I‘m curious if there is a chance to export (or find a documentation of) the Visual Paradigm Meta-Model. To be a bit more precise, I‘m looking for a way to find out the pre-defined object- and relationship types within VP. In know that due to the generic concept of VP a lot of the modeled content is „only“ connected via generic „relations“. Either as sub-structures within models, packages etc. and linked objects within the repository. So a great deal of the Meta-Model will only surface after the model is in place. Never the less at lest the diagram notations should reflect a basic Meta-Model structure.
If there is any way to visualize / export this meta Model structures in any way it would be great for ahead planning of any modeling activities.
If you are referring the desktop version of our product then it do have meta-model for most of the diagrams, and can be accessed via our Plugin API. Details about our Plugin API can be found at the links below.
Hello Rain, thanks for you help. Yes, I’m referring to the Desktop version. Can you elaborate the use of the API a bit more to extract the Meta-Model information for specific diagram types (e.g. BPMN or UML Activity Diagram)? I couldn’t find any detailed description how to access this information via the PluginAPI. In detail I’m looking for a way to extract the Meta-Model structures either as a pure list or (even better ) as a generated diagram (e.g. within a Class Diagram describing the elements and vertices used in the Meta-Model).
…one additional thought… it would be great if there is a functionality that would allow to generate the “real” and “used” part of the overall VP Meta-Modell in a diagram overarching way. The use case for this would be: think of a large VP project that has a lot of existing content in various models, packages, diagrams, object structures etc. and a clear approach of using the different modeling capabilities hasn’t been followed (as it should have been). So now there is a need to get this straight again. But where to start? To efficiently tackle this problem the first step is to get an overview of the current modeling structure. And this is best reflected in the “real” Meta-Model derived from the content itself. That’s where I’m heading…
Actually the Plugin API is a published layer of the meta model. It allows you to retrieve model data in a natural way. The knowhow pages in my previous reply contain lots of examples and sample code about how to access to manpiliate models and even generate diagrams via the API. I suggest you read couple of examples then you will understand how it works (it’s pretty straight forward).
we have created a small solution that allows us to access all the designed VisualParadigm content in an easy “graph based” way. The initial reason was to generate a visual representation of our “real” meta model expressed by the content that exists in our VP model. In other words we planed to use the designed content as basis to derive a picture of the specific meta model, represented exactly by this content. The use case for this is to “clean up” an existing VP model by finding out the different usage patterns in an existing model, find out areas that are densely populated, less used or plain wrong according to our specific modeling conventions. This is an easy task when you have just a few diagrams. But when there are hundreds of diagrams and many more objects it becomes a totally different animal…
So we came up with a little add-on that allows to access the VP content using graph technology and query it directly with Cypher. The fist picture below shows an example automatically generated and visualized.
For the example we have used some of the VP Demo diagrams as neutral content. The pre-labeling of the artifacts are added by our implementation. They mean O : relation that is “indirect” created based on objects linkt to other objects via attribute assignments or model hierarchy relations / AST : are relations that exists as graphical representations on diagrams and ET : entities that exist in the overall model. The latter part of the labels are converted VP labels. In the meta model picture some additional elements (e.g. diagrams etc.) have been skipped.
But when we we playing with the results of our solution it turned out it could be used for much more. As we made the whole VP content available for further analysis it could be used to set up a whole bunch of other things as well:
automatic methodology checks to improve model quality
generated overview diagrams that integrate the content of different diagrams into one view automatically
set up of a VP dashboard that allows to engage possible stakeholders that do not necessarily understand the VP notations
integrate VP content with live data to create additional value (e.g. Business-Impact monitors by connecting it with real time data from third parties)
link it with LLMs to use VP content in RAG based solutions
The following pictures show some of these solutions (again with demo content for confidentiality reasons).
This is an example how the VP data can be combined with real time system monitoring systems to create a live Business-Impact-Analysis. For the shown example we have just linked data from DigitalOcean’s ITSM system that can be queried at: https://status.digitalocean.com/api/v2/incidents.json
For the demo we combine the API JSON stream with the VP data to generate the above visulization.
This is a combination of VP data with the NIST IT vulnerability database. Based on the within VP described IT components the NIST DB is constantly monitored for possible vulnerabilities that affect a specific implementation. Here VP design content is directly linked with real time data as monitoring system.
) as a generated diagram (e.g. within a Class Diagram describing the elements and vertices used in the Meta-Model).
