X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?a=blobdiff_plain;f=mymail.c;h=2617d02a71c11603c5d0af165e8053aa0822c682;hb=17557f7799a9b15adab76086f1f22481f1d10bd1;hp=5b821bc56f859c722037fe7688edae953697dcd5;hpb=234fd9058ed477559401bd581d7dbf78ce08e3a5;p=mymail.git diff --git a/mymail.c b/mymail.c index 5b821bc..2617d02 100644 --- a/mymail.c +++ b/mymail.c @@ -21,13 +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. + 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. - It is low-tech, simple, light and fast. + It is low-tech, simple, light and fast. */ @@ -46,7 +46,7 @@ #define VERSION "0.1" -#define BUFFER_SIZE 16384 +#define BUFFER_SIZE 65536 struct parsable_field { char *name; @@ -88,12 +88,26 @@ void *safe_malloc(size_t n) { /*********************************************************************/ -void usage(FILE *out) { +void print_version(FILE *out) { fprintf(out, "mymail version %s (%s)\n", VERSION, UNAME); +} + +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, "\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, " -d , --db-file \n"); + fprintf(out, " set the data-base file\n"); + fprintf(out, " -s , --search \n"); + fprintf(out, " search for matching mails in the data-base file\n"); } /*********************************************************************/ @@ -134,11 +148,11 @@ void search_in_db(const char *search_name, const char *search_regexp_string, } else if(!already_written) { - if(strcmp(search_name, name) == 0 && regexec(®exp, value, 0, 0, 0) == 0) { + if(strcmp(search_name, name) == 0 && regexec(®exp, value, 0, 0, 0) == 0) { FILE *mail_file; mail_file = fopen(current_mail_filename, "r"); if(!mail_file) { - fprintf(stderr, "mymail: Can not open `%s'.\n", current_mail_filename); + fprintf(stderr, "mymail: Can not open '%s'.\n", current_mail_filename); exit(EXIT_FAILURE); } fseek(mail_file, current_position_in_mail, SEEK_SET); @@ -160,10 +174,24 @@ void search_in_db(const char *search_name, const char *search_regexp_string, /*********************************************************************/ -void read_file(const char *input_filename, - int nb_fields_to_parse, struct parsable_field *fields_to_parse, - FILE *db_file) { - char raw_line[BUFFER_SIZE]; +void index_one_mbox_line(int nb_fields_to_parse, struct parsable_field *fields_to_parse, + char *raw_line, FILE *db_file) { + regmatch_t matches; + int f; + for(f = 0; f < nb_fields_to_parse; f++) { + if(regexec(&fields_to_parse[f].regexp, raw_line, 1, &matches, 0) == 0) { + fprintf(db_file, "%s %s\n", + fields_to_parse[f].name, + raw_line + matches.rm_eo); + } + } +} + +void index_mbox(const char *input_filename, + int nb_fields_to_parse, struct parsable_field *fields_to_parse, + FILE *db_file) { + char raw_line[BUFFER_SIZE], full_line[BUFFER_SIZE]; + char *end_of_full_line; FILE *file; int in_header, new_header; unsigned long int position_in_file; @@ -171,7 +199,7 @@ void read_file(const char *input_filename, file = fopen(input_filename, "r"); if(!file) { - fprintf(stderr, "mymail: Can not open `%s'.\n", input_filename); + fprintf(stderr, "mymail: Can not open '%s'.\n", input_filename); if(paranoid) { exit(EXIT_FAILURE); } return; } @@ -180,6 +208,8 @@ void read_file(const char *input_filename, new_header = 0; position_in_file = 0; + end_of_full_line = 0; + full_line[0] = '\0'; while(fgets(raw_line, BUFFER_SIZE, file)) { if(strncmp(raw_line, "From ", 5) == 0) { @@ -196,20 +226,46 @@ void read_file(const char *input_filename, if(in_header) { in_header = 0; } } + /* printf("PARSE %d %s", in_header, raw_line); */ + if(in_header) { - int f; - regmatch_t matches; if(new_header) { fprintf(db_file, "mail %lu %s\n", position_in_file, input_filename); new_header = 0; } - for(f = 0; f < nb_fields_to_parse; f++) { - if(regexec(&fields_to_parse[f].regexp, raw_line, 1, &matches, 0) == 0) { - fprintf(db_file, "%s %s", - fields_to_parse[f].name, - raw_line + matches.rm_eo); + + if(raw_line[0] == ' ' || raw_line[0] == '\t') { + char *start = raw_line; + while(*start == ' ' || *start == '\t') start++; + *(end_of_full_line++) = ' '; + strcpy(end_of_full_line, start); + while(*end_of_full_line && *end_of_full_line != '\n') { + end_of_full_line++; } + *end_of_full_line = '\0'; } + + else { + /* if(!((raw_line[0] >= 'a' && raw_line[0] <= 'z') || */ + /* (raw_line[0] >= 'A' && raw_line[0] <= 'Z'))) { */ + /* fprintf(stderr, */ + /* "Header line syntax error %s:%lu.\n", */ + /* input_filename, position_in_file); */ + /* fprintf(stderr, "%s", raw_line); */ + /* } */ + + if(full_line[0]) { + index_one_mbox_line(nb_fields_to_parse, fields_to_parse, full_line, db_file); + } + + end_of_full_line = full_line; + strcpy(end_of_full_line, raw_line); + while(*end_of_full_line && *end_of_full_line != '\n') { + end_of_full_line++; + } + *end_of_full_line = '\0'; + } + } position_in_file += strlen(raw_line); @@ -253,7 +309,7 @@ void process_entry(const char *dir_name, } closedir(dir); } else { - read_file(dir_name, nb_fields_to_parse, fields_to_parse, db_file); + index_mbox(dir_name, nb_fields_to_parse, fields_to_parse, db_file); } } @@ -261,14 +317,14 @@ void process_entry(const char *dir_name, /* For long options that have no equivalent short option, use a non-character as a pseudo short option, starting with CHAR_MAX + 1. */ -enum -{ +enum { OPT_BASH_MODE = CHAR_MAX + 1 }; static struct option long_options[] = { { "help", no_argument, 0, 'h' }, - { "db-prefix", 1, 0, 'p' }, + { "version", no_argument, 0, 'v' }, + { "db-file", 1, 0, 'd' }, { "search-pattern", 1, 0, 's' }, { "index", 0, 0, 'i' }, { 0, 0, 0, 0 } @@ -300,7 +356,7 @@ int main(int argc, char **argv) { setlocale(LC_ALL, ""); - while ((c = getopt_long(argc, argv, "hip:s:", + while ((c = getopt_long(argc, argv, "hvip:s:", long_options, NULL)) != -1) { switch(c) { @@ -309,12 +365,17 @@ int main(int argc, char **argv) { show_help = 1; break; + case 'v': + print_version(stdout); + break; + case 'i': action_index = 1; break; case 'p': db_filename = strdup(optarg); + /* printf("db_filename=\"%s\"\n", db_filename); */ break; case 's': @@ -332,16 +393,18 @@ int main(int argc, char **argv) { } if(!db_filename) { - db_filename = strdup("/tmp/mymail"); + char *default_db_filename = getenv("MYMAIL_DB_FILE"); + if(!default_db_filename) { default_db_filename = "/tmp/mymail.db"; } + db_filename = strdup(default_db_filename); } if(error) { - usage(stderr); + print_usage(stderr); exit(EXIT_FAILURE); } if(show_help) { - usage(stdout); + print_usage(stdout); exit(EXIT_SUCCESS); } @@ -393,11 +456,6 @@ int main(int argc, char **argv) { exit(EXIT_FAILURE); } - /* printf("Starting search in %s for field \"%s\" matching \"%s\".\n", */ - /* db_filename, */ - /* search_name, */ - /* search_regexp_string); */ - db_file = fopen(db_filename, "r"); if(!db_file) {