On Tuesday 15 May 2007 10:52:34 klaas [dot] gadeyne [..] ... wrote:
> For more information about this bug, visit
> A comment was
> added:
> ------- Comment #2 from klaas [dot] gadeyne [..] ... 2007-05-15 10:52
>
> (In reply to comment #1)
>
> > Created an attachment (id=129)
>
> --> (http://www.fmtc.be/orocos-bugzilla/attachment.cgi?id=129&action=view)
> [details]
>
> > Patch provided by Erwin to build BFL for Windows.
>
> Probability(double p)
> {
> - assert( (std::isfinite(p) != 0) && (p >= 0));
> + assert( p >= 0 );
> _prob = p;
>
> isfinite seems to be valid C99, so I would rather see something like

>
> #ifdef _NONPORTABLE_COMPILER_
> assert( p >= 0 );
> #else
> assert( (std::isfinite(p) != 0) && (p >= 0));
> #endif
C99 is NOT C++. These standards are even conflicting sometimes. In fact
Microsoft calls std::isfinite an incompatibility. Documentation of g++
states that they sometimes offer more functions in order to be more
compatible with C99.

But whatever the truth is...
Here is how you can detect the compiler :

#ifdef _MSC_VER
assert( p >= 0 );
#else
assert( (std::isfinite(p) != 0) && (p >= 0));
#endif

>
> instead of just throwing away this check?
Is the reason behind this class only these checks ?

_______________________________________________
I hereby promise not to top-post on the
BFL mailing list
BFL [..] ...
http://lists.mech.kuleuven.be/mailman/listinfo/bfl

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm

For more information about this bug, visit
A new bug was added:
Summary: Building BFL in Windows
Product: BFL
Version: trunk
Platform: All
OS/Version: All
Status: NEW
Severity: normal
Priority: P2
Component: build system
AssignedTo: bfl [..] ...
ReportedBy: wim [dot] meeussen [..] ...

Since the port to CMake, it should be easier to use BFL in Windows. In this
bugreport we can collect all patches etc to make this possible.