36 |
int status_code; |
int status_code; |
37 |
char error_buffer[CURL_ERROR_SIZE]; |
char error_buffer[CURL_ERROR_SIZE]; |
38 |
regex_t re_status_line, re_token, re_token_value, re_content_type, re_ugly, |
regex_t re_status_line, re_token, re_token_value, re_content_type, re_ugly, |
39 |
re_absolute_uri, re_etag, re_server, re_transfer_coding, re_upgrade; |
re_absolute_uri, re_etag, re_server, re_transfer_coding, re_upgrade, |
40 |
|
re_rfc1123, re_rfc1036, re_asctime; |
41 |
|
|
42 |
|
|
43 |
void init(void); |
void init(void); |
48 |
void check_status_line(const char *s); |
void check_status_line(const char *s); |
49 |
void check_header(const char *name, const char *value); |
void check_header(const char *name, const char *value); |
50 |
bool parse_date(const char *s, struct tm *tm); |
bool parse_date(const char *s, struct tm *tm); |
51 |
|
int month(const char *s); |
52 |
|
time_t mktime_from_utc(struct tm *t); |
53 |
const char *skip_lws(const char *s); |
const char *skip_lws(const char *s); |
54 |
bool parse_list(const char *s, regex_t *preg, unsigned int n, unsigned int m, |
bool parse_list(const char *s, regex_t *preg, unsigned int n, unsigned int m, |
55 |
void (*callback)(const char *s, regmatch_t pmatch[])); |
void (*callback)(const char *s, regmatch_t pmatch[])); |
177 |
"^HTTP/([0-9]+)[.]([0-9]+) ([0-9][0-9][0-9]) ([\t -~€-ÿ]*)$", |
"^HTTP/([0-9]+)[.]([0-9]+) ([0-9][0-9][0-9]) ([\t -~€-ÿ]*)$", |
178 |
REG_EXTENDED); |
REG_EXTENDED); |
179 |
regcomp_wrapper(&re_token, |
regcomp_wrapper(&re_token, |
180 |
"^([-0-9a-zA-Z_.]+)", |
"^([-0-9a-zA-Z_.!]+)", |
181 |
REG_EXTENDED); |
REG_EXTENDED); |
182 |
regcomp_wrapper(&re_token_value, |
regcomp_wrapper(&re_token_value, |
183 |
"^([-0-9a-zA-Z_.]+)(=([-0-9a-zA-Z_.]+|\"([^\"]|[\\].)*\"))?", |
"^([-0-9a-zA-Z_.!]+)(=([-0-9a-zA-Z_.!]+|\"([^\"]|[\\].)*\"))?", |
184 |
REG_EXTENDED); |
REG_EXTENDED); |
185 |
regcomp_wrapper(&re_content_type, |
regcomp_wrapper(&re_content_type, |
186 |
"^([-0-9a-zA-Z_.]+)/([-0-9a-zA-Z_.]+)[ \t]*" |
"^([-0-9a-zA-Z_.]+)/([-0-9a-zA-Z_.]+)[ \t]*" |
194 |
"^(W/[ \t]*)?\"([^\"]|[\\].)*\"$", |
"^(W/[ \t]*)?\"([^\"]|[\\].)*\"$", |
195 |
REG_EXTENDED); |
REG_EXTENDED); |
196 |
regcomp_wrapper(&re_server, |
regcomp_wrapper(&re_server, |
197 |
"^((([-0-9a-zA-Z_.]+(/[-0-9a-zA-Z_.]+)?)|(\\(.*\\)))[ \t]*)+$", |
"^((([-0-9a-zA-Z_.!]+(/[-0-9a-zA-Z_.]+)?)|(\\(.*\\)))[ \t]*)+$", |
198 |
REG_EXTENDED); |
REG_EXTENDED); |
199 |
regcomp_wrapper(&re_transfer_coding, |
regcomp_wrapper(&re_transfer_coding, |
200 |
"^([-0-9a-zA-Z_.]+)[ \t]*" |
"^([-0-9a-zA-Z_.]+)[ \t]*" |
205 |
"^([-0-9a-zA-Z_.](/[-0-9a-zA-Z_.])?)+$", |
"^([-0-9a-zA-Z_.](/[-0-9a-zA-Z_.])?)+$", |
206 |
REG_EXTENDED); |
REG_EXTENDED); |
207 |
regcomp_wrapper(&re_ugly, |
regcomp_wrapper(&re_ugly, |
208 |
"^[a-zA-Z0-9]+://[^/]+[/a-zA-Z0-9-_]*$", |
"^[a-zA-Z0-9]+://[^/]+[-/a-zA-Z0-9_]*$", |
209 |
|
REG_EXTENDED); |
210 |
|
regcomp_wrapper(&re_rfc1123, |
211 |
|
"^(Mon|Tue|Wed|Thu|Fri|Sat|Sun), ([0123][0-9]) " |
212 |
|
"(Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec) ([0-9]{4}) " |
213 |
|
"([012][0-9]):([0-5][0-9]):([0-5][0-9]) GMT$", |
214 |
|
REG_EXTENDED); |
215 |
|
regcomp_wrapper(&re_rfc1036, |
216 |
|
"^(Monday|Tuesday|Wednesday|Thursday|Friday|Saturday|Sunday), " |
217 |
|
"([0123][0-9])-(Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec)-" |
218 |
|
"([0-9][0-9]) ([012][0-9]):([0-5][0-9]):([0-5][0-9]) GMT$", |
219 |
|
REG_EXTENDED); |
220 |
|
regcomp_wrapper(&re_asctime, |
221 |
|
"^(Mon|Tue|Wed|Thu|Fri|Sat|Sun) " |
222 |
|
"(Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec) ([ 12][0-9]) " |
223 |
|
"([012][0-9]):([0-5][0-9]):([0-5][0-9]) ([0-9]{4})$", |
224 |
REG_EXTENDED); |
REG_EXTENDED); |
225 |
} |
} |
226 |
|
|
398 |
if (header) { |
if (header) { |
399 |
header->count++; |
header->count++; |
400 |
header->handler(value); |
header->handler(value); |
401 |
} else |
} else if ((name[0] == 'X' || name[0] == 'x') && name[1] == '-') { |
402 |
|
lookup("xheader"); |
403 |
|
} else { |
404 |
lookup("nonstandard"); |
lookup("nonstandard"); |
405 |
|
} |
406 |
} |
} |
407 |
|
|
408 |
|
|
411 |
*/ |
*/ |
412 |
bool parse_date(const char *s, struct tm *tm) |
bool parse_date(const char *s, struct tm *tm) |
413 |
{ |
{ |
414 |
char *r; |
int r; |
415 |
int len = strlen(s); |
int len = strlen(s); |
416 |
|
regmatch_t pmatch[20]; |
417 |
|
|
418 |
if (len == 29) { |
if (len == 29) { |
419 |
/* RFC 1123 */ |
/* RFC 1123 */ |
420 |
r = strptime(s, "%a, %d %b %Y %H:%M:%S GMT", tm); |
r = regexec(&re_rfc1123, s, 20, pmatch, 0); |
421 |
if (r == s + len) |
if (r == 0) { |
422 |
|
tm->tm_mday = atoi(s + pmatch[2].rm_so); |
423 |
|
tm->tm_mon = month(s + pmatch[3].rm_so); |
424 |
|
tm->tm_year = atoi(s + pmatch[4].rm_so) - 1900; |
425 |
|
tm->tm_hour = atoi(s + pmatch[5].rm_so); |
426 |
|
tm->tm_min = atoi(s + pmatch[6].rm_so); |
427 |
|
tm->tm_sec = atoi(s + pmatch[7].rm_so); |
428 |
return true; |
return true; |
429 |
|
} |
430 |
|
|
431 |
} else if (len == 24) { |
} else if (len == 24) { |
432 |
/* asctime() format */ |
/* asctime() format */ |
433 |
r = strptime(s, "%a %b %d %H:%M:%S %Y", tm); |
r = regexec(&re_asctime, s, 20, pmatch, 0); |
434 |
if (r == s + len) { |
if (r == 0) { |
435 |
lookup("asctime"); |
if (s[pmatch[3].rm_so] == ' ') |
436 |
return true; |
tm->tm_mday = atoi(s + pmatch[3].rm_so + 1); |
437 |
} |
else |
438 |
r = strptime(s, "%a %b %d %H:%M:%S %Y", tm); |
tm->tm_mday = atoi(s + pmatch[3].rm_so); |
439 |
if (r == s + len) { |
tm->tm_mon = month(s + pmatch[2].rm_so); |
440 |
|
tm->tm_year = atoi(s + pmatch[7].rm_so) - 1900; |
441 |
|
tm->tm_hour = atoi(s + pmatch[4].rm_so); |
442 |
|
tm->tm_min = atoi(s + pmatch[5].rm_so); |
443 |
|
tm->tm_sec = atoi(s + pmatch[6].rm_so); |
444 |
lookup("asctime"); |
lookup("asctime"); |
445 |
return true; |
return true; |
446 |
} |
} |
447 |
|
|
448 |
} else { |
} else { |
449 |
/* RFC 1036 */ |
/* RFC 1036 */ |
450 |
r = strptime(s, "%a, %d-%b-%y %H:%M:%S GMT", tm); |
r = regexec(&re_rfc1036, s, 20, pmatch, 0); |
451 |
if (r == s + len) { |
if (r == 0) { |
452 |
|
tm->tm_mday = atoi(s + pmatch[2].rm_so); |
453 |
|
tm->tm_mon = month(s + pmatch[3].rm_so); |
454 |
|
tm->tm_year = 100 + atoi(s + pmatch[4].rm_so); |
455 |
|
tm->tm_hour = atoi(s + pmatch[5].rm_so); |
456 |
|
tm->tm_min = atoi(s + pmatch[6].rm_so); |
457 |
|
tm->tm_sec = atoi(s + pmatch[7].rm_so); |
458 |
lookup("rfc1036"); |
lookup("rfc1036"); |
459 |
return true; |
return true; |
460 |
} |
} |
467 |
|
|
468 |
|
|
469 |
/** |
/** |
470 |
|
* Convert a month name to the month number. |
471 |
|
*/ |
472 |
|
int month(const char *s) |
473 |
|
{ |
474 |
|
switch (s[0]) { |
475 |
|
case 'J': |
476 |
|
switch (s[1]) { |
477 |
|
case 'a': |
478 |
|
return 0; |
479 |
|
case 'u': |
480 |
|
return s[2] == 'n' ? 5 : 6; |
481 |
|
} |
482 |
|
case 'F': |
483 |
|
return 1; |
484 |
|
case 'M': |
485 |
|
return s[2] == 'r' ? 2 : 4; |
486 |
|
case 'A': |
487 |
|
return s[1] == 'p' ? 3 : 7; |
488 |
|
case 'S': |
489 |
|
return 8; |
490 |
|
case 'O': |
491 |
|
return 9; |
492 |
|
case 'N': |
493 |
|
return 10; |
494 |
|
case 'D': |
495 |
|
return 11; |
496 |
|
} |
497 |
|
return 0; |
498 |
|
} |
499 |
|
|
500 |
|
|
501 |
|
/** |
502 |
|
* UTC version of mktime, from |
503 |
|
* http://lists.debian.org/deity/2002/deity-200204/msg00082.html |
504 |
|
*/ |
505 |
|
time_t mktime_from_utc(struct tm *t) |
506 |
|
{ |
507 |
|
time_t tl, tb; |
508 |
|
struct tm *tg; |
509 |
|
|
510 |
|
tl = mktime (t); |
511 |
|
if (tl == -1) |
512 |
|
{ |
513 |
|
t->tm_hour--; |
514 |
|
tl = mktime (t); |
515 |
|
if (tl == -1) |
516 |
|
return -1; /* can't deal with output from strptime */ |
517 |
|
tl += 3600; |
518 |
|
} |
519 |
|
tg = gmtime (&tl); |
520 |
|
tg->tm_isdst = 0; |
521 |
|
tb = mktime (tg); |
522 |
|
if (tb == -1) |
523 |
|
{ |
524 |
|
tg->tm_hour--; |
525 |
|
tb = mktime (tg); |
526 |
|
if (tb == -1) |
527 |
|
return -1; /* can't deal with output from gmtime */ |
528 |
|
tb += 3600; |
529 |
|
} |
530 |
|
return (tl - (tb - tl)); |
531 |
|
} |
532 |
|
|
533 |
|
|
534 |
|
/** |
535 |
* Skip optional LWS (linear white space) [2.2] |
* Skip optional LWS (linear white space) [2.2] |
536 |
*/ |
*/ |
537 |
const char *skip_lws(const char *s) |
const char *skip_lws(const char *s) |
785 |
time0 = time(0); |
time0 = time(0); |
786 |
if (!parse_date(s, &tm)) |
if (!parse_date(s, &tm)) |
787 |
return; |
return; |
788 |
time1 = mktime(&tm); |
time1 = mktime_from_utc(&tm); |
789 |
|
|
790 |
diff = difftime(time0, time1); |
diff = difftime(time0, time1); |
791 |
if (10 < fabs(diff)) |
if (10 < fabs(diff)) |
820 |
time0 = time(0); |
time0 = time(0); |
821 |
if (!parse_date(s, &tm)) |
if (!parse_date(s, &tm)) |
822 |
return; |
return; |
823 |
time1 = mktime(&tm); |
time1 = mktime_from_utc(&tm); |
824 |
|
|
825 |
diff = difftime(time1, time0); |
diff = difftime(time1, time0); |
826 |
if (10 < diff) |
if (10 < diff) |
1089 |
{ "via", "This header was added by a proxy, cache or gateway." }, |
{ "via", "This header was added by a proxy, cache or gateway." }, |
1090 |
{ "wrongdate", "Warning: The server date-time differs from this system's " |
{ "wrongdate", "Warning: The server date-time differs from this system's " |
1091 |
"date-time by more than 10 seconds. Check that both the " |
"date-time by more than 10 seconds. Check that both the " |
1092 |
"system clocks are correct." } |
"system clocks are correct." }, |
1093 |
|
{ "xheader", "This is an extension header. I don't know how to check it." } |
1094 |
}; |
}; |
1095 |
|
|
1096 |
|
|