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 <stdbool.h>
28 #include <string.h>
29 #include <unistd.h>
30 
31 #include <diff/arraylist.h>
32 #include <diff/diff_main.h>
33 #include <diff/diff_output.h>
34 
35 #ifdef __linux__
36 /* stupid shims to compile and test on linux */
37 #define __dead
38 
39 static const char *getprogname()
40 {
41 	return "diff";
42 }
43 #endif
44 
45 __dead void	 usage(void);
46 int		 diffreg(char *, char *, int);
47 int		 openfile(const char *, char **, struct stat *);
48 
49 __dead void
50 usage(void)
51 {
52 	fprintf(stderr,
53 		"usage: %s [-p] file1 file2\n"
54 		"\n"
55 		"  -p   Use Patience Diff (slower but often nicer)\n"
56 		, getprogname());
57 	exit(1);
58 }
59 
60 static bool do_patience = false;
61 
62 int
63 main(int argc, char *argv[])
64 {
65 	int ch, rc;
66 
67 	while ((ch = getopt(argc, argv, "p")) != -1) {
68 		switch (ch) {
69 		case 'p':
70 			do_patience = true;
71 			break;
72 		default:
73 			usage();
74 		}
75 	}
76 
77 	argc -= optind;
78 	argv += optind;
79 
80 	if (argc != 2)
81 		usage();
82 
83 	rc = diffreg(argv[0], argv[1], 0);
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, int flags)
136 {
137 	char *str1, *str2;
138 	int fd1, fd2;
139 	struct stat st1, st2;
140 	struct diff_input_info info = {
141 		.left_path = file1,
142 		.right_path = file2,
143 	};
144 	struct diff_result *result;
145 	int rc;
146 	const struct diff_config *cfg;
147 	
148 	cfg = do_patience ? &diff_config_patience : &diff_config;
149 
150 	fd1 = openfile(file1, &str1, &st1);
151 	fd2 = openfile(file2, &str2, &st2);
152 
153 	result = diff_main(cfg, fd1, str1, st1.st_size, fd2, str2, st2.st_size);
154 #if 0
155 	rc = diff_output_plain(stdout, &info, result);
156 #else
157 	rc = diff_output_unidiff(stdout, &info, result, 3);
158 #endif
159 	diff_result_free(result);
160 
161 	if (str1)
162 		munmap(str1, st1.st_size);
163 	if (str2)
164 		munmap(str2, st2.st_size);
165 	close(fd1);
166 	close(fd2);
167 
168 	return rc;
169 }
170 
171 int
172 openfile(const char *path, char **p, struct stat *st)
173 {
174 	int			 fd;
175 
176 	fd = open(path, O_RDONLY);
177 	if (fd == -1)
178 		err(2, "%s", path);
179 
180 	if (fstat(fd, st) == -1)
181 		err(2, "%s", path);
182 
183 #ifndef DIFF_NO_MMAP
184 	*p = mmap(NULL, st->st_size, PROT_READ, MAP_PRIVATE, fd, 0);
185 	if (*p == MAP_FAILED)
186 #endif
187 		*p = NULL; /* fall back on file I/O */
188 
189 	return fd;
190 }