1 /* 2 * $OpenBSD: locate.code.c,v 1.15 2007/03/18 20:13:49 otto 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 * $Id: locate.code.c,v 1.15 2007/03/18 20:13:49 otto Exp $ 35 */ 36 37 #ifndef lint 38 static char copyright[] = 39 "@(#) Copyright (c) 1989, 1993\n\ 40 The Regents of the University of California. All rights reserved.\n"; 41 #endif /* not lint */ 42 43 #ifndef lint 44 #if 0 45 static char sccsid[] = "@(#)locate.code.c 8.1 (Berkeley) 6/6/93"; 46 #else 47 static char rcsid[] = "$OpenBSD: locate.code.c,v 1.15 2007/03/18 20:13:49 otto Exp $"; 48 #endif 49 #endif /* not lint */ 50 51 /* 52 * PURPOSE: sorted list compressor (works with a modified 'find' 53 * to encode/decode a filename database) 54 * 55 * USAGE: bigram < list > bigrams 56 * process bigrams (see updatedb) > common_bigrams 57 * code common_bigrams < list > squozen_list 58 * 59 * METHOD: Uses 'front compression' (see ";login:", Volume 8, Number 1 60 * February/March 1983, p. 8). Output format is, per line, an 61 * offset differential count byte followed by a partially bigram- 62 * encoded ascii residue. A bigram is a two-character sequence, 63 * the first 128 most common of which are encoded in one byte. 64 * 65 * EXAMPLE: For simple front compression with no bigram encoding, 66 * if the input is... then the output is... 67 * 68 * /usr/src 0 /usr/src 69 * /usr/src/cmd/aardvark.c 8 /cmd/aardvark.c 70 * /usr/src/cmd/armadillo.c 14 armadillo.c 71 * /usr/tmp/zoo 5 tmp/zoo 72 * 73 * The codes are: 74 * 75 * 0-28 likeliest differential counts + offset to make nonnegative 76 * 30 switch code for out-of-range count to follow in next word 77 * 31 an 8 bit char followed 78 * 128-255 bigram codes (128 most common, as determined by 'updatedb') 79 * 32-127 single character (printable) ascii residue (ie, literal) 80 * 81 * The locate database store any character except newline ('\n') 82 * and NUL ('\0'). The 8-bit character support don't wast extra 83 * space until you have characters in file names less than 32 84 * or greater than 127. 85 * 86 * 87 * SEE ALSO: updatedb.sh, ../bigram/locate.bigram.c 88 * 89 * AUTHOR: James A. Woods, Informatics General Corp., 90 * NASA Ames Research Center, 10/82 91 * 8-bit file names characters: 92 * Wolfram Schneider, Berlin September 1996 93 */ 94 95 #include <sys/param.h> 96 #include <err.h> 97 #include <errno.h> 98 #include <stdlib.h> 99 #include <string.h> 100 #include <stdio.h> 101 #include "locate.h" 102 103 #define BGBUFSIZE (NBG * 2) /* size of bigram buffer */ 104 105 u_char buf1[MAXPATHLEN] = " "; 106 u_char buf2[MAXPATHLEN]; 107 u_char bigrams[BGBUFSIZE + 1] = { 0 }; 108 109 #define LOOKUP 1 /* use a lookup array instead a function, 3x faster */ 110 111 #ifdef LOOKUP 112 #define BGINDEX(x) (big[(u_char)*x][(u_char)*(x + 1)]) 113 typedef short bg_t; 114 bg_t big[UCHAR_MAX + 1][UCHAR_MAX + 1]; 115 #else 116 #define BGINDEX(x) bgindex(x) 117 typedef int bg_t; 118 int bgindex(char *); 119 #endif /* LOOKUP */ 120 121 122 void usage(void); 123 extern int optind; 124 extern int optopt; 125 126 int 127 main(int argc, char *argv[]) 128 { 129 u_char *cp, *oldpath, *path; 130 int ch, code, count, diffcount, oldcount; 131 FILE *fp; 132 int i, j; 133 134 while ((ch = getopt(argc, argv, "")) != -1) 135 switch (ch) { 136 default: 137 usage(); 138 } 139 argc -= optind; 140 argv += optind; 141 142 if (argc != 1) 143 usage(); 144 145 if ((fp = fopen(argv[0], "r")) == NULL) 146 err(1, "%s", argv[0]); 147 148 /* First copy bigram array to stdout. */ 149 if (fgets(bigrams, sizeof(bigrams), fp) == NULL) 150 err(1, "fgets"); 151 152 if (strlen(bigrams) != BGBUFSIZE) 153 errx(1, "bigram array too small to build db, index more files"); 154 155 if (fputs(bigrams, stdout) == EOF) 156 err(1, "stdout"); 157 (void)fclose(fp); 158 159 #ifdef LOOKUP 160 /* init lookup table */ 161 for (i = 0; i < UCHAR_MAX + 1; i++) 162 for (j = 0; j < UCHAR_MAX + 1; j++) 163 big[i][j] = (bg_t)-1; 164 165 for (cp = bigrams, i = 0; *cp != '\0'; i += 2, cp += 2) 166 big[(u_char)*cp][(u_char)*(cp + 1)] = (bg_t)i; 167 168 #endif /* LOOKUP */ 169 170 oldpath = buf1; 171 path = buf2; 172 oldcount = 0; 173 174 while (fgets(path, sizeof(buf2), stdin) != NULL) { 175 176 /* skip empty lines */ 177 if (*path == '\n') 178 continue; 179 180 /* remove newline */ 181 for (cp = path; *cp != '\0'; cp++) { 182 /* chop newline */ 183 if (*cp == '\n') 184 *cp = '\0'; 185 } 186 187 /* Skip longest common prefix. */ 188 for (cp = path; *cp == *oldpath; cp++, oldpath++) 189 if (*cp == '\0') 190 break; 191 192 count = cp - path; 193 diffcount = count - oldcount + OFFSET; 194 oldcount = count; 195 if (diffcount < 0 || diffcount > 2 * OFFSET) { 196 if (putchar(SWITCH) == EOF || 197 putw(diffcount, stdout) == EOF) 198 err(1, "stdout"); 199 } else 200 if (putchar(diffcount) == EOF) 201 err(1, "stdout"); 202 203 while (*cp != '\0') { 204 /* print *two* characters */ 205 206 if ((code = BGINDEX(cp)) != (bg_t)-1) { 207 /* 208 * print *one* as bigram 209 * Found, so mark byte with 210 * parity bit. 211 */ 212 if (putchar((code / 2) | PARITY) == EOF) 213 err(1, "stdout"); 214 cp += 2; 215 } else { 216 for (i = 0; i < 2; i++) { 217 if (*cp == '\0') 218 break; 219 220 /* print umlauts in file names */ 221 if (*cp < ASCII_MIN || 222 *cp > ASCII_MAX) { 223 if (putchar(UMLAUT) == EOF || 224 putchar(*cp++) == EOF) 225 err(1, "stdout"); 226 } else { 227 /* normal character */ 228 if (putchar(*cp++) == EOF) 229 err(1, "stdout"); 230 } 231 } 232 233 } 234 } 235 236 if (path == buf1) { /* swap pointers */ 237 path = buf2; 238 oldpath = buf1; 239 } else { 240 path = buf1; 241 oldpath = buf2; 242 } 243 } 244 /* Non-zero status if there were errors */ 245 if (fflush(stdout) != 0 || ferror(stdout)) 246 exit(1); 247 exit(0); 248 } 249 250 #ifndef LOOKUP 251 int 252 bgindex(char *bg) /* Return location of bg in bigrams or -1. */ 253 { 254 char bg0, bg1, *p; 255 256 bg0 = bg[0]; 257 bg1 = bg[1]; 258 for (p = bigrams; *p != NULL; p++) 259 if (*p++ == bg0 && *p == bg1) 260 break; 261 return (*p == NULL ? -1 : (--p - bigrams)); 262 } 263 #endif /* !LOOKUP */ 264 265 void 266 usage(void) 267 { 268 (void)fprintf(stderr, 269 "usage: locate.code common_bigrams < list > squozen_list\n"); 270 exit(1); 271 } 272