I finally got around to testing bzp.
http://cs.fit.edu/~mmahoney/compression/text.html#2839
It is on the Pareto frontier for compression speed
but I have not updated the graphs yet.
Also I noticed in testing that if I compress a file in another folder like below then the path name is stored wrong. I compress \res\enwik8 and the name is stored as s\enwik8 with the first 3 characters cut off. Then when I decompress to . it creates a new subdirectory .s and saves .s\enwik8
Code:
C:\tmp>timer bzp c \res\enwik8 x8
Timer 3.01 Copyright (c) 2002-2003 Igor Pavlov 2003-07-10
BZP 0.1 Experimental archiver
(C)2008 By Nania Francesco Antonio(Italy - Messina)
-> 31563865
time 13.218 s.
Kernel Time = 0.296 = 00:00:00.296 = 2%
User Time = 12.093 = 00:00:12.093 = 91%
Process Time = 12.390 = 00:00:12.390 = 93%
Global Time = 13.219 = 00:00:13.219 = 100%
C:\tmp>bzp l x8
BZP 0.1 Experimental archiver
(C)2008 By Nania Francesco Antonio(Italy - Messina)
Archive list:
s\enwik8 size=100000000
Total n. 1 Files
There was no problem for the actual test where I compressed enwik8/9 in the current directory (to get proper file size).
Also, this is actually bzp 0.2 (according to readme.txt) although program says 0.1.