bzip2

Section: User Commands (1)
Page Index

NAME

bzme - recompress gziped, ziped, ... files into bzip2

SYNOPSIS

bzmz [ -fh ] [ filenames ... ]

DESCRIPTION

bzme recompresses files using the Burrows-Wheeler block sorting text compression algorithm, and Huffman coding. Compression is generally considerably better than that achieved by more conventional LZ77/LZ78-based compressors, and approaches the performance of the PPM family of statistical compressors.

bzme expects a list of file names to accompany the command-line flags. Each file is replaced by a recompressed version of itself, with the name as described in NEW NAMES section.

bzme won't overwrite by default existing files. If you want this to happen, specify the -f flag.

NEW NAMES

bzip2 attempts to guess the filename for the decompressed file from that of the compressed file as follows:

       filename.tgz    becomes   filename.tar.bz2
       filename.tar.gz becomes   filename.tar.bz2
       filename.zip    becomes   filename.tar.bz2
       filename.z      becomes   filename.bz2
       filename.Z      becomes   filename.bz2
       filename.gz     becomes   filename.bz2

If the file does not end in one of the recognised endings, .tgz,

or .zip, bzme complains that it cannot guess if the name of the recompressed file (ie it doesn't detect the original name to be a file compressed in a known format)

OPTIONS

-f: Force overwrite of output files, even if -k is used. Normally, bzip2 will not overwrite existing output files.
-k: Keep (don't delete) input files during compression or decompression.

SECURITY

bzme will keep source file if there's an error while decompressing source file or recompressing new file (or -k option is used of course).

bzme won't overwite the target file, even if -k option is used, if the source file doesn't exists.

As a self-check for your protection, bzip2 uses 32-bit CRCs to make sure that the decompressed version of a file is identical to the original. This offers a better protection against corruption of the compressed data than offered by gzip.

SPACE GAIN

Compression is only performed if the compressed file is smaller than the original: the original file is only removed if the newly compressed file is smaller, else the new recompressed file is deleted.

Text (aka non binary) files're quite nearly always better compressed by bzip2 rather than gzip.

MEMORY VS SPACE TRADEOFF

There're two things :

Consumed CPU time

The needed cpu time is reduced by decompressing only one time. Files to recompress were compressed through compress or gzip, used to be decompressed by gunzip -t in order to check that the original file was ok. This resulted in passing two times the data in the decompression process (one to check integrity, one to recompress). temporary space usage will be zero since bzme will use a pipe rather than a temporary file as it does in the early ages. Source error're detected through bash PIPESTATUS feature.

Occupied space

While recompressing files, if they were compressed through compress or gzip, temporary space usage will be zero since bzme will use a pipe rather than a temporary file as it did in the early ages. Source error're detected through bash PIPESTATUS feature. Zip files're still fully decompressed on disk.

As for the recompressed file and original file, only the smallest file is kept.

RETURN VALUES

0 for a normal exit. 1 will be returned if an unknown option is passed.

BUGS

Bash getopt (which is used to analyse options) isn't gnu style aware, ie cmd opt1 file1 file2 opt2 will result in ignoring opt2 option.

Solaris/SunOs du doesn't supports gnu option, and thus, bzme won't work on those OSes unless GNU fileutils got installed. I had once a day patched bzme to use right options for solaris but i had lost my changes.

So solaris remains unsupported.

AUTHOR

Thierry Vignaud <tvignaud@mandrakesoft.com>, 1999-2002