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
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 $
43 * look -- find lines in a sorted list.
45 * The man page said that TABs and SPACEs participate in -d comparisons.
46 * In fact, they were ignored. This implements historic practice, not
50 #include <sys/types.h>
65 #include "pathnames.h"
68 * FOLD and DICT convert characters to a normal form for comparison,
69 * according to the user specified flags.
71 * DICT expects integers because it uses a non-character value to
72 * indicate a character which should not participate in comparisons.
77 #define NO_COMPARE (-2)
79 #define FOLD(c) (isupper(c) ? tolower(c) : (unsigned char) (c))
80 #define DICT(c) (isalnum(c) ? (c) & 0xFF /* int */ : NO_COMPARE)
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 *);
90 static void usage(void);
93 main(int argc, char **argv)
96 int ch, fd, termchar, match;
97 unsigned char *back, *front, *string, *p;
98 const unsigned char *file;
101 setlocale(LC_CTYPE, "");
105 while ((ch = getopt(argc, argv, "dft:")) != -1)
125 if (argc == 1) /* But set -df by default. */
131 if (termchar != '\0' && (p = strchr(string, termchar)) != NULL)
136 if ((fd = open(file, O_RDONLY, 0)) < 0 || fstat(fd, &sb))
138 len = (size_t)sb.st_size;
139 if ((off_t)len != sb.st_size) {
143 if ((front = mmap(NULL, len, PROT_READ, MAP_SHARED, fd, (off_t)0)) == MAP_FAILED)
145 back = front + sb.st_size;
146 match *= (look(string, front, back));
148 } while (argc-- > 2 && (file = *argv++));
154 look(unsigned char *string, unsigned char *front, unsigned char *back)
157 unsigned char *readp, *writep;
159 /* Reformat string string to avoid doing it multiple times later. */
160 for (readp = writep = string; (ch = *readp++) != '\0';) {
165 if (ch != NO_COMPARE)
170 front = binary_search(string, front, back);
171 front = linear_search(string, front, back);
174 print_from(string, front, back);
175 return (front ? 0 : 1);
180 * Binary search for "string" in memory between "front" and "back".
182 * This routine is expected to return a pointer to the start of a line at
183 * *or before* the first word matching "string". Relaxing the constraint
184 * this way simplifies the algorithm.
187 * front points to the beginning of a line at or before the first
190 * back points to the beginning of a line at or after the first
193 * Base of the Invariants.
197 * Advancing the Invariants:
199 * p = first newline after halfway point from front to back.
201 * If the string at "p" is not greater than the string to match,
202 * p is the new front. Otherwise it is the new back.
206 * The definition of the routine allows it return at any point,
207 * since front is always at or before the line to print.
209 * In fact, it returns when the chosen "p" equals "back". This
210 * implies that there exists a string is least half as long as
211 * (back - front), which in turn implies that a linear search will
212 * be no more expensive than the cost of simply printing a string or two.
214 * Trying to continue with binary search at this point would be
215 * more trouble than it's worth.
217 #define SKIP_PAST_NEWLINE(p, back) \
218 while (p < back && *p++ != '\n');
221 binary_search(unsigned char *string, unsigned char *front,
226 p = front + (back - front) / 2;
227 SKIP_PAST_NEWLINE(p, back);
230 * If the file changes underneath us, make sure we don't
233 while (p < back && back > front) {
234 if (compare(string, p, back) == GREATER)
238 p = front + (back - front) / 2;
239 SKIP_PAST_NEWLINE(p, back);
245 * Find the first line that starts with string, linearly searching from front
248 * Return NULL for no such line.
250 * This routine assumes:
252 * o front points at the first character in a line.
253 * o front is before or at the first line to be printed.
256 linear_search(unsigned char *string, unsigned char *front, unsigned char *back)
258 while (front < back) {
259 switch (compare(string, front, back)) {
260 case EQUAL: /* Found it. */
263 case LESS: /* No such string. */
266 case GREATER: /* Keep going. */
269 SKIP_PAST_NEWLINE(front, back);
275 * Print as many lines as match string, starting at front.
278 print_from(unsigned char *string, unsigned char *front,
281 for (; front < back && compare(string, front, back) == EQUAL; ++front) {
282 for (; front < back && *front != '\n'; ++front)
283 if (putchar(*front) == EOF)
285 if (putchar('\n') == EOF)
291 * Return LESS, GREATER, or EQUAL depending on how the string1 compares with
292 * string2 (s1 ??? s2).
294 * o Matches up to len(s1) are EQUAL.
295 * o Matches up to len(s2) are GREATER.
297 * Compare understands about the -f and -d flags, and treats comparisons
300 * The string "s1" is null terminated. The string s2 is '\n' terminated (or
301 * "back" terminated).
304 compare(unsigned char *s1, unsigned char *s2,
309 for (; *s1 && s2 < back && *s2 != '\n'; ++s1, ++s2) {
316 if (ch == NO_COMPARE) {
317 ++s2; /* Ignore character in comparison. */
321 return (*s1 < ch ? LESS : GREATER);
323 return (*s1 ? GREATER : EQUAL);
329 fprintf(stderr, "usage: look [-df] [-t char] string [file ...]\n");