usr.bin/look/look.c

   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  * @(#) Copyright (c) 1991, 1993 The Regents of the University of California.  All rights reserved.
  37  * @(#)look.c   8.2 (Berkeley) 5/4/95
  38  * $FreeBSD: src/usr.bin/look/look.c,v 1.11 1999/08/28 01:03:14 peter Exp $
  39  * $DragonFly: src/usr.bin/look/look.c,v 1.4 2005/01/07 02:43:41 cpressey Exp $
  40  */
  41
  42 /*
  43  * look -- find lines in a sorted list.
  44  *
  45  * The man page said that TABs and SPACEs participate in -d comparisons.
  46  * In fact, they were ignored.  This implements historic practice, not
  47  * the manual page.
  48  */
  49
  50 #include <sys/types.h>
  51 #include <sys/mman.h>
  52 #include <sys/stat.h>
  53
  54 #include <ctype.h>
  55 #include <err.h>
  56 #include <errno.h>
  57 #include <fcntl.h>
  58 #include <limits.h>
  59 #include <locale.h>
  60 #include <stdio.h>
  61 #include <stdlib.h>
  62 #include <string.h>
  63 #include <unistd.h>
  64
  65 #include "pathnames.h"
  66
  67 /*
  68  * FOLD and DICT convert characters to a normal form for comparison,
  69  * according to the user specified flags.
  70  *
  71  * DICT expects integers because it uses a non-character value to
  72  * indicate a character which should not participate in comparisons.
  73  */
  74 #define EQUAL           0
  75 #define GREATER         1
  76 #define LESS            (-1)
  77 #define NO_COMPARE      (-2)
  78
  79 #define FOLD(c) (isupper(c) ? tolower(c) : (unsigned char) (c))
  80 #define DICT(c) (isalnum(c) ? (c) & 0xFF /* int */ : NO_COMPARE)
  81
  82 int dflag, fflag;
  83
  84 char    *binary_search(unsigned char *, unsigned char *, unsigned char *);
  85 int      compare(unsigned char *, unsigned char *, unsigned char *);
  86 char    *linear_search(unsigned char *, unsigned char *, unsigned char *);
  87 int      look(unsigned char *, unsigned char *, unsigned char *);
  88 void     print_from(unsigned char *, unsigned char *, unsigned char *);
  89
  90 static void usage(void);
  91
  92 int
  93 main(int argc, char **argv)
  94 {
  95         struct stat sb;
  96         int ch, fd, termchar, match;
  97         unsigned char *back, *front, *string, *p;
  98         const unsigned char *file;
  99
 100         setlocale(LC_CTYPE, "");
 101
 102         file = _PATH_WORDS;
 103         termchar = '\0';
 104         while ((ch = getopt(argc, argv, "dft:")) != -1)
 105                 switch(ch) {
 106                 case 'd':
 107                         dflag = 1;
 108                         break;
 109                 case 'f':
 110                         fflag = 1;
 111                         break;
 112                 case 't':
 113                         termchar = *optarg;
 114                         break;
 115                 case '?':
 116                 default:
 117                         usage();
 118                 }
 119         argc -= optind;
 120         argv += optind;
 121
 122         if (argc == 0)
 123                 usage();
 124         if (argc == 1)                  /* But set -df by default. */
 125                 dflag = fflag = 1;
 126         string = *argv++;
 127         if (argc >= 2)
 128                 file = *argv++;
 129
 130         if (termchar != '\0' && (p = strchr(string, termchar)) != NULL)
 131                 *++p = '\0';
 132         match = 1;
 133
 134         do {
 135                 if ((fd = open(file, O_RDONLY, 0)) < 0 || fstat(fd, &sb))
 136                         err(2, "%s", file);
 137                 if (sb.st_size > SIZE_T_MAX)
 138                         errx(2, "%s: %s", file, strerror(EFBIG));
 139                 if ((front = mmap(NULL, (size_t)sb.st_size, PROT_READ, MAP_SHARED, fd, (off_t)0)) == MAP_FAILED)
 140                         err(2, "%s", file);
 141                 back = front + sb.st_size;
 142                 match *= (look(string, front, back));
 143                 close(fd);
 144         } while (argc-- > 2 && (file = *argv++));
 145
 146         exit(match);
 147 }
 148
 149 int
 150 look(unsigned char *string, unsigned char *front, unsigned char *back)
 151 {
 152         int ch;
 153         unsigned char *readp, *writep;
 154
 155         /* Reformat string string to avoid doing it multiple times later. */
 156         for (readp = writep = string; (ch = *readp++) != '\0';) {
 157                 if (fflag)
 158                         ch = FOLD(ch);
 159                 if (dflag)
 160                         ch = DICT(ch);
 161                 if (ch != NO_COMPARE)
 162                         *(writep++) = ch;
 163         }
 164         *writep = '\0';
 165
 166         front = binary_search(string, front, back);
 167         front = linear_search(string, front, back);
 168
 169         if (front)
 170                 print_from(string, front, back);
 171         return (front ? 0 : 1);
 172 }
 173
 174
 175 /*
 176  * Binary search for "string" in memory between "front" and "back".
 177  *
 178  * This routine is expected to return a pointer to the start of a line at
 179  * *or before* the first word matching "string".  Relaxing the constraint
 180  * this way simplifies the algorithm.
 181  *
 182  * Invariants:
 183  *      front points to the beginning of a line at or before the first
 184  *      matching string.
 185  *
 186  *      back points to the beginning of a line at or after the first
 187  *      matching line.
 188  *
 189  * Base of the Invariants.
 190  *      front = NULL;
 191  *      back = EOF;
 192  *
 193  * Advancing the Invariants:
 194  *
 195  *      p = first newline after halfway point from front to back.
 196  *
 197  *      If the string at "p" is not greater than the string to match,
 198  *      p is the new front.  Otherwise it is the new back.
 199  *
 200  * Termination:
 201  *
 202  *      The definition of the routine allows it return at any point,
 203  *      since front is always at or before the line to print.
 204  *
 205  *      In fact, it returns when the chosen "p" equals "back".  This
 206  *      implies that there exists a string is least half as long as
 207  *      (back - front), which in turn implies that a linear search will
 208  *      be no more expensive than the cost of simply printing a string or two.
 209  *
 210  *      Trying to continue with binary search at this point would be
 211  *      more trouble than it's worth.
 212  */
 213 #define SKIP_PAST_NEWLINE(p, back) \
 214         while (p < back && *p++ != '\n');
 215
 216 char *
 217 binary_search(unsigned char *string, unsigned char *front,
 218               unsigned char *back)
 219 {
 220         unsigned char *p;
 221
 222         p = front + (back - front) / 2;
 223         SKIP_PAST_NEWLINE(p, back);
 224
 225         /*
 226          * If the file changes underneath us, make sure we don't
 227          * infinitely loop.
 228          */
 229         while (p < back && back > front) {
 230                 if (compare(string, p, back) == GREATER)
 231                         front = p;
 232                 else
 233                         back = p;
 234                 p = front + (back - front) / 2;
 235                 SKIP_PAST_NEWLINE(p, back);
 236         }
 237         return (front);
 238 }
 239
 240 /*
 241  * Find the first line that starts with string, linearly searching from front
 242  * to back.
 243  *
 244  * Return NULL for no such line.
 245  *
 246  * This routine assumes:
 247  *
 248  *      o front points at the first character in a line.
 249  *      o front is before or at the first line to be printed.
 250  */
 251 char *
 252 linear_search(unsigned char *string, unsigned char *front, unsigned char *back)
 253 {
 254         while (front < back) {
 255                 switch (compare(string, front, back)) {
 256                 case EQUAL:             /* Found it. */
 257                         return (front);
 258                         break;
 259                 case LESS:              /* No such string. */
 260                         return (NULL);
 261                         break;
 262                 case GREATER:           /* Keep going. */
 263                         break;
 264                 }
 265                 SKIP_PAST_NEWLINE(front, back);
 266         }
 267         return (NULL);
 268 }
 269
 270 /*
 271  * Print as many lines as match string, starting at front.
 272  */
 273 void
 274 print_from(unsigned char *string, unsigned char *front,
 275            unsigned char *back)
 276 {
 277         for (; front < back && compare(string, front, back) == EQUAL; ++front) {
 278                 for (; front < back && *front != '\n'; ++front)
 279                         if (putchar(*front) == EOF)
 280                                 err(2, "stdout");
 281                 if (putchar('\n') == EOF)
 282                         err(2, "stdout");
 283         }
 284 }
 285
 286 /*
 287  * Return LESS, GREATER, or EQUAL depending on how the string1 compares with
 288  * string2 (s1 ??? s2).
 289  *
 290  *      o Matches up to len(s1) are EQUAL.
 291  *      o Matches up to len(s2) are GREATER.
 292  *
 293  * Compare understands about the -f and -d flags, and treats comparisons
 294  * appropriately.
 295  *
 296  * The string "s1" is null terminated.  The string s2 is '\n' terminated (or
 297  * "back" terminated).
 298  */
 299 int
 300 compare(unsigned char *s1, unsigned char *s2,
 301         unsigned char *back)
 302 {
 303         int ch;
 304
 305         for (; *s1 && s2 < back && *s2 != '\n'; ++s1, ++s2) {
 306                 ch = *s2;
 307                 if (fflag)
 308                         ch = FOLD(ch);
 309                 if (dflag)
 310                         ch = DICT(ch);
 311
 312                 if (ch == NO_COMPARE) {
 313                         ++s2;           /* Ignore character in comparison. */
 314                         continue;
 315                 }
 316                 if (*s1 != ch)
 317                         return (*s1 < ch ? LESS : GREATER);
 318         }
 319         return (*s1 ? GREATER : EQUAL);
 320 }
 321
 322 static void
 323 usage(void)
 324 {
 325         fprintf(stderr, "usage: look [-df] [-t char] string [file ...]\n");
 326         exit(2);
 327 }