Re: [eigen] matrix exponential with c++0x

[Björn Piltz <bjornpiltz@xxxxxxxxxxxxxx>]

M_LOG2_E and M_PI (much less M_PIl) have never been defined in MSVC's stl afaik.

pow() has also been problematic since pow(double/float, int) wasn't defined in earlier versions, making "pow(2.0, 2)" not compile.

Right now some tests in unsupported don't compile. I'm attaching a patch.

2009/11/30 Benoit Jacob <jacob.benoit.1@xxxxxxxxx>

2009/11/30 Hauke Heibel <hauke.heibel@xxxxxxxxxxxxxx>:

> On Mon, Nov 30, 2009 at 2:37 PM, mmoll <Markus.Moll@xxxxxxxxxxxxxxxx> wrote:
>>
>> Hi
>>
>> Quoting Hauke Heibel <hauke.heibel@xxxxxxxxxxxxxx>:
>>
>> > First, M_LOG2_E suddenly was not defined anymore so I added
>> >
>> > #if defined(log2) && !defined(M_LOG2_E)
>> >   #define M_LOG2_E 0.693147180559945309417
>> > #endif
>>
>> As far as I know, M_LOG2_E was never standard? (not mentioned in C++98
>> or C99)
>
> Can you cite the exact draft and versions you have been looking at? I cannot
> even find M_PI.
>
> I was just looking since I found M_PIl (the last letter is an L for long) in
> one of our tests and as with M_LOG2_E it is not defined over here...

Everybody: Do *not* click on the following links,

http://download.tuxfamily.org/eigen/Cxx98.pdf
http://download.tuxfamily.org/eigen/Cxx03.pdf

as that *might* trigger an unauthorized download of copyrighted
material, which would obviously have the effect of putting the ISO out
of business and halting all further development of the C++ language
specification.

Thanks for your attention,
your friendly copyright enforcement agent.

Attachment: eigen-dev_rev2142.patch
Description: Binary data