FFT patch again

[Ulf Lorenz]

Hi,

as requested, I tried to add my latest fft patch from some time ago to
the current sources. It seems to work, and the patch is attached.

I would suggest to inspect the diff first, though. I think most of the
differences are formatting issues, and only test_fftw.cc is actually
important (also I applied the patch sent a few minutes ago). Still, I
wanted to meddle as little as possible with the file before sending it.


Ulf

[Juan Jose Garcia-Ripoll]

On Tue, Oct 22, 2013 at 9:15 PM, Ulf Lorenz <u...@wavepacket.org> wrote:

as requested, I tried to add my latest fft patch from some time ago to
the current sources. It seems to work, and the patch is attached.

It is now in github.

--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain) 
http://juanjose.garciaripoll.googlepages.com

[Ulf Lorenz]

On Mon, 28 Oct 2013 17:26:36 +0100
Juan Jose Garcia-Ripoll <juanjose.g...@gmail.com> wrote:

> On Tue, Oct 22, 2013 at 9:15 PM, Ulf Lorenz <u...@wavepacket.org>
> wrote:
>
> > as requested, I tried to add my latest fft patch from some time ago
> > to the current sources. It seems to work, and the patch is attached.
> >
>
> It is now in github.

Thanks. That is now exactly one day after my release, but at least the
next version of my program will not have a crutch of copy&paste code. :)

Regarding release, I do feel obliged now to spend on the order of 2-3
weeks of coding on the tensor library, so if you have something urgent
to do, I can try my best. Otherwise, I pick one of the tracker issues
that I submitted. :)


Ulf

[Juan Jose Garcia-Ripoll]

On Mon, Oct 28, 2013 at 8:26 PM, Ulf Lorenz <u...@wavepacket.org> wrote:

Regarding release, I do feel obliged now to spend on the order of 2-3
weeks of coding on the tensor library, so if you have something urgent
to do, I can try my best. Otherwise, I pick one of the tracker issues
that I submitted. :)

There is a problem with sparse matrices ---on my system one of the tests fails. Apart from that, I have no other priority than finding time to work on the webpage and producing a release.

[Ulf Lorenz]

On Tue, 29 Oct 2013 19:14:33 +0100

Juan Jose Garcia-Ripoll <juanjose.g...@gmail.com> wrote:

> There is a problem with sparse matrices ---on my system one of the
> tests fails. Apart from that, I have no other priority than finding
> time to work on the webpage and producing a release.

The problem was that the incorrect constructor was called. I added an
assert that should prevent this error, used the correct constructor, and
did a few layout fixes and documentation updates. Patch attached.


Ulf

[Juan Jose Garcia-Ripoll]

Thanks! It is now in github.