2 Representing DSL content as structured data
This chapter covers
- Learning to represent structured data using a notation for object-relation diagrams
- Representing the details in DSL content as tree-like structured data
- Learning common terms for the new things we encounter
In the previous chapter, we showed that one of the key aspects of a DSL is its structure, which underlies all DSL content. In this chapter, we’re going to make that structure explicitly visible, as shown in figure 2.1. More precisely, we’re going to represent DSL content as structured data. Data is good, data we can handle, right? This will allow us to start writing code for that structure in the next chapter.
Figure 2.1 What we’re going to do in this chapter
In section 2.1, you’ll learn how to write down structured data precisely, using the object-relation diagram (ORD) notation. This notation will be used throughout the book to graphically represent DSL content in an abstract syntax. In section 2.2, we’ll extract ORDs from example DSL content. This extraction process works for all notated DSL content, producing what software language engineers call abstract syntax trees (ASTs). Later, in chapter 4, we’ll reverse this process: given an AST represented in code, we’ll reconstruct the notated DSL content in what is also known as concrete syntax. The projection we’ll implement for this will be the basis for our DSL editor. In chapters 5 and 6, we’ll make this visualization editable.