X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?a=blobdiff_plain;f=finddup.c;h=e881fff9ba9f18527d0314c2cdf8765308bb6ab5;hb=ad1b1b341fdfabc708e174cbc29006988b1df342;hp=6b141428ecd77e25bd37a258456d0eb12ed666c7;hpb=87d05f158cb6c552c1de528628bbf10c65e0edb6;p=finddup.git diff --git a/finddup.c b/finddup.c index 6b14142..e881fff 100644 --- a/finddup.c +++ b/finddup.c @@ -23,7 +23,7 @@ * */ -#define VERSION_NUMBER "0.9" +#define VERSION_NUMBER "1.0" #define _BSD_SOURCE @@ -40,9 +40,6 @@ #include #include #include -#ifdef WITH_MD5 -#include -#endif /* 1M really helps compared to 64k */ #define READ_BUFFER_SIZE (1024 * 1024) @@ -74,10 +71,6 @@ int same_inodes_are_different = 0; /* 1 means that comparison between two files with same inode will always be false */ -#ifdef WITH_MD5 -int use_md5 = 0; /* 1 means we keep an MD5 signature for each file */ -#endif - /********************************************************************/ /* malloc with error checking. */ @@ -109,10 +102,6 @@ struct file_node { ino_t inode; int group_id; /* one per identical file content */ int dir_id; /* 1 for DIR1, and 2 for DIR2 */ -#ifdef WITH_MD5 - int md5_computed; - unsigned char md5[MD5_DIGEST_LENGTH]; -#endif }; void file_list_delete(struct file_node *head) { @@ -140,25 +129,6 @@ int same_content(struct file_node *f1, struct file_node *f2, char *buffer1, char *buffer2) { int fd1, fd2, s1, s2; -#ifdef WITH_MD5 - MD5_CTX c1, c2; - - if(use_md5) { - if(f1->md5_computed && f2->md5_computed) { - if(!memcmp(f1->md5, f2->md5, MD5_DIGEST_LENGTH)) { - return 0; - } - } else { - if(!f1->md5_computed) { - MD5_Init(&c1); - } - if(!f2->md5_computed) { - MD5_Init(&c2); - } - } - } -#endif - fd1 = open(f1->name, O_RDONLY); fd2 = open(f2->name, O_RDONLY); @@ -177,35 +147,14 @@ int same_content(struct file_node *f1, struct file_node *f2, if(s1 == 0) { close(fd1); close(fd2); -#ifdef WITH_MD5 - if(use_md5) { - if(!f1->md5_computed) { - MD5_Final(f1->md5, &c1); - f1->md5_computed = 1; - } - if(!f2->md5_computed) { - MD5_Final(f2->md5, &c2); - f2->md5_computed = 1; - } - } -#endif return 1; } else { if(memcmp(buffer1, buffer2, s1)) { + /* printf("size_to_read = %d\n", size_to_read); */ close(fd1); close(fd2); return 0; } -#ifdef WITH_MD5 - if(use_md5) { - if(!f1->md5_computed) { - MD5_Update(&c1, buffer1, s1); - } - if(!f2->md5_computed) { - MD5_Update(&c2, buffer2, s2); - } - } -#endif } } else { fprintf(stderr, @@ -281,9 +230,6 @@ struct file_node *scan_directory(struct file_node *tail, const char *name) { tmp->inode = sb.st_ino; tmp->group_id = -1; tmp->dir_id = -1; -#ifdef WITH_MD5 - tmp->md5_computed = 0; -#endif tail = tmp; } } @@ -403,8 +349,8 @@ void print_progress(struct progress_state *state) { if(state->bar_width <= 0 || position != state->last_position) { if(!ioctl (STDERR_FILENO, TIOCGWINSZ, (char *) &win)) { /* Something weird is going on if the previous test is wrong */ - if(win.ws_col >= PROGRESS_BUFFER_SIZE) { - state->bar_width = PROGRESS_BUFFER_SIZE - 8; + if(win.ws_col >= PROGRESS_BUFFER_SIZE - 3) { + state->bar_width = PROGRESS_BUFFER_SIZE - 10; } else { state->bar_width = win.ws_col - 7; } @@ -417,7 +363,9 @@ void print_progress(struct progress_state *state) { for(; k < state->bar_width; k++) { *(s++) = '-'; } - sprintf(s, " [%3d%%]\r", + + /* We need four % because of the fprintf that follows */ + sprintf(s, " [%3d%%%%]\r", (100 * state->value) / (state->nb_values - 1)); fprintf(stderr, buffer); @@ -440,13 +388,18 @@ void start(const char *dirname1, const char *dirname2) { not_in = 0; if(show_progress) { - fprintf(stderr, "Scanning %s ... ", dirname1); + fprintf(stderr, "Scanning '%s' ... ", dirname1); } list1 = scan_directory(0, dirname1); list1_length = file_list_length(list1); + if(show_progress) { + fprintf(stderr, "done (%d file%s).\n", + list1_length, (list1_length > 1 ? "s" : "")); + } + if(dirname2) { if(strncmp(dirname2, "not:", 4) == 0) { not_in = 1; @@ -457,25 +410,20 @@ void start(const char *dirname1, const char *dirname2) { dirname2 += 4; } if(show_progress) { - fprintf(stderr, "%s ... ", dirname2); + fprintf(stderr, "Scanning '%s' ... ", dirname2); } list2 = scan_directory(0, dirname2); + if(show_progress) { + list2_length = file_list_length(list2); + fprintf(stderr, "done (%d file%s).\n", + list2_length, (list2_length > 1 ? "s" : "")); + } } else { list2 = list1; } if(show_progress) { - fprintf(stderr, "done.\n"); - fprintf(stderr, - "%s: %d file%s.\n", - dirname1, list1_length, (list1_length > 1 ? "s" : "")); - if(dirname2) { - list2_length = file_list_length(list2); - fprintf(stderr, - "%s: %d file%s.\n", - dirname2, list2_length, (list2_length > 1 ? "s" : "")); - } - fprintf(stderr, "Now looking for identical files.\n"); + fprintf(stderr, "Now looking for identical files (this may take a while).\n"); } nb_groups = 0; @@ -555,9 +503,9 @@ void start(const char *dirname1, const char *dirname2) { } void usage(FILE *out) { - fprintf(out, "Usage: finddup [OPTION]... DIR1 [[and:|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. The and: prefix is the default and should be used only if you have a directory starting with 'not:'\n"); + fprintf(out, "Without DIR2, lists duplicated files found in DIR1, or the current directory if DIR1 is not provided. 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"); /* 01234567890123456789012345678901234567890123456789012345678901234567890123456789*/ fprintf(out, " -h, --help show this help\n"); @@ -570,9 +518,6 @@ void usage(FILE *out) { 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"); -#ifdef WITH_MD5 - fprintf(out, " -m, --md5 use MD5 hashing\n"); -#endif fprintf(out, "\n"); fprintf(out, "Report bugs and comments to .\n"); } @@ -588,7 +533,6 @@ static struct option long_options[] = { { "ignore-dots", no_argument, 0, 'd' }, { "ignore-empty", no_argument, 0, '0' }, { "show-progress", no_argument, 0, 'p' }, - { "md5", no_argument, 0, 'm' }, { 0, 0, 0, 0 } }; @@ -635,17 +579,6 @@ int main(int argc, char **argv) { show_hits = 0; break; - case 'm': -#ifdef WITH_MD5 - use_md5 = 1; -#else - fprintf(stderr, - "finddup has not been compiled with MD5 hashing.\n"); - usage(stderr); - exit(EXIT_FAILURE); -#endif - break; - default: usage(stderr); exit(EXIT_FAILURE); @@ -661,6 +594,9 @@ int main(int argc, char **argv) { } else if(optind + 1 == argc) { same_inodes_are_different = 1; start(argv[optind], 0); + } else if(optind == argc) { + same_inodes_are_different = 1; + start(".", 0); } else { usage(stderr); exit(EXIT_FAILURE);