1 |
|
|
/* $OpenBSD: look.c,v 1.21 2017/01/21 10:03:27 krw Exp $ */ |
2 |
|
|
/* $NetBSD: look.c,v 1.7 1995/08/31 22:41:02 jtc Exp $ */ |
3 |
|
|
|
4 |
|
|
/*- |
5 |
|
|
* Copyright (c) 1991, 1993 |
6 |
|
|
* The Regents of the University of California. All rights reserved. |
7 |
|
|
* |
8 |
|
|
* This code is derived from software contributed to Berkeley by |
9 |
|
|
* David Hitz of Auspex Systems, Inc. |
10 |
|
|
* |
11 |
|
|
* Redistribution and use in source and binary forms, with or without |
12 |
|
|
* modification, are permitted provided that the following conditions |
13 |
|
|
* are met: |
14 |
|
|
* 1. Redistributions of source code must retain the above copyright |
15 |
|
|
* notice, this list of conditions and the following disclaimer. |
16 |
|
|
* 2. Redistributions in binary form must reproduce the above copyright |
17 |
|
|
* notice, this list of conditions and the following disclaimer in the |
18 |
|
|
* documentation and/or other materials provided with the distribution. |
19 |
|
|
* 3. Neither the name of the University nor the names of its contributors |
20 |
|
|
* may be used to endorse or promote products derived from this software |
21 |
|
|
* without specific prior written permission. |
22 |
|
|
* |
23 |
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND |
24 |
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
25 |
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE |
26 |
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE |
27 |
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL |
28 |
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS |
29 |
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) |
30 |
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT |
31 |
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY |
32 |
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF |
33 |
|
|
* SUCH DAMAGE. |
34 |
|
|
*/ |
35 |
|
|
|
36 |
|
|
/* |
37 |
|
|
* look -- find lines in a sorted list. |
38 |
|
|
* |
39 |
|
|
* The man page said that TABs and SPACEs participate in -d comparisons. |
40 |
|
|
* In fact, they were ignored. This implements historic practice, not |
41 |
|
|
* the manual page. |
42 |
|
|
*/ |
43 |
|
|
|
44 |
|
|
#include <sys/types.h> |
45 |
|
|
#include <sys/mman.h> |
46 |
|
|
#include <sys/stat.h> |
47 |
|
|
|
48 |
|
|
#include <ctype.h> |
49 |
|
|
#include <errno.h> |
50 |
|
|
#include <fcntl.h> |
51 |
|
|
#include <stdint.h> |
52 |
|
|
#include <stdio.h> |
53 |
|
|
#include <stdlib.h> |
54 |
|
|
#include <string.h> |
55 |
|
|
#include <unistd.h> |
56 |
|
|
#include <err.h> |
57 |
|
|
|
58 |
|
|
#include "pathnames.h" |
59 |
|
|
|
60 |
|
|
#define EQUAL 0 |
61 |
|
|
#define GREATER 1 |
62 |
|
|
#define LESS (-1) |
63 |
|
|
|
64 |
|
|
int dflag, fflag; |
65 |
|
|
|
66 |
|
|
char *binary_search(char *, char *, char *); |
67 |
|
|
int compare(char *, char *, char *); |
68 |
|
|
char *linear_search(char *, char *, char *); |
69 |
|
|
int look(char *, char *, char *); |
70 |
|
|
void print_from(char *, char *, char *); |
71 |
|
|
void usage(void); |
72 |
|
|
|
73 |
|
|
int |
74 |
|
|
main(int argc, char *argv[]) |
75 |
|
|
{ |
76 |
|
|
struct stat sb; |
77 |
|
|
int ch, fd, termchar; |
78 |
|
|
char *back, *file, *front, *string, *p; |
79 |
|
|
|
80 |
|
|
if (pledge("stdio rpath flock cpath wpath", NULL) == -1) |
81 |
|
|
err(1, "pledge"); |
82 |
|
|
|
83 |
|
|
file = _PATH_WORDS; |
84 |
|
|
termchar = '\0'; |
85 |
|
|
while ((ch = getopt(argc, argv, "dft:")) != -1) |
86 |
|
|
switch(ch) { |
87 |
|
|
case 'd': |
88 |
|
|
dflag = 1; |
89 |
|
|
break; |
90 |
|
|
case 'f': |
91 |
|
|
fflag = 1; |
92 |
|
|
break; |
93 |
|
|
case 't': |
94 |
|
|
termchar = *optarg; |
95 |
|
|
break; |
96 |
|
|
case '?': |
97 |
|
|
default: |
98 |
|
|
usage(); |
99 |
|
|
} |
100 |
|
|
argc -= optind; |
101 |
|
|
argv += optind; |
102 |
|
|
|
103 |
|
|
switch (argc) { |
104 |
|
|
case 2: /* Don't set -df for user. */ |
105 |
|
|
string = *argv++; |
106 |
|
|
file = *argv; |
107 |
|
|
break; |
108 |
|
|
case 1: /* But set -df by default. */ |
109 |
|
|
dflag = fflag = 1; |
110 |
|
|
string = *argv; |
111 |
|
|
break; |
112 |
|
|
default: |
113 |
|
|
usage(); |
114 |
|
|
} |
115 |
|
|
|
116 |
|
|
if (termchar != '\0' && (p = strchr(string, termchar)) != NULL) |
117 |
|
|
*++p = '\0'; |
118 |
|
|
|
119 |
|
|
if ((fd = open(file, O_RDONLY, 0)) < 0 || fstat(fd, &sb)) |
120 |
|
|
err(2, "%s", file); |
121 |
|
|
if (sb.st_size > SIZE_MAX) |
122 |
|
|
errc(2, EFBIG, "%s", file); |
123 |
|
|
if ((front = mmap(NULL, |
124 |
|
|
(size_t)sb.st_size, PROT_READ, MAP_PRIVATE, fd, (off_t)0)) == MAP_FAILED) |
125 |
|
|
err(2, "%s", file); |
126 |
|
|
back = front + sb.st_size; |
127 |
|
|
exit(look(string, front, back)); |
128 |
|
|
} |
129 |
|
|
|
130 |
|
|
int |
131 |
|
|
look(char *string, char *front, char *back) |
132 |
|
|
{ |
133 |
|
|
int ch; |
134 |
|
|
char *readp, *writep; |
135 |
|
|
|
136 |
|
|
/* Reformat string to avoid doing it multiple times later. */ |
137 |
|
|
for (readp = writep = string; (ch = *readp++);) { |
138 |
|
|
if (fflag) |
139 |
|
|
ch = tolower((unsigned char)ch); |
140 |
|
|
if (!dflag || isalnum((unsigned char)ch)) |
141 |
|
|
*(writep++) = ch; |
142 |
|
|
} |
143 |
|
|
*writep = '\0'; |
144 |
|
|
|
145 |
|
|
front = binary_search(string, front, back); |
146 |
|
|
front = linear_search(string, front, back); |
147 |
|
|
|
148 |
|
|
if (front) |
149 |
|
|
print_from(string, front, back); |
150 |
|
|
return (front ? 0 : 1); |
151 |
|
|
} |
152 |
|
|
|
153 |
|
|
|
154 |
|
|
/* |
155 |
|
|
* Binary search for "string" in memory between "front" and "back". |
156 |
|
|
* |
157 |
|
|
* This routine is expected to return a pointer to the start of a line at |
158 |
|
|
* *or before* the first word matching "string". Relaxing the constraint |
159 |
|
|
* this way simplifies the algorithm. |
160 |
|
|
* |
161 |
|
|
* Invariants: |
162 |
|
|
* front points to the beginning of a line at or before the first |
163 |
|
|
* matching string. |
164 |
|
|
* |
165 |
|
|
* back points to the beginning of a line at or after the first |
166 |
|
|
* matching line. |
167 |
|
|
* |
168 |
|
|
* Base of the Invariants. |
169 |
|
|
* front = NULL; |
170 |
|
|
* back = EOF; |
171 |
|
|
* |
172 |
|
|
* Advancing the Invariants: |
173 |
|
|
* |
174 |
|
|
* p = first newline after halfway point from front to back. |
175 |
|
|
* |
176 |
|
|
* If the string at "p" is not greater than the string to match, |
177 |
|
|
* p is the new front. Otherwise it is the new back. |
178 |
|
|
* |
179 |
|
|
* Termination: |
180 |
|
|
* |
181 |
|
|
* The definition of the routine allows it return at any point, |
182 |
|
|
* since front is always at or before the line to print. |
183 |
|
|
* |
184 |
|
|
* In fact, it returns when the chosen "p" equals "back". This |
185 |
|
|
* implies that there exists a string is least half as long as |
186 |
|
|
* (back - front), which in turn implies that a linear search will |
187 |
|
|
* be no more expensive than the cost of simply printing a string or two. |
188 |
|
|
* |
189 |
|
|
* Trying to continue with binary search at this point would be |
190 |
|
|
* more trouble than it's worth. |
191 |
|
|
*/ |
192 |
|
|
#define SKIP_PAST_NEWLINE(p, back) \ |
193 |
|
|
while (p < back && *p++ != '\n'); |
194 |
|
|
|
195 |
|
|
char * |
196 |
|
|
binary_search(char *string, char *front, char *back) |
197 |
|
|
{ |
198 |
|
|
char *p; |
199 |
|
|
|
200 |
|
|
p = front + (back - front) / 2; |
201 |
|
|
SKIP_PAST_NEWLINE(p, back); |
202 |
|
|
|
203 |
|
|
/* |
204 |
|
|
* If the file changes underneath us, make sure we don't |
205 |
|
|
* infinitely loop. |
206 |
|
|
*/ |
207 |
|
|
while (p < back && back > front) { |
208 |
|
|
if (compare(string, p, back) == GREATER) |
209 |
|
|
front = p; |
210 |
|
|
else |
211 |
|
|
back = p; |
212 |
|
|
p = front + (back - front) / 2; |
213 |
|
|
SKIP_PAST_NEWLINE(p, back); |
214 |
|
|
} |
215 |
|
|
return (front); |
216 |
|
|
} |
217 |
|
|
|
218 |
|
|
/* |
219 |
|
|
* Find the first line that starts with string, linearly searching from front |
220 |
|
|
* to back. |
221 |
|
|
* |
222 |
|
|
* Return NULL for no such line. |
223 |
|
|
* |
224 |
|
|
* This routine assumes: |
225 |
|
|
* |
226 |
|
|
* o front points at the first character in a line. |
227 |
|
|
* o front is before or at the first line to be printed. |
228 |
|
|
*/ |
229 |
|
|
char * |
230 |
|
|
linear_search(char *string, char *front, char *back) |
231 |
|
|
{ |
232 |
|
|
while (front < back) { |
233 |
|
|
switch (compare(string, front, back)) { |
234 |
|
|
case EQUAL: /* Found it. */ |
235 |
|
|
return (front); |
236 |
|
|
break; |
237 |
|
|
case LESS: /* No such string. */ |
238 |
|
|
return (NULL); |
239 |
|
|
break; |
240 |
|
|
case GREATER: /* Keep going. */ |
241 |
|
|
break; |
242 |
|
|
} |
243 |
|
|
SKIP_PAST_NEWLINE(front, back); |
244 |
|
|
} |
245 |
|
|
return (NULL); |
246 |
|
|
} |
247 |
|
|
|
248 |
|
|
/* |
249 |
|
|
* Print as many lines as match string, starting at front. |
250 |
|
|
*/ |
251 |
|
|
void |
252 |
|
|
print_from(char *string, char *front, char *back) |
253 |
|
|
{ |
254 |
|
|
for (; front < back && compare(string, front, back) == EQUAL; ++front) { |
255 |
|
|
for (; front < back && *front != '\n'; ++front) |
256 |
|
|
if (putchar(*front) == EOF) |
257 |
|
|
err(2, "stdout"); |
258 |
|
|
if (putchar('\n') == EOF) |
259 |
|
|
err(2, "stdout"); |
260 |
|
|
} |
261 |
|
|
} |
262 |
|
|
|
263 |
|
|
/* |
264 |
|
|
* Return LESS, GREATER, or EQUAL depending on how the string1 compares with |
265 |
|
|
* string2 (s1 ??? s2). |
266 |
|
|
* |
267 |
|
|
* o Matches up to len(s1) are EQUAL. |
268 |
|
|
* o Matches up to len(s2) are GREATER. |
269 |
|
|
* |
270 |
|
|
* Compare understands about the -f and -d flags, and treats comparisons |
271 |
|
|
* appropriately. |
272 |
|
|
* |
273 |
|
|
* The string "s1" is null terminated. The string s2 is '\n' terminated (or |
274 |
|
|
* "back" terminated). |
275 |
|
|
*/ |
276 |
|
|
int |
277 |
|
|
compare(char *s1, char *s2, char *back) |
278 |
|
|
{ |
279 |
|
|
int ch; |
280 |
|
|
|
281 |
|
|
for (; *s1 && s2 < back && *s2 != '\n'; ++s1, ++s2) { |
282 |
|
|
ch = *s2; |
283 |
|
|
if (fflag) |
284 |
|
|
ch = tolower((unsigned char)ch); |
285 |
|
|
if (dflag && !isalnum((unsigned char)ch)) { |
286 |
|
|
++s2; /* Ignore character in comparison. */ |
287 |
|
|
continue; |
288 |
|
|
} |
289 |
|
|
if (*s1 != ch) |
290 |
|
|
return (*s1 < ch ? LESS : GREATER); |
291 |
|
|
} |
292 |
|
|
return (*s1 ? GREATER : EQUAL); |
293 |
|
|
} |
294 |
|
|
|
295 |
|
|
void |
296 |
|
|
usage(void) |
297 |
|
|
{ |
298 |
|
|
(void)fprintf(stderr, |
299 |
|
|
"usage: look [-df] [-t termchar] string [file]\n"); |
300 |
|
|
exit(2); |
301 |
|
|
} |