X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?p=mymail.git;a=blobdiff_plain;f=mymail.c;h=a04fde8eab17dba46399cac7ee8a1fd266628866;hp=fd3b4f5c88d80a28b7b9aa99792494bd279cd65e;hb=HEAD;hpb=a5c88357a01aeb45d646f0e70891fbed394e5344 diff --git a/mymail.c b/mymail.c index fd3b4f5..a04fde8 100644 --- a/mymail.c +++ b/mymail.c @@ -21,11 +21,13 @@ /* - This command is a dumb mail indexer. It can either (1) scan - directories containing mbox files, and create a db file containing - for each mail a list of fields computed from the header, or (2) - read such a db file and get all the mails matching regexp-defined - conditions on the fields, to create a resulting mbox file. + mymail is a simple mail indexer. It can: + + (1) scan mbox files, and create a db file containing for each mail a + list of fields computed from its header. + + (2) read such a db file, gets all the mails matching regexp-defined + conditions on the fields, and generates a resulting mbox file. It is low-tech, simple, light and fast. @@ -35,6 +37,7 @@ #include #include +#include #include #include #include @@ -46,7 +49,7 @@ #include #define MYMAIL_DB_MAGIC_TOKEN "mymail_index_file" -#define MYMAIL_VERSION "0.9.9" +#define MYMAIL_VERSION "0.9.10" #define MYMAIL_DB_FORMAT_VERSION 1 @@ -69,8 +72,10 @@ struct alias_node { int global_quiet; int global_use_leading_time; int global_nb_mails_max; +int global_discard_mail_from_the_future; regex_t global_leading_from_line_regexp; struct alias_node *global_alias_list; +time_t global_current_time; /********************************************************************/ @@ -84,6 +89,9 @@ enum { ID_PARTICIPANT, ID_BODY, ID_TIME_INTERVAL, + ID_MAIL_ID, + ID_REFERENCE_ID, + ID_THREAD_ID, MAX_ID }; @@ -96,7 +104,10 @@ static char *field_keys[] = { "date", "part", "body", - "interval" + "interval", + "mailid", + "reference", + "thread" }; /********************************************************************/ @@ -153,6 +164,20 @@ static struct parsable_field fields_to_parse[] = { { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } }, + { + ID_MAIL_ID, + REG_ICASE, + "^message-id: ", + { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } + }, + + { + ID_REFERENCE_ID, + REG_ICASE, + "^\\(in-reply-to\\|references\\): ", + { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } + }, + }; /********************************************************************/ @@ -240,6 +265,8 @@ void print_usage(FILE *out) { fprintf(out, " -t, --use-leading-time\n"); fprintf(out, " use the time stamp from the leading line of each mail and not the Date:\n"); fprintf(out, " field\n"); + fprintf(out, " -f, --do-not-discard-mails-from-the-future\n"); + fprintf(out, " do not ignore mails with a date more than 24h in the future\n"); fprintf(out, " -p , --db-pattern \n"); fprintf(out, " set the db filename pattern for recursive search\n"); fprintf(out, " -r , --db-root \n"); @@ -260,6 +287,7 @@ void print_usage(FILE *out) { fprintf(out, " set the maximum number of mails to extract\n"); fprintf(out, " -a , --default-search \n"); fprintf(out, " set the default search field\n"); + } /*********************************************************************/ @@ -293,6 +321,10 @@ int db_line_match_search(struct search_condition *condition, (condition->db_key == ID_FROM && db_key == ID_LEADING_LINE) + || + + (condition->db_key == ID_THREAD_ID && (db_key == ID_MAIL_ID || + db_key == ID_REFERENCE_ID)) ) && @@ -349,6 +381,7 @@ void extract_mail(const char *mail_filename, unsigned long int position_in_mail, /* printf("Extract\n"); */ mail_file = safe_fopen(mail_filename, "r", "mbox for mail extraction"); + /* fchmod(fileno(mail_file), 0x660); */ fseek(mail_file, position_in_mail, SEEK_SET); if(fgets(raw_mbox_line, BUFFER_SIZE, mail_file)) { @@ -423,7 +456,8 @@ void update_time(int db_key, const char *db_value, time_t *t) { if(db_key == ID_LEADING_LINE) { c = db_value; - while(*c && *c != ' ') c++; while(*c && *c == ' ') c++; + while(*c && *c != ' ') c++; + while(*c && *c == ' ') c++; /* printf("From %s", db_value); */ strptime(c, "%a %b %e %k:%M:%S %Y", &tm); *t = mktime(&tm); @@ -490,7 +524,7 @@ int search_in_db(const char *db_filename, for(n = 0; n < nb_search_conditions; n++) { hits[n] = 0; } nb_body_conditions = 0; - need_time = 0; + need_time = global_discard_mail_from_the_future; mail_time = 0; for(n = 0; n < nb_search_conditions; n++) { @@ -515,14 +549,14 @@ int search_in_db(const char *db_filename, db_value = parse_token(db_key_string, TOKEN_BUFFER_SIZE, ' ', raw_db_line); if(strcmp("mail", db_key_string) == 0) { - if(current_mail_filename[0]) { - if(check_full_mail_match(current_mail_filename, - mail_time, - nb_search_conditions, search_conditions, - nb_body_conditions, hits, current_position_in_mail)) { - extract_mail(current_mail_filename, current_position_in_mail, output_file); - nb_extracted_mails++; - } + if(current_mail_filename[0] && + (!global_discard_mail_from_the_future || mail_time < global_current_time + 3600 * 24) && + check_full_mail_match(current_mail_filename, + mail_time, + nb_search_conditions, search_conditions, + nb_body_conditions, hits, current_position_in_mail)) { + extract_mail(current_mail_filename, current_position_in_mail, output_file); + nb_extracted_mails++; } for(n = 0; n < nb_search_conditions; n++) { hits[n] = 0; } @@ -550,8 +584,12 @@ int search_in_db(const char *db_filename, } } + /* printf("global_discard_mail_from_the_future = %d\n", + global_discard_mail_from_the_future); */ + if(nb_extracted_mails < global_nb_mails_max && current_mail_filename[0] && + (!global_discard_mail_from_the_future || mail_time < global_current_time + 3600 * 24) && check_full_mail_match(current_mail_filename, mail_time, nb_search_conditions, search_conditions, @@ -773,6 +811,7 @@ static struct option long_options[] = { { "version", no_argument, 0, 'v' }, { "quiet", no_argument, 0, 'q' }, { "use-leading-time", no_argument, 0, 't' }, + { "do-not-discard-mails-from-the-future", no_argument, 0, 'f' }, { "db-file-output", 1, 0, 'd' }, { "db-pattern", 1, 0, 'p' }, { "db-root", 1, 0, 'r' }, @@ -797,11 +836,16 @@ struct time_criterion { static struct time_criterion time_criteria[] = { + { "1h", 0, 1, -1, -1 }, + { "2h", 0, 2, -1, -1 }, + { "4h", 0, 4, -1, -1 }, { "8h", 0, 8, -1, -1 }, { "24h", 0, 24, -1, -1 }, { "48h", 0, 48, -1, -1 }, { "week", 0, 24 * 7, -1, -1 }, + { "2weeks", 0, 24 * 14, -1, -1 }, { "month", 0, 24 * 31, -1, -1 }, + { "semester", 0, 24 * 185, -1, -1 }, { "trimester", 0, 24 * 92, -1, -1 }, { "year", 0, 24 * 365, -1, -1 }, @@ -821,17 +865,15 @@ static struct time_criterion time_criteria[] = { /*********************************************************************/ time_t time_for_past_day(int day) { - time_t t; struct tm *tm; int delta_day; - t = time(0); - tm = localtime(&t); + tm = localtime(&global_current_time); if(day > 0) { delta_day = (7 + tm->tm_wday - day) % 7; } else { delta_day = - day; } - return t - (delta_day * 3600 * 24 + tm->tm_sec + 60 * tm->tm_min + 3600 * tm->tm_hour); + return global_current_time - (delta_day * 3600 * 24 + tm->tm_sec + 60 * tm->tm_min + 3600 * tm->tm_hour); } void init_condition(struct search_condition *condition, const char *full_string, @@ -869,9 +911,9 @@ void init_condition(struct search_condition *condition, const char *full_string, condition->time_start = time_for_past_day(time_criteria[k].past_week_day); condition->time_stop = condition->time_start + 3600 * 24; } else { - condition->time_start = time(0) - 3600 * time_criteria[k].start_hour; + condition->time_start = global_current_time - 3600 * time_criteria[k].start_hour; if(time_criteria[k].end_hour >= 0) { - condition->time_stop = time(0) - 3600 * time_criteria[k].end_hour; + condition->time_stop = global_current_time - 3600 * time_criteria[k].end_hour; } else { condition->time_stop = 0; } @@ -1020,6 +1062,9 @@ int main(int argc, char **argv) { struct search_condition search_conditions[MAX_NB_SEARCH_CONDITIONS]; struct alias_node *a, *b; + /* Group and others have no access */ + umask(S_IRWXG | S_IRWXO); + if(regcomp(&global_leading_from_line_regexp, LEADING_FROM_LINE_REGEXP_STRING, 0)) { fprintf(stderr, "mymail: Cannot compile leading \"from\" line regexp. That is strange.\n"); @@ -1038,6 +1083,8 @@ int main(int argc, char **argv) { global_quiet = 0; global_use_leading_time = 0; global_nb_mails_max = 250; + global_discard_mail_from_the_future = 1; + global_current_time = time(0); default_search_field = 0; strncpy(output_filename, "", PATH_MAX); @@ -1060,7 +1107,7 @@ int main(int argc, char **argv) { nb_search_conditions = 0; - while ((c = getopt_long(argc, argv, "hvqip:s:d:r:l:o:a:m:", + while ((c = getopt_long(argc, argv, "hvqtfip:s:d:r:l:o:a:m:", long_options, NULL)) != -1) { switch(c) { @@ -1081,6 +1128,10 @@ int main(int argc, char **argv) { global_use_leading_time = 1; break; + case 'f': + global_discard_mail_from_the_future = 0; + break; + case 'i': action_index = 1; break; @@ -1152,6 +1203,16 @@ int main(int argc, char **argv) { } } + if(error) { + print_usage(stderr); + exit(EXIT_FAILURE); + } + + if(show_help) { + print_usage(stdout); + exit(EXIT_SUCCESS); + } + /* Set all the values that may defined in the arguments, through environment variables, or hard-coded */ @@ -1175,18 +1236,6 @@ int main(int argc, char **argv) { "MYMAIL_MBOX_PATTERN", 0); - /* Start the processing */ - - if(error) { - print_usage(stderr); - exit(EXIT_FAILURE); - } - - if(show_help) { - print_usage(stdout); - exit(EXIT_SUCCESS); - } - /* mbox indexing */ if(action_index) {