X-Git-Url: https://www.fleuret.org/cgi-bin/gitweb/gitweb.cgi?p=finddup.git;a=blobdiff_plain;f=finddup.c;h=a8847f5be4c4ac0602a510841b263807d5183ab0;hp=514a67644c153d7f9ab7bab54e01f4d4b2b270aa;hb=1621d18b51827a5c12b939f8e2573de58175d70f;hpb=776548b8b43cfaf709d8f501ebb697c16e8d7832 diff --git a/finddup.c b/finddup.c index 514a676..a8847f5 100644 --- a/finddup.c +++ b/finddup.c @@ -29,6 +29,7 @@ #include #include +#include #include #include #include @@ -41,6 +42,7 @@ #include #define BUFFER_SIZE 4096 +#define LARGE_BUFFER_SIZE 65536 typedef int64_t size_sum_t; @@ -49,18 +51,23 @@ typedef int64_t size_sum_t; int ignore_dotfiles = 0; /* 1 means ignore files and directories starting with a dot */ +int ignore_empty_files = 0; /* 1 means ignore empty files */ + int show_realpaths = 0; /* 1 means ignore files and directories starting with a dot */ -int show_progress = 1; /* 1 means show a progress bar when we are in a +int show_progress = 0; /* 1 means show a progress bar when we are in a tty */ int show_hits = 1; /* 1 means to show the files from dir2 corresponding to the ones from dir1 */ -int show_groups = 0; /* 1 means to show the group IDs when printing +int show_groups = 1; /* 1 means to show the group IDs when printing file names */ +int ignore_same_inode = 0; /* 1 means that comparison between two file + with same inode will always be false */ + /********************************************************************/ /* malloc with error checking. */ @@ -105,7 +112,7 @@ struct file_with_size { size_t size; ino_t inode; struct file_with_size *next; - int id; + int group_id; }; void file_list_delete(struct file_with_size *head) { @@ -131,15 +138,15 @@ int file_list_length(struct file_with_size *head) { int same_content(struct file_with_size *f1, struct file_with_size *f2) { int fd1, fd2, s1, s2; - char buffer1[BUFFER_SIZE], buffer2[BUFFER_SIZE]; + char buffer1[LARGE_BUFFER_SIZE], buffer2[LARGE_BUFFER_SIZE]; fd1 = open(f1->filename, O_RDONLY); fd2 = open(f2->filename, O_RDONLY); if(fd1 >= 0 && fd2 >= 0) { while(1) { - s1 = read(fd1, buffer1, BUFFER_SIZE); - s2 = read(fd2, buffer2, BUFFER_SIZE); + s1 = read(fd1, buffer1, LARGE_BUFFER_SIZE); + s2 = read(fd2, buffer2, LARGE_BUFFER_SIZE); if(s1 < 0 || s2 < 0) { close(fd1); @@ -153,7 +160,7 @@ int same_content(struct file_with_size *f1, struct file_with_size *f2) { close(fd2); return 1; } else { - if(strncmp(buffer1, buffer2, s1)) { + if(memcmp(buffer1, buffer2, s1)) { close(fd1); close(fd2); return 0; @@ -185,6 +192,9 @@ int same_content(struct file_with_size *f1, struct file_with_size *f2) { } int same_files(struct file_with_size *f1, struct file_with_size *f2) { + if(ignore_same_inode && f1->inode == f2->inode) { + return 0; + } return f1->size == f2->size && same_content(f1, f2); } @@ -196,7 +206,7 @@ struct file_with_size *scan_directory(struct file_with_size *tail, struct dirent *dir_e; struct stat sb; struct file_with_size *tmp; - char subname[PATH_MAX]; + char subname[PATH_MAX + 1]; if(lstat(name, &sb) != 0) { fprintf(stderr, "Can not stat \"%s\": %s\n", name, strerror(errno)); @@ -219,13 +229,17 @@ struct file_with_size *scan_directory(struct file_with_size *tail, closedir(dir); } else { if(S_ISREG(sb.st_mode)) { - tmp = safe_malloc(sizeof(struct file_with_size)); - tmp->next = tail; - tmp->filename = strdup(name); - tmp->size = sb.st_size; - tmp->inode = sb.st_ino; - tmp->id = -1; - tail = tmp; + if(!ignore_entry(name)) { + if(!ignore_empty_files || sb.st_size > 0) { + tmp = safe_malloc(sizeof(struct file_with_size)); + tmp->next = tail; + tmp->filename = strdup(name); + tmp->size = sb.st_size; + tmp->inode = sb.st_ino; + tmp->group_id = -1; + tail = tmp; + } + } } } @@ -233,62 +247,136 @@ struct file_with_size *scan_directory(struct file_with_size *tail, } void print_file(struct file_with_size *node) { + char tmp[PATH_MAX + 1]; if(show_realpaths) { if(show_groups) { - printf("%d %s\n", node->id, realpath(node->filename, 0)); + realpath(node->filename, tmp); + printf("%d %s\n", node->group_id, tmp); } else { - printf("%s\n", realpath(node->filename, 0)); + realpath(node->filename, tmp); + printf("%s\n", tmp); } } else { if(show_groups) { - printf("%d %s\n", node->id, node->filename); + printf("%d %s\n", node->group_id, node->filename); } else { printf("%s\n", node->filename); } } } -void start(const char *dirname1, const char *dirname2) { - struct file_with_size *list1, *list2; - struct file_with_size *node1, *node2; - struct stat sb1, sb2; - int not_in, found, same_dir; +int compare_nodes(const void *x1, const void *x2) { + const struct file_with_size **f1, **f2; - if(strncmp(dirname2, "not:", 4) == 0) { - not_in = 1; - dirname2 += 4; + f1 = (const struct file_with_size **) x1; + f2 = (const struct file_with_size **) x2; + + if((*f1)->group_id < (*f2)->group_id) { + return -1; + } else if((*f1)->group_id > (*f2)->group_id) { + return 1; } else { - not_in = 0; + return 0; } +} - if(lstat(dirname1, &sb1) != 0) { - fprintf(stderr, "Can not stat \"%s\": %s\n", dirname1, strerror(errno)); - exit(EXIT_FAILURE); + +void print_result(struct file_with_size *list1, struct file_with_size *list2) { + struct file_with_size *node1, *node2; + struct file_with_size **nodes; + int nb, n; + + nb = 0; + for(node1 = list1; node1; node1 = node1->next) { + if(node1->group_id >= 0) { nb++; } } - if(lstat(dirname2, &sb2) != 0) { - fprintf(stderr, "Can not stat \"%s\": %s\n", dirname2, strerror(errno)); - exit(EXIT_FAILURE); + if(list2) { + for(node2 = list2; node2; node2 = node2->next) { + if(node2->group_id >= 0) { nb++; } + } } - same_dir = (sb1.st_ino == sb2.st_ino); + nodes = safe_malloc(nb * sizeof(struct file_with_size *)); + + n = 0; + for(node1 = list1; node1; node1 = node1->next) { + if(node1->group_id >= 0) { + nodes[n++] = node1; + } + } + + if(list2) { + for(node2 = list2; node2; node2 = node2->next) { + if(node2->group_id >= 0) { + nodes[n++] = node2; + } + } + } + + qsort(nodes, nb, sizeof(struct file_with_size *), compare_nodes); + + for(n = 0; n < nb; n++) { + if(!show_groups && n > 0 && nodes[n]->group_id != nodes[n-1]->group_id) { + printf("\n"); + } + print_file(nodes[n]); + } + + free(nodes); +} + +void start(const char *dirname1, const char *dirname2) { + struct file_with_size *list1, *list2; + struct file_with_size *node1, *node2; + int not_in, found; + int k, p, pp, l1, n; + + not_in = 0; + + if(show_progress) { + fprintf(stderr, "Scanning %s ... ", dirname1); + } - fprintf(stderr, "Scanning %s ... ", dirname1); list1 = scan_directory(0, dirname1); - if(same_dir) { - list2 = list1; - } else { - fprintf(stderr, "%s ... ", dirname2); + + if(dirname2) { + if(strncmp(dirname2, "not:", 4) == 0) { + not_in = 1; + dirname2 += 4; + } + if(show_progress) { + fprintf(stderr, "%s ... ", dirname2); + } list2 = scan_directory(0, dirname2); + } else { + list2 = list1; } - fprintf(stderr, "done.\n"); + + if(show_progress) { + fprintf(stderr, "done.\n"); + } + + k = 0; + pp = -1; + n = 0; + l1 = file_list_length(list1); if(not_in) { for(node1 = list1; node1; node1 = node1->next) { + if(show_progress) { + p = (100 * n)/l1; + if(p > pp) { + fprintf(stderr, "%d%%\n", p); + pp = p; + } + n++; + } + found = 0; for(node2 = list2; !found && node2; node2 = node2->next) { - if(node1->inode != node2->inode && same_files(node1, node2)) { + if(same_files(node1, node2)) { found = 1; } } @@ -303,24 +391,29 @@ void start(const char *dirname1, const char *dirname2) { } } else { - int k = 0; - for(node1 = list1; node1; node1 = node1->next) { + if(show_progress) { + p = (100 * n)/l1; + if(p > pp) { + fprintf(stderr, "%d%%\n", p); + pp = p; + } + n++; + } + for(node2 = list2; node2; node2 = node2->next) { - if(node1->inode != node2->inode && same_files(node1, node2)) { - if(node1->id < 0) { - if(node2->id >= 0) { - node1->id = node2->id; - } else { - node1->id = k; - k++; + if(node1->group_id < 0 || node2->group_id < 0) { + if(same_files(node1, node2)) { + if(node1->group_id < 0) { + if(node2->group_id >= 0) { + node1->group_id = node2->group_id; + } else { + node1->group_id = k; + k++; + } } - print_file(node1); - } - if(node2->id < 0) { - node2->id = node1->id; - if(show_hits) { - print_file(node2); + if(node2->group_id < 0) { + node2->group_id = node1->group_id; } } } @@ -328,9 +421,13 @@ void start(const char *dirname1, const char *dirname2) { } } - file_list_delete(list1); - if(!same_dir) { + if(dirname2) { + print_result(list1, list2); + file_list_delete(list1); file_list_delete(list2); + } else { + print_result(list1, 0); + file_list_delete(list1); } } @@ -341,9 +438,12 @@ void print_help(FILE *out) { fprintf(out, "\n"); fprintf(out, " -h show this help\n"); fprintf(out, " -d ignore dot files and directories\n"); + fprintf(out, " -0 ignore empty files\n"); fprintf(out, " -c do not show which files in DIR2 corresponds to those in DIR1\n"); - fprintf(out, " -g show the file groups\n"); + fprintf(out, " -g do not show the file groups\n"); + fprintf(out, " -p show progress\n"); fprintf(out, " -r show the real file paths\n"); + fprintf(out, " -i consider files with same inode as different\n"); fprintf(out, "\n"); fprintf(out, "Report bugs and comments to \n"); } @@ -359,7 +459,7 @@ int main(int argc, char **argv) { setlocale (LC_ALL, ""); while (1) { - c = getopt(argc, argv, "hrcgd"); + c = getopt(argc, argv, "hrcgd0p"); if (c == -1) break; @@ -375,12 +475,24 @@ int main(int argc, char **argv) { ignore_dotfiles = 1; break; + case '0': + ignore_empty_files = 1; + break; + case 'r': show_realpaths = 1; break; + case 'i': + ignore_same_inode = 1; + break; + case 'g': - show_groups = 1; + show_groups = 0; + break; + + case 'p': + show_progress = 1; break; case 'c': @@ -392,10 +504,11 @@ int main(int argc, char **argv) { } } - if(optind + 1 < argc) { + if(optind + 2 == argc) { start(argv[optind], argv[optind + 1]); - } else if(optind < argc) { - start(argv[optind], argv[optind]); + } else if(optind + 1 == argc) { + ignore_same_inode = 1; + start(argv[optind], 0); } else { print_help(stderr); exit(EXIT_FAILURE);