X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?a=blobdiff_plain;f=finddup.c;h=688fb02a30a58742b6eff8b303a9f842202240cd;hb=9b4c525f5b820a47d15f99bcd07ca83a1847ee89;hp=150279166903355e63d76c9f95a4e2a115aedb66;hpb=5d880a48149f85f6d5ac7942a696248eac0b8acc;p=finddup.git diff --git a/finddup.c b/finddup.c index 1502791..688fb02 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 @@ -41,8 +41,8 @@ #include #include -#define BUFFER_SIZE 4096 -#define LARGE_BUFFER_SIZE 65536 +/* 1M really helps compared to 64k */ +#define READ_BUFFER_SIZE (1024 * 1024) typedef int64_t size_sum_t; @@ -68,6 +68,9 @@ int show_groups = 1; /* 1 means to show the group IDs when printing int ignore_same_inode = 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 */ + /********************************************************************/ /* malloc with error checking. */ @@ -136,17 +139,17 @@ 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_with_size *f1, struct file_with_size *f2, + char *buffer1, char *buffer2) { int fd1, fd2, s1, s2; - 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, LARGE_BUFFER_SIZE); - s2 = read(fd2, buffer2, LARGE_BUFFER_SIZE); + s1 = read(fd1, buffer1, READ_BUFFER_SIZE); + s2 = read(fd2, buffer2, READ_BUFFER_SIZE); if(s1 < 0 || s2 < 0) { close(fd1); @@ -191,11 +194,12 @@ 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) { +int same_files(struct file_with_size *f1, struct file_with_size *f2, + char *buffer1, char *buffer2) { if(ignore_same_inode && f1->inode == f2->inode) { return 0; } - return f1->size == f2->size && same_content(f1, f2); + return f1->size == f2->size && same_content(f1, f2, buffer1, buffer2); } /**********************************************************************/ @@ -333,11 +337,34 @@ void print_result(struct file_with_size *list1, struct file_with_size *list2) { free(nodes); } +void print_progress(int max, int n, int *pp) { + int p, k; + int width; + if(show_progress && tty_width > 0) { + width = tty_width - 7; + p = (width * n) / (max - 1); + if(p > *pp) { + for(k = 0; k < p; k++) { + fprintf(stderr, "+"); + } + for(; k < width; k++) { + fprintf(stderr, "-"); + } + *pp = 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; int not_in, found; - int k, p, pp, l1, n; + int k, pp, l1, n; + + char *buffer1 = safe_malloc(sizeof(char) * READ_BUFFER_SIZE); + char *buffer2 = safe_malloc(sizeof(char) * READ_BUFFER_SIZE); not_in = 0; @@ -350,6 +377,8 @@ void start(const char *dirname1, const char *dirname2) { if(dirname2) { if(strncmp(dirname2, "not:", 4) == 0) { not_in = 1; + /* groups are not computed in the not: mode */ + show_groups = 0; dirname2 += 4; } if(show_progress) { @@ -371,19 +400,13 @@ void start(const char *dirname1, const char *dirname2) { 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++; - } + print_progress(l1, n, &pp); + n++; found = 0; for(node2 = list2; !found && node2; node2 = node2->next) { - if(same_files(node1, node2)) { + if(same_files(node1, node2, buffer1, buffer2)) { found = 1; } } @@ -399,18 +422,12 @@ void start(const char *dirname1, const char *dirname2) { } else { 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++; - } + print_progress(l1, n, &pp); + n++; for(node2 = list2; node2; node2 = node2->next) { if(node1->group_id < 0 || node2->group_id < 0) { - if(same_files(node1, node2)) { + if(same_files(node1, node2, buffer1, buffer2)) { if(node1->group_id < 0) { if(node2->group_id >= 0) { node1->group_id = node2->group_id; @@ -430,6 +447,10 @@ void start(const char *dirname1, const char *dirname2) { } } + if(show_progress) { + fprintf(stderr, "\n"); + } + if(dirname2) { print_result(list1, list2); file_list_delete(list1); @@ -438,6 +459,9 @@ void start(const char *dirname1, const char *dirname2) { print_result(list1, 0); file_list_delete(list1); } + + free(buffer1); + free(buffer2); } void print_help(FILE *out) { @@ -462,6 +486,7 @@ void print_help(FILE *out) { int main(int argc, char **argv) { int c; struct file_with_size *root; + struct winsize win; root = 0; @@ -513,6 +538,12 @@ 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) {