[DateTime] from time_t to ptime and back - or not
[I am using Boost 1.43.0 on linux, gcc 4.4.1] I am having trouble with date_time's ptime. I am passing it a time_t, and store that in a ptime object. When I retrieve the time_t, I am getting a different value than the one I passed to it originally. (the difference seems to be 2 hours - in all probability due to UTC vs local time issues). Code: time_t now = time(NULL); boost::posix_time::ptime pt = boost::posix_time::from_time_t(now); struct tm tm = to_tm(pt); time_t should_be_now = mktime(&tm); BOOST_CHECK_EQUAL(now, should_be_now); fails for me: now is by 7200 seconds (i.e. 2 hours) ahead of. Is this a bug in date_time or am I just failing to use it properly? Thank you, and best regards Christoph
On Fri, Aug 20, 2010 at 11:23 AM, Christoph Duelli <duelli@melosgmbh.de> wrote:
Code: time_t now = time(NULL); boost::posix_time::ptime pt = boost::posix_time::from_time_t(now); struct tm tm = to_tm(pt); time_t should_be_now = mktime(&tm); BOOST_CHECK_EQUAL(now, should_be_now);
fails for me: now is by 7200 seconds (i.e. 2 hours) ahead of.
Is this a bug in date_time or am I just failing to use it properly?
I am by no means an expert, so take this with a grain of salt. I would guess the problem is with your assumptions. The documentation on to_tm (http://www.boost.org/doc/libs/1_44_0/doc/html/date_time/posix_time.html#ptim...) says nothing about timezones, and a ptime stores UTC, so I'd guess the struct tm members would be set to what that point in time would be in UTC. mktime, on the other hand, assumes the tm represents your local time (http://cplusplus.com/reference/clibrary/ctime/mktime/). Whether I'm right or wrong should be easily testable, especially for you (since you know what timezone your PC is set to). Add to your program a few output lines showing members of the tm and see if it looks right for your local time.
John B. Turpish wrote: Thank you.
On Fri, Aug 20, 2010 at 11:23 AM, Christoph Duelli <duelli@melosgmbh.de> wrote:
Code: time_t now = time(NULL); boost::posix_time::ptime pt = boost::posix_time::from_time_t(now); struct tm tm = to_tm(pt); time_t should_be_now = mktime(&tm); BOOST_CHECK_EQUAL(now, should_be_now);
fails for me: now is by 7200 seconds (i.e. 2 hours) ahead of.
Is this a bug in date_time or am I just failing to use it properly?
I am by no means an expert, so take this with a grain of salt. [snip] mktime, on the other hand, assumes the tm represents your local time (http://cplusplus.com/reference/clibrary/ctime/mktime/).
Whether I'm right or wrong should be easily testable, especially for you (since you know
what timezone your PC is set to). Add to your program a few output lines showing members of the tm and see if it looks right for your local time. However, that does not really help, because the struct tm that Boost ptime
Yes, with t = time(0) t == mktime(localtime(t)) but t != mktime(gmtime(t)) produces is different from both gmtime and localtime: DUMP gmtime t.tm_sec=50 t.tm_min=31 t.tm_hour=9 t.tm_mday=23 t.tm_mon=7 t.tm_year=110 t.tm_wday=1 t.tm_yday=234 t.tm_isdst=0 t.tm_gmtoff=0 t.tm_zone=GMT diff(time,mktime(gmtime(time)))=3600 DUMP localtime t.tm_sec=50 t.tm_min=31 t.tm_hour=11 t.tm_mday=23 t.tm_mon=7 t.tm_year=110 t.tm_wday=1 t.tm_yday=234 t.tm_isdst=1 t.tm_gmtoff=7200 t.tm_zone=CEST diff(time,mktime(localtime(time)))=0 DUMP boost from_time_t -> to_tm t.tm_sec=50 t.tm_min=31 t.tm_hour=9 t.tm_mday=23 t.tm_mon=7 t.tm_year=110 t.tm_wday=1 t.tm_yday=234 t.tm_isdst=-1 t.tm_gmtoff=0 t.tm_zone=NULL diff(time,mktime(to_tm(from_time_t(time))))=7200 It is not obvious how to fix the tm obtained from to_tm. Probably it is easier to fix the resulting time_t. I guess I will settle for the approach given in http://lists.boost.org/boost-users/2007/07/29567.php and stop using from_time_t and to_tm Best regards Christoph
On 2010-08-20 17:23, Christoph Duelli wrote:
[I am using Boost 1.43.0 on linux, gcc 4.4.1]
I am having trouble with date_time's ptime. I am passing it a time_t, and store that in a ptime object. When I retrieve the time_t, I am getting a different value than the one I passed to it originally. (the difference seems to be 2 hours - in all probability due to UTC vs local time issues).
Code: time_t now = time(NULL); boost::posix_time::ptime pt = boost::posix_time::from_time_t(now); struct tm tm = to_tm(pt); time_t should_be_now = mktime(&tm); BOOST_CHECK_EQUAL(now, should_be_now);
fails for me: now is by 7200 seconds (i.e. 2 hours) ahead of.
Is this a bug in date_time or am I just failing to use it properly?
Since struct tm is so weird, here is what I do: boost::posix_time::ptime epoch(boost::gregorian::date(1970,1,1)); time_t also_now = (pt - epoch).total_seconds(); HTH Roland
participants (1)
-
Christoph Duelli -
John B. Turpish -
Roland Bock