The following advanced example is a reproduction of an older example from the official C4 model examples created by Simon Brown (it is not there anymore, but similar ones).
The example uses pumla and C4-PlantUML and creates the C4 elements in a re-usable way.
Here you can see the reproduction of C4-PlantUML of that example:
And here is how it looks like:
And now here is the example coded in a re-usable manner with pumla:
pumla C4 Example Source Folder
pumla C4 example diagram Source
And here is how it looks like:
As you can see, there is no real difference in the diagram itself. The re-use had no impact on the layout. The only difference is, that I put the CheatSheet onto the diagram in addition, so that you can see how the C4 model elements are stored in the pumla model repository.
In order to make this example also work on your computer, you need to go
to the example folder (<pumla_root>/test/examples/C4example) and then
call the following:
pumla init
pumla update
Now, the core of the diagram is the "system" called techtribes.js. The
system boundary is modeled in a separate folder as a re-usable element.
Remember, we shall only have one re-usable element per file
(Modelling Guideline). The techtribes
element includes into its internals all possible children using the
powerful PUMLAInjectChildElements(...) macro. That way, all elements calling
techtribes their parent are included into the techtribes system boundary,
which is a C4 System_Boundary, made re-usable by pumla.
The elements that are part of the techtribes system (everything inside the
system boundary) are also modelled as separate files in separate folders:
web_appfilesystemrel_dbupdaternosql
Within the definition of these re-usable elements, they all call the techtribes
element their parent, using the corresponding pumla file marking
(PUMLAPARENT: techtribes).
The users are all modelled in the same users folder, each in their own file. Same
for the external systems (twitter, blogs, github), they are modelled in
their own folder ext_systems, each in their own file.
The relations are split up. The overall relations connecting the outside of the system with the system itself are modelled here in this file:
./relations_overall_system.puml
It contains the relations from the users to the system and from the system to the external systems.
The relations that combine the different "child" elements of the techtribes
system are stored in the techtribes.js folder, sorted in two files. One
contains the relations starting at the web_app element
./techtribes.js/relations_web_app.puml
In this file you can also see, that I added a tag/value to one of the relations. You can apply the pumla tagged values also to each C4 element and use the powerful macros to select elements depending on specific tag/value combinations. In addition, you can still use the C4-PlantUML tag/value mechanism, but they are only there for visualization and you cannot use them e.g. for filtering.
The relations starting at the updater element are stored in this file:
./techtribes.js/relations_updater.puml
That makes the example complete.
For the resulting diagram in this file...
pumla C4 example diagram Source
... first the users are added one by one. Then the techtribes system
element is added, and with that one, all it's children are also included.
So the elements inside the techtribes element must not be put onto
the diagram explicitly. Then the external systems are added one by one
(twitter, github, blogs). After that the relations are added. You could
add the relations one by one referencing them by their alias/ID, but I
made a shortcut and just added all relations in one macro call
(PUMLAPutAllRelations), as there were no relations defined that I don't
want to see on the diagram.
After that, just the CheatSheet was added and a C4 layout call (taken over from the original C4-PlantUML example) is there.