Funded by Fondecyt, as a “Regular” project, 2009 - 2013
Aspect languages have almost exclusively been of a general-purpose nature, but currently interest is growing in defining aspect languages that are specific to a particular concern, i.e. are Domain-Specific Languages (DSLs).
This project proposes to perform initial research on language engineering of such Domain-Specific Aspect Languages (DSALs). We wish to provide methodologies for their engineering, considering composition, as well as support for their implementation.
A first issue is that in a setting where we wish to use DSALs, we are confronted with the need to create these languages. However no methodology has been defined on how to best approach this, i.e., there are no guidelines on how to best evaluate the different trade-offs to be made in their design. Furthermore, we foresee the need for many such languages to be developed, and therefore reuse would yield important productivity benefits. In spite of this, there is no support for reuse in the development process of domain-specific languages, let alone DSALs.
Secondly, the issue of interactions between aspects needs to be addressed also when using DSALs. A significant amount of work on this topic has been performed, but only using general-purpose aspect languages. Due to the fundamentally different nature of general-purpose aspect languages and DSALs, this work is however not straightforwardly transferable from the former to the latter. Furthermore, it is envisioned that working with DSALs will bring some benefits for reasoning about interactions, as explicit domain information can be taken into account.
Lastly, because most of the research on aspects has been centered on general-purpose languages, no significant work has been undertaken on the use of aspects for programs written in domain-specific languages. While there is proof that support for this is required, there is no knowledge of the basic properties of this domain.