#define BUFFER_SIZE 65536
#define TOKEN_BUFFER_SIZE 1024
-#define LEADING_FROM_LINE_REGEXP "^From [^ ]* \\(Mon\\|Tue\\|Wed\\|Thu\\|Fri\\|Sat\\|Sun\\) \\(Jan\\|Feb\\|Mar\\|Apr\\|May\\|Jun\\|Jul\\|Aug\\|Sep\\|Oct\\|Nov\\|Dec\\) [ 123][0-9] [0-9][0-9]:[0-9][0-9]:[0-9][0-9] [0-9][0-9][0-9][0-9]\n$"
+#define LEADING_FROM_LINE_REGEXP_STRING "^From [^ ]* \\(Mon\\|Tue\\|Wed\\|Thu\\|Fri\\|Sat\\|Sun\\) \\(Jan\\|Feb\\|Mar\\|Apr\\|May\\|Jun\\|Jul\\|Aug\\|Sep\\|Oct\\|Nov\\|Dec\\) [ 123][0-9] [0-9][0-9]:[0-9][0-9]:[0-9][0-9] [0-9][0-9][0-9][0-9]\n$"
regex_t leading_from_line_regexp;
int paranoid;
int quiet;
-char *default_search_field;
int ignore_dot_files;
/********************************************************************/
return (a && !b) || (!a && b);
}
-char *parse_token(char *token_buffer, size_t token_buffer_size,
- char separator, char *string) {
+const char *parse_token(char *token_buffer, size_t token_buffer_size,
+ char separator, const char *string) {
char *u = token_buffer;
while(u < token_buffer + token_buffer_size - 1 && *string &&
*string != separator) {
return string;
}
-/********************************************************************/
-
-/* malloc with error checking. */
+char *default_value(char *current_value,
+ const char *env_variable,
+ const char *hard_default_value) {
+ if(current_value) {
+ return current_value;
+ } else {
+ char *env_value = getenv(env_variable);
+ if(env_value) {
+ return strdup(env_value);
+ } else if(hard_default_value) {
+ return strdup(hard_default_value);
+ } else {
+ return 0;
+ }
+ }
+}
-void *safe_malloc(size_t n) {
- void *p = malloc(n);
- if(!p && n != 0) {
+FILE *safe_fopen(const char *path, const char *mode, const char *comment) {
+ FILE *result = fopen(path, mode);
+ /* printf("Opening '%s' with mode \"%s\"\n", path, mode); */
+ if(result) {
+ return result;
+ } else {
fprintf(stderr,
- "mymail: cannot allocate memory: %s\n", strerror(errno));
+ "mymail: Cannot open file '%s' (%s) with mode \"%s\".\n",
+ path, comment, mode);
exit(EXIT_FAILURE);
}
- return p;
}
/*********************************************************************/
fprintf(out, " set the db root path for recursive search\n");
fprintf(out, " -l <db filename list>, --db-list <db filename list>\n");
fprintf(out, " set the semicolon-separated list of db files for search\n");
+ fprintf(out, " -m <mbox filename pattern>, --mbox-pattern <mbox filename pattern>\n");
+ fprintf(out, " set the mbox filename pattern for recursive search\n");
fprintf(out, " -s <search pattern>, --search <search pattern>\n");
fprintf(out, " search for matching mails in the db file\n");
- fprintf(out, " -d <db filename>, --db-file <db filename>\n");
+ fprintf(out, " -d <db filename>, --db-file-generate <db filename>\n");
fprintf(out, " set the db filename for indexing\n");
fprintf(out, " -i, --index\n");
fprintf(out, " index mails\n");
/*********************************************************************/
-time_t time_for_past_day(int day) {
- time_t t;
- struct tm *tm;
- int delta_day;
- t = time(0);
- tm = localtime(&t);
- delta_day = (7 + tm->tm_wday - day) % 7 + 1;
- return t - delta_day * 3600 * 24 + tm->tm_sec + 60 * tm->tm_min + 3600 * tm->tm_hour;
-}
-
-/*********************************************************************/
-
int ignore_entry(const char *name) {
return
strcmp(name, ".") == 0 ||
int is_a_leading_from_line(char *mbox_line) {
return
-
- /*
-
- The mbox man page in qmail documentation states:
-
- > The reader should not attempt to take advantage of the fact
- > that every From_ line (past the beginning of the file) is
- > preceded by a blank line.
-
- */
-
strncmp(mbox_line, "From ", 5) == 0 &&
regexec(&leading_from_line_regexp, mbox_line, 0, 0, 0) == 0;
}
int mbox_line_match_search(struct search_condition *condition,
- int mbox_id, char *mbox_value) {
+ int mbox_id, const char *mbox_value) {
if(condition->field_id == ID_INTERVAL) {
if(mbox_id == ID_LEADING_LINE) {
- char *c;
+ const char *c;
time_t t;
struct tm tm;
nb_body_hits = 0;
header = 1;
- mail_file = fopen(mail_filename, "r");
-
- if(!mail_file) {
- fprintf(stderr,
- "mymail: Cannot open mbox '%s' for body scan.\n",
- mail_filename);
- exit(EXIT_FAILURE);
- }
+ mail_file = safe_fopen(mail_filename, "r", "mbox for body scan");
fseek(mail_file, position_in_mail, SEEK_SET);
char raw_mbox_line[BUFFER_SIZE];
FILE *mail_file;
- mail_file = fopen(mail_filename, "r");
-
- if(!mail_file) {
- fprintf(stderr,
- "mymail: Cannot open mbox '%s' for mail extraction.\n",
- mail_filename);
- exit(EXIT_FAILURE);
- }
-
+ mail_file = safe_fopen(mail_filename, "r", "mbox for mail extraction");
fseek(mail_file, position_in_mail, SEEK_SET);
if(fgets(raw_mbox_line, BUFFER_SIZE, mail_file)) {
char raw_db_line[BUFFER_SIZE];
char current_mail_filename[PATH_MAX + 1];
unsigned long int current_position_in_mail;
- char mbox_name[TOKEN_BUFFER_SIZE], *mbox_value;
+ char mbox_name[TOKEN_BUFFER_SIZE];
+ const char *mbox_value;
int mbox_id;
int already_written, m, n;
int nb_body_conditions, nb_fulfilled_body_conditions;
fflush(stdout);
}
- db_file = fopen(db_filename, "r");
-
- if(!db_file) {
- fprintf(stderr,
- "mymail: Cannot open \"%s\" for reading: %s\n",
- db_filename,
- strerror(errno));
- exit(EXIT_FAILURE);
- }
+ db_file = safe_fopen(db_filename, "r", "index file for search");
/* First, check the db file leading line integrity */
exit(EXIT_FAILURE);
}
+ /* printf("recursive_search_in_db %s\n", entry_name); */
+
dir = opendir(entry_name);
if(dir) {
/*********************************************************************/
-void index_one_mbox_line(int nb_fields_to_parse, struct parsable_field *fields_to_parse,
+void index_one_mbox_line(unsigned int nb_fields_to_parse,
+ struct parsable_field *fields_to_parse,
char *raw_mbox_line, FILE *db_file) {
regmatch_t matches;
- int f;
+ unsigned int f;
for(f = 0; f < nb_fields_to_parse; f++) {
if(regexec(&fields_to_parse[f].regexp, raw_mbox_line, 1, &matches, 0) == 0) {
fprintf(db_file, "%s %s\n",
int in_header, new_header;
unsigned long int position_in_file;
- file = fopen(mbox_filename, "r");
-
- if(!file) {
- fprintf(stderr, "mymail: Cannot open '%s'.\n", mbox_filename);
- if(paranoid) { exit(EXIT_FAILURE); }
- return;
- }
+ file = safe_fopen(mbox_filename, "r", "mbox for indexing");
in_header = 0;
new_header = 0;
}
void recursive_index_mbox(FILE *db_file,
- const char *entry_name,
+ const char *entry_name, regex_t *mbox_filename_regexp,
int nb_fields_to_parse, struct parsable_field *fields_to_parse) {
DIR *dir;
struct dirent *dir_e;
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_index_mbox(db_file, subname, nb_fields_to_parse, fields_to_parse);
+ recursive_index_mbox(db_file, subname, mbox_filename_regexp,
+ nb_fields_to_parse, fields_to_parse);
}
}
closedir(dir);
} else {
- index_mbox(entry_name, nb_fields_to_parse, fields_to_parse, db_file);
+ const char *s = entry_name, *filename = s;
+ while(*s) { if(*s == '/') { filename = s+1; }; s++; }
+ if(!mbox_filename_regexp || regexec(mbox_filename_regexp, filename, 0, 0, 0) == 0) {
+ index_mbox(entry_name, nb_fields_to_parse, fields_to_parse, db_file);
+ }
}
}
{ "help", no_argument, 0, 'h' },
{ "version", no_argument, 0, 'v' },
{ "quiet", no_argument, 0, 'q' },
- { "db-file", 1, 0, 'd' },
+ { "db-file-generate", 1, 0, 'd' },
{ "db-pattern", 1, 0, 'p' },
{ "db-root", 1, 0, 'r' },
{ "db-list", 1, 0, 'l' },
+ { "mbox-pattern", 1, 0, 'm' },
{ "search", 1, 0, 's' },
{ "index", 0, 0, 'i' },
{ "output", 1, 0, 'o' },
{ 0, 0, 0, 0 }
};
+struct time_criterion {
+ char *label;
+ int start_hour, end_hour;
+ int past_week_day;
+};
+
/*********************************************************************/
-void init_condition(struct search_condition *condition, char *full_string) {
+static struct time_criterion time_criteria[] = {
+
+ { "8h", 8, -1, -1 },
+ { "today", 24, -1, -1 },
+ { "24h", 24, -1, -1 },
+ { "week", 24 * 7, -1, -1 },
+ { "month", 24 * 31, -1, -1 },
+ { "year", 24 * 365, -1, -1 },
+
+ { "yesterday", 48, 24, -1 },
+
+ { "monday", -1, -1, 1 },
+ { "tuesday", -1, -1, 2 },
+ { "wednesday", -1, -1, 3 },
+ { "thursday", -1, -1, 4 },
+ { "friday", -1, -1, 5 },
+ { "saturday", -1, -1, 6 },
+ { "sunday", -1, -1, 7 },
+
+};
+
+/*********************************************************************/
+
+time_t time_for_past_day(int day) {
+ time_t t;
+ struct tm *tm;
+ int delta_day;
+ t = time(0);
+ tm = localtime(&t);
+ delta_day = (7 + tm->tm_wday - day) % 7;
+ if(delta_day == 0) { delta_day = 7; }
+ return t - (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,
+ const char *default_search_field) {
char full_search_field[TOKEN_BUFFER_SIZE], *search_field;
- int m;
- char *string;
+ unsigned int k, m;
+ const char *string;
string = parse_token(full_search_field, TOKEN_BUFFER_SIZE, ' ', full_string);
search_field = full_search_field;
condition->negation = 0;
}
- /* Recently */
-
- if(strcmp(search_field, "8h") == 0) {
- condition->field_id = ID_INTERVAL;
- condition->interval_start = time(0) - 3600 * 8;
- condition->interval_stop = 0;
- }
-
- else if(strcmp(search_field, "24h") == 0 ||
- strcmp(search_field, "today") == 0) {
- condition->field_id = ID_INTERVAL;
- condition->interval_start = time(0) - 3600 * 24;
- condition->interval_stop = 0;
- }
-
- else if(strcmp(search_field, "week") == 0) {
- condition->field_id = ID_INTERVAL;
- condition->interval_start = time(0) - 3600 * 24 * 7;
- condition->interval_stop = 0;
- }
-
- else if(strcmp(search_field, "month") == 0) {
- condition->field_id = ID_INTERVAL;
- condition->interval_start = time(0) - 3600 * 24 * 31;
- condition->interval_stop = 0;
- }
-
- else if(strcmp(search_field, "year") == 0) {
- condition->field_id = ID_INTERVAL;
- condition->interval_start = time(0) - 3600 * 24 * 365;
- condition->interval_stop = 0;
- }
-
- /* Yesterday */
-
- else if(strcmp(search_field, "yesterday") == 0) {
- condition->field_id = ID_INTERVAL;
- condition->interval_start = time(0) - 2 * 3600 * 24;
- condition->interval_stop = condition->interval_start + 3600 * 24;
- }
-
- /* Week days */
+ condition->field_id = -1;
- else if(strcmp(search_field, "monday") == 0) {
- condition->field_id = ID_INTERVAL;
- condition->interval_start = time_for_past_day(1);
- condition->interval_stop = condition->interval_start + 3600 * 24;
- }
-
- else if(strcmp(search_field, "tuesday") == 0) {
- condition->field_id = ID_INTERVAL;
- condition->interval_start = time_for_past_day(2);
- condition->interval_stop = condition->interval_start + 3600 * 24;
- }
-
- else if(strcmp(search_field, "wednesday") == 0) {
- condition->field_id = ID_INTERVAL;
- condition->interval_start = time_for_past_day(3);
- condition->interval_stop = condition->interval_start + 3600 * 24;
- }
-
- else if(strcmp(search_field, "thursday") == 0) {
- condition->field_id = ID_INTERVAL;
- condition->interval_start = time_for_past_day(4);
- condition->interval_stop = condition->interval_start + 3600 * 24;
- }
-
- else if(strcmp(search_field, "friday") == 0) {
- condition->field_id = ID_INTERVAL;
- condition->interval_start = time_for_past_day(5);
- condition->interval_stop = condition->interval_start + 3600 * 24;
- }
-
- else if(strcmp(search_field, "saturday") == 0) {
- condition->field_id = ID_INTERVAL;
- condition->interval_start = time_for_past_day(6);
- condition->interval_stop = condition->interval_start + 3600 * 24;
- }
+ /* Recently */
- else if(strcmp(search_field, "sunday") == 0) {
- condition->field_id = ID_INTERVAL;
- condition->interval_start = time_for_past_day(7);
- condition->interval_stop = condition->interval_start + 3600 * 24;
+ for(k = 0; k < sizeof(time_criteria) / sizeof(struct time_criterion); k++) {
+ if(strcmp(time_criteria[k].label, search_field) == 0) {
+ condition->field_id = ID_INTERVAL;
+ if(time_criteria[k].past_week_day < 0) {
+ condition->interval_start = time(0) - 3600 * time_criteria[k].start_hour;
+ if(time_criteria[k].end_hour >= 0) {
+ condition->interval_stop = time(0) - 3600 * time_criteria[k].end_hour;
+ } else {
+ condition->interval_stop = 0;
+ }
+ } else {
+ condition->interval_start = time_for_past_day(time_criteria[k].past_week_day);
+ condition->interval_stop = condition->interval_start + 3600 * 24;
+ }
+ break;
+ }
}
- else {
-
- /* header-related conditions */
+ if(condition->field_id == -1) {
- condition->field_id = -1;
+ /* No time condition matched, look for the search fields */
for(m = 0; (m < MAX_ID) && condition->field_id == -1; m++) {
if(strncmp(field_names[m], search_field, strlen(search_field)) == 0) {
}
}
+ /* None match, if there is a default search field, re-run the search with it */
+
if(condition->field_id == -1) {
if(default_search_field) {
for(m = 0; (m < MAX_ID) && condition->field_id == -1; m++) {
/*********************************************************************/
int main(int argc, char **argv) {
- char *db_filename;
- char *db_filename_regexp_string;
- char *db_root_path;
- char *db_filename_list;
+ char *db_filename = 0;
+ char *db_filename_regexp_string = 0;
+ char *db_root_path = 0;
+ char *db_filename_list = 0;
+ char *mbox_filename_regexp_string = 0;
+ char *default_search_field;
char output_filename[PATH_MAX + 1];
- int action_index;
+ int action_index = 0;
int error = 0, show_help = 0;
- const int nb_fields_to_parse = sizeof(fields_to_parse) / sizeof(struct parsable_field);
+ const unsigned int nb_fields_to_parse =
+ sizeof(fields_to_parse) / sizeof(struct parsable_field);
char c;
- int f, n;
- int nb_search_conditions;
- FILE *output_file;
+ unsigned int f, n;
+ unsigned int nb_search_conditions;
struct search_condition search_conditions[MAX_NB_SEARCH_CONDITIONS];
- if(regcomp(&leading_from_line_regexp, LEADING_FROM_LINE_REGEXP, 0)) {
+ if(regcomp(&leading_from_line_regexp, LEADING_FROM_LINE_REGEXP_STRING, 0)) {
fprintf(stderr,
"mymail: Cannot compile leading \"from\" line regexp. That is strange.\n");
exit(EXIT_FAILURE);
}
paranoid = 0;
- action_index = 0;
- db_filename = 0;
- db_filename_regexp_string = 0;
- db_root_path = 0;
- db_filename_list = 0;
quiet = 0;
default_search_field = 0;
ignore_dot_files = 1;
+ strncpy(output_filename, "", PATH_MAX);
setlocale(LC_ALL, "");
nb_search_conditions = 0;
- while ((c = getopt_long(argc, argv, "hvqip:s:d:r:l:o:a:",
+ while ((c = getopt_long(argc, argv, "hvqip:s:d:r:l:o:a:m:",
long_options, NULL)) != -1) {
switch(c) {
break;
case 'd':
+ if(db_filename) {
+ fprintf(stderr, "mymail: Can not set the db filename twice.\n");
+ exit(EXIT_FAILURE);
+ }
db_filename = strdup(optarg);
break;
case 'p':
+ if(db_filename_regexp_string) {
+ fprintf(stderr, "mymail: Can not set the db filename pattern twice.\n");
+ exit(EXIT_FAILURE);
+ }
db_filename_regexp_string = strdup(optarg);
break;
+ case 'm':
+ if(mbox_filename_regexp_string) {
+ fprintf(stderr, "mymail: Can not set the mbox filename pattern twice.\n");
+ exit(EXIT_FAILURE);
+ }
+ mbox_filename_regexp_string = strdup(optarg);
+ break;
+
case 'o':
strncpy(output_filename, optarg, PATH_MAX);
break;
case 'r':
+ if(db_root_path) {
+ fprintf(stderr, "mymail: Can not set the db root path twice.\n");
+ exit(EXIT_FAILURE);
+ }
db_root_path = strdup(optarg);
break;
case 'l':
+ if(db_filename_list) {
+ fprintf(stderr, "mymail: Can not set the db filename list twice.\n");
+ exit(EXIT_FAILURE);
+ }
db_filename_list = strdup(optarg);
break;
fprintf(stderr, "mymail: Too many search patterns.\n");
exit(EXIT_FAILURE);
}
- init_condition(&search_conditions[nb_search_conditions], optarg);
+ init_condition(&search_conditions[nb_search_conditions], optarg, default_search_field);
nb_search_conditions++;
break;
}
}
- if(!db_filename) {
- char *default_db_filename = getenv("MYMAIL_DB_FILE");
+ /* Set all the values that may defined in the arguments, through
+ environment variables, or hard-coded */
- if(!default_db_filename) {
- default_db_filename = "mymail.db";
- }
+ db_filename = default_value(db_filename,
+ "MYMAIL_DB_FILE",
+ "mymail.db");
- db_filename = strdup(default_db_filename);
- }
+ db_filename_regexp_string = default_value(db_filename_regexp_string,
+ "MYMAIL_DB_FILE",
+ "\\.db$");
- if(!db_filename_regexp_string) {
- char *default_db_filename_regexp_string = getenv("MYMAIL_DB_PATTERN");
+ db_root_path = default_value(db_root_path,
+ "MYMAIL_DB_ROOT",
+ 0);
- if(!default_db_filename_regexp_string) {
- default_db_filename_regexp_string = "^mymail.db$";
- }
+ db_filename_list = default_value(db_filename_list,
+ "MYMAIL_DB_LIST",
+ 0);
- db_filename_regexp_string = strdup(default_db_filename_regexp_string);
- }
+ mbox_filename_regexp_string = default_value(mbox_filename_regexp_string,
+ "MYMAIL_MBOX_PATTERN",
+ 0);
- if(!db_root_path) {
- char *default_db_root_path = getenv("MYMAIL_DB_ROOT");
-
- if(default_db_root_path) {
- db_root_path = strdup(default_db_root_path);
- }
- }
-
- 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(output_filename[0]) {
- output_file = fopen(output_filename, "w");
-
- if(!output_file) {
- fprintf(stderr,
- "mymail: Cannot open result file \"%s\" for writing: %s\n",
- output_filename,
- strerror(errno));
- exit(EXIT_FAILURE);
- }
- } else {
- output_file = stdout;
- quiet = 1;
- }
+ /* Start the processing */
if(error) {
print_usage(stderr);
exit(EXIT_SUCCESS);
}
+ /* mbox indexing */
+
if(action_index) {
FILE *db_file;
+ regex_t mbox_filename_regexp_static;
+ regex_t *mbox_filename_regexp;
- db_file = fopen(db_filename, "w");
-
- if(!db_file) {
- fprintf(stderr,
- "mymail: Cannot open \"%s\" for writing: %s\n",
- db_filename,
- strerror(errno));
- exit(EXIT_FAILURE);
+ if(mbox_filename_regexp_string) {
+ if(regcomp(&mbox_filename_regexp_static,
+ mbox_filename_regexp_string,
+ 0)) {
+ fprintf(stderr,
+ "mymail: Syntax error in regexp \"%s\".\n",
+ mbox_filename_regexp_string);
+ exit(EXIT_FAILURE);
+ }
+ mbox_filename_regexp = &mbox_filename_regexp_static;
+ } else {
+ mbox_filename_regexp = 0;
}
+ db_file = safe_fopen(db_filename, "w", "index file for indexing");
+
for(f = 0; f < nb_fields_to_parse; f++) {
if(regcomp(&fields_to_parse[f].regexp,
fields_to_parse[f].regexp_string,
while(optind < argc) {
recursive_index_mbox(db_file,
- argv[optind],
+ argv[optind], mbox_filename_regexp,
nb_fields_to_parse, fields_to_parse);
optind++;
}
fflush(db_file);
fclose(db_file);
+ if(mbox_filename_regexp) {
+ regfree(mbox_filename_regexp);
+ }
+
for(f = 0; f < nb_fields_to_parse; f++) {
regfree(&fields_to_parse[f].regexp);
}
}
+ /* Mail search */
+
else {
+
+ FILE *output_file;
int nb_extracted_mails = 0;
+ if(output_filename[0]) {
+ output_file = safe_fopen(output_filename, "w", "result mbox");
+ } else {
+ output_file = stdout;
+ quiet = 1;
+ }
+
if(nb_search_conditions > 0) {
/* Recursive search if db_root_path is set */
if(db_filename_list) {
char db_filename[PATH_MAX + 1];
- char *s;
+ const char *s;
s = db_filename_list;
}
}
+ fflush(output_file);
+
+ if(output_file != stdout) {
+ fclose(output_file);
+ }
}
for(n = 0; n < nb_search_conditions; n++) {
free_condition(&search_conditions[n]);
}
- if(output_file != stdout) {
- fflush(output_file);
- fclose(output_file);
- }
-
free(db_filename);
free(db_filename_regexp_string);
free(db_root_path);
free(db_filename_list);
+ free(mbox_filename_regexp_string);
regfree(&leading_from_line_regexp);
projects / mymail.git / blobdiff