that didn't have a cast already.  Per Stefan.

suppressing Sun Studio compiler warnings.  Per Stefan.

when built on a 64-bit machine.  Per buildfarm results extracted by Stefan.

enabled, because the only Kerberos library supported always contains it.

such as OpenBSD (possibly all Heimdal).

Stefan Kaltenbrunner

free function provided in the Kerberos library.
This fixes a very hard to track down heap corruption on windows
when using debug runtimes.

Per suggestion from Alvaro.

* use elog not ereport for debug
* fix debug levels for some output
* properly check for memory allocation errors in a couple of missed places

building from a distribution tarball does not.

Documentation still being written, will be committed later.

Henry B. Hotz and Magnus Hagander

literally, whether quoted or not.  Since we allow $ as a character within
identifiers, this behavior is useful, whereas the previous behavior of
treating it as the regexp ending anchor was nearly useless given that the
pattern is automatically anchored anyway.  This affects the arguments of
psql's \d commands as well as pg_dump's -n and -t switches.  Per discussion.

the same outputs as stddev_samp() and var_samp() respectively.

SIGQUIT) will be recognized and processed while waiting for input,
rather than only after something has been typed.  Also make SIGQUIT
do the same thing as SIGTERM in single-user mode, ie, do a normal
shutdown and exit.  Since it's relatively easy to provoke SIGQUIT
from the keyboard, people may try that instead of control-D, and we'd
rather this leads to orderly shutdown.  Per report from Leon Mergen
and subsequent discussion.

we don't know at that point which relation OID to tell pgstat to forget.
The code was passing the relfilenode, which is incorrect, and could possibly
cause some other relation's stats to be zeroed out.  While we could try to
clean this up, it seems much simpler and more reliable to let the next
invocation of pgstat_vacuum_tabstat() fix things; which indeed is how it
worked before I introduced the buggy code into 8.1.3 and later :-(.
Problem noticed by Itagaki Takahiro, fix is per subsequent discussion.