1 /* 2 * SPDX-License-Identifier: BSD-4-Clause 3 * 4 * Copyright (c) 1995 Wolfram Schneider <wosch@FreeBSD.org>. Berlin. 5 * Copyright (c) 1989, 1993 6 * The Regents of the University of California. All rights reserved. 7 * 8 * This code is derived from software contributed to Berkeley by 9 * James A. Woods. 10 * 11 * Redistribution and use in source and binary forms, with or without 12 * modification, are permitted provided that the following conditions 13 * are met: 14 * 1. Redistributions of source code must retain the above copyright 15 * notice, this list of conditions and the following disclaimer. 16 * 2. Redistributions in binary form must reproduce the above copyright 17 * notice, this list of conditions and the following disclaimer in the 18 * documentation and/or other materials provided with the distribution. 19 * 3. All advertising materials mentioning features or use of this software 20 * must display the following acknowledgement: 21 * This product includes software developed by the University of 22 * California, Berkeley and its contributors. 23 * 4. Neither the name of the University nor the names of its contributors 24 * may be used to endorse or promote products derived from this software 25 * without specific prior written permission. 26 * 27 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 28 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 29 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 30 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 31 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 32 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 33 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 35 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 36 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 37 * SUCH DAMAGE. 38 * 39 * $FreeBSD$ 40 */ 41 42 #if 0 43 #ifndef lint 44 static char copyright[] = 45 "@(#) Copyright (c) 1989, 1993\n\ 46 The Regents of the University of California. All rights reserved.\n"; 47 #endif /* not lint */ 48 49 #ifndef lint 50 static char sccsid[] = "@(#)locate.code.c 8.1 (Berkeley) 6/6/93"; 51 #endif /* not lint */ 52 #endif 53 54 /* 55 * PURPOSE: sorted list compressor (works with a modified 'find' 56 * to encode/decode a filename database) 57 * 58 * USAGE: bigram < list > bigrams 59 * process bigrams (see updatedb) > common_bigrams 60 * code common_bigrams < list > squozen_list 61 * 62 * METHOD: Uses 'front compression' (see ";login:", Volume 8, Number 1 63 * February/March 1983, p. 8). Output format is, per line, an 64 * offset differential count byte followed by a partially bigram- 65 * encoded ascii residue. A bigram is a two-character sequence, 66 * the first 128 most common of which are encoded in one byte. 67 * 68 * EXAMPLE: For simple front compression with no bigram encoding, 69 * if the input is... then the output is... 70 * 71 * /usr/src 0 /usr/src 72 * /usr/src/cmd/aardvark.c 8 /cmd/aardvark.c 73 * /usr/src/cmd/armadillo.c 14 armadillo.c 74 * /usr/tmp/zoo 5 tmp/zoo 75 * 76 * The codes are: 77 * 78 * 0-28 likeliest differential counts + offset to make nonnegative 79 * 30 switch code for out-of-range count to follow in next word 80 * 31 an 8 bit char followed 81 * 128-255 bigram codes (128 most common, as determined by 'updatedb') 82 * 32-127 single character (printable) ascii residue (ie, literal) 83 * 84 * The locate database store any character except newline ('\n') 85 * and NUL ('\0'). The 8-bit character support don't wast extra 86 * space until you have characters in file names less than 32 87 * or greather than 127. 88 * 89 * 90 * SEE ALSO: updatedb.sh, ../bigram/locate.bigram.c 91 * 92 * AUTHOR: James A. Woods, Informatics General Corp., 93 * NASA Ames Research Center, 10/82 94 * 8-bit file names characters: 95 * Wolfram Schneider, Berlin September 1996 96 */ 97 98 #include <sys/param.h> 99 #include <err.h> 100 #include <errno.h> 101 #include <stdlib.h> 102 #include <string.h> 103 #include <stdio.h> 104 #include <unistd.h> 105 #include "locate.h" 106 107 #define BGBUFSIZE (NBG * 2) /* size of bigram buffer */ 108 109 u_char buf1[MAXPATHLEN] = " "; 110 u_char buf2[MAXPATHLEN]; 111 u_char bigrams[BGBUFSIZE + 1] = { 0 }; 112 113 #define LOOKUP 1 /* use a lookup array instead a function, 3x faster */ 114 115 #ifdef LOOKUP 116 #define BGINDEX(x) (big[(u_char)*x][(u_char)*(x + 1)]) 117 typedef short bg_t; 118 bg_t big[UCHAR_MAX + 1][UCHAR_MAX + 1]; 119 #else 120 #define BGINDEX(x) bgindex(x) 121 typedef int bg_t; 122 int bgindex(char *); 123 #endif /* LOOKUP */ 124 125 126 void usage(void); 127 128 int 129 main(int argc, char *argv[]) 130 { 131 u_char *cp, *oldpath, *path; 132 int ch, code, count, diffcount, oldcount; 133 u_int i, j; 134 FILE *fp; 135 136 while ((ch = getopt(argc, argv, "")) != -1) 137 switch(ch) { 138 default: 139 usage(); 140 } 141 argc -= optind; 142 argv += optind; 143 144 if (argc != 1) 145 usage(); 146 147 if ((fp = fopen(argv[0], "r")) == NULL) 148 err(1, "%s", argv[0]); 149 150 /* First copy bigram array to stdout. */ 151 (void)fgets(bigrams, BGBUFSIZE + 1, fp); 152 153 if (fwrite(bigrams, 1, BGBUFSIZE, stdout) != BGBUFSIZE) 154 err(1, "stdout"); 155 (void)fclose(fp); 156 157 #ifdef LOOKUP 158 /* init lookup table */ 159 for (i = 0; i < UCHAR_MAX + 1; i++) 160 for (j = 0; j < UCHAR_MAX + 1; j++) 161 big[i][j] = (bg_t)-1; 162 163 for (cp = bigrams, i = 0; *cp != '\0'; i += 2, cp += 2) 164 big[(u_char)*cp][(u_char)*(cp + 1)] = (bg_t)i; 165 166 #endif /* LOOKUP */ 167 168 oldpath = buf1; 169 path = buf2; 170 oldcount = 0; 171 172 while (fgets(path, sizeof(buf2), stdin) != NULL) { 173 174 /* skip empty lines */ 175 if (*path == '\n') 176 continue; 177 178 /* remove newline */ 179 for (cp = path; *cp != '\0'; cp++) { 180 #ifndef LOCATE_CHAR30 181 /* old locate implementations core'd for char 30 */ 182 if (*cp == SWITCH) 183 *cp = '?'; 184 else 185 #endif /* !LOCATE_CHAR30 */ 186 187 /* chop newline */ 188 if (*cp == '\n') 189 *cp = '\0'; 190 } 191 192 /* Skip longest common prefix. */ 193 for (cp = path; *cp == *oldpath; cp++, oldpath++) 194 if (*cp == '\0') 195 break; 196 197 count = cp - path; 198 diffcount = count - oldcount + OFFSET; 199 oldcount = count; 200 if (diffcount < 0 || diffcount > 2 * OFFSET) { 201 if (putchar(SWITCH) == EOF || 202 putw(diffcount, stdout) == EOF) 203 err(1, "stdout"); 204 } else 205 if (putchar(diffcount) == EOF) 206 err(1, "stdout"); 207 208 while (*cp != '\0') { 209 /* print *two* characters */ 210 211 if ((code = BGINDEX(cp)) != (bg_t)-1) { 212 /* 213 * print *one* as bigram 214 * Found, so mark byte with 215 * parity bit. 216 */ 217 if (putchar((code / 2) | PARITY) == EOF) 218 err(1, "stdout"); 219 cp += 2; 220 } 221 222 else { 223 for (i = 0; i < 2; i++) { 224 if (*cp == '\0') 225 break; 226 227 /* print umlauts in file names */ 228 if (*cp < ASCII_MIN || 229 *cp > ASCII_MAX) { 230 if (putchar(UMLAUT) == EOF || 231 putchar(*cp++) == EOF) 232 err(1, "stdout"); 233 } 234 235 else { 236 /* normal character */ 237 if(putchar(*cp++) == EOF) 238 err(1, "stdout"); 239 } 240 } 241 242 } 243 } 244 245 if (path == buf1) { /* swap pointers */ 246 path = buf2; 247 oldpath = buf1; 248 } else { 249 path = buf1; 250 oldpath = buf2; 251 } 252 } 253 /* Non-zero status if there were errors */ 254 if (fflush(stdout) != 0 || ferror(stdout)) 255 exit(1); 256 exit(0); 257 } 258 259 #ifndef LOOKUP 260 int 261 bgindex(char *bg) /* Return location of bg in bigrams or -1. */ 262 { 263 char bg0, bg1, *p; 264 265 bg0 = bg[0]; 266 bg1 = bg[1]; 267 for (p = bigrams; *p != NULL; p++) 268 if (*p++ == bg0 && *p == bg1) 269 break; 270 return (*p == NULL ? -1 : (--p - bigrams)); 271 } 272 #endif /* !LOOKUP */ 273 274 void 275 usage(void) 276 { 277 (void)fprintf(stderr, 278 "usage: locate.code common_bigrams < list > squozen_list\n"); 279 exit(1); 280 } 281