Re: Geometry Chapter of the manual

[John W. Eaton]

On 24-Jul-2007, Quentin Spencer wrote:

| John W. Eaton wrote:
| 
| >On 16-Jul-2007, David Bateman wrote:
| >
| >| Thanking about what to do with the Geometry chapter of the manual that
| >| only contains the function polyarea, it seems to me the logical thing to
| >| do would be to port the octave-forge functions convhull, convhulln,
| >| delaunay, delaunay3, delaunayn, griddata, tsearch, voronoi and voronoin
| >| and make a real chapter out of it. These are all core functions of
| >| matlab. However, the reason I didn't do this in the past is that this
| >| would require a dependency on the QHULL package to be added to Octave.
| >| Are we willing to accept an additional dependency on QHULL? If so I'm
| >| willing to do the conversion work..
| >
| >I don't have a problem with the additional dependency provided that it
| >is optional.
| >  
| >
| 
| The addition of the sparse matrix dependencies seemed to cause a fair 
| amount of confusion on the help list. I have no problem with adding more 
| dependencies, but maybe we need to make it clearer to people who want to 
| compile octave themselves that they are optional. I don't know how to do 
| that--any ideas?

The current messages look something like

  WARNING: PCRE library not found.  This will result in some loss of
  functionality for the regular expression matching functions.

I don't know what else to add.  To me, "loss of functionality" doesn't
imply that the build will fail.  If someone has an idea for better
wordking, then submit a patch for configure.in.

| Maybe there should be a document somewhere telling 
| people how to install the development packages for common 
| distributions--that one seems to trip people up a lot.

I think install.txi is the right place for this information (the
INSTALL.OCTAVE file is generated from install.txi).  I'd consider a
patch.

jwe