look.c revision 99112
1/*- 2 * Copyright (c) 1991, 1993 3 * The Regents of the University of California. All rights reserved. 4 * 5 * This code is derived from software contributed to Berkeley by 6 * David Hitz of Auspex Systems, Inc. 7 * 8 * Redistribution and use in source and binary forms, with or without 9 * modification, are permitted provided that the following conditions 10 * are met: 11 * 1. Redistributions of source code must retain the above copyright 12 * notice, this list of conditions and the following disclaimer. 13 * 2. Redistributions in binary form must reproduce the above copyright 14 * notice, this list of conditions and the following disclaimer in the 15 * documentation and/or other materials provided with the distribution. 16 * 3. All advertising materials mentioning features or use of this software 17 * must display the following acknowledgement: 18 * This product includes software developed by the University of 19 * California, Berkeley and its contributors. 20 * 4. Neither the name of the University nor the names of its contributors 21 * may be used to endorse or promote products derived from this software 22 * without specific prior written permission. 23 * 24 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 25 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 26 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 27 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 28 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 29 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 30 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 31 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 32 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 33 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 34 * SUCH DAMAGE. 35 */ 36 37#ifndef lint 38static const char copyright[] = 39"@(#) Copyright (c) 1991, 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 45static char sccsid[] = "@(#)look.c 8.2 (Berkeley) 5/4/95"; 46#endif 47#endif /* not lint */ 48#include <sys/cdefs.h> 49__FBSDID("$FreeBSD: head/usr.bin/look/look.c 99112 2002-06-30 05:25:07Z obrien $"); 50 51/* 52 * look -- find lines in a sorted list. 53 * 54 * The man page said that TABs and SPACEs participate in -d comparisons. 55 * In fact, they were ignored. This implements historic practice, not 56 * the manual page. 57 */ 58 59#include <sys/types.h> 60#include <sys/mman.h> 61#include <sys/stat.h> 62 63#include <ctype.h> 64#include <err.h> 65#include <errno.h> 66#include <fcntl.h> 67#include <limits.h> 68#include <locale.h> 69#include <stdio.h> 70#include <stdlib.h> 71#include <string.h> 72#include <unistd.h> 73 74#include "pathnames.h" 75 76static char _path_words[] = _PATH_WORDS; 77 78/* 79 * FOLD and DICT convert characters to a normal form for comparison, 80 * according to the user specified flags. 81 * 82 * DICT expects integers because it uses a non-character value to 83 * indicate a character which should not participate in comparisons. 84 */ 85#define EQUAL 0 86#define GREATER 1 87#define LESS (-1) 88#define NO_COMPARE (-2) 89 90#define FOLD(c) (isupper(c) ? tolower(c) : (unsigned char) (c)) 91#define DICT(c) (isalnum(c) ? (c) & 0xFF /* int */ : NO_COMPARE) 92 93int dflag, fflag; 94 95char *binary_search(unsigned char *, unsigned char *, unsigned char *); 96int compare(unsigned char *, unsigned char *, unsigned char *); 97char *linear_search(unsigned char *, unsigned char *, unsigned char *); 98int look(unsigned char *, unsigned char *, unsigned char *); 99void print_from(unsigned char *, unsigned char *, unsigned char *); 100 101static void usage(void); 102 103int 104main(argc, argv) 105 int argc; 106 char *argv[]; 107{ 108 struct stat sb; 109 int ch, fd, termchar, match; 110 unsigned char *back, *front, *string, *p; 111 unsigned const char *file; 112 113 (void) setlocale(LC_CTYPE, ""); 114 115 file = _path_words; 116 termchar = '\0'; 117 while ((ch = getopt(argc, argv, "dft:")) != -1) 118 switch(ch) { 119 case 'd': 120 dflag = 1; 121 break; 122 case 'f': 123 fflag = 1; 124 break; 125 case 't': 126 termchar = *optarg; 127 break; 128 case '?': 129 default: 130 usage(); 131 } 132 argc -= optind; 133 argv += optind; 134 135 if (argc == 0) 136 usage(); 137 if (argc == 1) /* But set -df by default. */ 138 dflag = fflag = 1; 139 string = *argv++; 140 if (argc >= 2) 141 file = *argv++; 142 143 if (termchar != '\0' && (p = strchr(string, termchar)) != NULL) 144 *++p = '\0'; 145 match = 1; 146 147 do { 148 if ((fd = open(file, O_RDONLY, 0)) < 0 || fstat(fd, &sb)) 149 err(2, "%s", file); 150 if (sb.st_size > SIZE_T_MAX) 151 errx(2, "%s: %s", file, strerror(EFBIG)); 152 if ((front = mmap(NULL, (size_t)sb.st_size, PROT_READ, MAP_SHARED, fd, (off_t)0)) == MAP_FAILED) 153 err(2, "%s", file); 154 back = front + sb.st_size; 155 match *= (look(string, front, back)); 156 close(fd); 157 } while (argc-- > 2 && (file = *argv++)); 158 159 exit(match); 160} 161 162int 163look(string, front, back) 164 unsigned char *string, *front, *back; 165{ 166 register int ch; 167 register unsigned char *readp, *writep; 168 169 /* Reformat string string to avoid doing it multiple times later. */ 170 for (readp = writep = string; (ch = *readp++);) { 171 if (fflag) 172 ch = FOLD(ch); 173 if (dflag) 174 ch = DICT(ch); 175 if (ch != NO_COMPARE) 176 *(writep++) = ch; 177 } 178 *writep = '\0'; 179 180 front = binary_search(string, front, back); 181 front = linear_search(string, front, back); 182 183 if (front) 184 print_from(string, front, back); 185 return (front ? 0 : 1); 186} 187 188 189/* 190 * Binary search for "string" in memory between "front" and "back". 191 * 192 * This routine is expected to return a pointer to the start of a line at 193 * *or before* the first word matching "string". Relaxing the constraint 194 * this way simplifies the algorithm. 195 * 196 * Invariants: 197 * front points to the beginning of a line at or before the first 198 * matching string. 199 * 200 * back points to the beginning of a line at or after the first 201 * matching line. 202 * 203 * Base of the Invariants. 204 * front = NULL; 205 * back = EOF; 206 * 207 * Advancing the Invariants: 208 * 209 * p = first newline after halfway point from front to back. 210 * 211 * If the string at "p" is not greater than the string to match, 212 * p is the new front. Otherwise it is the new back. 213 * 214 * Termination: 215 * 216 * The definition of the routine allows it return at any point, 217 * since front is always at or before the line to print. 218 * 219 * In fact, it returns when the chosen "p" equals "back". This 220 * implies that there exists a string is least half as long as 221 * (back - front), which in turn implies that a linear search will 222 * be no more expensive than the cost of simply printing a string or two. 223 * 224 * Trying to continue with binary search at this point would be 225 * more trouble than it's worth. 226 */ 227#define SKIP_PAST_NEWLINE(p, back) \ 228 while (p < back && *p++ != '\n'); 229 230char * 231binary_search(string, front, back) 232 register unsigned char *string, *front, *back; 233{ 234 register unsigned char *p; 235 236 p = front + (back - front) / 2; 237 SKIP_PAST_NEWLINE(p, back); 238 239 /* 240 * If the file changes underneath us, make sure we don't 241 * infinitely loop. 242 */ 243 while (p < back && back > front) { 244 if (compare(string, p, back) == GREATER) 245 front = p; 246 else 247 back = p; 248 p = front + (back - front) / 2; 249 SKIP_PAST_NEWLINE(p, back); 250 } 251 return (front); 252} 253 254/* 255 * Find the first line that starts with string, linearly searching from front 256 * to back. 257 * 258 * Return NULL for no such line. 259 * 260 * This routine assumes: 261 * 262 * o front points at the first character in a line. 263 * o front is before or at the first line to be printed. 264 */ 265char * 266linear_search(string, front, back) 267 unsigned char *string, *front, *back; 268{ 269 while (front < back) { 270 switch (compare(string, front, back)) { 271 case EQUAL: /* Found it. */ 272 return (front); 273 break; 274 case LESS: /* No such string. */ 275 return (NULL); 276 break; 277 case GREATER: /* Keep going. */ 278 break; 279 } 280 SKIP_PAST_NEWLINE(front, back); 281 } 282 return (NULL); 283} 284 285/* 286 * Print as many lines as match string, starting at front. 287 */ 288void 289print_from(string, front, back) 290 register unsigned char *string, *front, *back; 291{ 292 for (; front < back && compare(string, front, back) == EQUAL; ++front) { 293 for (; front < back && *front != '\n'; ++front) 294 if (putchar(*front) == EOF) 295 err(2, "stdout"); 296 if (putchar('\n') == EOF) 297 err(2, "stdout"); 298 } 299} 300 301/* 302 * Return LESS, GREATER, or EQUAL depending on how the string1 compares with 303 * string2 (s1 ??? s2). 304 * 305 * o Matches up to len(s1) are EQUAL. 306 * o Matches up to len(s2) are GREATER. 307 * 308 * Compare understands about the -f and -d flags, and treats comparisons 309 * appropriately. 310 * 311 * The string "s1" is null terminated. The string s2 is '\n' terminated (or 312 * "back" terminated). 313 */ 314int 315compare(s1, s2, back) 316 register unsigned char *s1, *s2, *back; 317{ 318 register int ch; 319 320 for (; *s1 && s2 < back && *s2 != '\n'; ++s1, ++s2) { 321 ch = *s2; 322 if (fflag) 323 ch = FOLD(ch); 324 if (dflag) 325 ch = DICT(ch); 326 327 if (ch == NO_COMPARE) { 328 ++s2; /* Ignore character in comparison. */ 329 continue; 330 } 331 if (*s1 != ch) 332 return (*s1 < ch ? LESS : GREATER); 333 } 334 return (*s1 ? GREATER : EQUAL); 335} 336 337static void 338usage() 339{ 340 (void)fprintf(stderr, "usage: look [-df] [-t char] string [file ...]\n"); 341 exit(2); 342} 343