Unverified Commit 1c39128d authored by Daniel Stenberg's avatar Daniel Stenberg
Browse files

parsedate: fix date parsing for systems with 32 bit long 

Make curl_getdate() handle dates before 1970 as well (returning negative
values).

Make test 517 test dates for 64 bit time_t.

This fixes bug (3) mentioned in #2238

Closes #2250
parent 1433e464

configure.ac

+18 −1
Original line number Diff line number Diff line
@@ -5,7 +5,7 @@ 
#                            | (__| |_| |  _ <| |___

#                             \___|\___/|_| \_\_____|

#

# Copyright (C) 1998 - 2017, Daniel Stenberg, <daniel@haxx.se>, et al.

# Copyright (C) 1998 - 2018, Daniel Stenberg, <daniel@haxx.se>, et al.

#

# This software is licensed as described in the file COPYING, which

# you should have received as part of this distribution. The terms
@@ -3483,6 +3483,23 @@ AC_CHECK_TYPE([bool],[ 
#endif

])



AC_MSG_CHECKING([if time_t is unsigned])

AC_RUN_IFELSE([

  AC_LANG_SOURCE([[

  #include <time.h>

  #include <limits.h>

  time_t t = ULONG_MAX;

  return (t > 0);

  ]])] ,[

  AC_MSG_RESULT([yes])

  AC_DEFINE(HAVE_TIME_T_UNSIGNED, 1, [Define this if time_t is unsigned])

],[

  AC_MSG_RESULT([no])

],[

  dnl cross-compiling, most systems are unsigned

  AC_MSG_RESULT([no])

])



CURL_CONFIGURE_CURL_SOCKLEN_T



CURL_CONFIGURE_PULL_SYS_POLL

docs/libcurl/curl_getdate.3

+9 −7
Original line number Diff line number Diff line
@@ -5,7 +5,7 @@ 
.\" *                            | (__| |_| |  _ <| |___

.\" *                             \___|\___/|_| \_\_____|

.\" *

.\" * Copyright (C) 1998 - 2015, Daniel Stenberg, <daniel@haxx.se>, et al.

.\" * Copyright (C) 1998 - 2018, Daniel Stenberg, <daniel@haxx.se>, et al.

.\" *

.\" * This software is licensed as described in the file COPYING, which

.\" * you should have received as part of this distribution. The terms
@@ -96,13 +96,15 @@ only ones RFC 7231 says HTTP applications may use. 
This function returns -1 when it fails to parse the date string. Otherwise it

returns the number of seconds as described.



If the year is larger than 2037 on systems with 32 bit time_t, this function

will return 0x7fffffff (since that is the largest possible signed 32 bit

number).

On systems with a signed 32 bit time_t: if the year is larger than 2037 or

less than 1903, this function will return -1.



Having a 64 bit time_t is not a guarantee that dates beyond 03:14:07 UTC,

January 19, 2038 will work fine. On systems with a 64 bit time_t but with a

crippled mktime(), \fIcurl_getdate(3)\fP will return -1 in this case.

On systems with an unsigned 32 bit time_t: if the year is larger than 2106 or

less than 1970, this function will return -1.



On systems with 64 bit time_t: if the year is less than 1583, this function

will return -1. (The Gregorian calendar was first introduced 1582 so no "real"

dates in this way of doing dates existed before then.)

.SH "SEE ALSO"

.BR curl_easy_escape "(3), " curl_easy_unescape "(3), "

.BR CURLOPT_TIMECONDITION "(3), " CURLOPT_TIMEVALUE "(3) "

lib/curl_setup.h

+18 −0
Original line number Diff line number Diff line
@@ -424,6 +424,24 @@ 
#endif

#define CURL_OFF_T_MIN (-CURL_OFF_T_MAX - CURL_OFF_T_C(1))



#if (SIZEOF_TIME_T == 4)

#  ifdef HAVE_TIME_T_UNSIGNED

#  define TIME_T_MAX UINT_MAX

#  define TIME_T_MIN 0

#  else

#  define TIME_T_MAX INT_MAX

#  define TIME_T_MIN INT_MIN

#  endif

#else

#  ifdef HAVE_TIME_T_UNSIGNED

#  define TIME_T_MAX 0xFFFFFFFFFFFFFFFF

#  define TIME_T_MIN 0

#  else

#  define TIME_T_MAX 0x7FFFFFFFFFFFFFFF

#  define TIME_T_MIN -0x10000000000000000

#  endif

#endif



/*

 * Arg 2 type for gethostname in case it hasn't been defined in config file.

 */

lib/parsedate.c

+55 −39
Original line number Diff line number Diff line
@@ -5,7 +5,7 @@ 
 *                            | (__| |_| |  _ <| |___

 *                             \___|\___/|_| \_\_____|

 *

 * Copyright (C) 1998 - 2017, Daniel Stenberg, <daniel@haxx.se>, et al.

 * Copyright (C) 1998 - 2018, Daniel Stenberg, <daniel@haxx.se>, et al.

 *

 * This software is licensed as described in the file COPYING, which

 * you should have received as part of this distribution. The terms
@@ -276,26 +276,23 @@ struct my_tm { 
  int tm_hour;

  int tm_mday;

  int tm_mon;

  int tm_year;

  int tm_year; /* full year */

};



/* struct tm to time since epoch in GMT time zone.

 * This is similar to the standard mktime function but for GMT only, and

 * doesn't suffer from the various bugs and portability problems that

 * some systems' implementations have.

 *

 * Returns 0 on success, otherwise non-zero.

 */

static time_t my_timegm(struct my_tm *tm)

static void my_timegm(struct my_tm *tm, time_t *t)

{

  static const int month_days_cumulative [12] =

    { 0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334 };

  int month, year, leap_days;



  if(tm->tm_year < 70)

    /* we don't support years before 1970 as they will cause this function

       to return a negative value */

    return -1;



  year = tm->tm_year + 1900;

  year = tm->tm_year;

  month = tm->tm_mon;

  if(month < 0) {

    year += (11 - month) / 12;
@@ -310,7 +307,7 @@ static time_t my_timegm(struct my_tm *tm) 
  leap_days = ((leap_days / 4) - (leap_days / 100) + (leap_days / 400)

               - (1969 / 4) + (1969 / 100) - (1969 / 400));



  return ((((time_t) (year - 1970) * 365

  *t = ((((time_t) (year - 1970) * 365

          + leap_days + month_days_cumulative[month] + tm->tm_mday - 1) * 24

         + tm->tm_hour) * 60 + tm->tm_min) * 60 + tm->tm_sec;

}
@@ -462,7 +459,7 @@ static int parsedate(const char *date, time_t *output) 
        if(!found && (dignext == DATE_YEAR) && (yearnum == -1)) {

          yearnum = val;

          found = TRUE;

          if(yearnum < 1900) {

          if(yearnum < 100) {

            if(yearnum > 70)

              yearnum += 1900;

            else
@@ -491,18 +488,39 @@ static int parsedate(const char *date, time_t *output) 
    /* lacks vital info, fail */

    return PARSEDATE_FAIL;



#if SIZEOF_TIME_T < 5

  /* 32 bit time_t can only hold dates to the beginning of 2038 */

  if(yearnum > 2037) {

    *output = 0x7fffffff;

    return PARSEDATE_LATER;

#ifdef HAVE_TIME_T_UNSIGNED

  if(yearnum < 1970) {

    /* only positive numbers cannot return earlier */

    *output = TIME_T_MIN;

    return PARSEDATE_SOONER;

  }

#endif



  if(yearnum < 1970) {

    *output = 0;

#if (SIZEOF_TIME_T < 5)



#if HAVE_TIME_T_UNSIGNED

  /* an unsigned 32 bit time_t can only hold dates to 2106 */

  if(yearnum > 2105) {

    *output = TIME_T_MAX;

    return PARSEDATE_LATER;

  }

#else

  /* a signed 32 bit time_t can only hold dates to the beginning of 2038 */

  if(yearnum > 2037) {

    *output = TIME_T_MAX;

    return PARSEDATE_LATER;

  }

  if(yearnum < 1903) {

    *output = TIME_T_MIN;

    return PARSEDATE_SOONER;

  }

#endif



#else

  /* The Gregorian calendar was introduced 1582 */

  if(yearnum < 1583)

    return PARSEDATE_FAIL;

#endif



  if((mdaynum > 31) || (monnum > 11) ||

     (hournum > 23) || (minnum > 59) || (secnum > 60))
@@ -513,30 +531,24 @@ static int parsedate(const char *date, time_t *output) 
  tm.tm_hour = hournum;

  tm.tm_mday = mdaynum;

  tm.tm_mon = monnum;

  tm.tm_year = yearnum - 1900;



  /* my_timegm() returns a time_t. time_t is often 32 bits, even on many

     architectures that feature 64 bit 'long'.

  tm.tm_year = yearnum;



     Some systems have 64 bit time_t and deal with years beyond 2038. However,

     even on some of the systems with 64 bit time_t mktime() returns -1 for

     dates beyond 03:14:07 UTC, January 19, 2038. (Such as AIX 5100-06)

  /* my_timegm() returns a time_t. time_t is often 32 bits, sometimes even on

     architectures that feature 64 bit 'long' but ultimately time_t is the

     correct data type to use.

  */

  t = my_timegm(&tm);



  /* time zone adjust (cast t to int to compare to negative one) */

  if(-1 != (int)t) {

  my_timegm(&tm, &t);



  /* Add the time zone diff between local time zone and GMT. */

    long delta = (long)(tzoff!=-1?tzoff:0);

  if(tzoff == -1)

    tzoff = 0;



    if((delta>0) && (t > LONG_MAX - delta)) {

      *output = 0x7fffffff;

  if((tzoff > 0) && (t > TIME_T_MAX - tzoff)) {

    *output = TIME_T_MAX;

    return PARSEDATE_LATER; /* time_t overflow */

  }



    t += delta;

  }

  t += tzoff;



  *output = t;
@@ -551,6 +563,10 @@ time_t curl_getdate(const char *p, const time_t *now) 


  switch(rc) {

  case PARSEDATE_OK:

    if(parsed == -1)

      /* avoid returning -1 for a working scenario */

      parsed++;

    /* fallthrough */

  case PARSEDATE_LATER:

  case PARSEDATE_SOONER:

    return parsed;

tests/data/test517

+0 −90
Original line number Diff line number Diff line
@@ -32,96 +32,6 @@ nothing 
# Verify data after the test has been "shot"

<verify>

<stdout>

0: Sun, 06 Nov 1994 08:49:37 GMT => 784111777

1: Sunday, 06-Nov-94 08:49:37 GMT => 784111777

2: Sun Nov  6 08:49:37 1994 => 784111777

3: 06 Nov 1994 08:49:37 GMT => 784111777

4: 06-Nov-94 08:49:37 GMT => 784111777

5: Nov  6 08:49:37 1994 => 784111777

6: 06 Nov 1994 08:49:37 => 784111777

7: 06-Nov-94 08:49:37 => 784111777

8: 1994 Nov 6 08:49:37 => 784111777

9: GMT 08:49:37 06-Nov-94 Sunday => 784111777

10: 94 6 Nov 08:49:37 => 784111777

11: 1994 Nov 6 => 784080000

12: 06-Nov-94 => 784080000

13: Sun Nov 6 94 => 784080000

14: 1994.Nov.6 => 784080000

15: Sun/Nov/6/94/GMT => 784080000

16: Sun, 06 Nov 1994 08:49:37 CET => 784108177

17: 06 Nov 1994 08:49:37 EST => 784129777

18: Sun, 12 Sep 2004 15:05:58 -0700 => 1095026758

19: Sat, 11 Sep 2004 21:32:11 +0200 => 1094931131

20: 20040912 15:05:58 -0700 => 1095026758

21: 20040911 +0200 => 1094853600

22: Thu, 01-Jan-1970 00:59:59 GMT => 3599

23: Thu, 01-Jan-1970 01:00:00 GMT => 3600

24: Sat, 15-Apr-17 21:01:22 GMT => 1492290082

25: Thu, 19-Apr-2007 16:00:00 GMT => 1176998400

26: Wed, 25 Apr 2007 21:02:13 GMT => 1177534933

27: Thu, 19/Apr\2007 16:00:00 GMT => 1176998400

28: Fri, 1 Jan 2010 01:01:50 GMT => 1262307710

29: Wednesday, 1-Jan-2003 00:00:00 GMT => 1041379200

30: , 1-Jan-2003 00:00:00 GMT => 1041379200

31:  1-Jan-2003 00:00:00 GMT => 1041379200

32: 1-Jan-2003 00:00:00 GMT => 1041379200

33: Wed,18-Apr-07 22:50:12 GMT => 1176936612

34: WillyWonka  , 18-Apr-07 22:50:12 GMT => -1

35: WillyWonka  , 18-Apr-07 22:50:12 => -1

36: WillyWonka  ,  18-apr-07   22:50:12 => -1

37: Mon, 18-Apr-1977 22:50:13 GMT => 230251813

38: Mon, 18-Apr-77 22:50:13 GMT => 230251813

39: "Sat, 15-Apr-17\"21:01:22\"GMT" => 1492290082

40: Partyday, 18- April-07 22:50:12 => -1

41: Partyday, 18 - Apri-07 22:50:12 => -1

42: Wednes, 1-Januar-2003 00:00:00 GMT => -1

43: Sat, 15-Apr-17 21:01:22 => 1492290082

44: Sat, 15-Apr-17 21:01:22 GMT-2 => 1492290082

45: Sat, 15-Apr-17 21:01:22 GMT BLAH => 1492290082

46: Sat, 15-Apr-17 21:01:22 GMT-0400 => 1492290082

47: Sat, 15-Apr-17 21:01:22 GMT-0400 (EDT) => 1492290082

48: Sat, 15-Apr-17 21:01:22 DST => -1

49: Sat, 15-Apr-17 21:01:22 -0400 => 1492304482

50: Sat, 15-Apr-17 21:01:22 (hello there) => -1

51: Sat, 15-Apr-17 21:01:22 11:22:33 => -1

52: Sat, 15-Apr-17 ::00 21:01:22 => -1

53: Sat, 15-Apr-17 boink:z 21:01:22 => -1

54: Sat, 15-Apr-17 91:22:33 21:01:22 => -1

55: Thu Apr 18 22:50:12 2007 GMT => 1176936612

56: 22:50:12 Thu Apr 18 2007 GMT => 1176936612

57: Thu 22:50:12 Apr 18 2007 GMT => 1176936612

58: Thu Apr 22:50:12 18 2007 GMT => 1176936612

59: Thu Apr 18 22:50:12 2007 GMT => 1176936612

60: Thu Apr 18 2007 22:50:12 GMT => 1176936612

61: Thu Apr 18 2007 GMT 22:50:12 => 1176936612

62: Sat, 15-Apr-17 21:01:22 GMT => 1492290082

63: 15-Sat, Apr-17 21:01:22 GMT => 1492290082

64: 15-Sat, Apr 21:01:22 GMT 17 => 1492290082

65: 15-Sat, Apr 21:01:22 GMT 2017 => 1492290082

66: 15 Apr 21:01:22 2017 => 1492290082

67: 15 17 Apr 21:01:22 => 1492290082

68: Apr 15 17 21:01:22 => 1492290082

69: Apr 15 21:01:22 17 => 1492290082

70: 2017 April 15 21:01:22 => -1

71: 15 April 2017 21:01:22 => -1

72: 98 April 17 21:01:22 => -1

73: Thu, 012-Aug-2008 20:49:07 GMT => 1218574147

74: Thu, 999999999999-Aug-2007 20:49:07 GMT => -1

75: Thu, 12-Aug-2007 20:61:99999999999 GMT => -1

76: IAintNoDateFool => -1

77: Thu Apr 18 22:50 2007 GMT => 1176936600

78: 20110623 12:34:56 => 1308832496

79: 20110632 12:34:56 => -1

80: 20110623 56:34:56 => -1

81: 20111323 12:34:56 => -1

82: 20110623 12:34:79 => -1

83: Wed, 31 Dec 2008 23:59:60 GMT => 1230768000

84: 20110623 12:3 => 1308830580

85: 20110623 1:3 => 1308790980

86: 20110623 1:30 => 1308792600

87: 20110623 12:12:3 => 1308831123

88: 20110623 01:12:3 => 1308791523

89: 20110623 01:99:30 => -1

</stdout>



# This test case previously tested an overflow case ("2094 Nov 6 =>