packJPG, packMP3, packPNM, packARI, packHUF, packLZH and unpackJPG are now updated with new contact information and some minor bugfixes. This step was required because the original home of packJPG on the HTW Aalen University website is closing soon (granted, that website is severely outdated now, too). With the original
Archive for packLZH
Needless to say, this step was long overdue. You may now download all of my source codes, follow my development, fork my software and collaborate with me on my official GitHub repo.
This blog will still get updated, but only with public releases.
Well finally! This is a big update featuring various code improvements and also much better compression ratios:
- Complete rewrite of all coding routines
- Added some developer options for you to fiddle around
The compression routines are a bit more complex now and feature a combination of huffman, LZ and VLI (variable length integer) coding. Be aware that this still has to be thoroughly tested. If you happen to find a bug, please contact me.
The new developer options are the following:
- Maximum LZ count [-lc?]: decrease for more speed, but also worse compression.
- Maximum LZ distance [-ld?]: increasing slows down, but (typically!) improves compression.
- LZH block size [-lb?]: decrease for less memory consumption and more adaptivity (which might not always be a good thing), increase for more memory consumption and less adaptivity.
A rather minor update, functionality wise:
- Major source code cleanup
- Huffman coder improved to better handle special cases
- LZ coder handling improved
- Some minor bug fixes
This update was mainly done to prepare for the next version which will actually have better compression ratio. For the developers among you, this will be a far better base for your own projects than v0.4 was before.
And now for something completely new:
packLZH is based on packHUF and adds the well known LZ77 algorithm to the mix. Compression ratios highly depend on file content. In most cases, they should be a little worse than those reached by packARI, but at faster compression speeds and much faster decompression speeds.
This has basically not been tested at all, so proceed at your own risk.