X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?a=blobdiff_plain;f=mymail.c;h=03ecfb8d7ae61187725bea8c7ba27da5ba3e6662;hb=27dce4e10266dd47c80b9a7ac8d345396b665f4c;hp=27b48f4fe8a5f2f481d46932e4ae4b266eba78be;hpb=b18f24afc58a95f309f33e06ba7b2e67db06bea8;p=mymail.git diff --git a/mymail.c b/mymail.c index 27b48f4..03ecfb8 100644 --- a/mymail.c +++ b/mymail.c @@ -45,14 +45,16 @@ #include #define MYMAIL_DB_MAGIC_TOKEN "mymail_index_file" -#define VERSION "0.9" +#define VERSION "0.9.1" -#define MAX_NB_SEARCH_REQUESTS 10 +#define MAX_NB_SEARCH_CONDITIONS 10 #define BUFFER_SIZE 65536 char *db_filename; +char *db_filename_regexp_string; char *db_root_path; +char *db_filename_list; int paranoid; int action_index; @@ -62,23 +64,27 @@ int action_index; enum { ID_MAIL = 0, ID_FROM, - ID_DEST, + ID_TO, ID_SUBJECT, + ID_DATE, ID_PARTICIPANT, + ID_BODY, MAX_ID }; static char *field_names[] = { "mail", "from", - "dest", + "to", "subject", - "part" + "date", + "part", + "body" }; /********************************************************************/ -struct search_request { +struct search_condition { int field_id; int negation; regex_t regexp; @@ -100,7 +106,7 @@ static struct parsable_field fields_to_parse[] = { }, { - ID_DEST, + ID_TO, "^\\([Tt][Oo]\\|[Cc][Cc]\\|[Bb][Cc][Cc]\\): ", { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } }, @@ -111,6 +117,12 @@ static struct parsable_field fields_to_parse[] = { { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } }, + { + ID_DATE, + "^[Dd][Aa][Tt][Ee]: ", + { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } + }, + }; /********************************************************************/ @@ -155,20 +167,24 @@ void print_usage(FILE *out) { print_version(out); fprintf(out, "Written by Francois Fleuret .\n"); fprintf(out, "\n"); - fprintf(out, "Usage: mymail [options] [ [ ...]]\n"); + fprintf(out, "Usage: mymail [options] [ [ ...]| [ ...]]\n"); fprintf(out, "\n"); fprintf(out, " -h, --help\n"); fprintf(out, " show this help\n"); fprintf(out, " -v, --version\n"); fprintf(out, " print the version number\n"); - fprintf(out, " -i, --index\n"); - fprintf(out, " index mails\n"); + fprintf(out, " -p , --db-pattern \n"); + fprintf(out, " set the db filename pattern for recursive search\n"); + fprintf(out, " -r , --db-root \n"); + fprintf(out, " set the db root path for recursive search\n"); + fprintf(out, " -l , --db-list \n"); + fprintf(out, " set the semicolon-separated list of db files for search\n"); fprintf(out, " -s , --search \n"); - fprintf(out, " search for matching mails in the data-base file\n"); + fprintf(out, " search for matching mails in the db file\n"); fprintf(out, " -d , --db-file \n"); - fprintf(out, " set the data-base file\n"); - fprintf(out, " -r , --db-root \n"); - fprintf(out, " set the data-base root path for recursive search\n"); + fprintf(out, " set the db filename for indexing\n"); + fprintf(out, " -i, --index\n"); + fprintf(out, " index mails\n"); } /*********************************************************************/ @@ -180,19 +196,19 @@ int ignore_entry(const char *name) { (name[0] == '.' && name[1] != '/'); } -int mbox_line_match_search(struct search_request *request, +int mbox_line_match_search(struct search_condition *condition, int mbox_id, char *mbox_value) { return - (request->field_id == mbox_id || - (request->field_id == ID_PARTICIPANT && (mbox_id == ID_FROM || mbox_id == ID_DEST))) + (condition->field_id == mbox_id || + (condition->field_id == ID_PARTICIPANT && (mbox_id == ID_FROM || mbox_id == ID_TO))) && - regexec(&request->regexp, mbox_value, 0, 0, 0) == 0; + regexec(&condition->regexp, mbox_value, 0, 0, 0) == 0; } -void search_in_db(int nb_search_requests, - struct search_request *search_requests, +void search_in_db(int nb_search_conditions, + struct search_condition *search_conditions, FILE *db_file) { - int hits[MAX_NB_SEARCH_REQUESTS]; + int hits[MAX_NB_SEARCH_CONDITIONS]; char raw_db_line[BUFFER_SIZE]; char raw_mbox_line[BUFFER_SIZE]; char current_mail_filename[PATH_MAX + 1]; @@ -201,11 +217,21 @@ void search_in_db(int nb_search_requests, int mbox_id; int already_written, m, n; int last_mbox_line_was_empty; + int nb_body_conditions, nb_fulfilled_body_conditions; current_position_in_mail = 0; already_written = 0; - for(n = 0; n < nb_search_requests; n++) { hits[n] = 0; } + for(n = 0; n < nb_search_conditions; n++) { hits[n] = 0; } + + nb_body_conditions = 0; + for(n = 0; n < nb_search_conditions; n++) { + if(search_conditions[n].field_id == ID_BODY) { + nb_body_conditions++; + } + } + + strcpy(current_mail_filename, ""); while(fgets(raw_db_line, BUFFER_SIZE, db_file)) { mbox_name = raw_db_line; @@ -215,39 +241,96 @@ void search_in_db(int nb_search_requests, char *position_in_file_string; char *mail_filename; - for(n = 0; n < nb_search_requests && xor(hits[n], search_requests[n].negation); n++); + if(current_mail_filename[0]) { - /* for(n = 0; n < nb_search_requests && */ - /* ((hits[n] && !search_requests[n].negation) || */ - /* (!hits[n] && search_requests[n].negation)); n++); */ + /* We first check all conditions but the body ones */ - if(n == nb_search_requests) { - FILE *mail_file; + for(n = 0; n < nb_search_conditions && + ((search_conditions[n].field_id == ID_BODY) || + xor(hits[n], search_conditions[n].negation)); n++); - mail_file = fopen(current_mail_filename, "r"); + if(n == nb_search_conditions) { - if(!mail_file) { - fprintf(stderr, "mymail: Cannot open mbox '%s'.\n", current_mail_filename); - exit(EXIT_FAILURE); - } + /* all conditions but the body ones are fine, check the body + ones */ + + nb_fulfilled_body_conditions = 0; + + if(nb_body_conditions > 0) { + FILE *mail_file; + int header; + + header = 1; + mail_file = fopen(current_mail_filename, "r"); - fseek(mail_file, current_position_in_mail, SEEK_SET); + if(!mail_file) { + fprintf(stderr, + "mymail: Cannot open mbox '%s' for body scan.\n", + current_mail_filename); + exit(EXIT_FAILURE); + } - if(fgets(raw_mbox_line, BUFFER_SIZE, mail_file)) { - last_mbox_line_was_empty = 1; - printf("%s", raw_mbox_line); - while(1) { - if(!fgets(raw_mbox_line, BUFFER_SIZE, mail_file) || - (last_mbox_line_was_empty && strncmp(raw_mbox_line, "From ", 5) == 0)) break; - last_mbox_line_was_empty = (raw_mbox_line[0] == '\n'); - printf("%s", raw_mbox_line); + fseek(mail_file, current_position_in_mail, SEEK_SET); + + if(fgets(raw_mbox_line, BUFFER_SIZE, mail_file)) { + while(nb_fulfilled_body_conditions < nb_body_conditions) { + last_mbox_line_was_empty = (raw_mbox_line[0] == '\n'); + + if(last_mbox_line_was_empty) { header = 0; } + + if(!header) { + for(n = 0; n < nb_search_conditions; n++) { + if(search_conditions[n].field_id == ID_BODY && !hits[n]) { + hits[n] = + (regexec(&search_conditions[n].regexp, raw_mbox_line, 0, 0, 0) == 0); + if(hits[n]) { + nb_fulfilled_body_conditions++; + } + } + } + } + + if(!fgets(raw_mbox_line, BUFFER_SIZE, mail_file) || + (last_mbox_line_was_empty && strncmp(raw_mbox_line, "From ", 5) == 0)) + break; + } + } + + fclose(mail_file); } - } - fclose(mail_file); + if(nb_body_conditions == nb_fulfilled_body_conditions) { + FILE *mail_file; + + mail_file = fopen(current_mail_filename, "r"); + + if(!mail_file) { + fprintf(stderr, + "mymail: Cannot open mbox '%s' for mail extraction.\n", + current_mail_filename); + exit(EXIT_FAILURE); + } + + fseek(mail_file, current_position_in_mail, SEEK_SET); + + if(fgets(raw_mbox_line, BUFFER_SIZE, mail_file)) { + last_mbox_line_was_empty = 1; + printf("%s", raw_mbox_line); + while(1) { + if(!fgets(raw_mbox_line, BUFFER_SIZE, mail_file) || + (last_mbox_line_was_empty && strncmp(raw_mbox_line, "From ", 5) == 0)) + break; + last_mbox_line_was_empty = (raw_mbox_line[0] == '\n'); + printf("%s", raw_mbox_line); + } + } + + fclose(mail_file); + } + } } - for(n = 0; n < nb_search_requests; n++) { hits[n] = 0; } + for(n = 0; n < nb_search_conditions; n++) { hits[n] = 0; } position_in_file_string = mbox_value; mail_filename = segment_next_field(mbox_value); @@ -265,17 +348,17 @@ void search_in_db(int nb_search_requests, mbox_id = m; } } - for(n = 0; n < nb_search_requests; n++) { - hits[n] |= mbox_line_match_search(&search_requests[n], + for(n = 0; n < nb_search_conditions; n++) { + hits[n] |= mbox_line_match_search(&search_conditions[n], mbox_id, mbox_value); } } } } -void recursive_search_in_db(const char *entry_name, - int nb_search_requests, - struct search_request *search_requests) { +void recursive_search_in_db(const char *entry_name, regex_t *db_filename_regexp, + int nb_search_conditions, + struct search_condition *search_conditions) { DIR *dir; struct dirent *dir_e; struct stat sb; @@ -296,17 +379,18 @@ void recursive_search_in_db(const char *entry_name, while((dir_e = readdir(dir))) { if(!ignore_entry(dir_e->d_name)) { snprintf(subname, PATH_MAX, "%s/%s", entry_name, dir_e->d_name); - recursive_search_in_db(subname, - nb_search_requests, - search_requests); + recursive_search_in_db(subname, db_filename_regexp, + nb_search_conditions, search_conditions); } } closedir(dir); - } else { + } + + else { const char *s = entry_name, *filename = entry_name; while(*s) { if(*s == '/') { filename = s+1; } s++; } - if(strcmp(filename, db_filename) == 0) { + if(regexec(db_filename_regexp, filename, 0, 0, 0) == 0) { FILE *db_file = fopen(entry_name, "r"); if(!db_file) { @@ -331,7 +415,7 @@ void recursive_search_in_db(const char *entry_name, exit(EXIT_FAILURE); } - search_in_db(nb_search_requests, search_requests, db_file); + search_in_db(nb_search_conditions, search_conditions, db_file); fclose(db_file); } @@ -486,7 +570,9 @@ static struct option long_options[] = { { "help", no_argument, 0, 'h' }, { "version", no_argument, 0, 'v' }, { "db-file", 1, 0, 'd' }, + { "db-pattern", 1, 0, 'p' }, { "db-root", 1, 0, 'r' }, + { "db-list", 1, 0, 'l' }, { "search", 1, 0, 's' }, { "index", 0, 0, 'i' }, { 0, 0, 0, 0 } @@ -499,8 +585,8 @@ int main(int argc, char **argv) { const int nb_fields_to_parse = sizeof(fields_to_parse) / sizeof(struct parsable_field); char c; int f; - int nb_search_requests; - char *search_request_strings[MAX_NB_SEARCH_REQUESTS]; + int nb_search_conditions; + char *search_condition_strings[MAX_NB_SEARCH_CONDITIONS]; /* for(f = 0; f < argc; f++) { */ /* printf("arg %d \"%s\"\n", f, argv[f]); */ @@ -510,12 +596,13 @@ int main(int argc, char **argv) { action_index = 0; db_filename = 0; db_root_path = 0; + db_filename_list = 0; setlocale(LC_ALL, ""); - nb_search_requests = 0; + nb_search_conditions = 0; - while ((c = getopt_long(argc, argv, "hvip:s:d:r:", + while ((c = getopt_long(argc, argv, "hvip:s:d:r:l:", long_options, NULL)) != -1) { switch(c) { @@ -536,16 +623,24 @@ int main(int argc, char **argv) { db_filename = strdup(optarg); break; + case 'p': + db_filename_regexp_string = strdup(optarg); + break; + case 'r': db_root_path = strdup(optarg); break; + case 'l': + db_filename_list = strdup(optarg); + break; + case 's': - if(nb_search_requests == MAX_NB_SEARCH_REQUESTS) { + if(nb_search_conditions == MAX_NB_SEARCH_CONDITIONS) { fprintf(stderr, "mymail: Too many search patterns.\n"); exit(EXIT_FAILURE); } - search_request_strings[nb_search_requests++] = strdup(optarg); + search_condition_strings[nb_search_conditions++] = strdup(optarg); break; default: @@ -564,6 +659,16 @@ int main(int argc, char **argv) { db_filename = strdup(default_db_filename); } + if(!db_filename_regexp_string) { + char *default_db_filename_regexp_string = getenv("MYMAIL_DB_PATTERN"); + + if(!default_db_filename_regexp_string) { + default_db_filename_regexp_string = "^mymail.db$"; + } + + db_filename_regexp_string = strdup(default_db_filename_regexp_string); + } + if(!db_root_path) { char *default_db_root_path = getenv("MYMAIL_DB_ROOT"); @@ -572,6 +677,14 @@ int main(int argc, char **argv) { } } + if(!db_filename_list) { + char *default_db_filename_list = getenv("MYMAIL_DB_LIST"); + + if(default_db_filename_list) { + db_filename_list = strdup(default_db_filename_list); + } + } + if(error) { print_usage(stderr); exit(EXIT_FAILURE); @@ -625,65 +738,129 @@ int main(int argc, char **argv) { else { - if(!db_root_path) { - fprintf(stderr, - "mymail: db root path is not set\n"); - exit(EXIT_FAILURE); - } - - if(nb_search_requests > 0) { - struct search_request search_requests[MAX_NB_SEARCH_REQUESTS]; + if(nb_search_conditions > 0) { + struct search_condition search_conditions[MAX_NB_SEARCH_CONDITIONS]; char *search_field, *search_regexp_string; int m, n; - for(n = 0; n < nb_search_requests; n++) { - search_field = search_request_strings[n]; - search_regexp_string = segment_next_field(search_request_strings[n]); + for(n = 0; n < nb_search_conditions; n++) { + search_field = search_condition_strings[n]; + search_regexp_string = segment_next_field(search_condition_strings[n]); if(search_field[0] == '!') { search_field++; - search_requests[n].negation = 1; + search_conditions[n].negation = 1; } else { - search_requests[n].negation = 0; + search_conditions[n].negation = 0; } - search_requests[n].field_id = -1; - for(m = 0; (m < MAX_ID) && search_requests[n].field_id == -1; m++) { + search_conditions[n].field_id = -1; + for(m = 0; (m < MAX_ID) && search_conditions[n].field_id == -1; m++) { if(strncmp(field_names[m], search_field, strlen(search_field)) == 0) { - search_requests[n].field_id = m; + search_conditions[n].field_id = m; } } - if(search_requests[n].field_id == -1) { + if(search_conditions[n].field_id == -1) { fprintf(stderr, "mymail: Syntax error in field name \"%s\".\n", search_field); exit(EXIT_FAILURE); } - if(regcomp(&search_requests[n].regexp, + if(regcomp(&search_conditions[n].regexp, search_regexp_string, REG_ICASE)) { fprintf(stderr, "mymail: Syntax error in regexp \"%s\" for field \"%s\".\n", search_regexp_string, - field_names[search_requests[n].field_id]); + field_names[search_conditions[n].field_id]); exit(EXIT_FAILURE); } } - recursive_search_in_db(db_root_path, - nb_search_requests, search_requests); + /* Recursive search if db_root_path is set */ + + if(db_root_path) { + regex_t db_filename_regexp; + if(regcomp(&db_filename_regexp, + db_filename_regexp_string, + 0)) { + fprintf(stderr, + "mymail: Syntax error in regexp \"%s\".\n", + db_filename_regexp_string); + exit(EXIT_FAILURE); + } + + recursive_search_in_db(db_root_path, &db_filename_regexp, + nb_search_conditions, search_conditions); + + regfree(&db_filename_regexp); + } + + /* Search in all db files listed in db_filename_list */ + + if(db_filename_list) { + char db_filename[PATH_MAX + 1]; + char *s, *t; + FILE *db_file; + + s = db_filename_list; + + while(*s) { + t = db_filename; + while(*s == ';') { s++; } + while(*s && *s != ';') { *t++ = *s++; } + *t++ = '\0'; + + if(db_filename[0]) { + db_file = fopen(db_filename, "r"); + + if(!db_file) { + fprintf(stderr, + "mymail: Cannot open \"%s\" for reading: %s\n", + argv[optind], + strerror(errno)); + exit(EXIT_FAILURE); + } + + search_in_db(nb_search_conditions, search_conditions, db_file); + + fclose(db_file); + } + } + } + + /* Search in all db files listed in the command arguments */ + + while(optind < argc) { + FILE *db_file = fopen(argv[optind], "r"); + + if(!db_file) { + fprintf(stderr, + "mymail: Cannot open \"%s\" for reading: %s\n", + argv[optind], + strerror(errno)); + exit(EXIT_FAILURE); + } + + search_in_db(nb_search_conditions, search_conditions, db_file); + + fclose(db_file); + optind++; + } - for(n = 0; n < nb_search_requests; n++) { - regfree(&search_requests[n].regexp); - free(search_request_strings[n]); + for(n = 0; n < nb_search_conditions; n++) { + regfree(&search_conditions[n].regexp); + free(search_condition_strings[n]); } } } free(db_filename); + free(db_filename_regexp_string); free(db_root_path); + free(db_filename_list); exit(EXIT_SUCCESS); }