X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?a=blobdiff_plain;f=finddup.c;h=98008f4a631f3123059218ed46e5b1849f8d977b;hb=46181230584702e9d59e54890ee761d9373c4dd3;hp=ccf21c6ba572537e80418cd6651692afe7feee90;hpb=a61c9478f31b957e0d4007df9feddd6f0139ccf8;p=finddup.git diff --git a/finddup.c b/finddup.c index ccf21c6..98008f4 100644 --- a/finddup.c +++ b/finddup.c @@ -40,7 +40,9 @@ #include #include #include +#ifdef WITH_MD5 #include +#endif /* 1M really helps compared to 64k */ #define READ_BUFFER_SIZE (1024 * 1024) @@ -66,13 +68,16 @@ int show_hits = 1; /* 1 means to show the files from dir2 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 */ +int same_inodes_are_different = 0; /* 1 means that comparison between + two file with same inode will + always be false */ int tty_width = -1; /* Positive value means what width to use to show the progress bar */ +#ifdef WITH_MD5 int use_md5 = 0; /* 1 means we keep an MD5 signature for each file */ +#endif /********************************************************************/ @@ -105,8 +110,10 @@ 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) { @@ -133,12 +140,15 @@ int file_list_length(struct file_node *head) { 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 && - !memcmp(f1->md5, f2->md5, MD5_DIGEST_LENGTH)) { - return 0; + 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); @@ -148,6 +158,7 @@ int same_content(struct file_node *f1, struct file_node *f2, } } } +#endif fd1 = open(f1->name, O_RDONLY); fd2 = open(f2->name, O_RDONLY); @@ -167,6 +178,7 @@ 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); @@ -177,8 +189,15 @@ int same_content(struct file_node *f1, struct file_node *f2, f2->md5_computed = 1; } } +#endif return 1; } else { + if(memcmp(buffer1, buffer2, s1)) { + close(fd1); + close(fd2); + return 0; + } +#ifdef WITH_MD5 if(use_md5) { if(!f1->md5_computed) { MD5_Update(&c1, buffer1, s1); @@ -187,11 +206,7 @@ int same_content(struct file_node *f1, struct file_node *f2, MD5_Update(&c2, buffer2, s2); } } - if(memcmp(buffer1, buffer2, s1)) { - close(fd1); - close(fd2); - return 0; - } +#endif } } else { fprintf(stderr, @@ -221,9 +236,10 @@ int same_content(struct file_node *f1, struct file_node *f2, int same_files(struct file_node *f1, struct file_node *f2, char *buffer1, char *buffer2) { - if(ignore_same_inode && f1->inode == f2->inode) { + if(same_inodes_are_different && f1->inode == f2->inode) { return 0; } + return f1->size == f2->size && same_content(f1, f2, buffer1, buffer2); } @@ -266,7 +282,9 @@ 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; } } @@ -502,7 +520,7 @@ void start(const char *dirname1, const char *dirname2) { free(buffer2); } -void print_help(FILE *out) { +void usage(FILE *out) { 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"); @@ -518,7 +536,9 @@ void print_help(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"); } @@ -548,7 +568,7 @@ int main(int argc, char **argv) { switch (c) { case 'h': - print_help(stdout); + usage(stdout); exit(EXIT_SUCCESS); break; @@ -566,7 +586,7 @@ int main(int argc, char **argv) { break; case 'i': - ignore_same_inode = 1; + same_inodes_are_different = 1; break; case 'g': @@ -582,10 +602,18 @@ int main(int argc, char **argv) { 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); } } @@ -593,10 +621,10 @@ int main(int argc, char **argv) { if(optind + 2 == argc) { start(argv[optind], argv[optind + 1]); } else if(optind + 1 == argc) { - ignore_same_inode = 1; + same_inodes_are_different = 1; start(argv[optind], 0); } else { - print_help(stderr); + usage(stderr); exit(EXIT_FAILURE); }