[Webkit-unassigned] [Bug 9678] fmod function of MS CRT returns NaN for x / Inf

[bugzilla-daemon at opendarwin.org]

http://bugzilla.opendarwin.org/show_bug.cgi?id=9678


darin at apple.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |darin at apple.com
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1




------- Comment #5 from darin at apple.com  2006-07-04 10:43 PDT -------
(In reply to comment #4)
> I think Darin's suggestion was to just use an alternate function name besides
> fmod at the call sites, and for non-MSVC compilers it could just expand inline
> to fmod. It could just be called _fmod, or fremainder or something. I actually
> think this solution is fine though, so I'll just r+ it.

Maciej's right about what I was suggesting. But I think it's also OK to use a
trick so it can still be fmod at the call site. I would not have called the
other function _fmod, though, because those identifiers are specifically
reserved for the implementation! Maybe fmodWithWorkaround or something like
that.

Also, if we are doing a macro, a macro that takes parameters is safer than one
that just does the bare name. But I also think this is OK as-is, so I'll just
leave Maciej's review flag intact. We can always enhance the technique used in
MathExtras.h when and if we run into trouble.


-- 
Configure bugmail: http://bugzilla.opendarwin.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.