Generators Map Models to a Solution Domain Generators close the gap
between the model and code worlds. The generator speci?¬?es how information is
extracted from the models and transformed into code. In the simplest cases, each
modeling symbol produces certain ?¬?xed code, including the values entered into the
symbol as arguments. The generator can also generate different code depending on the
values in the symbol, the relationships it has with other symbols, or other information
in the model.
The generator itself is usually invisible to modelers and the construction and
modi?¬?cation of the generator is done by just a few experienced programmers. At this
point, we should note that the generator is not usually solely responsible for providing
the mapping to the code since the supporting domain framework and target platform
bring the implementation world closer to the problem domain. For instance, libraries,
components, frameworks (Fayad and Johnson, 1999), and domain-speci?¬?c
architectures ( Q2 Duffy, 2004) already make the work of generators easier by raising
the level of abstraction on the code side.
3.1.3 Full Code Generation
In DSM, full code is generated from the application developer??™s point of view and
manual rewriting of the generated code is not needed. This completeness has been
the cornerstone of other successful shifts made with programming languages.
Pages:
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111