Modeling history to analyze software evolution

Abstract

The histories of software systems hold useful information when reasoning about the systems at hand or when reasoning about general laws of software evolution. Over the past 30 years, research has been increasingly spent on understanding software evolution. However, the approaches developed so far do not rely on an explicit meta-model and, thus, they make it difficult to reuse or compare their results. We argue that there is a need for an explicit meta-model for software evolution analysis. We present a survey of the evolution analyses and deduce a set of requirements that an evolution meta-model should have. We define Hismo, a meta-model in which history is modeled as an explicit entity. Hismo adds a time layer on top of structural information, and provides a common infrastructure for expressing and combining evolution analyses and structural analyses. We validate the usefulness of our meta-model by presenting how different analyses are expressed on it.what changes appeared in the system [2]. Some basic services were also added to model extra or meta information such as who changed files and why. However, only very rudimentary models were used to represent this information-typically a few unstructured lines of text to be inserted in a log file.Whereas versioning systems enabled the history of each source file to be recorded independently, configuration management systems attempted to record the history of software products as a collection of versioned source files. Research on configuration management was very active in the 1980s and 1990s, but the emphasis was still on controlling and recording software evolution.The importance of modeling and analyzing software evolution began to be recognized in the early 1970s with the work of Lehman and Belady [3]. However, it was not until recent years that extensive research was spent on exploiting the wealth of information residing in versioning repositories for different purposes such as reverse engineering or cost prediction. Problems such as software aging [4] and code decaying [5] gained increasing recognition both in academia and in industry.Various approaches have been proposed to analyze aspects of software evolution for purposes such as identifying driving forces in software evolution or reverse engineering. Each of these approaches typically focuses on only some traits of software evolution, and most of these approaches rely on ad hoc models (i.e., models that are not described by an explicit meta-model) or their meta-model is specific to the goals of the supported analysis.By a meta-model we mean a specification model for a class of systems under study where each system in the class is itself a valid model expressed in a certain modeling language [6]. By model we mean a simplification of a system built with an intended goal in mind [7]. A meta-model describes the way the domain can be represented by the model; that is, it provides bricks for the analysis. An explicit meta-model allows those bricks to be understood. Understanding the bricks allows for the compari...

ProMeTA: a taxonomy for program metamodels in program reverse engineering

“…15 During the SLR, several history-aware metamodels were found to explicitly address the version history: Ring (Gȯmez and Ducasse 2012), Hismo (Gîrba and Ducasse 2006;Gȯmez et al 2009), FAMIX-based RHDB code model (Antoniol et al 2005) and FAMIX-based ArchEvoDB schema . If practitioners and researchers conduct reverse engineering in which history analysis is taken into account, selecting a history-aware metamodel, especially the RHDB code model and the ArchEvoDB schema, may be better since these are defined as extensions of FAMIX, which is a widely accepted popular metamodel.…”

A logic foundation for a general-purpose history querying tool

“…Ring stores information about the structural entities (i.e., packages, classes, interfaces, fields and methods) declared in the source code and information about how they are related (i.e., inheritance, containment). Similar to the HISMO meta-model [10], each structural entity has a reference to the version of the software in which it was defined.…”

Reasoning over the Evolution of Source Code Using Quantified Regular Path Expressions

“…Time warp is a prototype extension of SOUL that has a similar goal as ABSINTHE. It offers an ad-hoc specification language for quantifying over instances of HISMO [16] models. While this work has inspired the development of the approach detailed in this paper, ABSINTHE uses a custom repository representation for scalability reasons and offers a more structured approach to quantify over its information.…”