[GRASS-user] starspan with grass support compilation failure

[Jonathan Greenberg]

Folks:

    I would LOVE to see someone take over development on starspan -- I 
did submit a OSGEO application but it came back denied, more or less, 
because of lack of community support.  There are a lot of bugs and user 
interface issues still remaining that would be nice to get cleaned up, 
but I'm not a C programmer and, while I developed the algorithm itself, 
I was not the coder.

    If anyone is interested in working on the development, please email 
me!  If you all are not sure what STARSPAN does, it is a flexible way to 
get raster data (raw or summarized) associated with vector coverages 
(similar to Arc's zonalstats, although with more output types).  E.g., 
if you want to get the average pixel values from a SET of raster images 
underlying a single vector coverage, starspan will spit out a CSV 
containing the vector's original database, along with a column for the 
average pixel value for that polygon.  The algorithm tends to be very 
fast, especially in cases where the polygon coverage is a fraction of 
the raster coverage (which tends to be the case in many problems).

--j

Dylan Beaudette wrote:
> On Thu, Jan 29, 2009 at 4:22 AM, Nikos Alexandris
> <nikos.alexandris at felis.uni-freiburg.de> wrote:
>   
>> On Wed, 2009-01-28 at 18:05 +0100, Nikos Alexandris wrote:
>>     
>>> Not sure if I should post this here. I have a problem compiling starspan
>>> with support for GRASS.
>>>
>>> # configuring reports no problems
>>> ./configure --with-grass=/usr/local/grass-6.5.svn
>>>
>>> # attempt to compile...
>>> make
>>>
>>> [...]
>>> src/stats/Stats.cc:232: error: 'sort' was not declared in this scope
>>> make: *** [Stats.o] Error 1
>>>
>>> The same problem appears also with a bare "./configure".
>>>       
>> The same error appears with all of the three versions of starspan
>> 1.2.01, 1.2.02, 1.2.03.
>>
>> Perhaps it's not starspan-related.
>>
>>     
>
> Hmm. Check the archives for Jonathan Greenberg -- he is one of the
> lead developers. It would be nice to get starspan into OSGEO so that
> questions / code patches ect. can be handled in an efficient manner.
> Starspan is an emmensly important GIS tool, but suffers from a lack of
> developers, documentation, and bug handling.
>
> Perhaps I should take these comments to the OSGEO mailing list... What
> do you think Jonathan?
>
> Cheers,
>
> Dylan
>