1 /*
2 * $OpenBSD: locate.code.c,v 1.21 2019/01/17 06:15:44 tedu Exp $
3 *
4 * Copyright (c) 1989, 1993
5 * The Regents of the University of California. All rights reserved.
6 *
7 * This code is derived from software contributed to Berkeley by
8 * James A. Woods.
9 *
10 * Redistribution and use in source and binary forms, with or without
11 * modification, are permitted provided that the following conditions
12 * are met:
13 * 1. Redistributions of source code must retain the above copyright
14 * notice, this list of conditions and the following disclaimer.
15 * 2. Redistributions in binary form must reproduce the above copyright
16 * notice, this list of conditions and the following disclaimer in the
17 * documentation and/or other materials provided with the distribution.
18 * 3. Neither the name of the University nor the names of its contributors
19 * may be used to endorse or promote products derived from this software
20 * without specific prior written permission.
21 *
22 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
23 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
24 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
25 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
26 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
27 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
28 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
29 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
30 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
31 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32 * SUCH DAMAGE.
33 */
34
35 /*
36 * PURPOSE: sorted list compressor (works with a modified 'find'
37 * to encode/decode a filename database)
38 *
39 * USAGE: bigram < list > bigrams
40 * process bigrams (see updatedb) > common_bigrams
41 * code common_bigrams < list > squozen_list
42 *
43 * METHOD: Uses 'front compression' (see ";login:", Volume 8, Number 1
44 * February/March 1983, p. 8). Output format is, per line, an
45 * offset differential count byte followed by a partially bigram-
46 * encoded ascii residue. A bigram is a two-character sequence,
47 * the first 128 most common of which are encoded in one byte.
48 *
49 * EXAMPLE: For simple front compression with no bigram encoding,
50 * if the input is... then the output is...
51 *
52 * /usr/src 0 /usr/src
53 * /usr/src/cmd/aardvark.c 8 /cmd/aardvark.c
54 * /usr/src/cmd/armadillo.c 14 armadillo.c
55 * /usr/tmp/zoo 5 tmp/zoo
56 *
57 * The codes are:
58 *
59 * 0-28 likeliest differential counts + offset to make nonnegative
60 * 30 switch code for out-of-range count to follow in next word
61 * 31 an 8 bit char followed
62 * 128-255 bigram codes (128 most common, as determined by 'updatedb')
63 * 32-127 single character (printable) ascii residue (ie, literal)
64 *
65 * The locate database store any character except newline ('\n')
66 * and NUL ('\0'). The 8-bit character support don't wast extra
67 * space until you have characters in file names less than 32
68 * or greater than 127.
69 *
70 *
71 * SEE ALSO: updatedb.sh, ../bigram/locate.bigram.c
72 *
73 * AUTHOR: James A. Woods, Informatics General Corp.,
74 * NASA Ames Research Center, 10/82
75 * 8-bit file names characters:
76 * Wolfram Schneider, Berlin September 1996
77 */
78
79 #include <err.h>
80 #include <errno.h>
81 #include <stdio.h>
82 #include <stdlib.h>
83 #include <string.h>
84 #include <unistd.h>
85 #include <limits.h>
86
87 #include "locate.h"
88
89 #define BGBUFSIZE (NBG * 2) /* size of bigram buffer */
90
91 u_char buf1[PATH_MAX] = " ";
92 u_char buf2[PATH_MAX];
93 u_char bigrams[BGBUFSIZE + 1] = { 0 };
94
95 #define LOOKUP 1 /* use a lookup array instead a function, 3x faster */
96
97 #ifdef LOOKUP
98 #define BGINDEX(x) (big[(u_char)*x][(u_char)*(x + 1)])
99 typedef short bg_t;
100 bg_t big[UCHAR_MAX + 1][UCHAR_MAX + 1];
101 #else
102 #define BGINDEX(x) bgindex(x)
103 typedef int bg_t;
104 int bgindex(char *);
105 #endif /* LOOKUP */
106
107
108 void usage(void);
109 extern int optind;
110 extern int optopt;
111
112 int
main(int argc,char * argv[])113 main(int argc, char *argv[])
114 {
115 u_char *cp, *oldpath, *path;
116 int ch, code, count, diffcount, oldcount;
117 FILE *fp;
118 int i, j;
119
120 if (pledge("stdio rpath", NULL) == -1)
121 err(1, "pledge");
122
123 while ((ch = getopt(argc, argv, "")) != -1)
124 switch (ch) {
125 default:
126 usage();
127 }
128 argc -= optind;
129 argv += optind;
130
131 if (argc != 1)
132 usage();
133
134 if ((fp = fopen(argv[0], "r")) == NULL)
135 err(1, "%s", argv[0]);
136
137 if (pledge("stdio", NULL) == -1)
138 err(1, "pledge");
139
140 /* First copy bigram array to stdout. */
141 if (fgets(bigrams, sizeof(bigrams), fp) == NULL) {
142 if (ferror(fp)) {
143 err(1, "fgets on %s", argv[0]);
144 } else {
145 errx(1, "premature end of bigram file %s", argv[0]);
146 }
147 }
148
149 if (strlen(bigrams) != BGBUFSIZE)
150 errx(1, "bigram array too small to build db, index more files");
151
152 if (fputs(bigrams, stdout) == EOF)
153 err(1, "stdout");
154 (void)fclose(fp);
155
156 #ifdef LOOKUP
157 /* init lookup table */
158 for (i = 0; i < UCHAR_MAX + 1; i++)
159 for (j = 0; j < UCHAR_MAX + 1; j++)
160 big[i][j] = (bg_t)-1;
161
162 for (cp = bigrams, i = 0; *cp != '\0'; i += 2, cp += 2)
163 big[(u_char)*cp][(u_char)*(cp + 1)] = (bg_t)i;
164
165 #endif /* LOOKUP */
166
167 oldpath = buf1;
168 path = buf2;
169 oldcount = 0;
170
171 while (fgets(path, sizeof(buf2), stdin) != NULL) {
172
173 /* skip empty lines */
174 if (*path == '\n')
175 continue;
176
177 /* remove newline */
178 for (cp = path; *cp != '\0'; cp++) {
179 /* chop newline */
180 if (*cp == '\n')
181 *cp = '\0';
182 }
183
184 /* Skip longest common prefix. */
185 for (cp = path; *cp == *oldpath; cp++, oldpath++)
186 if (*cp == '\0')
187 break;
188
189 count = cp - path;
190 diffcount = count - oldcount + OFFSET;
191 oldcount = count;
192 if (diffcount < 0 || diffcount > 2 * OFFSET) {
193 if (putchar(SWITCH) == EOF ||
194 putw(diffcount, stdout) == EOF)
195 err(1, "stdout");
196 } else
197 if (putchar(diffcount) == EOF)
198 err(1, "stdout");
199
200 while (*cp != '\0') {
201 /* print *two* characters */
202
203 if ((code = BGINDEX(cp)) != (bg_t)-1) {
204 /*
205 * print *one* as bigram
206 * Found, so mark byte with
207 * parity bit.
208 */
209 if (putchar((code / 2) | PARITY) == EOF)
210 err(1, "stdout");
211 cp += 2;
212 } else {
213 for (i = 0; i < 2; i++) {
214 if (*cp == '\0')
215 break;
216
217 /* print umlauts in file names */
218 if (*cp < ASCII_MIN ||
219 *cp > ASCII_MAX) {
220 if (putchar(UMLAUT) == EOF ||
221 putchar(*cp++) == EOF)
222 err(1, "stdout");
223 } else {
224 /* normal character */
225 if (putchar(*cp++) == EOF)
226 err(1, "stdout");
227 }
228 }
229
230 }
231 }
232
233 if (path == buf1) { /* swap pointers */
234 path = buf2;
235 oldpath = buf1;
236 } else {
237 path = buf1;
238 oldpath = buf2;
239 }
240 }
241 /* Non-zero status if there were errors */
242 if (fflush(stdout) != 0 || ferror(stdout))
243 exit(1);
244 exit(0);
245 }
246
247 #ifndef LOOKUP
248 int
bgindex(char * bg)249 bgindex(char *bg) /* Return location of bg in bigrams or -1. */
250 {
251 char bg0, bg1, *p;
252
253 bg0 = bg[0];
254 bg1 = bg[1];
255 for (p = bigrams; *p != NULL; p++)
256 if (*p++ == bg0 && *p == bg1)
257 break;
258 return (*p == NULL ? -1 : (--p - bigrams));
259 }
260 #endif /* !LOOKUP */
261
262 void
usage(void)263 usage(void)
264 {
265 (void)fprintf(stderr,
266 "usage: locate.code common_bigrams < list > squozen_list\n");
267 exit(1);
268 }
269