6.2 Anchor
The Anchor grouping node retrieves the content of a URL when the user activates (e.g., clicks) some geometry contained within the Anchor node's children. If the URL points to a valid VRML file, that world replaces the world of which the Anchor node is a part (except when the parameter field, described below, alters this behaviour). If non-VRML data is retrieved, the browser shall determine how to handle that data; typically, it will be passed to an appropriate non-VRML browser.
Exactly how a user activates geometry contained by the Anchor node depends on the pointing device and is determined by the VRML browser. Typically, clicking with the pointing device will result in the new scene replacing the current scene. An Anchor node with an empty url does nothing when its children are chosen. A description of how multiple Anchors and pointing-device sensors are resolved on activation is contained in 4.6.7, Sensor nodes.
More details on the children, addChildren, and removeChildren fields and eventIns can be found in 4.6.5, Grouping and children nodes.
The description field in the Anchor node specifies a textual description of the Anchor node. This may be used by browser-specific user interfaces that wish to present users with more detailed information about the Anchor.
The parameter exposed field may be used to supply any additional information to be interpreted by the browser. Each string shall consist of "keyword=value" pairs. For example, some browsers allow the specification of a 'target' for a link to display a link in another part of an HTML document. The parameter field is then:
Anchor {
parameter [ "target=name_of_frame" ]
...
}
An Anchor node may be used to bind the initial Viewpoint node in a world by specifying a URL ending with "#ViewpointName" where "ViewpointName" is the name of a viewpoint defined in the VRML file. For example:
Anchor {
url "http://www.school.edu/vrml/someScene.wrl#OverView"
children Shape { geometry Box {} }
}
specifies an anchor that loads the VRML file "someScene.wrl" and binds the initial user view to the Viewpoint node named "OverView" when the Anchor node's geometry (Box) is activated. If the named Viewpoint node is not found in the VRML file, the VRML file is loaded using the default Viewpoint node binding stack rules (see 6.53, Viewpoint).
If the url field is specified in the form "#ViewpointName" (i.e. no file name), the Viewpoint node with the given name ("ViewpointName") in the Anchor's run-time name scope(s) shall be bound (set_bind TRUE). The results are undefined if there are multiple Viewpoints with the same name in the Anchor's run-time name scope(s). The results are undefined if the Anchor node is not part of any run-time name scope or is part of more than one run-time name scope. See 4.4.6, Run-time name scope, for a description of run-time name scopes. See 6.53, Viewpoint, for the Viewpoint transition rules that specify how browsers shall interpret the transition from the old Viewpoint node to the new one. For example:
Anchor {
url "#Doorway"
children Shape { geometry Sphere {} }
}
binds the viewer to the viewpoint defined by the "Doorway" viewpoint in the current world when the sphere is activated. In this case, if the Viewpoint is not found, no action occurs on activation.
More details on the url field are contained in 4.5, VRML and the World Wide Web.
The bboxCenter and bboxSize fields specify a bounding box that encloses the Anchor's children. This is a hint that may be used for optimization purposes. The results are undefined if the specified bounding box is smaller than the actual bounding box of the children at any time. The default bboxSize value, (-1, -1, -1), implies that the bounding box is not specified and if needed shall be calculated by the browser. More details on the bboxCenter and bboxSize fields can be found in 4.6.4, Bounding boxes.
