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  /* 1999-02-22 Arkadiusz Miśkiewicz <misiek@pld.ORG.PL>
  38   * - added Native Language Support
  39   */
  40
  41 /*
  42  * look -- find lines in a sorted list.
  43  *
  44  * The man page said that TABs and SPACEs participate in -d comparisons.
  45  * In fact, they were ignored.  This implements historic practice, not
  46  * the manual page.
  47  */
  48
  49 #include <sys/mman.h>
  50 #include <sys/stat.h>
  51 #include <errno.h>
  52 #include <fcntl.h>
  53 #include <stdio.h>
  54 #include <stdlib.h>
  55 #include <stddef.h>
  56 #include <string.h>
  57 #include <ctype.h>
  58 #include <getopt.h>
  59
  60 #include "nls.h"
  61 #include "xalloc.h"
  62 #include "pathnames.h"
  63 #include "closestream.h"
  64
  65 #define EQUAL           0
  66 #define GREATER         1
  67 #define LESS            (-1)
  68
  69 static int dflag, fflag;
  70 /* uglified the source a bit with globals, so that we only need
  71    to allocate comparbuf once */
  72 static int stringlen;
  73 static char *string;
  74 static char *comparbuf;
  75
  76 static char *binary_search (char *, char *);
  77 static int compare (char *, char *);
  78 static char *linear_search (char *, char *);
  79 static int look (char *, char *);
  80 static void print_from (char *, char *);
  81 static void __attribute__((__noreturn__)) usage(void);
  82
  83 int
  84 main(int argc, char *argv[])
  85 {
  86         struct stat sb;
  87         int ch, fd, termchar;
  88         char *back, *file, *front, *p;
  89
  90         static const struct option longopts[] = {
  91                 {"alternative", no_argument, NULL, 'a'},
  92                 {"alphanum", no_argument, NULL, 'd'},
  93                 {"ignore-case", no_argument, NULL, 'f'},
  94                 {"terminate", required_argument, NULL, 't'},
  95                 {"version", no_argument, NULL, 'V'},
  96                 {"help", no_argument, NULL, 'h'},
  97                 {NULL, 0, NULL, 0}
  98         };
  99
 100         setlocale(LC_ALL, "");
 101         bindtextdomain(PACKAGE, LOCALEDIR);
 102         textdomain(PACKAGE);
 103         close_stdout_atexit();
 104
 105         setlocale(LC_ALL, "");
 106
 107         if ((file = getenv("WORDLIST")) && !access(file, R_OK))
 108                 /* use the WORDLIST */;
 109         else
 110                 file = _PATH_WORDS;
 111
 112         termchar = '\0';
 113         string = NULL;          /* just for gcc */
 114
 115         while ((ch = getopt_long(argc, argv, "adft:Vh", longopts, NULL)) != -1)
 116                 switch(ch) {
 117                 case 'a':
 118                         file = _PATH_WORDS_ALT;
 119                         break;
 120                 case 'd':
 121                         dflag = 1;
 122                         break;
 123                 case 'f':
 124                         fflag = 1;
 125                         break;
 126                 case 't':
 127                         termchar = *optarg;
 128                         break;
 129                 case 'V':
 130                         print_version(EXIT_SUCCESS);
 131                 case 'h':
 132                         usage();
 133                 default:
 134                         errtryhelp(EXIT_FAILURE);
 135                 }
 136         argc -= optind;
 137         argv += optind;
 138
 139         switch (argc) {
 140         case 2:                         /* Don't set -df for user. */
 141                 string = *argv++;
 142                 file = *argv;
 143                 break;
 144         case 1:                         /* But set -df by default. */
 145                 dflag = fflag = 1;
 146                 string = *argv;
 147                 break;
 148         default:
 149                 warnx(_("bad usage"));
 150                 errtryhelp(EXIT_FAILURE);
 151         }
 152
 153         if (termchar != '\0' && (p = strchr(string, termchar)) != NULL)
 154                 *++p = '\0';
 155
 156         if ((fd = open(file, O_RDONLY, 0)) < 0 || fstat(fd, &sb))
 157                 err(EXIT_FAILURE, "%s", file);
 158         front = mmap(NULL, (size_t) sb.st_size, PROT_READ,
 159                      MAP_SHARED, fd, (off_t) 0);
 160         if
 161 #ifdef MAP_FAILED
 162                 (front == MAP_FAILED)
 163 #else
 164                 ((void *)(front) <= (void *)0)
 165 #endif
 166                         err(EXIT_FAILURE, "%s", file);
 167         back = front + sb.st_size;
 168         return look(front, back);
 169 }
 170
 171 static int
 172 look(char *front, char *back)
 173 {
 174         int ch;
 175         char *readp, *writep;
 176
 177         /* Reformat string to avoid doing it multiple times later. */
 178         if (dflag) {
 179                 for (readp = writep = string; (ch = *readp++) != 0;) {
 180                         if (isalnum(ch) || isblank(ch))
 181                                 *(writep++) = ch;
 182                 }
 183                 *writep = '\0';
 184                 stringlen = writep - string;
 185         } else
 186                 stringlen = strlen(string);
 187
 188         comparbuf = xmalloc(stringlen+1);
 189
 190         front = binary_search(front, back);
 191         front = linear_search(front, back);
 192
 193         if (front)
 194                 print_from(front, back);
 195
 196         free(comparbuf);
 197
 198         return (front ? 0 : 1);
 199 }
 200
 201
 202 /*
 203  * Binary search for "string" in memory between "front" and "back".
 204  *
 205  * This routine is expected to return a pointer to the start of a line at
 206  * *or before* the first word matching "string".  Relaxing the constraint
 207  * this way simplifies the algorithm.
 208  *
 209  * Invariants:
 210  *      front points to the beginning of a line at or before the first
 211  *      matching string.
 212  *
 213  *      back points to the beginning of a line at or after the first
 214  *      matching line.
 215  *
 216  * Advancing the Invariants:
 217  *
 218  *      p = first newline after halfway point from front to back.
 219  *
 220  *      If the string at "p" is not greater than the string to match,
 221  *      p is the new front.  Otherwise it is the new back.
 222  *
 223  * Termination:
 224  *
 225  *      The definition of the routine allows it return at any point,
 226  *      since front is always at or before the line to print.
 227  *
 228  *      In fact, it returns when the chosen "p" equals "back".  This
 229  *      implies that there exists a string is least half as long as
 230  *      (back - front), which in turn implies that a linear search will
 231  *      be no more expensive than the cost of simply printing a string or two.
 232  *
 233  *      Trying to continue with binary search at this point would be
 234  *      more trouble than it's worth.
 235  */
 236 #define SKIP_PAST_NEWLINE(p, back) \
 237         while (p < back && *p++ != '\n')
 238
 239 static char *
 240 binary_search(char *front, char *back)
 241 {
 242         char *p;
 243
 244         p = front + (back - front) / 2;
 245         SKIP_PAST_NEWLINE(p, back);
 246
 247         /*
 248          * If the file changes underneath us, make sure we don't
 249          * infinitely loop.
 250          */
 251         while (p < back && back > front) {
 252                 if (compare(p, back) == GREATER)
 253                         front = p;
 254                 else
 255                         back = p;
 256                 p = front + (back - front) / 2;
 257                 SKIP_PAST_NEWLINE(p, back);
 258         }
 259         return (front);
 260 }
 261
 262 /*
 263  * Find the first line that starts with string, linearly searching from front
 264  * to back.
 265  *
 266  * Return NULL for no such line.
 267  *
 268  * This routine assumes:
 269  *
 270  *      o front points at the first character in a line.
 271  *      o front is before or at the first line to be printed.
 272  */
 273 static char *
 274 linear_search(char *front, char *back)
 275 {
 276         while (front < back) {
 277                 switch (compare(front, back)) {
 278                 case EQUAL:             /* Found it. */
 279                         return (front);
 280                 case LESS:              /* No such string. */
 281                         return (NULL);
 282                 case GREATER:           /* Keep going. */
 283                         break;
 284                 }
 285                 SKIP_PAST_NEWLINE(front, back);
 286         }
 287         return (NULL);
 288 }
 289
 290 /*
 291  * Print as many lines as match string, starting at front.
 292  */
 293 static void
 294 print_from(char *front, char *back)
 295 {
 296         int eol;
 297
 298         while (front < back && compare(front, back) == EQUAL) {
 299                 if (compare(front, back) == EQUAL) {
 300                         eol = 0;
 301                         while (front < back && !eol) {
 302                                 if (putchar(*front) == EOF)
 303                                         err(EXIT_FAILURE, "stdout");
 304                                 if (*front++ == '\n')
 305                                         eol = 1;
 306                         }
 307                 } else
 308                         SKIP_PAST_NEWLINE(front, back);
 309         }
 310 }
 311
 312 /*
 313  * Return LESS, GREATER, or EQUAL depending on how  string  compares with
 314  * string2 (s1 ??? s2).
 315  *
 316  *      o Matches up to len(s1) are EQUAL.
 317  *      o Matches up to len(s2) are GREATER.
 318  *
 319  * Compare understands about the -f and -d flags, and treats comparisons
 320  * appropriately.
 321  *
 322  * The string "string" is null terminated.  The string "s2" is '\n' terminated
 323  * (or "s2end" terminated).
 324  *
 325  * We use strcasecmp etc, since it knows how to ignore case also
 326  * in other locales.
 327  */
 328 static int
 329 compare(char *s2, char *s2end) {
 330         int i;
 331         char *p;
 332
 333         /* copy, ignoring things that should be ignored */
 334         p = comparbuf;
 335         i = stringlen;
 336         while(s2 < s2end && *s2 != '\n' && i) {
 337                 if (!dflag || isalnum(*s2) || isblank(*s2))
 338                 {
 339                         *p++ = *s2;
 340                         i--;
 341                 }
 342                 s2++;
 343         }
 344         *p = 0;
 345
 346         /* and compare */
 347         if (fflag)
 348                 i = strncasecmp(comparbuf, string, stringlen);
 349         else
 350                 i = strncmp(comparbuf, string, stringlen);
 351
 352         return ((i > 0) ? LESS : (i < 0) ? GREATER : EQUAL);
 353 }
 354
 355 static void __attribute__((__noreturn__)) usage(void)
 356 {
 357         FILE *out = stdout;
 358         fputs(USAGE_HEADER, out);
 359         fprintf(out, _(" %s [options] <string> [<file>...]\n"), program_invocation_short_name);
 360
 361         fputs(USAGE_SEPARATOR, out);
 362         fputs(_("Display lines beginning with a specified string.\n"), out);
 363
 364         fputs(USAGE_OPTIONS, out);
 365         fputs(_(" -a, --alternative        use the alternative dictionary\n"), out);
 366         fputs(_(" -d, --alphanum           compare only blanks and alphanumeric characters\n"), out);
 367         fputs(_(" -f, --ignore-case        ignore case differences when comparing\n"), out);
 368         fputs(_(" -t, --terminate <char>   define the string-termination character\n"), out);
 369
 370         fputs(USAGE_SEPARATOR, out);
 371         printf(USAGE_HELP_OPTIONS(26));
 372         printf(USAGE_MAN_TAIL("look(1)"));
 373
 374         exit(EXIT_SUCCESS);
 375 }