octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Compiling with Qhull


From: Ben Abbott
Subject: Re: Compiling with Qhull
Date: Sat, 17 Dec 2011 16:55:04 -0500

On Dec 17, 2011, at 1:09 PM, John W. Eaton wrote:

> On 17-Dec-2011, Ben Abbott wrote:
> 
> | After building with gcc 4.5, I deactivated qhull 2010.1 and activated 
> 2011.2. Then I ran "make check". I encountered 13 failures related to qhull.
> 
> What exactly did you do?  Did you remove the old version?
> 
> | Looking at the Macports port-file for qhull ...
> | 
> |     https://trac.macports.org/browser/trunk/dports/math/qhull/Portfile
> | 
> | ... I see that the standard "qhull.h" is being used (not 
> qhull.h-deprecated) by macports.
> 
> What do you mean by standard?  If macports already renamed libqhull.h
> to qhull.h, then I think you are OK, except that you probably need to
> modify the src/DLD-FUNCTIONS to have
> 
>  #define qh_QHpointer 1
> 
> before including any Qhull header files.  I don't know how to write a
> configure test so that this can happen automatically.
> 
> Or is the qhull.h file that you have from some earlier release?  I
> don't think we can hope to handle the situation of multiple versions
> of qhull installed at once.
> 
> | Rik, I don't recall which package management system you are using. Do you 
> know if your qhull 2011 uses the standard qhull.h or the deprecated version?
> 
> I don't see any point in bothering with qhull.h-deprecated.  All it
> does is include libqhull.h, which Octave's configure script already
> checks for.
> 
> jwe

ah-ha ... I should have read the emails a bit more carefully. I made the change 
below.

$ hg diff
diff --git a/src/DLD-FUNCTIONS/oct-qhull.h b/src/DLD-FUNCTIONS/oct-qhull.h
--- a/src/DLD-FUNCTIONS/oct-qhull.h
+++ b/src/DLD-FUNCTIONS/oct-qhull.h
@@ -20,6 +20,8 @@
 
 */
 
+#define qh_QHpointer 1
+
 #if !defined (octave_oct_qhull_h)
 #define octave_oct_qhull_h 1

Now, I only have one test fail.

  src/DLD-FUNCTIONS/convhulln.cc ......................... PASS    2/3    FAIL 1

>From fntests.log ...

  29   ***** testif HAVE_QHULL
  30  cube = [0 0 0;1 0 0;1 1 0;0 1 0;0 0 1;1 0 1;1 1 1;0 1 1];
  31  [h, v] = convhulln (cube);
  32  assert (size (h), [6 4]);
  33  h = sortrows (sort (h, 2), [1:4]);
  34  assert (h, [1 2 3 4; 1 2 5 6; 1 4 5 8; 2 3 6 7; 3 4 7 8; 5 6 7 8]);
  35  assert (v, 1, 10*eps);
  36 !!!!! test failed
  37 assert (size (h),[6, 4]) expected
  38    6   4
  39 but got
  40    12    3
  41 values do not match

Ben



reply via email to

[Prev in Thread] Current Thread [Next in Thread]