Mercurial > libpst

comparison src/readpst.c @ 345:a8577226f7a9

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
fixes from AJ Shankar for attachment processing and body encodings that contain embedded null chars
author Carl Byington <carl@five-ten-sg.com>
date Mon, 09 Mar 2015 10:47:58 -0700
parents aedcf979f439
children a57c15b3108a
comparison
equal deleted inserted replaced
344:aedcf979f439 345:a8577226f7a9
52 void header_has_field(char *header, char *field, int *flag); 52 void header_has_field(char *header, char *field, int *flag);
53 void header_get_subfield(char *field, const char *subfield, char *body_subfield, size_t size_subfield); 53 void header_get_subfield(char *field, const char *subfield, char *body_subfield, size_t size_subfield);
54 char* header_get_field(char *header, char *field); 54 char* header_get_field(char *header, char *field);
55 char* header_end_field(char *field); 55 char* header_end_field(char *field);
56 void header_strip_field(char *header, char *field); 56 void header_strip_field(char *header, char *field);
57 int test_base64(char *body); 57 int test_base64(char *body, size_t len);
58 void find_html_charset(char *html, char *charset, size_t charsetlen); 58 void find_html_charset(char *html, char *charset, size_t charsetlen);
59 void find_rfc822_headers(char** extra_mime_headers); 59 void find_rfc822_headers(char** extra_mime_headers);
60 void write_body_part(FILE* f_output, pst_string *body, char *mime, char *charset, char *boundary, pst_file* pst); 60 void write_body_part(FILE* f_output, pst_string *body, char *mime, char *charset, char *boundary, pst_file* pst);
61 void write_schedule_part_data(FILE* f_output, pst_item* item, const char* sender, const char* method); 61 void write_schedule_part_data(FILE* f_output, pst_item* item, const char* sender, const char* method);
62 void write_schedule_part(FILE* f_output, pst_item* item, const char* sender, const char* boundary); 62 void write_schedule_part(FILE* f_output, pst_item* item, const char* sender, const char* boundary);
130 int contact_mode = CMODE_VCARD; 130 int contact_mode = CMODE_VCARD;
131 int deleted_mode = DMODE_EXCLUDE; 131 int deleted_mode = DMODE_EXCLUDE;
132 int output_type_mode = 0xff; // Default to all. 132 int output_type_mode = 0xff; // Default to all.
133 int contact_mode_specified = 0; 133 int contact_mode_specified = 0;
134 int overwrite = 0; 134 int overwrite = 0;
135 int prefer_utf8 = 0;
135 int save_rtf_body = 1; 136 int save_rtf_body = 1;
136 int file_name_len = 10; // enough room for MODE_SPEARATE file name 137 int file_name_len = 10; // enough room for MODE_SPEARATE file name
137 pst_file pstfile; 138 pst_file pstfile;
138 regex_t meta_charset_pattern; 139 regex_t meta_charset_pattern;
139 char* default_charset = NULL; 140 char* default_charset = NULL;
450 printf("cannot compile regex pattern to find content charset in html bodies\n"); 451 printf("cannot compile regex pattern to find content charset in html bodies\n");
451 exit(3); 452 exit(3);
452 } 453 }
453 454
454 // command-line option handling 455 // command-line option handling
455 while ((c = getopt(argc, argv, "a:bC:c:Dd:emhj:kMo:qrSt:uVw"))!= -1) { 456 while ((c = getopt(argc, argv, "a:bC:c:Dd:emhj:kMo:qrSt:uVw8"))!= -1) {
456 switch (c) { 457 switch (c) {
457 case 'a': 458 case 'a':
458 if (optarg) { 459 if (optarg) {
459 int n = strlen(optarg); 460 int n = strlen(optarg);
460 acceptable_extensions = (char*)pst_malloc(n+2); 461 acceptable_extensions = (char*)pst_malloc(n+2);
584 version(); 585 version();
585 exit(0); 586 exit(0);
586 break; 587 break;
587 case 'w': 588 case 'w':
588 overwrite = 1; 589 overwrite = 1;
590 break;
591 case '8':
592 prefer_utf8 = 1;
589 break; 593 break;
590 default: 594 default:
591 usage(); 595 usage();
592 exit(1); 596 exit(1);
593 break; 597 break;
756 printf("\t-q\t- Quiet. Only print error messages\n"); 760 printf("\t-q\t- Quiet. Only print error messages\n");
757 printf("\t-r\t- Recursive. Output in a recursive format\n"); 761 printf("\t-r\t- Recursive. Output in a recursive format\n");
758 printf("\t-t[eajc]\t- Set the output type list. e = email, a = attachment, j = journal, c = contact\n"); 762 printf("\t-t[eajc]\t- Set the output type list. e = email, a = attachment, j = journal, c = contact\n");
759 printf("\t-u\t- Thunderbird mode. Write two extra .size and .type files\n"); 763 printf("\t-u\t- Thunderbird mode. Write two extra .size and .type files\n");
760 printf("\t-w\t- Overwrite any output mbox files\n"); 764 printf("\t-w\t- Overwrite any output mbox files\n");
765 printf("\t-8\t- Output bodies in UTF-8, rather than original encoding, if UTF-8 version is available\n");
761 printf("\n"); 766 printf("\n");
762 printf("Only one of -M -S -e -k -m -r should be specified\n"); 767 printf("Only one of -M -S -e -k -m -r should be specified\n");
763 DEBUG_RET(); 768 DEBUG_RET();
764 } 769 }
765 770
1265 if (e && f && (f < e)) e = f; 1270 if (e && f && (f < e)) e = f;
1266 } 1271 }
1267 if (!e || (e > n)) e = n; // use the trailing lf as terminator if nothing better 1272 if (!e || (e > n)) e = n; // use the trailing lf as terminator if nothing better
1268 save = *e; 1273 save = *e;
1269 *e = '\0'; 1274 *e = '\0';
1270 snprintf(body_subfield, size_subfield, "%s", s); // copy the subfield to our buffer 1275 snprintf(body_subfield, size_subfield, "%s", s); // copy the subfield to our buffer
1271 *e = save; 1276 *e = save;
1272 DEBUG_INFO(("body %s %s from headers\n", subfield, body_subfield)); 1277 DEBUG_INFO(("body %s %s from headers\n", subfield, body_subfield));
1273 } 1278 }
1274 DEBUG_RET(); 1279 DEBUG_RET();
1275 } 1280 }
1314 } 1319 }
1315 } 1320 }
1316 } 1321 }
1317 1322
1318 1323
1319 int test_base64(char *body) 1324 int test_base64(char *body, size_t len)
1320 { 1325 {
1321 int b64 = 0; 1326 int b64 = 0;
1322 uint8_t *b = (uint8_t *)body; 1327 uint8_t *b = (uint8_t *)body;
1323 DEBUG_ENT("test_base64"); 1328 DEBUG_ENT("test_base64");
1324 while (*b) { 1329 while (len--) {
1325 if ((*b < 32) && (*b != 9) && (*b != 10)) { 1330 if ((*b < 32) && (*b != 9) && (*b != 10)) {
1326 DEBUG_INFO(("found base64 byte %d\n", (int)*b)); 1331 DEBUG_INFO(("found base64 byte %d\n", (int)*b));
1327 DEBUG_HEXDUMPC(body, strlen(body), 0x10); 1332 DEBUG_HEXDUMPC(body, strlen(body), 0x10);
1328 b64 = 1; 1333 b64 = 1;
1329 break; 1334 break;
1399 1404
1400 1405
1401 void write_body_part(FILE* f_output, pst_string *body, char *mime, char *charset, char *boundary, pst_file* pst) 1406 void write_body_part(FILE* f_output, pst_string *body, char *mime, char *charset, char *boundary, pst_file* pst)
1402 { 1407 {
1403 DEBUG_ENT("write_body_part"); 1408 DEBUG_ENT("write_body_part");
1409 removeCR(body->str);
1410 size_t body_len = strlen(body->str);
1411
1404 if (body->is_utf8 && (strcasecmp("utf-8", charset))) { 1412 if (body->is_utf8 && (strcasecmp("utf-8", charset))) {
1405 // try to convert to the specified charset since the target 1413 if (prefer_utf8) {
1406 // is not utf-8, and the data came from a unicode (utf16) field
1407 // and is now in utf-8.
1408 size_t rc;
1409 DEBUG_INFO(("Convert %s utf-8 to %s\n", mime, charset));
1410 pst_vbuf *newer = pst_vballoc(2);
1411 rc = pst_vb_utf8to8bit(newer, body->str, strlen(body->str), charset);
1412 if (rc == (size_t)-1) {
1413 // unable to convert, change the charset to utf8
1414 free(newer->b);
1415 DEBUG_INFO(("Failed to convert %s utf-8 to %s\n", mime, charset));
1416 charset = "utf-8"; 1414 charset = "utf-8";
1417 } 1415 } else {
1418 else { 1416 // try to convert to the specified charset since the target
1419 // null terminate the output string 1417 // is not utf-8, and the data came from a unicode (utf16) field
1420 pst_vbgrow(newer, 1); 1418 // and is now in utf-8.
1421 newer->b[newer->dlen] = '\0'; 1419 size_t rc;
1422 free(body->str); 1420 DEBUG_INFO(("Convert %s utf-8 to %s\n", mime, charset));
1423 body->str = newer->b; 1421 pst_vbuf *newer = pst_vballoc(2);
1424 } 1422 rc = pst_vb_utf8to8bit(newer, body->str, body_len, charset);
1425 free(newer); 1423 if (rc == (size_t)-1) {
1426 } 1424 // unable to convert, change the charset to utf8
1427 removeCR(body->str); 1425 free(newer->b);
1428 int base64 = test_base64(body->str); 1426 DEBUG_INFO(("Failed to convert %s utf-8 to %s\n", mime, charset));
1427 charset = "utf-8";
1428 } else {
1429 // null terminate the output string
1430 pst_vbgrow(newer, 1);
1431 newer->b[newer->dlen] = '\0';
1432 free(body->str);
1433 body->str = newer->b;
1434 body_len = newer->dlen;
1435 }
1436 free(newer);
1437 }
1438 }
1439 int base64 = test_base64(body->str, body_len);
1429 fprintf(f_output, "\n--%s\n", boundary); 1440 fprintf(f_output, "\n--%s\n", boundary);
1430 fprintf(f_output, "Content-Type: %s; charset=\"%s\"\n", mime, charset); 1441 fprintf(f_output, "Content-Type: %s; charset=\"%s\"\n", mime, charset);
1431 if (base64) fprintf(f_output, "Content-Transfer-Encoding: base64\n"); 1442 if (base64) fprintf(f_output, "Content-Transfer-Encoding: base64\n");
1432 fprintf(f_output, "\n"); 1443 fprintf(f_output, "\n");
1444 // Any body that uses an encoding with NULLs, e.g. UTF16, will be base64-encoded here.
1433 if (base64) { 1445 if (base64) {
1434 char *enc = pst_base64_encode(body->str, strlen(body->str)); 1446 char *enc = pst_base64_encode(body->str, body_len);
1435 if (enc) { 1447 if (enc) {
1436 write_email_body(f_output, enc); 1448 write_email_body(f_output, enc);
1437 fprintf(f_output, "\n"); 1449 fprintf(f_output, "\n");
1438 free(enc); 1450 free(enc);
1439 } 1451 }
1533 em_time = pst_fileTimeToUnixTime(item->email->sent_date); 1545 em_time = pst_fileTimeToUnixTime(item->email->sent_date);
1534 c_time = ctime(&em_time); 1546 c_time = ctime(&em_time);
1535 if (c_time) 1547 if (c_time)
1536 c_time[strlen(c_time)-1] = '\0'; //remove end \n 1548 c_time[strlen(c_time)-1] = '\0'; //remove end \n
1537 else 1549 else
1538 c_time = "Fri Dec 28 12:06:21 2001"; 1550 c_time = "Thu Jan 1 00:00:00 1970";
1539 } else 1551 } else
1540 c_time = "Fri Dec 28 12:06:21 2001"; 1552 c_time = "Thu Jan 1 00:00:00 1970";
1541 1553
1542 // create our MIME boundaries here. 1554 // create our MIME boundaries here.
1543 snprintf(boundary, sizeof(boundary), "--boundary-LibPST-iamunique-%i_-_-", rand()); 1555 snprintf(boundary, sizeof(boundary), "--boundary-LibPST-iamunique-%i_-_-", rand());
1544 snprintf(altboundary, sizeof(altboundary), "alt-%s", boundary); 1556 snprintf(altboundary, sizeof(altboundary), "alt-%s", boundary);
1545 1557