X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?a=blobdiff_plain;f=finddup.c;h=350afa06586793aa312b096cc82119b9c5213c73;hb=587468f168cc9532c9c8ed3e4c9e5c6ac0fa2e2a;hp=fb8b5d24846309bad5ae68ea7f939bdcc3640efe;hpb=d52ddc7b1fce5529c3987bfa44b8d827c94c4d0e;p=finddup.git diff --git a/finddup.c b/finddup.c index fb8b5d2..350afa0 100644 --- a/finddup.c +++ b/finddup.c @@ -4,7 +4,7 @@ * to several directories, or files present in one directory and not * in another one. * - * Copyright (c) 2010 Francois Fleuret + * Copyright (c) 2010, 2011 Francois Fleuret * Written by Francois Fleuret * * This file is part of finddup. @@ -23,13 +23,14 @@ * */ -#define VERSION_NUMBER "1.1" +#define VERSION_NUMBER "1.2" #define _BSD_SOURCE #include #include #include +#include #include #include #include @@ -74,6 +75,11 @@ int same_inodes_are_different = 0; /* 1 means that comparison between int sort_by_time = 0; /* 1 means to sort files in each group according to the modification time */ +char *command_to_exec = 0; /* the name of the command to exec for each + group of identical files */ + +char *result_file_prefix = 0; /* The prefix to use to write result */ + /********************************************************************/ /* malloc with error checking. */ @@ -267,14 +273,14 @@ struct file_node *scan_directory(struct file_node *tail, const char *name) { return result; } -void print_file(struct file_node *node) { +void print_file(FILE *out, struct file_node *node) { char tmp[PATH_MAX + 1]; if(show_realpaths) { if(realpath(node->name, tmp)) { if(show_groups) { - printf("%d %s\n", node->group_id, tmp); + fprintf(out, "%d %s\n", node->group_id, tmp); } else { - printf("%s\n", tmp); + fprintf(out, "%s\n", tmp); } } else { fprintf(stderr, @@ -285,9 +291,9 @@ void print_file(struct file_node *node) { } } else { if(show_groups) { - printf("%d %s\n", node->group_id, node->name); + fprintf(out, "%d %s\n", node->group_id, node->name); } else { - printf("%s\n", node->name); + fprintf(out, "%s\n", node->name); } } } @@ -323,6 +329,68 @@ int compare_nodes(const void *x1, const void *x2) { } } +void exec_command(int nb, struct file_node **nodes) { + char **args; + int max_group_size = 0, group_size, m, n, status; + pid_t pid; + + for(n = 0; n < nb; n++) { + if(n == 0 || nodes[n]->group_id != nodes[n-1]->group_id) { + group_size = 0; + } + group_size++; + if(group_size > max_group_size) { + max_group_size = group_size; + } + } + + args = safe_malloc((max_group_size + 2) * sizeof(char *)); + args[0] = command_to_exec; + + n = 0; + while(n < nb) { + m = n; + while(n < nb && nodes[n]->group_id == nodes[m]->group_id) { + args[n - m + 1] = nodes[n]->name; + n++; + } + args[n - m + 1] = 0; + pid = fork(); + if(pid < 0) { + } else if(pid == 0) { + if(execvp(command_to_exec, args) < 0) { + perror("execvp"); + exit(EXIT_FAILURE); + } + } else { + while(wait(&status) != pid); + if(status > 0) { + exit(EXIT_FAILURE); + } + } + } + + free(args); +} + +void write_groups_in_files(int nb, struct file_node **nodes) { + FILE *file = 0; + int current_group = -1, n; + char filename[PATH_MAX + 1]; + + for(n = 0; n < nb; n++) { + if(nodes[n]->group_id != current_group) { + if(file) { fclose(file); } + sprintf(filename, "%s%06d", result_file_prefix, nodes[n]->group_id); + file = fopen(filename, "w"); + current_group = nodes[n]->group_id; + printf("Writing %s.\n" , filename); + } + print_file(file, nodes[n]); + } + + if(file) { fclose(file); } +} void print_result(struct file_node *list1, struct file_node *list2) { struct file_node *node1, *node2; @@ -359,11 +427,17 @@ void print_result(struct file_node *list1, struct file_node *list2) { 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) { - printf("\n"); + if(command_to_exec) { + exec_command(nb, nodes); + } else if(result_file_prefix) { + write_groups_in_files(nb, nodes); + } else { + for(n = 0; n < nb; n++) { + if(!show_groups && n > 0 && nodes[n]->group_id != nodes[n-1]->group_id) { + printf("\n"); + } + print_file(stdout, nodes[n]); } - print_file(nodes[n]); } free(nodes); @@ -406,10 +480,10 @@ void print_progress(struct progress_state *state) { } /* We need four % because of the fprintf that follows */ - sprintf(s, " [%3d%%%%]\r", + sprintf(s, " [%3d%%]\r", (100 * state->value) / normalizer); - fprintf(stderr, buffer); + fprintf(stderr, "%s", buffer); } } } @@ -532,6 +606,7 @@ void usage(FILE *out) { 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, " -v, --version prints the version number and exit\n"); 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"); @@ -543,6 +618,14 @@ 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"); + fprintf(out, " -e , --exec \n"); + fprintf(out, " execute the provided command for each group of\n"); + fprintf(out, " identical files, with their names as arguments\n"); + fprintf(out, " -f , --result-prefix \n"); + fprintf(out, " for each group of identical files, write one\n"); + fprintf(out, " result file whose name is the given prefix string\n"); + fprintf(out, " followed the group number, and containing one\n"); + fprintf(out, " filename per line\n"); fprintf(out, "\n"); fprintf(out, "Report bugs and comments to .\n"); } @@ -550,6 +633,7 @@ void usage(FILE *out) { /**********************************************************************/ static struct option long_options[] = { + { "version", no_argument, 0, 'v' }, { "help", no_argument, 0, 'h' }, { "same-inodes-are-different", no_argument, 0, 'i' }, { "real-paths", no_argument, 0, 'r' }, @@ -559,6 +643,8 @@ static struct option long_options[] = { { "ignore-dots", no_argument, 0, 'd' }, { "ignore-empty", no_argument, 0, '0' }, { "show-progress", no_argument, 0, 'p' }, + { "exec", 1, 0, 'e' }, + { "result-prefix", 1, 0, 'f' }, { 0, 0, 0, 0 } }; @@ -567,10 +653,15 @@ int main(int argc, char **argv) { setlocale (LC_ALL, ""); - while ((c = getopt_long(argc, argv, "hircgtd0pm", + while ((c = getopt_long(argc, argv, "vhircgtd0pme:f:", long_options, NULL)) != -1) { switch (c) { + case 'v': + printf("finddup version %s (%s)\n", VERSION_NUMBER, UNAME); + exit(EXIT_SUCCESS); + break; + case 'h': usage(stdout); exit(EXIT_SUCCESS); @@ -609,6 +700,21 @@ int main(int argc, char **argv) { show_hits = 0; break; + case 'e': + if(command_to_exec != 0) { + free(command_to_exec); + } + command_to_exec = strdup(optarg); + break; + + case 'f': + if(result_file_prefix != 0) { + free(result_file_prefix); + } + result_file_prefix = strdup(optarg); + show_groups = 0; + break; + default: usage(stderr); exit(EXIT_FAILURE);