[GRASS-dev] [GRASS GIS] #2442: np.float128 does not always exist

[GRASS GIS]

#2442: np.float128 does not always exist
-----------------------+----------------------------------------------------
 Reporter:  jradinger  |       Owner:  grass-dev@…              
     Type:  defect     |      Status:  new                      
 Priority:  normal     |   Milestone:                           
Component:  Default    |     Version:  svn-trunk                
 Keywords:             |    Platform:  Unspecified              
      Cpu:  x86-32     |  
-----------------------+----------------------------------------------------
 When compiling the latest SVN (62197) of GRASS7 trunk I run into following
 error (first among the error list, mostly in the temporal dir):


 {{{
 Traceback (most recent call last):
   File "/usr/local/src/grass7_trunk/dist.i686-pc-linux-
 gnu/scripts/t.vect.univar", line 66, in <module>
     import grass.temporal as tgis
   File "/usr/local/src/grass7_trunk/dist.i686-pc-linux-
 gnu/etc/python/grass/temporal/__init__.py", line 29, in <module>
     from temporal_vector_algebra import *
   File "/usr/local/src/grass7_trunk/dist.i686-pc-linux-
 gnu/etc/python/grass/temporal/temporal_vector_algebra.py", line 420, in
 <module>
     import grass.pygrass.modules as pygrass
   File "/usr/local/src/grass7_trunk/dist.i686-pc-linux-
 gnu/etc/python/grass/pygrass/__init__.py", line 15, in <module>
     from . import raster
   File "/usr/local/src/grass7_trunk/dist.i686-pc-linux-
 gnu/etc/python/grass/pygrass/raster/__init__.py", line 35, in <module>
     from grass.pygrass.raster.buffer import Buffer
   File "/usr/local/src/grass7_trunk/dist.i686-pc-linux-
 gnu/etc/python/grass/pygrass/raster/buffer.py", line 14, in <module>
     DCELL = (np.float64, np.float128)
 AttributeError: 'module' object has no attribute 'float128'
 make[1]: *** [t.vect.univar.tmp.html] Error 1
 }}}

 Here I am running Python 2.7.3 with numpy 1.9.0.

 Here [http://stackoverflow.com/questions/9062562/what-is-the-internal-
 precision-of-numpy-float128] it says that is is related to the
 architecture. In my case I am on a 32 bit system.

 Changing from np.float128() to np.longdouble() (which exists on my system)
 should solve the problem.

-- 
Ticket URL: <http://trac.osgeo.org/grass/ticket/2442>
GRASS GIS <http://grass.osgeo.org>