X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?p=mymail.git;a=blobdiff_plain;f=mymail.c;h=a04fde8eab17dba46399cac7ee8a1fd266628866;hp=0fb694a5d7436137f4e8cf4473998d80bc11a72d;hb=HEAD;hpb=c7dd0337a78674576cbda563bc45c998e82bdbe0 diff --git a/mymail.c b/mymail.c index 0fb694a..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 +49,9 @@ #include #define MYMAIL_DB_MAGIC_TOKEN "mymail_index_file" -#define VERSION "0.9.7" +#define MYMAIL_VERSION "0.9.10" + +#define MYMAIL_DB_FORMAT_VERSION 1 #define MAX_NB_SEARCH_CONDITIONS 32 @@ -55,12 +60,22 @@ #define LEADING_FROM_LINE_REGEXP_STRING "^From .*\\(Mon\\|Tue\\|Wed\\|Thu\\|Fri\\|Sat\\|Sun\\) \\(Jan\\|Feb\\|Mar\\|Apr\\|May\\|Jun\\|Jul\\|Aug\\|Sep\\|Oct\\|Nov\\|Dec\\) [ 0123][0-9] [0-9][0-9]:[0-9][0-9]:[0-9][0-9] [0-9][0-9][0-9][0-9]\n$" +/********************************************************************/ + +struct alias_node { + char *alias, *value; + struct alias_node *next; +}; + /* Global variables! */ 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; /********************************************************************/ @@ -74,6 +89,9 @@ enum { ID_PARTICIPANT, ID_BODY, ID_TIME_INTERVAL, + ID_MAIL_ID, + ID_REFERENCE_ID, + ID_THREAD_ID, MAX_ID }; @@ -86,7 +104,10 @@ static char *field_keys[] = { "date", "part", "body", - "interval" + "interval", + "mailid", + "reference", + "thread" }; /********************************************************************/ @@ -143,6 +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 } + }, + }; /********************************************************************/ @@ -184,14 +219,27 @@ char *default_value(char *current_value, } } +/********************************************************************/ + +void *safe_malloc(size_t n) { + void *p = malloc(n); + if(!p && n != 0) { + fprintf(stderr, + "mymail: cannot allocate memory: %s\n", strerror(errno)); + exit(EXIT_FAILURE); + } + return p; +} + FILE *safe_fopen(const char *path, const char *mode, const char *comment) { FILE *result = fopen(path, mode); if(result) { return result; } else { fprintf(stderr, - "mymail: Cannot open file '%s' (%s) with mode \"%s\".\n", - path, comment, mode); + "mymail: Cannot open file '%s' (%s) with mode \"%s\": %s\n", + path, comment, mode, + strerror(errno)); exit(EXIT_FAILURE); } } @@ -199,7 +247,7 @@ FILE *safe_fopen(const char *path, const char *mode, const char *comment) { /*********************************************************************/ void print_version(FILE *out) { - fprintf(out, "mymail version %s (%s)\n", VERSION, UNAME); + fprintf(out, "mymail version %s (%s)\n", MYMAIL_VERSION, UNAME); } void print_usage(FILE *out) { @@ -217,6 +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"); @@ -227,14 +277,17 @@ void print_usage(FILE *out) { fprintf(out, " set the mbox filename pattern for recursive search\n"); fprintf(out, " -s , --search \n"); fprintf(out, " search for matching mails in the db file\n"); - fprintf(out, " -d , --db-file-generate \n"); + fprintf(out, " -d , --db-file-output \n"); fprintf(out, " set the db filename for indexing\n"); fprintf(out, " -i, --index\n"); fprintf(out, " index mails\n"); fprintf(out, " -o , --output \n"); fprintf(out, " set the result file, use stdout if unset\n"); + fprintf(out, " -n , --nb-mails-max \n"); + 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"); + } /*********************************************************************/ @@ -268,6 +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)) ) && @@ -324,6 +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)) { @@ -394,9 +452,12 @@ void update_time(int db_key, const char *db_value, time_t *t) { const char *c; struct tm tm; + memset(&tm, 0, sizeof(struct tm)); + 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); @@ -414,6 +475,7 @@ void update_time(int db_key, const char *db_value, time_t *t) { } int search_in_db(const char *db_filename, + int nb_extracted_mails, int nb_search_conditions, struct search_condition *search_conditions, FILE *output_file) { @@ -428,13 +490,10 @@ int search_in_db(const char *db_filename, int db_key; int hits[MAX_NB_SEARCH_CONDITIONS]; int nb_body_conditions, need_time; - int nb_extracted_mails; time_t mail_time; int m, n; - nb_extracted_mails = 0; - if(!global_quiet) { printf("Searching in '%s' ... ", db_filename); fflush(stdout); @@ -465,7 +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++) { @@ -479,23 +538,30 @@ int search_in_db(const char *db_filename, strcpy(current_mail_filename, ""); - while(fgets(raw_db_line, BUFFER_SIZE, db_file)) { + while(nb_extracted_mails < global_nb_mails_max && + fgets(raw_db_line, BUFFER_SIZE, db_file)) { + + /* Removes the CR */ + char *s = raw_db_line; + while(*s && *s != '\n') { s++; } + *s = '\0'; + 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; } db_value = parse_token(position_in_file_string, TOKEN_BUFFER_SIZE, ' ', db_value); - db_value = parse_token(current_mail_filename, PATH_MAX+1, '\n', db_value); + strncpy(current_mail_filename, db_value, PATH_MAX + 1); current_position_in_mail = atol(position_in_file_string); } @@ -518,14 +584,18 @@ int search_in_db(const char *db_filename, } } - 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++; - } + /* 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, + nb_body_conditions, hits, current_position_in_mail)) { + extract_mail(current_mail_filename, current_position_in_mail, output_file); + nb_extracted_mails++; } fclose(db_file); @@ -539,6 +609,7 @@ int search_in_db(const char *db_filename, } int recursive_search_in_db(const char *entry_name, regex_t *db_filename_regexp, + int nb_extracted_mails, int nb_search_conditions, struct search_condition *search_conditions, FILE *output_file) { @@ -546,7 +617,6 @@ int recursive_search_in_db(const char *entry_name, regex_t *db_filename_regexp, struct dirent *dir_e; struct stat sb; char subname[PATH_MAX + 1]; - int nb_extracted_mails = 0; if(lstat(entry_name, &sb) != 0) { fprintf(stderr, @@ -561,12 +631,14 @@ int recursive_search_in_db(const char *entry_name, regex_t *db_filename_regexp, dir = opendir(entry_name); if(dir) { - while((dir_e = readdir(dir))) { + while((dir_e = readdir(dir)) && + nb_extracted_mails < global_nb_mails_max) { if(!ignore_entry(dir_e->d_name)) { snprintf(subname, PATH_MAX, "%s/%s", entry_name, dir_e->d_name); - nb_extracted_mails += recursive_search_in_db(subname, db_filename_regexp, - nb_search_conditions, search_conditions, - output_file); + nb_extracted_mails = recursive_search_in_db(subname, db_filename_regexp, + nb_extracted_mails, + nb_search_conditions, search_conditions, + output_file); } } closedir(dir); @@ -577,8 +649,10 @@ int recursive_search_in_db(const char *entry_name, regex_t *db_filename_regexp, while(*s) { if(*s == '/') { filename = s+1; } s++; } if(regexec(db_filename_regexp, filename, 0, 0, 0) == 0) { - nb_extracted_mails += - search_in_db(entry_name, nb_search_conditions, search_conditions, output_file); + nb_extracted_mails = + search_in_db(entry_name, + nb_extracted_mails, + nb_search_conditions, search_conditions, output_file); } } @@ -737,7 +811,8 @@ static struct option long_options[] = { { "version", no_argument, 0, 'v' }, { "quiet", no_argument, 0, 'q' }, { "use-leading-time", no_argument, 0, 't' }, - { "db-file-generate", 1, 0, 'd' }, + { "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' }, { "db-list", 1, 0, 'l' }, @@ -746,6 +821,7 @@ static struct option long_options[] = { { "index", 0, 0, 'i' }, { "output", 1, 0, 'o' }, { "default-search", 1, 0, 'a' }, + { "nb-mails-max", 1, 0, 'n' }, { 0, 0, 0, 0 } }; @@ -760,11 +836,17 @@ 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 }, { "yesterday", 1, -1, -1, -1 }, @@ -783,17 +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, @@ -801,6 +881,14 @@ void init_condition(struct search_condition *condition, const char *full_string, char full_search_field[TOKEN_BUFFER_SIZE], *search_field; unsigned int k, m; const char *string; + struct alias_node *a; + + for(a = global_alias_list; a; a = a->next) { + if(strcmp(full_string, a->alias) == 0) { + full_string = a->value; + break; + } + } string = parse_token(full_search_field, TOKEN_BUFFER_SIZE, ' ', full_string); search_field = full_search_field; @@ -823,9 +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; } @@ -885,6 +973,72 @@ void free_condition(struct search_condition *condition) { } } +const char *eat_space(const char *s) { + while(*s == ' ' || *s == '\t') { s++; } + return s; +} + +void read_rc_file(const char *rc_filename) { + char raw_line[BUFFER_SIZE]; + char command[TOKEN_BUFFER_SIZE], tmp_token[TOKEN_BUFFER_SIZE]; + + FILE *rc_file; + int line_number; + const char *s; + char *t; + + rc_file = fopen(rc_filename, "r"); + + if(rc_file) { + line_number = 1; + while(fgets(raw_line, BUFFER_SIZE, rc_file)) { + t = raw_line; + while(*t) { if(*t == '\n') { *t = '\0'; }; t++; } + + s = raw_line; + s = eat_space(s); + + if(*s && *s != '#') { + s = parse_token(command, TOKEN_BUFFER_SIZE, ' ', s); + + if(strcmp(command, "alias") == 0) { + struct alias_node *a = safe_malloc(sizeof(struct alias_node)); + a->next = global_alias_list; + global_alias_list = a; + if(s) { + s = eat_space(s); + s = parse_token(tmp_token, TOKEN_BUFFER_SIZE, '=', s); + a->alias = strdup(tmp_token); + if(s) { + s = eat_space(s); + a->value = strdup(s); + } else { + fprintf(stderr, "%s:%d syntax error, missing alias value.\n", + rc_filename, + line_number); + exit(EXIT_FAILURE); + } + } else { + fprintf(stderr, "%s:%d syntax error, missing alias key.\n", + rc_filename, + line_number); + exit(EXIT_FAILURE); + } + } else { + fprintf(stderr, "%s:%d syntax error, unknown command '%s'.\n", + rc_filename, + line_number, + command); + exit(EXIT_FAILURE); + } + } + + line_number++; + } + fclose(rc_file); + } +} + /*********************************************************************/ /*********************************************************************/ /*********************************************************************/ @@ -897,6 +1051,7 @@ int main(int argc, char **argv) { char *mbox_filename_regexp_string = 0; char *default_search_field; char output_filename[PATH_MAX + 1]; + char rc_filename[PATH_MAX + 1]; int action_index = 0; int error = 0, show_help = 0; const unsigned int nb_fields_to_parse = @@ -905,6 +1060,10 @@ int main(int argc, char **argv) { unsigned int f, n; unsigned int nb_search_conditions; 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, @@ -912,16 +1071,43 @@ int main(int argc, char **argv) { exit(EXIT_FAILURE); } + if(getenv("MYMAILRC")) { + sprintf(rc_filename, "%s", getenv("MYMAILRC")); + } else if(getenv("HOME")) { + sprintf(rc_filename, "%s/.mymailrc", getenv("HOME")); + } else { + rc_filename[0] = '\0'; + } + + global_alias_list = 0; 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); + if(rc_filename[0]) { + read_rc_file(rc_filename); + } + + /* + { + #warning Test code added on 2013 May 02 11:17:01 + struct alias_node *a; + for(a = global_alias_list; a; a = a->next) { + printf ("ALIAS [%s] [%s]\n", a->alias, a->value); + } + } + */ + setlocale(LC_ALL, ""); 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) { @@ -942,6 +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; @@ -1003,12 +1193,26 @@ int main(int argc, char **argv) { default_search_field = optarg; break; + case 'n': + global_nb_mails_max = atoi(optarg); + break; + default: error = 1; break; } } + 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 */ @@ -1032,18 +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) { @@ -1079,7 +1271,11 @@ int main(int argc, char **argv) { } } - fprintf(db_file, "%s version_%s raw\n", MYMAIL_DB_MAGIC_TOKEN, VERSION); + fprintf(db_file, + "%s version_%s format_%d raw\n", + MYMAIL_DB_MAGIC_TOKEN, + MYMAIL_VERSION, + MYMAIL_DB_FORMAT_VERSION); while(optind < argc) { recursive_index_mbox(db_file, @@ -1129,9 +1325,10 @@ int main(int argc, char **argv) { exit(EXIT_FAILURE); } - nb_extracted_mails += recursive_search_in_db(db_root_path, &db_filename_regexp, - nb_search_conditions, search_conditions, - output_file); + nb_extracted_mails = recursive_search_in_db(db_root_path, &db_filename_regexp, + nb_extracted_mails, + nb_search_conditions, search_conditions, + output_file); regfree(&db_filename_regexp); } @@ -1148,8 +1345,10 @@ int main(int argc, char **argv) { s = parse_token(db_filename, PATH_MAX + 1, ';', s); if(db_filename[0]) { - nb_extracted_mails += - search_in_db(db_filename, nb_search_conditions, search_conditions, output_file); + nb_extracted_mails = + search_in_db(db_filename, + nb_extracted_mails, + nb_search_conditions, search_conditions, output_file); } } } @@ -1157,8 +1356,10 @@ int main(int argc, char **argv) { /* Search in all db files listed in the command arguments */ while(optind < argc) { - nb_extracted_mails += - search_in_db(argv[optind], nb_search_conditions, search_conditions, output_file); + nb_extracted_mails = + search_in_db(argv[optind], + nb_extracted_mails, + nb_search_conditions, search_conditions, output_file); optind++; } } @@ -1182,6 +1383,15 @@ int main(int argc, char **argv) { free_condition(&search_conditions[n]); } + a = global_alias_list; + while(a) { + b = a->next; + free(a->alias); + free(a->value); + free(a); + a = b; + } + free(db_filename); free(db_filename_regexp_string); free(db_root_path);