The app uses a custom data model for representing the rules to create conditional checklists. Each rule is bound to a certain condition which activates items for the packlist and questions for activation of other rules. More on this can be found in the rules documentation.
When creating the application, I decided to use a text base format for storing those rules. This also means, that the data entered in the UI is repeatedly parsed and serialized from a string representation to an object graph and vice versa.
After writing everything from scratch in the beginning, I decided to use a parser generator to create a more robust and maintainable solution. I chose Peggy for this task. The grammar for the rules can be found here.
For developing on the data model, it is useful to run the following two commands in parallel:
nx run model:generate --watch
nx run model:test --watch