[GRASS-dev] Re: [GRASS GIS] #1351: v.build error in split_q.c

[GRASS GIS]

#1351: v.build error in split_q.c
-------------------------------+--------------------------------------------
 Reporter:  epatton            |       Owner:  grass-dev@…              
     Type:  defect             |      Status:  new                      
 Priority:  normal             |   Milestone:  6.4.2                    
Component:  Vector             |     Version:  svn-develbranch6         
 Keywords:  v.build, topology  |    Platform:  Linux                    
      Cpu:  x86-32             |  
-------------------------------+--------------------------------------------

Comment(by mmetz):

 Replying to [ticket:1351 epatton]:
 > Hello,
 >
 > I'm trying to rebuild topology on a vector, and when doing so, I run
 into the attached error (full backtrace from gdb).
 >
 > Aside from the error itself, it's quite odd that the vector's topology
 is damaged in the first place; I guess that's a separate issue.
 >
 I guess that the import did not finish successfully and left behind a
 broken vector, indicated by that warning about a coor file size mismatch.

 This assertion error

 {{{
 split_q.c:69: RTreeClassify: Assertion `!p->taken[i]' failed.
 }}}

 occurred in the past in some cases of reprojection. Since the vector name
 is GSHHS_land, I guess that it has been imported with v.in.gshhs which
 does projection on the fly but does not check if the reprojection is
 legal. If this GSHHS_Land vector has been imported with v.in.gshhs into a
 non-latlon projection, then it needs to be confirmed first that the extent
 chosen for import can actually be represented in the target location. If
 not, v.build is actually right in refusing to build topology, although the
 error message is as cryptic as it can get.

 Markus M

-- 
Ticket URL: <http://trac.osgeo.org/grass/ticket/1351#comment:1>
GRASS GIS <http://grass.osgeo.org>