Mercurial > hg > Applications > mh

comparison miscellany/compress-4.0/compress.1 @ 0:bce86c4163a3

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Initial revision
author kono
date Mon, 18 Apr 2005 23:46:02 +0900
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:bce86c4163a3
1 .PU
2 .TH COMPRESS 1 local
3 .SH NAME
4 compress, uncompress, zcat \- compress and expand data
5 .SH SYNOPSIS
6 .ll +8
7 .B compress
8 [
9 .B \-f
10 ] [
11 .B \-v
12 ] [
13 .B \-c
14 ] [
15 .B \-V
16 ] [
17 .B \-b
18 .I bits
19 ] [
20 .I "name \&..."
21 ]
22 .ll -8
23 .br
24 .B uncompress
25 [
26 .B \-f
27 ] [
28 .B \-v
29 ] [
30 .B \-c
31 ] [
32 .B \-V
33 ] [
34 .I "name \&..."
35 ]
36 .br
37 .B zcat
38 [
39 .B \-V
40 ] [
41 .I "name \&..."
42 ]
43 .SH DESCRIPTION
44 .I Compress
45 reduces the size of the named files using adaptive Lempel-Ziv coding.
46 Whenever possible,
47 each file is replaced by one with the extension
48 .B "\&.Z,"
49 while keeping the same ownership modes, access and modification times.
50 If no files are specified, the standard input is compressed to the
51 standard output.
52 Compressed files can be restored to their original form using
53 .I uncompress
54 or
55 .I zcat.
56 .PP
57 The
58 .B \-f
59 option will force compression of
60 .I name.
61 This is useful for compressing an entire directory,
62 even if some of the files do not actually shrink.
63 If
64 .B \-f
65 is not given and
66 .I compress
67 is run in the foreground,
68 the user is prompted as to whether an existing file should be overwritten.
69 .PP
70 The
71 .B \-c
72 option makes
73 .I compress/uncompress
74 write to the standard output; no files are changed.
75 The nondestructive behavior of
76 .I zcat
77 is identical to that of
78 .I uncompress
79 .B \-c.
80 .PP
81 .I Compress
82 uses the modified Lempel-Ziv algorithm popularized in
83 "A Technique for High Performance Data Compression",
84 Terry A. Welch,
85 .I "IEEE Computer,"
86 vol. 17, no. 6 (June 1984), pp. 8-19.
87 Common substrings in the file are first replaced by 9-bit codes 257 and up.
88 When code 512 is reached, the algorithm switches to 10-bit codes and
89 continues to use more bits until the
90 limit specified by the
91 .B \-b
92 flag is reached (default 16).
93 .I Bits
94 must be between 9 and 16. The default can be changed in the source to allow
95 .I compress
96 to be run on a smaller machine.
97 .PP
98 After the
99 .I bits
100 limit is attained,
101 .I compress
102 periodically checks the compression ratio. If it is increasing,
103 .I compress
104 continues to use the existing code dictionary. However,
105 if the compression ratio decreases,
106 .I compress
107 discards the table of substrings and rebuilds it from scratch. This allows
108 the algorithm to adapt to the next "block" of the file.
109 .PP
110 Note that the
111 .B \-b
112 flag is omitted for
113 .I uncompress,
114 since the
115 .I bits
116 parameter specified during compression
117 is encoded within the output, along with
118 a magic number to ensure that neither decompression of random data nor
119 recompression of compressed data is attempted.
120 .PP
121 .ne 8
122 The amount of compression obtained depends on the size of the
123 input, the number of
124 .I bits
125 per code, and the distribution of common substrings.
126 Typically, text such as source code or English
127 is reduced by 50\-60%.
128 Compression is generally much better than that achieved by
129 Huffman coding (as used in
130 .IR pack ),
131 or adaptive Huffman coding
132 .RI ( compact ),
133 and takes less time to compute.
134 .PP
135 Under the
136 .B \-v
137 option,
138 a message is printed yielding the percentage of
139 reduction for each file compressed.
140 .PP
141 If the
142 .B \-V
143 option is specified, the current version and compile options are printed on
144 stderr.
145 .PP
146 Exit status is normally 0;
147 if the last file is larger after (attempted) compression, the status is 2;
148 if an error occurs, exit status is 1.
149 .SH "SEE ALSO"
150 pack(1), compact(1)
151 .SH "DIAGNOSTICS"
152 Usage: compress [\-dfvcV] [\-b maxbits] [file ...]
153 .in +8
154 Invalid options were specified on the command line.
155 .in -8
156 Missing maxbits
157 .in +8
158 Maxbits must follow
159 .BR \-b \.
160 .in -8
161 .IR file :
162 not in compressed format
163 .in +8
164 The file specified to
165 .I uncompress
166 has not been compressed.
167 .in -8
168 .IR file :
169 compressed with
170 .I xx
171 bits, can only handle
172 .I yy
173 bits
174 .in +8
175 .I File
176 was compressed by a program that could deal with
177 more
178 .I bits
179 than the compress code on this machine.
180 Recompress the file with smaller
181 .IR bits \.
182 .in -8
183 .IR file :
184 already has .Z suffix -- no change
185 .in +8
186 The file is assumed to be already compressed.
187 Rename the file and try again.
188 .in -8
189 .IR file :
190 filename too long to tack on .Z
191 .in +8
192 The file cannot be compressed because its name is longer than
193 12 characters.
194 Rename and try again.
195 This message does not occur on BSD systems.
196 .in -8
197 .I file
198 already exists; do you wish to overwrite (y or n)?
199 .in +8
200 Respond "y" if you want the output file to be replaced; "n" if not.
201 .in -8
202 uncompress: corrupt input
203 .in +8
204 A SIGSEGV violation was detected which usually means that the input file has
205 been corrupted.
206 .in -8
207 Compression:
208 .I "xx.xx%"
209 .in +8
210 Percentage of the input saved by compression.
211 (Relevant only for
212 .BR \-v \.)
213 .in -8
214 -- not a regular file: unchanged
215 .in +8
216 When the input file is not a regular file,
217 (e.g. a directory), it is
218 left unaltered.
219 .in -8
220 -- has
221 .I xx
222 other links: unchanged
223 .in +8
224 The input file has links; it is left unchanged. See
225 .IR ln "(1)"
226 for more information.
227 .in -8
228 -- file unchanged
229 .in +8
230 No savings is achieved by
231 compression. The input remains virgin.
232 .in -8
233 .SH "BUGS"
234 Although compressed files are compatible between machines with large memory,
235 .BR \-b \12
236 should be used for file transfer to architectures with
237 a small process data space (64KB or less, as exhibited by the DEC PDP
238 series, the Intel 80286, etc.)