X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?p=finddup.git;a=blobdiff_plain;f=finddup.c;h=7f223136d4f3e06904e5669ffa1546c3041a6118;hp=c0a51a1bd4826f360f4ad1e88b2bec4aea108007;hb=HEAD;hpb=069e440749c66d6fd7a9aa89ef7c31df1b3f34ea diff --git a/finddup.c b/finddup.c index c0a51a1..7d15689 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. @@ -23,9 +23,9 @@ * */ -#define VERSION_NUMBER "1.1" +#define VERSION_NUMBER "1.2.1" -#define _BSD_SOURCE +#define _DEFAULT_SOURCE #include #include @@ -56,8 +56,8 @@ int ignore_dotfiles = 0; /* 1 means ignore files and directories int ignore_empty_files = 0; /* 1 means ignore empty files */ -int show_realpaths = 0; /* 1 means ignore files and directories - starting with a dot */ +int show_realpaths = 0; /* 1 show the canonical absolute pathname for + printed files */ int show_progress = 0; /* 1 means show a progress bar when we are in a tty */ @@ -75,9 +75,13 @@ 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 */ +int trim_first = 0; /* remove the first entry in each group */ + 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. */ @@ -271,14 +275,14 @@ struct file_node *scan_directory(struct file_node *tail, const char *name) { return result; } -void print_file(struct file_node *node) { +void write_one_entry_to_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, @@ -289,9 +293,9 @@ 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); } } } @@ -329,11 +333,11 @@ 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; + int max_group_size = 0, group_size = 0, m, n, status; pid_t pid; for(n = 0; n < nb; n++) { - if(n == 0 || nodes[n]->group_id != nodes[n-1]->group_id) { + if(n > 0 && nodes[n]->group_id != nodes[n-1]->group_id) { group_size = 0; } group_size++; @@ -348,8 +352,11 @@ void exec_command(int nb, struct file_node **nodes) { n = 0; while(n < nb) { m = n; + if(trim_first) { m++; } while(n < nb && nodes[n]->group_id == nodes[m]->group_id) { - args[n - m + 1] = nodes[n]->name; + if(n >= m) { + args[n - m + 1] = nodes[n]->name; + } n++; } args[n - m + 1] = 0; @@ -371,10 +378,34 @@ void exec_command(int nb, struct file_node **nodes) { free(args); } +void write_groups_in_files(int nb, struct file_node **nodes) { + FILE *file = 0; + int current_group = -1, n, first_of_group; + char filename[PATH_MAX + 1]; + + for(n = 0; n < nb; n++) { + first_of_group = (n == 0); + 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); + first_of_group = 1; + } + + if(!trim_first || !first_of_group) { + write_one_entry_to_file(file, nodes[n]); + } + } + + if(file) { fclose(file); } +} + void print_result(struct file_node *list1, struct file_node *list2) { struct file_node *node1, *node2; struct file_node **nodes; - int nb, n; + int nb, n, first_of_group; nb = 0; for(node1 = list1; node1; node1 = node1->next) { @@ -408,12 +439,20 @@ void print_result(struct file_node *list1, struct file_node *list2) { 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"); + first_of_group = (n == 0); + if(n > 0 && nodes[n]->group_id != nodes[n-1]->group_id) { + if(!show_groups) { + printf("\n"); + } + first_of_group = 1; + } + if(!trim_first || !first_of_group) { + write_one_entry_to_file(stdout, nodes[n]); } - print_file(nodes[n]); } } @@ -457,10 +496,10 @@ 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); } } } @@ -472,7 +511,7 @@ void start(const char *dirname1, const char *dirname2) { struct progress_state progress_state; int not_in, found; int nb_groups, nb_nodes; - int list1_length, previous_progress; + int list1_length; char *buffer1 = safe_malloc(sizeof(char) * READ_BUFFER_SIZE); char *buffer2 = safe_malloc(sizeof(char) * READ_BUFFER_SIZE); @@ -502,7 +541,6 @@ void start(const char *dirname1, const char *dirname2) { } nb_groups = 0; - previous_progress = -1; nb_nodes = 0; progress_state.bar_width = -1; @@ -587,10 +625,11 @@ void usage(FILE *out) { 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, " -c, --hide-matchings do not show which files in DIR2 correspond to\n"); fprintf(out, " those in DIR1\n"); fprintf(out, " -g, --no-group-ids do not show the file groups\n"); fprintf(out, " -t, --time-sort sort according to modification time in each group\n"); + fprintf(out, " -q, --trim-first do not show the first file in each group\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"); @@ -598,6 +637,11 @@ void usage(FILE *out) { 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 by the group number, and containing\n"); + fprintf(out, " one filename per line\n"); fprintf(out, "\n"); fprintf(out, "Report bugs and comments to .\n"); } @@ -612,10 +656,12 @@ static struct option long_options[] = { { "hide-matchings", no_argument, 0, 'c' }, { "no-group-ids", no_argument, 0, 'g' }, { "time-sort", no_argument, 0, 't' }, + { "trim-first", no_argument, 0, 'q' }, { "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 } }; @@ -624,7 +670,7 @@ int main(int argc, char **argv) { setlocale (LC_ALL, ""); - while ((c = getopt_long(argc, argv, "vhircgtd0pme:", + while ((c = getopt_long(argc, argv, "vhircgtqd0pme:f:", long_options, NULL)) != -1) { switch (c) { @@ -663,6 +709,10 @@ int main(int argc, char **argv) { sort_by_time = 1; break; + case 'q': + trim_first = 1; + break; + case 'p': show_progress = 1; break; @@ -678,6 +728,14 @@ int main(int argc, char **argv) { 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);