1 /* Commandline diff utility to test diff implementations. */
2 /*
3  * Copyright (c) 2018 Martin Pieuchot
4  * Copyright (c) 2020 Neels Hofmeyr <neels@hofmeyr.de>
5  *
6  * Permission to use, copy, modify, and distribute this software for any
7  * purpose with or without fee is hereby granted, provided that the above
8  * copyright notice and this permission notice appear in all copies.
9  *
10  * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
11  * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
12  * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
13  * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
14  * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
15  * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
16  * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
17  */
18 
19 #include <sys/mman.h>
20 #include <sys/stat.h>
21 #include <sys/types.h>
22 
23 #include <err.h>
24 #include <fcntl.h>
25 #include <inttypes.h>
26 #include <stdio.h>
27 #include <stdlib.h>
28 #include <stdbool.h>
29 #include <string.h>
30 #include <unistd.h>
31 
32 #include <diff/arraylist.h>
33 #include <diff/diff_main.h>
34 #include <diff/diff_output.h>
35 
36 __dead void	 usage(void);
37 int		 diffreg(char *, char *, bool, bool, int);
38 int		 openfile(const char *, char **, struct stat *);
39 
40 __dead void
41 usage(void)
42 {
43 	fprintf(stderr,
44 		"usage: %s [-pw] [-C n] file1 file2\n"
45 		"\n"
46 		"  -p   Use Patience Diff (slower but often nicer)\n"
47 		"  -w   Ignore Whitespace\n"
48 		"  -C n Number of Context Lines\n"
49 		, getprogname());
50 	exit(1);
51 }
52 
53 int
54 main(int argc, char *argv[])
55 {
56 	int ch, rc;
57 	bool do_patience = false, ignore_whitespace = false;
58 	int context_lines = 3;
59 
60 	while ((ch = getopt(argc, argv, "pwC:")) != -1) {
61 		switch (ch) {
62 		case 'p':
63 			do_patience = true;
64 			break;
65 		case 'w':
66 			ignore_whitespace = true;
67 			break;
68 		case 'C':
69 			context_lines = atoi(optarg);
70 			break;
71 		default:
72 			usage();
73 		}
74 	}
75 
76 	argc -= optind;
77 	argv += optind;
78 
79 	if (argc != 2)
80 		usage();
81 
82 	rc = diffreg(argv[0], argv[1], do_patience, ignore_whitespace,
83 	    context_lines);
84 	if (rc != DIFF_RC_OK) {
85 		fprintf(stderr, "diff: %s\n", strerror(rc));
86 		return 1;
87 	}
88 	return 0;
89 }
90 
91 const struct diff_algo_config myers_then_patience;
92 const struct diff_algo_config myers_then_myers_divide;
93 const struct diff_algo_config patience;
94 const struct diff_algo_config myers_divide;
95 
96 const struct diff_algo_config myers_then_patience = (struct diff_algo_config){
97 	.impl = diff_algo_myers,
98 	.permitted_state_size = 1024 * 1024 * sizeof(int),
99 	.fallback_algo = &patience,
100 };
101 
102 const struct diff_algo_config myers_then_myers_divide =
103 	(struct diff_algo_config){
104 	.impl = diff_algo_myers,
105 	.permitted_state_size = 1024 * 1024 * sizeof(int),
106 	.fallback_algo = &myers_divide,
107 };
108 
109 const struct diff_algo_config patience = (struct diff_algo_config){
110 	.impl = diff_algo_patience,
111 	/* After subdivision, do Patience again: */
112 	.inner_algo = &patience,
113 	/* If subdivision failed, do Myers Divide et Impera: */
114 	.fallback_algo = &myers_then_myers_divide,
115 };
116 
117 const struct diff_algo_config myers_divide = (struct diff_algo_config){
118 	.impl = diff_algo_myers_divide,
119 	/* When division succeeded, start from the top: */
120 	.inner_algo = &myers_then_myers_divide,
121 	/* (fallback_algo = NULL implies diff_algo_none). */
122 };
123 
124 const struct diff_config diff_config = {
125 	.atomize_func = diff_atomize_text_by_line,
126 	.algo = &myers_then_myers_divide,
127 };
128 
129 const struct diff_config diff_config_patience = {
130 	.atomize_func = diff_atomize_text_by_line,
131 	.algo = &myers_then_patience,
132 };
133 
134 int
135 diffreg(char *file1, char *file2, bool do_patience, bool ignore_whitespace,
136     int context_lines)
137 {
138 	char *str1, *str2;
139 	int fd1, fd2;
140 	struct stat st1, st2;
141 	struct diff_input_info info = {
142 		.left_path = file1,
143 		.right_path = file2,
144 	};
145 	struct diff_result *result;
146 	int rc;
147 	const struct diff_config *cfg;
148 	
149 	cfg = do_patience ? &diff_config_patience : &diff_config;
150 
151 	fd1 = openfile(file1, &str1, &st1);
152 	fd2 = openfile(file2, &str2, &st2);
153 
154 	result = diff_main(cfg, fd1, str1, st1.st_size, fd2, str2, st2.st_size,
155 	    ignore_whitespace);
156 #if 0
157 	rc = diff_output_plain(stdout, &info, result);
158 #else
159 	rc = diff_output_unidiff(stdout, &info, result, context_lines);
160 #endif
161 	diff_result_free(result);
162 
163 	if (str1)
164 		munmap(str1, st1.st_size);
165 	if (str2)
166 		munmap(str2, st2.st_size);
167 	close(fd1);
168 	close(fd2);
169 
170 	return rc;
171 }
172 
173 int
174 openfile(const char *path, char **p, struct stat *st)
175 {
176 	int			 fd;
177 
178 	fd = open(path, O_RDONLY);
179 	if (fd == -1)
180 		err(2, "%s", path);
181 
182 	if (fstat(fd, st) == -1)
183 		err(2, "%s", path);
184 
185 #ifndef DIFF_NO_MMAP
186 	*p = mmap(NULL, st->st_size, PROT_READ, MAP_PRIVATE, fd, 0);
187 	if (*p == MAP_FAILED)
188 #endif
189 		*p = NULL; /* fall back on file I/O */
190 
191 	return fd;
192 }