Changes:
* add "-S", "-D" and "-m" to provide a way to show the amount of
difference relative to the total sizes of the files.
* add "-q" (suppress the "0 files changed" message for empty diffs)
From changelog:
04-Sep-2007
diffstat 1.45
add "-b" option (Debian #379380, patch by Greg Norris).
26-Aug-2007
diffstat 1.44
add a check for default-diff output, e.g., "diff foo bar|diffstat",
which has no "diff" command in the output for diffstat to guess that
a diff is beginning.
add a check after completing a chunk for unified diff to ensure that
the next line either begins another chunk for the diff, or starts
another diff, e.g., a context diff (report by Adrian Bunk).
add configure options for leak-checking valgrind, etc.
use install-sh rather than install.sh
updated config.guess and config.sub
updated configure macros CF_GCC_ATTRIBUTES CF_GCC_WARNINGS CF_MSG_LOG
CF_PATH_SYNTAX CF_VERBOSE CF_XOPEN_SOURCE
Changes since 1.41:
- fix to avoid modifying data which is being used by tsearch() for
ordering the binary tree (report by Adrian Bunk).
- do not ignore pathnames in /tmp/, since some tools create usable
pathnames for both old/new files there (Debian #376086).
- correct ifdef for fgetc_unlocked().
- updated configure macros CF_GCC_VERSION, CF_PATH_SYNTAX and
CF_XOPEN_SOURCE
- add configure check for compress, gzip and bzip2 programs that may be
used to decompress files.
* add options -l, -r and -t. add to -f option
* improve I/O on Linux using fgetc_locked().
* improve searching of long sorted lists using tsearch()
* change order of merging and prefix-stripping so stripping all prefixes,
e.g., with -p9, will be sorted as expected
* correct a reference to freed memory after merging found with valgrind.
* fix a different case for data beginning with "--" which was treated as
a header line.
* Fix allocation problems. Open files in binary mode for reading.
Getopt returns -1, not necessarily EOF. Add const where useful. Use
NO_IDENT where necessary. malloc() comes from <stdlib.h> in standard
systems
* minor fix for resync of unified diffs checks for range (line beginning
with '@' without header lines (successive lines beginning with "---"
and "+++")
* Improve check for certain lines (not those that begin a chunk) that
begin with '-', treating all except "---" as a delete (report by
James Rowe <Jay@jnrowe.uklinux.net>).
* Minor updates to configure script.
From the readme:
This program is a simple filter that reads the output of the 'diff' program,
and produces a histogram of the total number of lines that were changed. It is
useful for scanning a patch file to see which files were changed.
