GeometryModel

class cee.geo.GeometryModel()

The GeometryModel implements a client-side model that can handle a large number of parts efficiently.

The model can be created and modified in the client application and is not dependent on a server (unlike ug.RemoteModel).

Each view can contain many GeometryModels. A GeometryModel is a collection of parts. Each Part is defined by its Mesh, which describes the triangles, lines and points, and by its PartSettings, which specify how the part is rendered.

A GeometryModel can have a transformation matrix (transformationMatrix) useful for scene composition or for doing rigid body animations.

Picking is supported via rayIntersect which returns a HitItem containing the intersection point, part index and primitive index within that part, and regionIntersect which returns an array of parts (partially) within a given rectangle in screen coordinates.

See the example in Examples/BuildYourFirstApp/1-HelloEnvision for how to use the GeometryModel.

Here is a simple example that creates a geometry model with one part and one triangle and adds it to the current view:

var geoModel = new cee.geo.GeometryModel();

// Create a simple one triangle one part model
var indices = [0, 1, 2];
var triangleVertices = [0,0,0,  1,0,0,  0,1,0];

let part = geoModel.addPart();
let mesh = new cee.geo.MeshIndexedTriangles(triangleVertices, indices);
part.mesh = mesh;

part.settings.color = new cee.Color3(0, 1, 0);

var view = myViewer.getViewAt(0);
view.addModel(geoModel);
view.requestRedraw();

Constructors

constructor

Accessors

ignoreViewClipping
name
partCount
transformationMatrix

Methods

addPart
deleteAllParts
deletePartAt
deletePartsAt
getBoundingBox
getDefaultCameraConfig
getPartArray
getPartAt
rayIntersect
regionIntersect
setColorPerPartFromMapper

Constructors

GeometryModel.constructor()

Constructor

Return type:: GeometryModel

Accessors

cee.geo.ignoreViewClipping()

Enable or disable option to ignore view clipping in this model

Return type:: boolean

cee.geo.ignoreViewClipping(ignore)

Arguments:

ignore (boolean) – None

Return type:

void

cee.geo.name()

The name of the geometry model. Mainly used for debugging.

Return type:: string

cee.geo.name(name)

Arguments:

name (string) – None

Return type:

void

cee.geo.partCount()

The number of parts in the model.

Return type:: number

cee.geo.transformationMatrix()

The transformation matrix to use for this model.

If specified, all parts in the model will be transformed by this matrix. This can be useful for scene composition and for rigid body type animations. The default is null.

Return type:: Mat4

cee.geo.transformationMatrix(transformationMatrix)

Arguments:

transformationMatrix (Mat4) – None

Return type:

void

Methods

addPart

GeometryModel.addPart()

Creates a new part and adds it to the model.

Returns the newly created part.

Return type:: Part

deleteAllParts

GeometryModel.deleteAllParts()

Deletes all parts in model

Return type:: void

deletePartAt

GeometryModel.deletePartAt(partIndex)

Arguments:

partIndex (number) – None

Deletes the part at the given (zero based) index.

Return type:: void

deletePartsAt

GeometryModel.deletePartsAt(partIndicesArr)

Arguments:

partIndicesArr ([number]) – None

Deletes the parts at the given indices.

Return type:: void

getBoundingBox

GeometryModel.getBoundingBox([options])

Arguments:

options (ModelBoundingBoxOptions) – optional None

Returns the BoundingBox (in world coordinates) of the model.

By default, the returned bounding box will only include visible parts. If you want the bounding box to include all parts, regardless of visibility, set the ModelBoundingBoxOptions.includeHiddenParts option to true. Please note that there might be a significant performance penalty for including hidden parts.

Return type:: BoundingBox

getDefaultCameraConfig

GeometryModel.getDefaultCameraConfig()

Returns default camera configuration, which is always null for this model.

Return type:: unknown

getPartArray

GeometryModel.getPartArray()

Returns a read only array with all parts

Return type:: unknown

getPartAt

GeometryModel.getPartAt(partIndex)

Arguments:

partIndex (number) – None

Returns an active reference to the part at the given (zero based) index.

Return type:: Part

rayIntersect

GeometryModel.rayIntersect(ray)

Arguments:

ray (Ray) – None

Performs picking on the model.

If something was hit, returns a HitItem containing information about the part and primitive that was hit.

If nothing was hit, returns null.

Return type:: HitItem

regionIntersect

GeometryModel.regionIntersect(x, y, width, height, view, acceptPartiallyContainedParts)

Arguments:

x (number) – None
y (number) – None
width (number) – None
height (number) – None
view (View) – None
acceptPartiallyContainedParts (boolean) – None

Returns the parts that are (partially) inside the given region.

The returned list of parts will contain all parts that are inside the given region. If acceptPartiallyContainedParts is set to true, parts will be considered inside if they are partially (at least one of the vertices) inside the region. If false, the entire part needs to be completely inside the region.

Please note that this method works at the vertex level. This means that for a part to be considered partially inside the region, at least one of its vertices must be within the specified region. If acceptPartiallyContainedParts is set to false, all vertices must be inside the region. If this method is used in combination with view clipping planes, only vertices that are visible with regards to the specified clipping planes will be considered.

The x and y coordinates must be specified in OpenGL style coordinates, which means a right handed coordinate system with the origin in the lower left corner of the window. The HTML coordinate system is with origin in top left, so if this is your input (e.g. MouseEvent.offsetY, clientY, pageY, etc.) you will have to flip the Y coordinate. The x and y are specified in native pixels, so you will have to adjust the input for the current devicePixelRatio (window.devicePixelRatio).

The width and height are specified in native pixels. So you will have to adjust the input for the current devicePixelRatio (window.devicePixelRatio).

NOTE: This is the only method using OpenGL/WebGL style coordinates. All other relevant methods use Canvas/Viewer local CSS coordinates. This behavior will change in an upcoming major release.

Example: Change the color of all parts within the rectangle defined by a rubber band

// this.m_startX/this.m_startY is the event.offsetX/event.offsetY at the
// start of the region definition
endRegionSelection(event: MouseEvent) {
    this.m_regionSelectActive = false;

    let pixelScaleFactor = window.devicePixelRatio || 1;
    let startXPixels = this.m_startX * pixelScaleFactor;
    let startYPixels = this.m_startY * pixelScaleFactor;
    let endXPixels = event.offsetX * pixelScaleFactor;
    let endYPixels = event.offsetY * pixelScaleFactor;

    // Note: Canvas height is in native pixels, so no scaling with pixelScaleFactor
    let canvasHeightPixels = this.m_canvas.height;

    let leftPixels = Math.min(startXPixels, endXPixels);
    let rightPixels = Math.max(startXPixels, endXPixels);
    let topPixels = Math.max(canvasHeightPixels - startYPixels,
                             canvasHeightPixels - endYPixels);
    let bottomPixels = Math.min(canvasHeightPixels - startYPixels,
                             canvasHeightPixels - endYPixels);
    let widthPixels = (rightPixels - leftPixels);
    let heightPixels = (topPixels - bottomPixels);

    let parts = this.m_model.regionIntersect(leftPixels, bottomPixels,
                                  widthPixels, heightPixels, this.m_view, true);

    let selectColor = new cee.Color3(1,0,1);
    for(let part of parts) {
        part.settings.color = selectColor;
    }

    this.m_regionSelectionDiv.hidden = true;
}

Return type:: [Part]

setColorPerPartFromMapper

GeometryModel.setColorPerPartFromMapper(scalarMapper, perPartResult)

Arguments:

scalarMapper (ScalarMapper) – None
perPartResult (ArrayLike) – None

Sets the color of the parts in the model based on the given scalarMapper and a scalar per part.

This is useful for showing per-part scalar results on a GeometryModel.

To show a color legend representing the scalar mapper in a View, use the View.overlay.addCustomColorLegendForScalarMapper() method.

This method is just a helper for doing the following:

const partCount = this.partCount;
for (let i = 0; i < partCount; ++i) {
    let color = scalarMapper.mapToColor(perPartResult[i]);
    let part = this.getPartAt(i);
    part.settings.color = color.toColor3();
    part.settings.opacity = color.a;
}

Return type:: void