JojoDiff - diff utility for binary files

Copyright © 2020 Joris Heirbaut
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program. If not, see https://www.gnu.org/licenses/gpl-3.0.html.

This software is hosted by:

1. Purpose

JDIFF differentiates two files so that the second file can be recreated from the first by "undiffing".
JDIFF aims for the smallest possible diff-file.
For example:
• jdiff -j old-file.tar new-file.tar dif-file.jdf
• jdiff -u old-file.tar dif-file.jdf new-file2.tar
  
will recreate new-file2.tar identical to new-file.tar.
Possible applications include:
• incremental backups, or
  
• synchronising files between two computers over a slow network.
Download JojoDiff from Jojo's Binary Diff Download Page.

2. Version and history

The current version of this utility is bèta 0.8.4c from September 2020.
Change history:

v0.8.5	October 2020	Bugfix in Windows binary, improved internal logic.
v0.8.4	September 2020	Improved progress feedback, support for sequential files.
v0.8	December 2011	Conversion to C++.
v0.7	November 2009	Index table collision strategy for uniform distribution.
v0.6	April 2005	Large files support.
v0.5	September 2003	Revision of file-buffering logic.
v0.4c	January 2003	Selection algorithm between multiple matches.
v0.4b	October 2002	Optimize matches.
v0.4a	September 2002	Select "best" of multiple matches.
v0.3a	July 2002	Copy/insert algorithm.
v0.2c	July 2002	Bugfix on divide-by-zero in verbose mode.
v0.2b	July 2002	Bugfix on code-length of 252.
v0.2a	June 2002	Optimized patch files.
v0.1	June 2002	Insert/delete algorithm.

3. Installation

On Windows systems:
• Compiled executables are within the "win" directory. You can run them from a command prompt.
On Linux systems:
• Compile the source by running "make" within the "src" directory or use the compiled executable within the linux directory.
• Copy the resulting executable to your /usr/local/bin.

4. Usage

• jdiff -j [options] source_file destination_file [diff_file]
• jdiff -u [options] source_file diff_file [destination_file]

Options:

-j		JDiff: create a difference file.
-u		Undiff: undiff a difference file.
-v	--verbose	Verbose: greeting, results and tips.
-vv		Extra Verbose: progress info and statistics.
-vvv		Ultra Verbose: all info, including help and details.
-h -hh	--help	Help, additional help (-hh) and exit.
-l	--listing	Detailed human readable output.
-r	--regions	Grouped human readable output.
-c	--console	Write verbose and debug info to stdout.
-b	--better	Better: use more memory, search more.
-bb		Best: even more memory, search more.
-f	--lazy	Lazy: no unbuffered searching (often slower).
-ff		Lazier: no full index table.
-p	--sequential-source	Sequential source (to avoid !) (with - for stdin).
-q	--sequential-dest	Sequential destination (with - for stdin).
-s	--stdio	Use stdio files (for testing).
-a	--search-size	Size (in KB) to search (default=buffer-size).
-i	--index-size	Size (in MB) for index table (default 64MB).
-k	--block-size	Block size in bytes for reading (default 8192).
-m	--buffer-size	Size (in KB) for search buffers (default 2MB).
-n	--search-min	Minimum number of matches to search (default 2).
-x	--search-max	Maximum number of matches to search (default 512).

Hint: Do not use jdiff on compressed files. Rather use jdiff first and compress afterwards, e.g.: jdiff -j old new | gzip >dif.jdf.gz (or 7z with -si)
Principles:
JDIFF tries to find equal regions between two binary files using a heuristic hash algorithm and outputs the differences between both files. Heuristics are generally used for improving performance and memory usage, at the cost of accuracy. Therefore, this program may not find a minimal set of differences between files.

Notes:
• Options -b, -bb, -f and -ff should be used before other options.
• Accuracy may be improved by increasing the index table size (-i) or the buffer size (-m).
• The index table size is always lowered to the nearest lower prime number.
• Output is sent to standard output if no output file is specified.