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 *, bool, bool);
47 int		 openfile(const char *, char **, struct stat *);
48 
49 __dead void
50 usage(void)
51 {
52 	fprintf(stderr,
53 		"usage: %s [-pw] file1 file2\n"
54 		"\n"
55 		"  -p   Use Patience Diff (slower but often nicer)\n"
56 		"  -w   Ignore Whitespace\n"
57 		, getprogname());
58 	exit(1);
59 }
60 
61 int
62 main(int argc, char *argv[])
63 {
64 	int ch, rc;
65 	bool do_patience = false, ignore_whitespace = false;
66 
67 	while ((ch = getopt(argc, argv, "pw")) != -1) {
68 		switch (ch) {
69 		case 'p':
70 			do_patience = true;
71 			break;
72 		case 'w':
73 			ignore_whitespace = true;
74 			break;
75 		default:
76 			usage();
77 		}
78 	}
79 
80 	argc -= optind;
81 	argv += optind;
82 
83 	if (argc != 2)
84 		usage();
85 
86 	rc = diffreg(argv[0], argv[1], do_patience, ignore_whitespace);
87 	if (rc != DIFF_RC_OK) {
88 		fprintf(stderr, "diff: %s\n", strerror(rc));
89 		return 1;
90 	}
91 	return 0;
92 }
93 
94 const struct diff_algo_config myers_then_patience;
95 const struct diff_algo_config myers_then_myers_divide;
96 const struct diff_algo_config patience;
97 const struct diff_algo_config myers_divide;
98 
99 const struct diff_algo_config myers_then_patience = (struct diff_algo_config){
100 	.impl = diff_algo_myers,
101 	.permitted_state_size = 1024 * 1024 * sizeof(int),
102 	.fallback_algo = &patience,
103 };
104 
105 const struct diff_algo_config myers_then_myers_divide =
106 	(struct diff_algo_config){
107 	.impl = diff_algo_myers,
108 	.permitted_state_size = 1024 * 1024 * sizeof(int),
109 	.fallback_algo = &myers_divide,
110 };
111 
112 const struct diff_algo_config patience = (struct diff_algo_config){
113 	.impl = diff_algo_patience,
114 	/* After subdivision, do Patience again: */
115 	.inner_algo = &patience,
116 	/* If subdivision failed, do Myers Divide et Impera: */
117 	.fallback_algo = &myers_then_myers_divide,
118 };
119 
120 const struct diff_algo_config myers_divide = (struct diff_algo_config){
121 	.impl = diff_algo_myers_divide,
122 	/* When division succeeded, start from the top: */
123 	.inner_algo = &myers_then_myers_divide,
124 	/* (fallback_algo = NULL implies diff_algo_none). */
125 };
126 
127 const struct diff_config diff_config = {
128 	.atomize_func = diff_atomize_text_by_line,
129 	.algo = &myers_then_myers_divide,
130 };
131 
132 const struct diff_config diff_config_patience = {
133 	.atomize_func = diff_atomize_text_by_line,
134 	.algo = &myers_then_patience,
135 };
136 
137 int
138 diffreg(char *file1, char *file2, bool do_patience, bool ignore_whitespace)
139 {
140 	char *str1, *str2;
141 	int fd1, fd2;
142 	struct stat st1, st2;
143 	struct diff_input_info info = {
144 		.left_path = file1,
145 		.right_path = file2,
146 	};
147 	struct diff_result *result;
148 	int rc;
149 	const struct diff_config *cfg;
150 	
151 	cfg = do_patience ? &diff_config_patience : &diff_config;
152 
153 	fd1 = openfile(file1, &str1, &st1);
154 	fd2 = openfile(file2, &str2, &st2);
155 
156 	result = diff_main(cfg, fd1, str1, st1.st_size, fd2, str2, st2.st_size,
157 	    ignore_whitespace);
158 #if 0
159 	rc = diff_output_plain(stdout, &info, result);
160 #else
161 	rc = diff_output_unidiff(stdout, &info, result, 3);
162 #endif
163 	diff_result_free(result);
164 
165 	if (str1)
166 		munmap(str1, st1.st_size);
167 	if (str2)
168 		munmap(str2, st2.st_size);
169 	close(fd1);
170 	close(fd2);
171 
172 	return rc;
173 }
174 
175 int
176 openfile(const char *path, char **p, struct stat *st)
177 {
178 	int			 fd;
179 
180 	fd = open(path, O_RDONLY);
181 	if (fd == -1)
182 		err(2, "%s", path);
183 
184 	if (fstat(fd, st) == -1)
185 		err(2, "%s", path);
186 
187 #ifndef DIFF_NO_MMAP
188 	*p = mmap(NULL, st->st_size, PROT_READ, MAP_PRIVATE, fd, 0);
189 	if (*p == MAP_FAILED)
190 #endif
191 		*p = NULL; /* fall back on file I/O */
192 
193 	return fd;
194 }