2 * Copyright (c) 1991, 1993
3 * The Regents of the University of California. All rights reserved.
5 * This code is derived from software contributed to Berkeley by
6 * David Hitz of Auspex Systems, Inc.
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
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.
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
38 static const char copyright[] =
39 "@(#) Copyright (c) 1991, 1993\n\
40 The Regents of the University of California. All rights reserved.\n";
45 static char sccsid[] = "@(#)look.c 8.2 (Berkeley) 5/4/95";
48 #include <sys/cdefs.h>
49 __FBSDID("$FreeBSD$");
52 * look -- find lines in a sorted list.
54 * The man page said that TABs and SPACEs participate in -d comparisons.
55 * In fact, they were ignored. This implements historic practice, not
59 #include <sys/types.h>
75 #include "pathnames.h"
77 static char _path_words[] = _PATH_WORDS;
85 char *binary_search(wchar_t *, unsigned char *, unsigned char *);
86 int compare(wchar_t *, unsigned char *, unsigned char *);
87 char *linear_search(wchar_t *, unsigned char *, unsigned char *);
88 int look(wchar_t *, unsigned char *, unsigned char *);
89 wchar_t *prepkey(const char *, wchar_t);
90 void print_from(wchar_t *, unsigned char *, unsigned char *);
92 static void usage(void);
95 main(int argc, char *argv[])
100 unsigned char *back, *front;
101 unsigned const char *file;
104 (void) setlocale(LC_CTYPE, "");
108 while ((ch = getopt(argc, argv, "dft:")) != -1)
117 if (mbrtowc(&termchar, optarg, MB_LEN_MAX, NULL) !=
119 errx(2, "invalid termination character");
130 if (argc == 1) /* But set -df by default. */
132 key = prepkey(*argv++, termchar);
139 if ((fd = open(file, O_RDONLY, 0)) < 0 || fstat(fd, &sb))
141 if (sb.st_size > SIZE_T_MAX)
142 errx(2, "%s: %s", file, strerror(EFBIG));
143 if (sb.st_size == 0) {
147 if ((front = mmap(NULL, (size_t)sb.st_size, PROT_READ, MAP_SHARED, fd, (off_t)0)) == MAP_FAILED)
149 back = front + sb.st_size;
150 match *= (look(key, front, back));
152 } while (argc-- > 2 && (file = *argv++));
158 prepkey(const char *string, wchar_t termchar)
161 wchar_t *key, *writep;
166 * Reformat search string and convert to wide character representation
167 * to avoid doing it multiple times later.
169 if ((key = malloc(sizeof(wchar_t) * (strlen(string) + 1))) == NULL)
173 while ((clen = mbrtowc(&ch, readp, MB_LEN_MAX, NULL)) != 0) {
174 if (clen == (size_t)-1 || clen == (size_t)-2)
175 errc(2, EILSEQ, NULL);
178 if (!dflag || iswalnum(ch))
183 if (termchar != L'\0' && (writep = wcschr(key, termchar)) != NULL)
189 look(wchar_t *string, unsigned char *front, unsigned char *back)
192 front = binary_search(string, front, back);
193 front = linear_search(string, front, back);
196 print_from(string, front, back);
197 return (front ? 0 : 1);
202 * Binary search for "string" in memory between "front" and "back".
204 * This routine is expected to return a pointer to the start of a line at
205 * *or before* the first word matching "string". Relaxing the constraint
206 * this way simplifies the algorithm.
209 * front points to the beginning of a line at or before the first
212 * back points to the beginning of a line at or after the first
215 * Base of the Invariants.
219 * Advancing the Invariants:
221 * p = first newline after halfway point from front to back.
223 * If the string at "p" is not greater than the string to match,
224 * p is the new front. Otherwise it is the new back.
228 * The definition of the routine allows it return at any point,
229 * since front is always at or before the line to print.
231 * In fact, it returns when the chosen "p" equals "back". This
232 * implies that there exists a string is least half as long as
233 * (back - front), which in turn implies that a linear search will
234 * be no more expensive than the cost of simply printing a string or two.
236 * Trying to continue with binary search at this point would be
237 * more trouble than it's worth.
239 #define SKIP_PAST_NEWLINE(p, back) \
240 while (p < back && *p++ != '\n');
243 binary_search(wchar_t *string, unsigned char *front, unsigned char *back)
247 p = front + (back - front) / 2;
248 SKIP_PAST_NEWLINE(p, back);
251 * If the file changes underneath us, make sure we don't
254 while (p < back && back > front) {
255 if (compare(string, p, back) == GREATER)
259 p = front + (back - front) / 2;
260 SKIP_PAST_NEWLINE(p, back);
266 * Find the first line that starts with string, linearly searching from front
269 * Return NULL for no such line.
271 * This routine assumes:
273 * o front points at the first character in a line.
274 * o front is before or at the first line to be printed.
277 linear_search(wchar_t *string, unsigned char *front, unsigned char *back)
279 while (front < back) {
280 switch (compare(string, front, back)) {
281 case EQUAL: /* Found it. */
283 case LESS: /* No such string. */
285 case GREATER: /* Keep going. */
288 SKIP_PAST_NEWLINE(front, back);
294 * Print as many lines as match string, starting at front.
297 print_from(wchar_t *string, unsigned char *front, unsigned char *back)
299 for (; front < back && compare(string, front, back) == EQUAL; ++front) {
300 for (; front < back && *front != '\n'; ++front)
301 if (putchar(*front) == EOF)
303 if (putchar('\n') == EOF)
309 * Return LESS, GREATER, or EQUAL depending on how the string1 compares with
310 * string2 (s1 ??? s2).
312 * o Matches up to len(s1) are EQUAL.
313 * o Matches up to len(s2) are GREATER.
315 * Compare understands about the -f and -d flags, and treats comparisons
318 * The string "s1" is null terminated. The string s2 is '\n' terminated (or
319 * "back" terminated).
322 compare(wchar_t *s1, unsigned char *s2, unsigned char *back)
327 for (; *s1 && s2 < back && *s2 != '\n'; ++s1, s2 += len2) {
329 len2 = mbrtowc(&ch2, s2, back - s2, NULL);
330 if (len2 == (size_t)-1 || len2 == (size_t)-2) {
336 if (dflag && !iswalnum(ch2)) {
337 /* Ignore character in comparison. */
342 return (ch1 < ch2 ? LESS : GREATER);
344 return (*s1 ? GREATER : EQUAL);
350 (void)fprintf(stderr, "usage: look [-df] [-t char] string [file ...]\n");