make all negative handles invalid, not just -1

Since opposite_halfedge_handle for a handle with idx -1 produces a handle with idx -2, it can be helpful to detect handles with -2 as index as invalid as well.

See merge request !96

Gilab windows ci

Windows Ci via gitlab

See merge request !95

added missing ifdef for debug



See merge request !94

Fix for handling negative indices in OBJ loader

This fix resolves an issue that occurrs with the current code that uses two passes to parse the OBJ - see my previous merge request:
https://graphics.rwth-aachen.de:9000/OpenMesh/OpenMesh/merge_requests/84

Sorry for the trouble! :-(

To verify that the negative indices were really not working any more, and to check that this fix resolves the issue, I used the Cornell Box example from the tiny OBJ loader project:
https://github.com/syoyo/tinyobjloader/blob/master/models/cornell_box.obj

See merge request !93

range check for vertex colors and normals in OBJ loader

This merge request adds explicit range checks and error reporting for the reading of colors and normals. The code is copied from texture coordinates, where a similar check already existed.

See merge request !92

Fixed comment in OBJ loader



See merge request !91

Fixed broken indices for some cases of OBJ parsing by using two passes

(Continuing fixing / discussion from https://graphics.rwth-aachen.de:9000/OpenMesh/OpenMesh/merge_requests/51)

This merge request is fixing an issue that came up as a follow-up of the following merge request:
 https://graphics.rwth-aachen.de:9000/OpenMesh/OpenMesh/merge_requests/51

The problem it fixes occurs for some kinds of OBJ data, where vertices and face data are stored interleaved and, at the same time, the mesh contains some non-manifold geometry.  As non-manifold geometry triggers the on-the-fly creation of new vertices during parsing, the indices from the OBJ file do not match any more and the result gets messed up.

The straightforward solution applied on this branch is to parse to whole OBJ file in two passes, first the vertices and then the material data and faces. This way, possible extra vertices will always be added to the back, not interfering with indices from the OBJ file

I tested the performance on a 1.1 GB OBJ file. The loading time on my machine increased from ~59 seconds to ~65 seconds, so I believe the performance overhead could probably be considered acceptable.

Looking forward to your feedback!


See merge request !84

Updated documentation of BaseKernel's property functions to hint at PropertyManager.



See merge request !90

Build python windows

Fixes a linker problem as well as issue #29.

See merge request !89

# Conflicts:
#   src/Python/Vector.hh

Added a little safeguard against GCC optimizer bug to CMakeLists.txt.



See merge request !88

g++ 4.7 compile fix in vector11

With g++ 4.7, the following error occurs when trying to compile code that uses Vector11:

```
../OpenMesh/Core/Geometry/Vector11T.hh: In function ‘decltype ((_v1 % _v2)) OpenMesh::cross(const OpenMesh::VectorT<LScalar, DIM>&, const OpenMesh::VectorT<RScalar, DIM>&) [with LScalar = float; RScalar = float; int DIM = 3; decltype ((_v1 % _v2)) = OpenMesh::VectorT<float, 3>]’:
../OpenMesh/Core/Geometry/Vector11T.hh:89:19: error: ‘OpenMesh::VectorT<float, 3>::container OpenMesh::VectorT<float, 3>::values_’ is private
```

It looks like this version of g++ would not be able to deal with the decltype statement used when overriding the "%" operator, where the "this->values_" part is not working properly. This issue should be fixed with this merge request.

See merge request !87

Added unit test to trigger GCC alignment bug. See #32.

Merge this even if CI unit tests fail. Failed unit test expose a compiler bug.

See merge request !86

Propmgr

Added some more convenience to the PropertyManager class.

See merge request !85

Made member function pointer types explicit. This gets rid of a preprocessor conditional and sidesteps a bug in msvc.