We apply mathematical concept analysis to the problem of reengineering configurations. Concept analysis will reconstruct a taxonomy of concepts from a relation between objects and attributes. We use concept analysis to infer configuration structures from existing source code. Our tool NORA/RECS will accept source code, where configuration-specific code pieces are controlled by the preprocessor. The algorithm will compute a so-called concept lattice, which —when visually displayed — offers remarkable insight into the structure and properties of possible configurations. The lattice not only displays tine-grained dependencies between configurations, but also visualizes the overall quality of configuration structures according to software engineering principles. In a second step, interferences between configurations can be analyzed in order to restructure or simplify configurations. Interferences showing up in the lattice indicate high coupling and low cohesion between configuration concepts. Source files can then be simplified according to the lattice structure. Finally, we show how governing expressions can be simplified by utilizing an isomorphism theorem of mathematical concept analysis.

Implementing software configuration management (SCM) in an organization raises various integration problems. We present the Incremental Configuration Environment (ICE), a novel SCM system providing smooth integration with both the software process and the development environment. ICE is based on the version set model, where versions, components, and aggregates are grouped into sets according to their features, using feature logic as a formal base to denote sets and operations and to deduce consistency. Version sets generalize well-known SCM concepts such as components, repositories, workspaces, aggregates, or configurations and allow for flexibility in combining these concepts. For integration in software development environments, ICE provides a featured file system (FFS), where version sets are represented as files and directories. In the FFS, arbitrary programs can incrementally access and explore version sets. Using the well-known C preprocessor (CPP) representation, users can vi...

Integration of configuration management (CM) tools into software development environments raises the need for CM models to interoperate through a unified CM model. We present the version set model, where versions, components, and aggregates are grouped into sets according to their features, using feature logic as a formal base to denote sets and operations and deduce consistency. Version sets generalize well-known CM concepts such as components, repositories, workspaces, aggregates, or configurations. Arbitrary revision/variant combinations of components and aggregates are modeled in a uniform and orthogonal way. We show how the concepts of four central configuration management models---the checkin/ checkout model, the change set model, the composition model, and the long transaction model--- are encompassed and extended by the version set model, making it a unified basis for modeling, realizing and integrating configuration management tasks. Finally, some conditions for efficient real...

Software configuration management (SCM) is the discipline for controlling the evolution of software systems. The central problems of SCM are closely related to central artificial intelligence (AI) topics, such as knowledge representation (how do we represent the features of versions and components, and how does this knowledge involve in time?), configuration (how do we compose a consistent configuration from components, and how do we express constraints ?), and planning (how do we construct a software product from a source configuration, and what are the features of this product?). Although the research communities of both SCM and AI work on configuration topics, the knowledge about the mutual problems and methods is still small. We show how feature logic, a description logic with boolean operations, can be used to represent both knowledge about versions and components, as well as to infer the consistency of possible configurations and thus solve configuration problems in SCM. This i...

We apply feature logic to the problem of incremental configuration management. Feature logic has originally been developed in computer linguistics as a knowledge representation and inference mechanism. It offers a uniform formalism for the description of variants and revisions, where sets of versions rather than single versions are the basic units of reasoning. Feature logic thus opens a whole algebra of version sets, which includes specific configurations as special cases. Our approach allows for interactive configuration management, where a configuration thread is constructed by adding or modifying configuration constraints until either a complete configuration or an inconsistency can be deduced. A set of versions of a software component can be represented and processed as a single source file enriched with preprocessor statements. Thus, our tool can be used as an intelligent front end to more traditional techniques.