Browse Prior Art Database

Incremental Loading of Modeling Diagrams Disclosure Number: IPCOM000225196D
Publication Date: 2013-Jan-30
Document File: 2 page(s) / 38K

Publishing Venue

The Prior Art Database


Disclosed is a mechanism applied to modeling tools that allows the tool to load diagrams in a scalable fashion. The mechanism decides which sub-graph of the diagram's graph of shapes and edges to load in order to allow the current visible window on the diagram to correctly render. As the window moves, the mechanism loads more of the graph as needed. The idea can generically work with any diagram type and/or persistence scheme.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Incremental Loading of Modeling Diagrams

Modeling tools often represent diagrams using some sort of a schema/metamodel/ontology that preserves the notational aspects of the diagram (e.g., bounds of shapes, colors, fonts etc.) separate from what the diagram represents (e.g., Unified Modeling Language (UML) class diagram, Business Process Modeling Notation (BPMN) process diagram, etc.). Examples of such representation are GMF's DI metamodel, OMG's DI metamodel, etc. This representation of a diagram is then persisted in a suitable persistence scheme (e.g., Extensible Markup Language-Based Metadata Interchange (XMI), Resource Description Framework (RDF), etc.).

It is also often the case that a diagram is represented as a graph of shapes and edges. Shapes are constrained by their bounds (although when auto-resize is used the bounds are not explicitly provided), while edges (polylines connecting the shapes) are constrained by their attach points (or waypoints), that is, the set of points that make up the polyline of the edge.

This invention addresses the problem of loading such a diagram in a scalable fashion. When diagrams are small, the whole diagram can be loaded efficiently in one shot. However, when diagrams are large (and this is case of many kinds of diagrams like UML class, interaction and activity diagrams), it can be considerably inefficient to load the entire diagram in one shot, especially when only a small part of it is visible at a time. This is especially true when memory or processing power is limited or when a diagram has to be sent across the wire from a server to a client (as

with the DM web-based tool).

In most cases, large diagrams are not viewed in their entirety due to the limited real estate of the viewer's window; therefore, a method is needed to load the diagram incrementally based on the sliding window of the viewer. In other words, only the visible parts of the diagram need to be loaded and the other parts incrementally load on demand as they are being viewed.

Most desktop modeling tools in the market today lack such a mechanism , which

causes their large diagrams to be expensive to load. The problem is more aggravated for the new generation of web-based modeling tools.

The idea of this invention is a mechanism to decide which sub -graph of the diagram's gra...