Re: [Boost-users] [date_time/posix_time] [Fwd: Warnings with microsec_time_clock.hpp with VC8.0]

From: Jeff Garland (jeff_at_[hidden])
Date: 2006-10-01 12:58:04

Bill Somerville wrote:
> 
> ------------------------------------------------------------------------
> 
> Subject:
> Warnings with microsec_time_clock.hpp with VC8.0
> From:
> "David Wyles" <dwyles_at_[hidden]>
> Date:
> Thu, 28 Sep 2006 15:33:48 +0100
> To:
> "Bill Somerville" <bsomerville_at_[hidden]>
> 
> To:
> "Bill Somerville" <bsomerville_at_[hidden]>
> 
> 
> I have an issue with a cast warning between a uint64_t and a uint32_t
> 
>  
> 
> Line: 156---
> 
>       boost::uint64_t filetime = ft.dwHighDateTime;
> 
>       filetime = filetime << 32;
> 
>       filetime += ft.dwLowDateTime;
> 
>       filetime -= OFFSET;
> 
>       // filetime now holds 100-nanoseconds since 1970-Jan-01
> 
>  
> 
>       boost::uint32_t sub_sec = (filetime % 10000000) / 10; // microseconds
> 
> ^^^^^^^^^^^^^^^^^^^^^^^^
> 
> cast warning here as filetime is a uint64_t and subsec is not.
> 
>  
> 
> Is this already fixed?

Yes, the microsec_time_clock is fixed in CVS and will be in 1.34.  I don't see 
  any unit32 calcs in from ftime.  Maybe you can try out the latest cvs or 
1.34 rc and see?

Jeff