X-Git-Url: https://www.fleuret.org/cgi-bin/gitweb/gitweb.cgi?a=blobdiff_plain;f=finddup.c;h=bf59852a5274fe626d10cbaad100c3fc525096df;hb=e1996bdfbf277c5ff0c0328c402efdde0dbe6110;hp=ab345d3ddb2804d1b9566466a442d7ba8f0ab547;hpb=dcc6728402ad901e945dbf2f8be6f3eae35f8895;p=finddup.git diff --git a/finddup.c b/finddup.c index ab345d3..bf59852 100644 --- a/finddup.c +++ b/finddup.c @@ -1,8 +1,8 @@ /* - * finddup is a simple utility find duplicated files, files common to - * several directories, or files present in one directory and not in - * another one. + * finddup is a simple utility to find duplicated files, files common + * to several directories, or files present in one directory and not + * in another one. * * Copyright (c) 2010 Francois Fleuret * Written by Francois Fleuret @@ -23,7 +23,7 @@ * */ -#define VERSION_NUMBER "0.6" +#define VERSION_NUMBER "0.8" #define _BSD_SOURCE @@ -93,42 +93,28 @@ int ignore_entry(const char *name) { (ignore_dotfiles && name[0] == '.'); } -void print_size_sum(size_sum_t s) { - char tmp[100]; - char *a = tmp + sizeof(tmp)/sizeof(char); - *(--a) = '\0'; - if(s) { - while(s) { - *(--a) = s%10 + '0'; - s /= 10; - } - } else { - *(--a) = '0'; - } - printf(a); -} - /**********************************************************************/ -struct file_with_size { - char *filename; +struct file_node { + struct file_node *next; + char *name; size_t size; ino_t inode; - struct file_with_size *next; - int group_id, dir_id; + int group_id; /* one per identical file content */ + int dir_id; /* 1 for DIR1, and 2 for DIR2 */ }; -void file_list_delete(struct file_with_size *head) { - struct file_with_size *next; +void file_list_delete(struct file_node *head) { + struct file_node *next; while(head) { next = head->next; - free(head->filename); + free(head->name); free(head); head = next; } } -int file_list_length(struct file_with_size *head) { +int file_list_length(struct file_node *head) { int l = 0; while(head) { l++; @@ -139,12 +125,12 @@ int file_list_length(struct file_with_size *head) { /**********************************************************************/ -int same_content(struct file_with_size *f1, struct file_with_size *f2, +int same_content(struct file_node *f1, struct file_node *f2, char *buffer1, char *buffer2) { int fd1, fd2, s1, s2; - fd1 = open(f1->filename, O_RDONLY); - fd2 = open(f2->filename, O_RDONLY); + fd1 = open(f1->name, O_RDONLY); + fd2 = open(f2->name, O_RDONLY); if(fd1 >= 0 && fd2 >= 0) { while(1) { @@ -180,21 +166,21 @@ int same_content(struct file_with_size *f1, struct file_with_size *f2, if(fd1 < 0) { fprintf(stderr, "Can not open \"%s\" error: %s\n", - f1->filename, + f1->name, strerror(errno)); } if(fd2 < 0) { fprintf(stderr, "Can not open \"%s\" error: %s\n", - f2->filename, + f2->name, strerror(errno)); } exit(EXIT_FAILURE); } } -int same_files(struct file_with_size *f1, struct file_with_size *f2, +int same_files(struct file_node *f1, struct file_node *f2, char *buffer1, char *buffer2) { if(ignore_same_inode && f1->inode == f2->inode) { return 0; @@ -204,12 +190,12 @@ int same_files(struct file_with_size *f1, struct file_with_size *f2, /**********************************************************************/ -struct file_with_size *scan_directory(struct file_with_size *tail, +struct file_node *scan_directory(struct file_node *tail, const char *name) { DIR *dir; struct dirent *dir_e; struct stat sb; - struct file_with_size *tmp; + struct file_node *tmp; char subname[PATH_MAX + 1]; if(lstat(name, &sb) != 0) { @@ -235,9 +221,9 @@ struct file_with_size *scan_directory(struct file_with_size *tail, if(S_ISREG(sb.st_mode)) { if(!ignore_entry(name)) { if(!ignore_empty_files || sb.st_size > 0) { - tmp = safe_malloc(sizeof(struct file_with_size)); + tmp = safe_malloc(sizeof(struct file_node)); tmp->next = tail; - tmp->filename = strdup(name); + tmp->name = strdup(name); tmp->size = sb.st_size; tmp->inode = sb.st_ino; tmp->group_id = -1; @@ -251,30 +237,35 @@ struct file_with_size *scan_directory(struct file_with_size *tail, return tail; } -void print_file(struct file_with_size *node) { +void print_file(struct file_node *node) { char tmp[PATH_MAX + 1]; if(show_realpaths) { - if(show_groups) { - realpath(node->filename, tmp); - printf("%d %s\n", node->group_id, tmp); + if(realpath(node->name, tmp)) { + if(show_groups) { + printf("%d %s\n", node->group_id, tmp); + } else { + printf("%s\n", tmp); + } } else { - realpath(node->filename, tmp); - printf("%s\n", tmp); + printf("Can not get the realpath of \"%s\": %s\n", + node->name, + strerror(errno)); + exit(EXIT_FAILURE); } } else { if(show_groups) { - printf("%d %s\n", node->group_id, node->filename); + printf("%d %s\n", node->group_id, node->name); } else { - printf("%s\n", node->filename); + printf("%s\n", node->name); } } } int compare_nodes(const void *x1, const void *x2) { - const struct file_with_size **f1, **f2; + const struct file_node **f1, **f2; - f1 = (const struct file_with_size **) x1; - f2 = (const struct file_with_size **) x2; + f1 = (const struct file_node **) x1; + f2 = (const struct file_node **) x2; if((*f1)->group_id < (*f2)->group_id) { return -1; @@ -292,9 +283,9 @@ int compare_nodes(const void *x1, const void *x2) { } -void print_result(struct file_with_size *list1, struct file_with_size *list2) { - struct file_with_size *node1, *node2; - struct file_with_size **nodes; +void print_result(struct file_node *list1, struct file_node *list2) { + struct file_node *node1, *node2; + struct file_node **nodes; int nb, n; nb = 0; @@ -308,7 +299,7 @@ void print_result(struct file_with_size *list1, struct file_with_size *list2) { } } - nodes = safe_malloc(nb * sizeof(struct file_with_size *)); + nodes = safe_malloc(nb * sizeof(struct file_node *)); n = 0; for(node1 = list1; node1; node1 = node1->next) { @@ -325,7 +316,7 @@ void print_result(struct file_with_size *list1, struct file_with_size *list2) { } } - qsort(nodes, nb, sizeof(struct file_with_size *), compare_nodes); + qsort(nodes, nb, sizeof(struct file_node *), compare_nodes); for(n = 0; n < nb; n++) { if(!show_groups && n > 0 && nodes[n]->group_id != nodes[n-1]->group_id) { @@ -342,7 +333,7 @@ void print_progress(int max, int n, int *pp) { int width; if(show_progress && tty_width > 0) { width = tty_width - 7; - p = (width * n) / max; + p = (width * n) / (max - 1); if(p > *pp) { for(k = 0; k < p; k++) { fprintf(stderr, "+"); @@ -351,17 +342,19 @@ void print_progress(int max, int n, int *pp) { fprintf(stderr, "-"); } *pp = p; - p = (100 * n) / max; - fprintf(stderr, " [% 3d%%]\r", p); + p = (100 * n) / (max - 1); + fprintf(stderr, " [%3d%%]\r", p); } } } void start(const char *dirname1, const char *dirname2) { - struct file_with_size *list1, *list2; - struct file_with_size *node1, *node2; + struct file_node *list1, *list2; + struct file_node *node1, *node2; int not_in, found; - int k, pp, l1, n; + int nb_groups, nb_nodes; + int list1_length, previous_progress; + struct winsize win; char *buffer1 = safe_malloc(sizeof(char) * READ_BUFFER_SIZE); char *buffer2 = safe_malloc(sizeof(char) * READ_BUFFER_SIZE); @@ -369,6 +362,10 @@ void start(const char *dirname1, const char *dirname2) { not_in = 0; if(show_progress) { + if(isatty(STDOUT_FILENO) && + !ioctl (STDOUT_FILENO, TIOCGWINSZ, (char *) &win)) { + tty_width = win.ws_col; + } fprintf(stderr, "Scanning %s ... ", dirname1); } @@ -377,10 +374,11 @@ void start(const char *dirname1, const char *dirname2) { if(dirname2) { if(strncmp(dirname2, "not:", 4) == 0) { not_in = 1; - /* We should show groups even in that mode. However they are not - properly calculated for now, so we force it off. */ + /* groups are not computed in the not: mode */ show_groups = 0; dirname2 += 4; + } else if(strncmp(dirname2, "and:", 4) == 0) { + dirname2 += 4; } if(show_progress) { fprintf(stderr, "%s ... ", dirname2); @@ -394,15 +392,15 @@ void start(const char *dirname1, const char *dirname2) { fprintf(stderr, "done.\n"); } - k = 0; - pp = -1; - n = 0; - l1 = file_list_length(list1); + nb_groups = 0; + previous_progress = -1; + nb_nodes = 0; + list1_length = file_list_length(list1); if(not_in) { for(node1 = list1; node1; node1 = node1->next) { - print_progress(l1, n, &pp); - n++; + print_progress(list1_length, nb_nodes, &previous_progress); + nb_nodes++; found = 0; @@ -414,17 +412,17 @@ void start(const char *dirname1, const char *dirname2) { if(!found) { if(show_realpaths) { - printf("%s\n", realpath(node1->filename, 0)); + printf("%s\n", realpath(node1->name, 0)); } else { - printf("%s\n", node1->filename); + printf("%s\n", node1->name); } } } } else { for(node1 = list1; node1; node1 = node1->next) { - print_progress(l1, n, &pp); - n++; + print_progress(list1_length, nb_nodes, &previous_progress); + nb_nodes++; for(node2 = list2; node2; node2 = node2->next) { if(node1->group_id < 0 || node2->group_id < 0) { @@ -433,9 +431,9 @@ void start(const char *dirname1, const char *dirname2) { if(node2->group_id >= 0) { node1->group_id = node2->group_id; } else { - node1->group_id = k; + node1->group_id = nb_groups; node1->dir_id = 1; - k++; + nb_groups++; } } if(node2->group_id < 0) { @@ -466,38 +464,46 @@ void start(const char *dirname1, const char *dirname2) { } void print_help(FILE *out) { - fprintf(out, "Usage: finddup [OPTION]... DIR1 [[not:]DIR2]\n"); + fprintf(out, "Usage: finddup [OPTION]... DIR1 [[and:|not:]DIR2]\n"); fprintf(out, "Version %s (%s)\n", VERSION_NUMBER, UNAME); - 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"); + 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. The and: prefix is the default and should be used only if you have a directory starting with 'not:'\n"); 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 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"); + /* 01234567890123456789012345678901234567890123456789012345678901234567890123456789*/ + fprintf(out, " -h, --help show this help\n"); + fprintf(out, " -d, --ignore-dots ignore dot files and directories\n"); + fprintf(out, " -0, --ignore-empty ignore empty files\n"); + fprintf(out, " -c, --hide-matchings do not show which files in DIR2 corresponds to\n"); + fprintf(out, " those in DIR1\n"); + fprintf(out, " -g, --no-group-ids do not show the file groups\n"); + fprintf(out, " -p, --show-progress show progress\n"); + fprintf(out, " -r, --real-paths show the real file paths\n"); + fprintf(out, " -i, --same-inodes-are-different\n"); + fprintf(out, " consider files with same inode as different\n"); fprintf(out, "\n"); - fprintf(out, "Report bugs and comments to \n"); + fprintf(out, "Report bugs and comments to .\n"); } /**********************************************************************/ +static struct option long_options[] = { + { "help", no_argument, 0, 'h' }, + { "same-inodes-are-different", no_argument, 0, 'i' }, + { "real-paths", no_argument, 0, 'r' }, + { "hide-matchings", no_argument, 0, 'c' }, + { "no-group-ids", no_argument, 0, 'g' }, + { "ignore-dots", no_argument, 0, 'd' }, + { "ignore-empty", no_argument, 0, '0' }, + { "show-progress", no_argument, 0, 'p' }, + { 0, 0, 0, 0 } +}; + int main(int argc, char **argv) { int c; - struct file_with_size *root; - struct winsize win; - - root = 0; setlocale (LC_ALL, ""); - while (1) { - c = getopt(argc, argv, "hircgd0p"); - if (c == -1) - break; - + while ((c = getopt_long(argc, argv, "hircgd0p", + long_options, NULL)) != -1) { switch (c) { case 'h': @@ -539,12 +545,6 @@ int main(int argc, char **argv) { } } - if(show_progress && - isatty(STDOUT_FILENO) && - !ioctl (STDOUT_FILENO, TIOCGWINSZ, (char *) &win)) { - tty_width = win.ws_col; - } - if(optind + 2 == argc) { start(argv[optind], argv[optind + 1]); } else if(optind + 1 == argc) {