The release of boost 1.67 fixes an issue where floating point values are automatically converted into integers for the constructors of boost::posix_time::time_duration, and helper functions like boost::posix_time::microseconds(long). This causes errors when special values (NaN, -∞, etc) are passed to it.
Currently the value double limit value in Baz Burster get passed to boost::posix_time::microseconds(limit * 1e6), which isn't auto converting anymore.
This is the current compilation error.
/home/user/projects/gr-baz/src/gr-baz/lib/baz_burster.cc: In member function ‘virtual int baz_burster::general_work(int, gr_vector_int&, gr_vector_const_void_star&, gr_vector_void_star&)’:
/home/user/projects/gr-baz/src/gr-baz/lib/baz_burster.cc:225:62: error: no matching function for call to ‘boost::date_time::subsecond_duration<boost::posix_time::time_duration, 1000000>::subsecond_duration(double)’
if (diff >= boost::posix_time::microseconds(limit * 1e6))
The release of boost 1.67 fixes an issue where floating point values are automatically converted into integers for the constructors of
boost::posix_time::time_duration
, and helper functions likeboost::posix_time::microseconds(long)
. This causes errors when special values (NaN, -∞, etc) are passed to it.Currently the value
double limit
value in Baz Burster get passed toboost::posix_time::microseconds(limit * 1e6)
, which isn't auto converting anymore.This is the current compilation error.