Copy stable/9 to releng/9.3 as part of the 9.3-RELEASE cycle.

Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

Copy head to stable/9 as part of 9.0-RELEASE release cycle.

Approved by: re (implicit)

Change the internal buffer used to store input lines from a static buffer
to a dynamically allocated one in order to support input lines of
arbitrary length.

Approved by: kan (mentor)
MFC after: 1 month

Prevent stack-smashing buffer overflows in -D and -r options by using
buffer-safe string functions. The rest of the code is still probably
unsafe.

MFC after: 1 week

Add many prototypes and ANSI-ize 'patch. Not a complete job of it, but it gets
our patch to look a little more like NetBSD's, and has the nice characteristic
that the object code is exactly the same after the change as before it (even in
patch.c and pch.c, which have pesky 'assert' statements in them).

Reviewed by: /sbin/md5 on i386, alpha, sparc64
MFC after: 3 days

Add -i option (SUSv3) to patch(1). This allows one to specify a
patch file on command line instead of stdin.

Submitted by: Joseph Mallett <jmallett@xMach.org>
MFC after: 1 month

Fix previous commit such that only -S/--skip ignores errors when applying
a patch, returning -f/--force and -t/--batch to their previous semantics.

Pointed out by: asami

Fix patch such that skipping files does not count as a failure.
Previously, using -S/--skip, -f/--force, or -t/--batch to skip a patch in
a patchset still registers a failure which causes patch to return a
non-zero exit code. This is particularly undesirable with regards to
ports as there is no way to ignore the non-zero code. (Luckily, we don't
currently have any ports that make use of any of these options.)

The PR (yes, my own) is slightly incorrect: It states that -f does indeed
properly skip patches. It does, but it still sets the failure flag causing
patch to return non-zero.

PR: 19638
Submitted by: kbyanc@posi.net

Clean up some $Log$ debris, including some real messes.

$Header$ -> $FreeBSD$

Add PATCH_INDEX_FIRST environment variable to do the same as -I option

Update usage line with new option

Add -I or --index-first option to take Index: precedence over context diff,
as it was in hacked FreeBSD version

Resurrect patch 2.1 without FreeBSD Index: hack