On 2010-01-13 18:40-0800 David MacMahon wrote:
>
> On Jan 13, 2010, at 17:37 , Alan W. Irwin wrote:
>
>> That is clearly in error for qhull (it should point to my special install
>> location of /home/software/qhull/install/lib for qhull-2010.1) which
>> confirms the issue you have reported. I hop
On Jan 13, 2010, at 17:37 , Alan W. Irwin wrote:
> That is clearly in error for qhull (it should point to my special
> install
> location of /home/software/qhull/install/lib for qhull-2010.1) which
> confirms the issue you have reported. I hope to fix that shortly.
Great, thanks!
I also foun
On 2010-01-13 09:46-0800 Alan W. Irwin wrote:
> Hi David:
>
> As a temporary workaround, please try the new CMake-based build system
> for the installed examples. It's undocumented for now, but the steps
> (after "make install" for the core build sets up that build system) is
> to cd to an initia
On 2010-01-12 23:12-0800 David MacMahon wrote:
> I recently built and installed plplot from subversion. I then tried
> to rebuild the examples and discovered that "pkg-config --libs"
> outputs insufficient information...
>
> $ make x01c
> /usr/bin/cc x01c.c -o x01c `PKG_CONFIG_PATH=/Users/davidm
The attached patch (plus an "uncrustify" patch, I'll remember next
time!) clarifies (I hope!) comments in plplot.h for the plf2eval-like
functions. It also defines a typedef for pointers to such
functions. The typedef can be used to simplify parameters lists for
functions that accept such
Humbly offered for your consideration...
After the attached patch (plus the other attached "uncrustify"
patch), plAlloc2dGrid will allocate one block of memory that is big
enough for nx pointers plus nx*ny PLFLTs. This means only one call
to malloc and only one call to free no matter how
