The following code was obtained from the Interface.ImportMenu module and is responsible for loading an .IGES file into the application. This part of the code uses a previously created Reader object to call an internal method named TransferList() to obtain all the loaded entities of the model, including all Manifold Solids, Faces, Surfaces, Edges, Vertices and other entities:

# Checking if the file were read successfully and creating a OpenCASCADE shape object:
shape = None
if status == IFSelect_RetDone:
    Reader.TransferList(Reader.GiveList('xst-model-all'))
    for i in range(1, Reader.NbShapes()+1):
        parent.shapeList.append(Reader.Shape(i))
    shape = Reader.Shape(1)

However, the shape variable only stores the first loaded shape: Reader.Shape(1) and thus, the following code:

# Displaying the loaded entity as an OpenCASCADE shape:
parent.canvas._display.DisplayShape(shape, update=True)

Just display the first entity, even if the loaded .IGES file contains multiple Manifold Solids.

The aim of this Issue is to implement a logic to load as many Manifold Solids as needed, so all of them can be used by the discretization methods.

The application does not support .IGES files that are based on entities which differ from the Manifold Solid entities. Due to that limitation, a warning message must be implemented to notify users every time they open a non-maifold solid based .IGES file.

Some CAD models in IGES file format uses different units to specify vertices, edges and faces. These units needs to be considered during the evaluation of the discrete surfaces.

When trying to auto-discretize a surface using the Discretization.DiscretizeModel.discretizeFace() method, if the default operation consists in using a points/cm value for the density argument, then the generated points will suffer from a minor displacement from some of the edges of the face.

Maybe this is mainly caused by the "Point in Polygon" algorithm that is used, but this is most likely to be a mathematical error while computing the edges.

The auto discretization process needs to be implemented in two different versions: one that considers only a grid of n x n points laid on a surface and another version that allows a consistent density of points. At this time, only the second version of the auto discretization process is implemented.

The implementation needs to affects how the Discretization.DiscretizeModel module works.

The Menu Bar has some options for changing the default selection method. The selection method determines what kind of object will be selected after a mouse click. The possible selection methods are:

• Manifold Solid Selection.
• Face Selection.
• Edge Selection.
• Vertex Selection.

Each selection method needs to be implemented in the menu bar by calling the following methods:

parent.canvas._display.setSelectionModeNeutral()
parent.canvas._display.setSelectionModeFace()
parent.canvas._display.setSelectionModeEdge()
parent.canvas._display.setSelectionModeVertex()

The tmp/ folder is a directory used for store information about non-planar discretized surfaces and loops. An implementation that doesn't require such directory will faster the discretization and will allow the software to run into folders that doesn't require administrative privileges on Windows.

In the Actions.ActionList module, the implemented closeActionProcedure() method needs to fix some unknown error, most common after changing the default visualization method of the CAD.

After closing some opened files, the actual visualization does not clean properly, usually, showing a "ghost" or faded image from the previous IGES file that was used.

More info:
If a user tries to open another IGES file, the new CAD model will be displayed over the preview one, which is, indeed, a notable bug.

After implementing the auto-discretization process, new methods for individual discretizations will need to be implemented for handling other kinds of geometries.

Such geometries include:

• Individual Plane Surfaces
• Cylindrical and Conical Elements
• The Sphere IGES Entity (Unsupported in the current state of the project)

Some of the implementations need only to be adapted since the auto-discretization process already has some functions for dealing with some of the individual surfaces.

The random defects needs to be implemented.

The main goal here is to provide a useful sidebar/interface that allow users to apply random manufacturing erros in their cloud points data. This random errors will be generated by selecting a random direction and a random displacement that will be applied to all points in that specific direction. Further enhancements will be implemented, as selecting a normal or fixed direction.

The translational defects functionality for individual faces needs to be implemented.

The main issue in this is how to detect the normal direction given an IGES surface entity and a point that lies on that surface. Some methods using derivatives was presented in "The NURBS Book". These implementations may solve the present issue.

The current implementation of the Rotational Defects Side Widget allows the user to input three values (x, y and z) for the desired rotation angle. These values are stored as fields of the rotationalDefectsMenu class, avaliable at the Interface.RotationalDefectsMenu module. These values need to be verified before the rotatePoints() method converts the self.xAngle, self.yAngle and self.zAngle to a floating point number, otherwise, an exception is generated and the software crashes.

We need a side widget displaying the currently generated points on each face.
Each point needs to be selectable and must contain information about their x, y and z coordinates.

In the same way as the Issue #3, the rotational defects needs to be implemented.

The main goal here is to create a way for selecting individual edges or individual vetices to be used as references for the rotation axis. This can be implemented using the compute_bbox() function that is avaliable in the core_display_callbacks.py example of the pythonocc-core package.

Note that this function can be implemented directly in the source code of this project, without the need of importing content from the example module provided by the pythonocc-core package.