|-b bits||The code size (see below) is limited to bits, which must be in the range 9..16. The default is 16.|
|-c||Compressed or uncompressed output is written to the standard output. No files are modified. The -v option is ignored. Compression is attempted even if the results will be larger than the original.|
|-f||Files are overwritten without prompting for confirmation. Also, for compress, files are compressed even if they are not actually reduced in size.|
|-v||Print the percentage reduction of each file. Ignored by uncompress or if the -c option is also used.|
The compress utility uses a modified Lempel-Ziv algorithm. Common substrings in the file are first replaced by 9-bit codes 257 and up. When code 512 is reached, the algorithm switches to 10-bit codes and continues to use more bits until the limit specified by the -b option or its default is reached.
After the limit is reached, compress periodically checks the compression ratio. If it is increasing, compress continues to use the existing code dictionary. However, if the compression ratio decreases, compress discards the table of substrings and rebuilds it from scratch. This allows the algorithm to adapt to the next "block" of the file.
The -b option is unavailable for uncompress since the bits parameter specified during compression is encoded within the output, along with a magic number to ensure that neither decompression of random data nor recompression of compressed data is attempted.
The amount of compression obtained depends on the size of the input, the number of bits per code, and the distribution of common substrings. Typically, text such as source code or English is reduced by 50-60%. Compression is generally much better than that achieved by Huffman coding (as used in the historical command pack), or adaptive Huffman coding (as used in the historical command compact), and takes less time to compute.
.Ex -std compress uncompress
The compress utility exits 2 if attempting to compress a file would not reduce its size and the -f option was not specified and if no other error occurs.
gunzip(1), gzexe(1), gzip(1), zcat(1), zmore(1), znew(1)
.Rs A Technique for High Performance Data Compression
The compress and uncompress utilities conform to -p1003.1-2001.
The compress command appeared in BSD 4.3 .
Some of these might be considered otherwise-undocumented features.
compress: If the utility does not compress a file because doing so would not reduce its size, and a file of the same name except with an .Z extension exists, the named file is not really ignored as stated above; it causes a prompt to confirm the overwriting of the file with the extension. If the operation is confirmed, that file is deleted.
uncompress: If an empty file is compressed (using -f ), the resulting .Z file is also empty. That seems right, but if uncompress is then used on that file, an error will occur.
Both utilities: If a '' argument is used and the utility prompts the user, the standard input is taken as the users reply to the prompt.
Both utilities: If the specified file does not exist, but a similarly-named one with (for compress) or without (for uncompress) a .Z extension does exist, the utility will waste the users time by not immediately emitting an error message about the missing file and continuing. Instead, it first asks for confirmation to overwrite the existing file and then does not overwrite it.