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 
22 #include <err.h>
23 #include <fcntl.h>
24 #include <inttypes.h>
25 #include <stdio.h>
26 #include <stdlib.h>
27 #include <unistd.h>
28 
29 #include <diff/diff_main.h>
30 #include <diff/diff_output.h>
31 
32 #ifdef __linux__
33 /* stupid shims to compile and test on linux */
34 #define __dead
35 
36 static const char *getprogname()
37 {
38 	return "diff";
39 }
40 #endif
41 
42 __dead void	 usage(void);
43 int		 diffreg(char *, char *, int);
44 char		*mmapfile(const char *, struct stat *);
45 
46 __dead void
47 usage(void)
48 {
49 	fprintf(stderr, "usage: %s file1 file2\n", getprogname());
50 	exit(1);
51 }
52 
53 int
54 main(int argc, char *argv[])
55 {
56 	int ch;
57 
58 	while ((ch = getopt(argc, argv, "")) != -1) {
59 		switch (ch) {
60 		default:
61 			usage();
62 		}
63 	}
64 
65 	argc -= optind;
66 	argv += optind;
67 
68 	if (argc != 2)
69 		usage();
70 
71 	return diffreg(argv[0], argv[1], 0);
72 }
73 
74 const struct diff_algo_config myers, patience, myers_divide;
75 
76 const struct diff_algo_config myers = (struct diff_algo_config){
77 	.impl = diff_algo_myers,
78 	.permitted_state_size = 1024 * 1024 * sizeof(int),
79 	.fallback_algo = &patience,
80 };
81 
82 const struct diff_algo_config patience = (struct diff_algo_config){
83 	.impl = diff_algo_patience,
84 	.inner_algo = &patience,	// After subdivision, do Patience again.
85 	.fallback_algo = &myers_divide, // If subdivision failed, do Myers Divide et Impera.
86 };
87 
88 const struct diff_algo_config myers_divide = (struct diff_algo_config){
89 	.impl = diff_algo_myers_divide,
90 	.inner_algo = &myers,		// When division succeeded, start from the top.
91 					// (fallback_algo = NULL implies diff_algo_none).
92 };
93 
94 const struct diff_config diff_config = {
95 	.atomize_func = diff_atomize_text_by_line,
96 	.algo = &myers,
97 };
98 
99 int
100 diffreg(char *file1, char *file2, int flags)
101 {
102 	char *str1, *str2;
103 	struct stat st1, st2;
104 	struct diff_input_info info = {
105 		.left_path = file1,
106 		.right_path = file2,
107 	};
108 	struct diff_result *result;
109 	enum diff_rc rc;
110 
111 	str1 = mmapfile(file1, &st1);
112 	str2 = mmapfile(file2, &st2);
113 
114 	result = diff_main(&diff_config, str1, st1.st_size, str2, st2.st_size);
115 #if 0
116 	rc = diff_output_plain(stdout, &info, result);
117 #else
118 	rc = diff_output_unidiff(stdout, &info, result, 3);
119 #endif
120 	diff_result_free(result);
121 
122 	munmap(str1, st1.st_size);
123 	munmap(str2, st2.st_size);
124 
125 	return rc;
126 }
127 
128 char *
129 mmapfile(const char *path, struct stat *st)
130 {
131 	int			 fd;
132 	char			*p;
133 
134 	fd = open(path, O_RDONLY);
135 	if (fd == -1)
136 		err(2, "%s", path);
137 
138 	if (fstat(fd, st) == -1)
139 		err(2, "%s", path);
140 
141 	if ((uintmax_t)st->st_size > SIZE_MAX)
142 		errx(2, "%s: file too big to fit memory", path);
143 
144 	p = mmap(NULL, st->st_size, PROT_READ, MAP_PRIVATE, fd, 0);
145 	if (p == MAP_FAILED)
146 		err(2, "mmap");
147 
148 	close(fd);
149 
150 	return p;
151 }