On Sat, Mar 8, 2014 at 4:22 PM, Francesco P. Lovergine <[email protected]>wrote:
> On Sat, Mar 08, 2014 at 03:05:22PM +0100, Rashad M wrote: > > > > > > About Ossim (the lib) I'm not sure if supporting the MPI flavor should > > > be a priority, for instance. I would avoid having still another > > > horror story as in the HDF5 case. > > > > > Can you give your advice? > > > > > > No supporting MPI could be avoided for now. I too had such bad stories > in > > packaging ossimpreadator and also I think OTB doesnt rely on OSSIM's MPI > > code. But I would ask OTB devs to have a word on it. > > > > About Ossim vs Geos, I wonder if the C++ interface is truly required. > I actually linked both C/C++ libraries without a deeper analysis. > Officially, > Geos deprecates the use of the C++ interface by third parties, but > I'm not sure about that. If you did verify its status it would help. > > I found that Geos is used in ossimPolyArea2d class which inturn is used ossimPolyCutter and ossimTrimFilter which are truly required. Although C++ code of geos that is used and I didnt find any issues when compiling with gcc. If the C++ interface is deprecated by Geos that make it intolerable for any existing libraries using that part and I think ossim will also be replacing geos like all others. Copy paste from geos page[1]: "" The recommended low-level interface to the GEOS library is the simplified C wrapper interface<http://geos.refractions.net/ro/doxygen_docs/html/c_iface.html>. This will ensure stability of the API and the ABI of the library during performance improvements that will likely change classes definitions. If you prefer troubles you can use the C++ interface<http://geos.refractions.net/ro/doxygen_docs/html/cpp_iface.html> "" This is not that comfortable in the main page of documentation and keep people away. I really dont understand what they mean by "troubles". Is it incorrectly spelled some how? ( no offense. just a joke) [1] http://geos.refractions.net/ro/doxygen_docs/html/index.html > > -- > Francesco P. Lovergine > -- Regards, Rashad
