misc-utils/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
  37 #ifndef lint
  38 static 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 static char sccsid[] = "@(#)look.c      8.1 (Berkeley) 6/14/93";
  45 #endif /* not lint */
  46
  47 /*
  48  * look -- find lines in a sorted list.
  49  *
  50  * The man page said that TABs and SPACEs participate in -d comparisons.
  51  * In fact, they were ignored.  This implements historic practice, not
  52  * the manual page.
  53  */
  54
  55 #include <sys/types.h>
  56 #include <sys/mman.h>
  57 #include <sys/stat.h>
  58
  59 #include <limits.h>
  60 #include <errno.h>
  61 #include <fcntl.h>
  62 #include <stdio.h>
  63 #include <stdlib.h>
  64 #include <string.h>
  65 #include <ctype.h>
  66 #include <getopt.h>
  67 #include "pathnames.h"
  68
  69 /*
  70  * FOLD and DICT convert characters to a normal form for comparison,
  71  * according to the user specified flags.
  72  *
  73  * DICT expects integers because it uses a non-character value to
  74  * indicate a character which should not participate in comparisons.
  75  */
  76 #define EQUAL           0
  77 #define GREATER         1
  78 #define LESS            (-1)
  79 #define NO_COMPARE      (-2)
  80
  81 #define FOLD(c) (isascii(c) && isupper(c) ? tolower(c) : (c))
  82 #define DICT(c) (isascii(c) && isalnum(c) ? (c) : NO_COMPARE)
  83
  84 int dflag, fflag;
  85
  86 char    *binary_search __P((char *, char *, char *));
  87 int      compare __P((char *, char *, char *));
  88 void     err __P((const char *fmt, ...));
  89 char    *linear_search __P((char *, char *, char *));
  90 int      look __P((char *, char *, char *));
  91 void     print_from __P((char *, char *, char *));
  92
  93 static void usage __P((void));
  94
  95 void
  96 main(argc, argv)
  97         int argc;
  98         char *argv[];
  99 {
 100         struct stat sb;
 101         int ch, fd, termchar;
 102         char *back, *file, *front, *string, *p;
 103
 104         file = _PATH_WORDS;
 105         termchar = '\0';
 106         string = NULL;          /* just for gcc */
 107
 108         while ((ch = getopt(argc, argv, "adft:")) != EOF)
 109                 switch(ch) {
 110                 case 'a':
 111                         file = _PATH_WORDS_ALT;
 112                         break;
 113                 case 'd':
 114                         dflag = 1;
 115                         break;
 116                 case 'f':
 117                         fflag = 1;
 118                         break;
 119                 case 't':
 120                         termchar = *optarg;
 121                         break;
 122                 case '?':
 123                 default:
 124                         usage();
 125                 }
 126         argc -= optind;
 127         argv += optind;
 128
 129         switch (argc) {
 130         case 2:                         /* Don't set -df for user. */
 131                 string = *argv++;
 132                 file = *argv;
 133                 break;
 134         case 1:                         /* But set -df by default. */
 135                 dflag = fflag = 1;
 136                 string = *argv;
 137                 break;
 138         default:
 139                 usage();
 140         }
 141
 142         if (termchar != '\0' && (p = strchr(string, termchar)) != NULL)
 143                 *++p = '\0';
 144
 145         if ((fd = open(file, O_RDONLY, 0)) < 0 || fstat(fd, &sb))
 146                 err("%s: %s", file, strerror(errno));
 147         if ((void *)(front = mmap(NULL,
 148                                   (size_t)sb.st_size,
 149                                   PROT_READ,
 150                                   MAP_FILE|MAP_SHARED,
 151                                   fd,
 152                                   (off_t)0)) <= (void *)0)
 153                 err("%s: %s", file, strerror(errno));
 154         back = front + sb.st_size;
 155         exit(look(string, front, back));
 156 }
 157
 158 int
 159 look(string, front, back)
 160         char *string, *front, *back;
 161 {
 162         register int ch;
 163         register char *readp, *writep;
 164
 165         /* Reformat string string to avoid doing it multiple times later. */
 166         for (readp = writep = string; (ch = *readp++) != 0;) {
 167                 if (fflag)
 168                         ch = FOLD(ch);
 169                 if (dflag)
 170                         ch = DICT(ch);
 171                 if (ch != NO_COMPARE)
 172                         *(writep++) = ch;
 173         }
 174         *writep = '\0';
 175
 176         front = binary_search(string, front, back);
 177         front = linear_search(string, front, back);
 178
 179         if (front)
 180                 print_from(string, front, back);
 181         return (front ? 0 : 1);
 182 }
 183
 184
 185 /*
 186  * Binary search for "string" in memory between "front" and "back".
 187  *
 188  * This routine is expected to return a pointer to the start of a line at
 189  * *or before* the first word matching "string".  Relaxing the constraint
 190  * this way simplifies the algorithm.
 191  *
 192  * Invariants:
 193  *      front points to the beginning of a line at or before the first
 194  *      matching string.
 195  *
 196  *      back points to the beginning of a line at or after the first
 197  *      matching line.
 198  *
 199  * Base of the Invariants.
 200  *      front = NULL;
 201  *      back = EOF;
 202  *
 203  * Advancing the Invariants:
 204  *
 205  *      p = first newline after halfway point from front to back.
 206  *
 207  *      If the string at "p" is not greater than the string to match,
 208  *      p is the new front.  Otherwise it is the new back.
 209  *
 210  * Termination:
 211  *
 212  *      The definition of the routine allows it return at any point,
 213  *      since front is always at or before the line to print.
 214  *
 215  *      In fact, it returns when the chosen "p" equals "back".  This
 216  *      implies that there exists a string is least half as long as
 217  *      (back - front), which in turn implies that a linear search will
 218  *      be no more expensive than the cost of simply printing a string or two.
 219  *
 220  *      Trying to continue with binary search at this point would be
 221  *      more trouble than it's worth.
 222  */
 223 #define SKIP_PAST_NEWLINE(p, back) \
 224         while (p < back && *p++ != '\n');
 225
 226 char *
 227 binary_search(string, front, back)
 228         register char *string, *front, *back;
 229 {
 230         register char *p;
 231
 232         p = front + (back - front) / 2;
 233         SKIP_PAST_NEWLINE(p, back);
 234
 235         /*
 236          * If the file changes underneath us, make sure we don't
 237          * infinitely loop.
 238          */
 239         while (p < back && back > front) {
 240                 if (compare(string, p, back) == GREATER)
 241                         front = p;
 242                 else
 243                         back = p;
 244                 p = front + (back - front) / 2;
 245                 SKIP_PAST_NEWLINE(p, back);
 246         }
 247         return (front);
 248 }
 249
 250 /*
 251  * Find the first line that starts with string, linearly searching from front
 252  * to back.
 253  *
 254  * Return NULL for no such line.
 255  *
 256  * This routine assumes:
 257  *
 258  *      o front points at the first character in a line.
 259  *      o front is before or at the first line to be printed.
 260  */
 261 char *
 262 linear_search(string, front, back)
 263         char *string, *front, *back;
 264 {
 265         while (front < back) {
 266                 switch (compare(string, front, back)) {
 267                 case EQUAL:             /* Found it. */
 268                         return (front);
 269                         break;
 270                 case LESS:              /* No such string. */
 271                         return (NULL);
 272                         break;
 273                 case GREATER:           /* Keep going. */
 274                         break;
 275                 }
 276                 SKIP_PAST_NEWLINE(front, back);
 277         }
 278         return (NULL);
 279 }
 280
 281 /*
 282  * Print as many lines as match string, starting at front.
 283  */
 284 void
 285 print_from(string, front, back)
 286         register char *string, *front, *back;
 287 {
 288         for (; front < back && compare(string, front, back) == EQUAL; ++front) {
 289                 for (; front < back && *front != '\n'; ++front)
 290                         if (putchar(*front) == EOF)
 291                                 err("stdout: %s", strerror(errno));
 292                 if (putchar('\n') == EOF)
 293                         err("stdout: %s", strerror(errno));
 294         }
 295 }
 296
 297 /*
 298  * Return LESS, GREATER, or EQUAL depending on how the string1 compares with
 299  * string2 (s1 ??? s2).
 300  *
 301  *      o Matches up to len(s1) are EQUAL.
 302  *      o Matches up to len(s2) are GREATER.
 303  *
 304  * Compare understands about the -f and -d flags, and treats comparisons
 305  * appropriately.
 306  *
 307  * The string "s1" is null terminated.  The string s2 is '\n' terminated (or
 308  * "back" terminated).
 309  */
 310 int
 311 compare(s1, s2, back)
 312         register char *s1, *s2, *back;
 313 {
 314         register int ch;
 315
 316         for (; *s1 && s2 < back && *s2 != '\n';) {
 317                 ch = *s2;
 318                 if (fflag)
 319                         ch = FOLD(ch);
 320                 if (dflag)
 321                         ch = DICT(ch);
 322
 323                 if (ch == NO_COMPARE) {
 324                         ++s2;           /* Ignore character in comparison. */
 325                         continue;
 326                 }
 327                 if (*s1 != ch)
 328                         return (*s1 < ch ? LESS : GREATER);
 329                 ++s1;
 330                 ++s2;
 331         }
 332         return (*s1 ? GREATER : EQUAL);
 333 }
 334
 335 static void
 336 usage()
 337 {
 338         (void)fprintf(stderr, "usage: look [-dfa] [-t char] string [file]\n");
 339         exit(2);
 340 }
 341
 342 #if __STDC__
 343 #include <stdarg.h>
 344 #else
 345 #include <varargs.h>
 346 #endif
 347
 348 void
 349 #if __STDC__
 350 err(const char *fmt, ...)
 351 #else
 352 err(fmt, va_alist)
 353         char *fmt;
 354         va_dcl
 355 #endif
 356 {
 357         va_list ap;
 358 #if __STDC__
 359         va_start(ap, fmt);
 360 #else
 361         va_start(ap);
 362 #endif
 363         (void)fprintf(stderr, "look: ");
 364         (void)vfprintf(stderr, fmt, ap);
 365         va_end(ap);
 366         (void)fprintf(stderr, "\n");
 367         exit(2);
 368         /* NOTREACHED */
 369 }