3 * finddup is a simple utility find duplicated files, files common to
4 * several directories, or files present in one directory and not in
7 * Copyright (c) 2010 Francois Fleuret
8 * Written by Francois Fleuret <francois@fleuret.org>
10 * This file is part of finddup.
12 * finddup is free software: you can redistribute it and/or modify it
13 * under the terms of the GNU General Public License version 3 as
14 * published by the Free Software Foundation.
16 * finddup is distributed in the hope that it will be useful, but WITHOUT
17 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
18 * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
19 * License for more details.
21 * You should have received a copy of the GNU General Public License
22 * along with finddup. If not, see <http://www.gnu.org/licenses/>.
26 #define VERSION_NUMBER "0.6"
30 #include <sys/types.h>
32 #include <sys/param.h>
39 #include <sys/ioctl.h>
44 #define BUFFER_SIZE 4096
46 typedef int64_t size_sum_t;
48 /* Yeah, global variables! */
50 int ignore_dotfiles = 0; /* 1 means ignore files and directories
51 starting with a dot */
53 int show_realpaths = 0; /* 1 means ignore files and directories
54 starting with a dot */
56 int show_progress = 0; /* 1 means show a progress bar when we are in a
59 int show_hits = 1; /* 1 means to show the files from dir2
60 corresponding to the ones from dir1 */
62 int show_groups = 1; /* 1 means to show the group IDs when printing
65 /********************************************************************/
67 /* malloc with error checking. */
69 void *safe_malloc(size_t n) {
72 printf("Can not allocate memory: %s\n", strerror(errno));
78 /********************************************************************/
80 int ignore_entry(const char *name) {
82 strcmp(name, ".") == 0 ||
83 strcmp(name, "..") == 0 ||
84 (ignore_dotfiles && name[0] == '.');
87 void print_size_sum(size_sum_t s) {
89 char *a = tmp + sizeof(tmp)/sizeof(char);
102 /**********************************************************************/
104 struct file_with_size {
108 struct file_with_size *next;
112 void file_list_delete(struct file_with_size *head) {
113 struct file_with_size *next;
116 free(head->filename);
122 int file_list_length(struct file_with_size *head) {
131 /**********************************************************************/
133 int same_content(struct file_with_size *f1, struct file_with_size *f2) {
134 int fd1, fd2, s1, s2;
135 char buffer1[BUFFER_SIZE], buffer2[BUFFER_SIZE];
137 fd1 = open(f1->filename, O_RDONLY);
138 fd2 = open(f2->filename, O_RDONLY);
140 if(fd1 >= 0 && fd2 >= 0) {
142 s1 = read(fd1, buffer1, BUFFER_SIZE);
143 s2 = read(fd2, buffer2, BUFFER_SIZE);
145 if(s1 < 0 || s2 < 0) {
157 if(strncmp(buffer1, buffer2, s1)) {
165 "Different read size without error on files of same size.\n");
173 "Can not open \"%s\" error: %s\n",
180 "Can not open \"%s\" error: %s\n",
188 int same_files(struct file_with_size *f1, struct file_with_size *f2) {
189 return f1->size == f2->size && same_content(f1, f2);
192 /**********************************************************************/
194 struct file_with_size *scan_directory(struct file_with_size *tail,
197 struct dirent *dir_e;
199 struct file_with_size *tmp;
200 char subname[PATH_MAX + 1];
202 if(lstat(name, &sb) != 0) {
203 fprintf(stderr, "Can not stat \"%s\": %s\n", name, strerror(errno));
207 if(S_ISLNK(sb.st_mode)) {
214 while((dir_e = readdir(dir))) {
215 if(!ignore_entry(dir_e->d_name)) {
216 snprintf(subname, PATH_MAX, "%s/%s", name, dir_e->d_name);
217 tail = scan_directory(tail, subname);
222 if(S_ISREG(sb.st_mode)) {
223 tmp = safe_malloc(sizeof(struct file_with_size));
225 tmp->filename = strdup(name);
226 tmp->size = sb.st_size;
227 tmp->inode = sb.st_ino;
236 void print_file(struct file_with_size *node) {
237 char tmp[PATH_MAX + 1];
240 realpath(node->filename, tmp);
241 printf("%d %s\n", node->group_id, tmp);
243 realpath(node->filename, tmp);
248 printf("%d %s\n", node->group_id, node->filename);
250 printf("%s\n", node->filename);
255 int compare_nodes(const void *x1, const void *x2) {
256 const struct file_with_size **f1, **f2;
258 f1 = (const struct file_with_size **) x1;
259 f2 = (const struct file_with_size **) x2;
261 if((*f1)->group_id < (*f2)->group_id) {
263 } else if((*f1)->group_id > (*f2)->group_id) {
271 void print_result(struct file_with_size *list1, struct file_with_size *list2) {
272 struct file_with_size *node1, *node2;
273 struct file_with_size **nodes;
277 for(node1 = list1; node1; node1 = node1->next) {
278 if(node1->group_id >= 0) { nb++; }
282 for(node2 = list2; node2; node2 = node2->next) {
283 if(node2->group_id >= 0) { nb++; }
287 nodes = safe_malloc(nb * sizeof(struct file_with_size *));
290 for(node1 = list1; node1; node1 = node1->next) {
291 if(node1->group_id >= 0) {
297 for(node2 = list2; node2; node2 = node2->next) {
298 if(node2->group_id >= 0) {
304 qsort(nodes, nb, sizeof(struct file_with_size *), compare_nodes);
306 for(n = 0; n < nb; n++) {
307 print_file(nodes[n]);
313 void start(const char *dirname1, const char *dirname2) {
314 struct file_with_size *list1, *list2;
315 struct file_with_size *node1, *node2;
316 struct stat sb1, sb2;
317 int not_in, found, same_dir;
320 if(strncmp(dirname2, "not:", 4) == 0) {
327 if(lstat(dirname1, &sb1) != 0) {
328 fprintf(stderr, "Can not stat \"%s\": %s\n", dirname1, strerror(errno));
332 if(lstat(dirname2, &sb2) != 0) {
333 fprintf(stderr, "Can not stat \"%s\": %s\n", dirname2, strerror(errno));
337 same_dir = (sb1.st_ino == sb2.st_ino);
340 fprintf(stderr, "Scanning %s ... ", dirname1);
343 list1 = scan_directory(0, dirname1);
349 fprintf(stderr, "%s ... ", dirname2);
351 list2 = scan_directory(0, dirname2);
355 fprintf(stderr, "done.\n");
361 l1 = file_list_length(list1);
364 for(node1 = list1; node1; node1 = node1->next) {
368 fprintf(stderr, "%d%%\n", p);
376 for(node2 = list2; !found && node2; node2 = node2->next) {
377 if(node1->inode != node2->inode && same_files(node1, node2)) {
384 printf("%s\n", realpath(node1->filename, 0));
386 printf("%s\n", node1->filename);
392 for(node1 = list1; node1; node1 = node1->next) {
396 fprintf(stderr, "%d%%\n", p);
402 for(node2 = list2; node2; node2 = node2->next) {
403 if(node1->inode != node2->inode && same_files(node1, node2)) {
404 if(node1->group_id < 0) {
405 if(node2->group_id >= 0) {
406 node1->group_id = node2->group_id;
412 if(node2->group_id < 0) {
413 node2->group_id = node1->group_id;
420 print_result(list1, list2);
422 file_list_delete(list1);
424 file_list_delete(list2);
428 void print_help(FILE *out) {
429 fprintf(out, "Usage: finddup [OPTION]... DIR1 [[not:]DIR2]\n");
430 fprintf(out, "Version %s (%s)\n", VERSION_NUMBER, UNAME);
431 fprintf(out, "Without DIR2, lists duplicated files found in DIR1. With DIR2, lists files common to both directories. With the not: prefix, lists files found in DIR1 which do not exist in DIR2.\n");
433 fprintf(out, " -h show this help\n");
434 fprintf(out, " -d ignore dot files and directories\n");
435 fprintf(out, " -c do not show which files in DIR2 corresponds to those in DIR1\n");
436 fprintf(out, " -g do not show the file groups\n");
437 fprintf(out, " -p show progress\n");
438 fprintf(out, " -r show the real file paths\n");
440 fprintf(out, "Report bugs and comments to <francois@fleuret.org>\n");
443 /**********************************************************************/
445 int main(int argc, char **argv) {
447 struct file_with_size *root;
451 setlocale (LC_ALL, "");
454 c = getopt(argc, argv, "hrcgdp");
491 if(optind + 1 < argc) {
492 start(argv[optind], argv[optind + 1]);
493 } else if(optind < argc) {
494 start(argv[optind], argv[optind]);