In recent years many tools and techniques have been proposed and even implemented which support maintenance programmers to gain an understanding of existing source code. However, studies have found that some tools are of marginal benefit and not appropriate for all maintenance tasks. This may be due to tools having been designed with little regard to the common behavioural practices that occur during software maintenance tasks because the current state of knowledge of such practices is quite limited. The aim of this study is to alleviate the above situation by performing a detailed behavioural analysis of software maintenance tasks undertaken under laboratory experiment conditions. To support this aim an experiment was conducted in which twenty-nine student subjects experienced in C programming independently undertook one of two maintenance tasks on the same C program in a university environment. The data produced from the experiment included subjects??? interactions with documents and code and their responses to program comprehension questions at the end of the task. The results indicate that maintenance task success is associated with a heightened concentration on high-level procedures that control the execution of lower level procedures. Among subjects, IT industry experience was linked to efficient comprehension of program and task documentation and the performance of more code searches. In general, code searches mainly involved searches for procedure name references. Compared to subjects undertaking the enhancement task, subjects performing the complex corrective task utilised program documentation more but also obtained a greater understanding of the elementary operations and control-flow of the program. In addition, this study illustrates a method to identify common information usage strategies from the recorded maintenance behaviour of subjects. Largely unsuccessful attempts were also made to sequentially analyse maintenance behaviour. This lack of success is possibly attributable to the as-needed comprehension approach adopted by subjects. The results of this research suggest that cost-effective maintenance of software is best supported using tools implementing facilities for feature location and visualisation of the control-flow of a program. The former facility would be most helpful to maintainers undertaking corrections, whereas the latter would particularly assist maintainers undertaking enhancements.

In recent years many tools and techniques have been proposed and even implemented which support maintenance programmers to gain an understanding of existing source code. However, studies have found that some tools are of marginal benefit and not appropriate for all maintenance tasks. This may be due to tools having been designed with little regard to the common behavioural practices that occur during software maintenance tasks because the current state of knowledge of such practices is quite limited. The aim of this study is to alleviate the above situation by performing a detailed behavioural analysis of software maintenance tasks undertaken under laboratory experiment conditions. To support this aim an experiment was conducted in which twenty-nine student subjects experienced in C programming independently undertook one of two maintenance tasks on the same C program in a university environment. The data produced from the experiment included subjects??? interactions with documents and code and their responses to program comprehension questions at the end of the task. The results indicate that maintenance task success is associated with a heightened concentration on high-level procedures that control the execution of lower level procedures. Among subjects, IT industry experience was linked to efficient comprehension of program and task documentation and the performance of more code searches. In general, code searches mainly involved searches for procedure name references. Compared to subjects undertaking the enhancement task, subjects performing the complex corrective task utilised program documentation more but also obtained a greater understanding of the elementary operations and control-flow of the program. In addition, this study illustrates a method to identify common information usage strategies from the recorded maintenance behaviour of subjects. Largely unsuccessful attempts were also made to sequentially analyse maintenance behaviour. This lack of success is possibly attributable to the as-needed comprehension approach adopted by subjects. The results of this research suggest that cost-effective maintenance of software is best supported using tools implementing facilities for feature location and visualisation of the control-flow of a program. The former facility would be most helpful to maintainers undertaking corrections, whereas the latter would particularly assist maintainers undertaking enhancements.

Report (M.S.)--Virginia Polytechnic Institute and State University, 1994. / Abstract. Includes bibliographical references (leaves 147-152). Also available via the Internet.

Thesis (Ph. D.)--Virginia Polytechnic Institute and State University, 1993. / Vita. Abstract. Includes bibliographical references (leaves 105-109). Also available via the Internet.

Thesis (M.S.)--Virginia Polytechnic Institute and State University, 1990. / Vita. Abstract. Includes bibliographical references (leaves 96-101). Also available via the Internet.

Thesis (Ph. D.)--University of Washington, 1996. / Vita. Includes bibliographical references (p. [140]-149).

Report (M.S.)--Virginia Polytechnic Institute and State University. M.S. 1992. / Vita. Abstract. Includes bibliographical references (leaves 93-94). Also available via the Internet.

Thesis (Ph. D.)--Oregon State University, 2006. / Printout. Includes bibliographical references (leaves 187-208). Also available on the World Wide Web.

Program Comprehension is a key factor in providing effective software maintenance and enabling successful evolution of software systems. The objective of this research is to provide a framework and mechanism to facilitate the understanding of large software systems. There exist a number of theories and models of Program Comprehension where each favours a different approach to comprehension. It is evident that there is no real consensus on how maintainers understand software systems. The disparities in the comprehension strategies are largely dependent on the personal and circumstantial factors. Factors such as the level of technical competence of the maintainers, the size and complexity of the piece of software, and the types and goals of the maintenance activities can influence the process of comprehension. This research proposes an alternative approach to Program Comprehension. It acknowledges that the process of comprehension is opportunistic, and that the current comprehension theories are inadequate in addressing this. There is a need for a more flexible approach towards comprehension, and the Integrated Approach proposed in this thesis provides a way for the utilisation of the various comprehension theories under a single environment. It recognises that any one of the comprehension theories may become active during comprehension. Under the Integrated Approach, maintainers have the option of selecting and executing the various comprehension strategies as they see fit. The Integrated Approach to comprehension is based on a matrix of Program Relationships between Program Elements of a programming language. In this thesis, these Program Relationships are derived for the C programming language constructs. This work also involves the investigation of the roles of both textual and graphical representations during the comprehension process. Both representations are commonly used to alleviate the problem of information overloading when maintainers trying to understand and maintain a software system. The Integrated Approach is realised in a tool named PUI (program understanding implements) which provides an environment enabling the utilisation of various comprehension theories.

This thesis presents and analyses a solution to the problem of formally re- engineering program data structures, allowing new representations of a program to be developed. The work is based around Ward's theory of program transformations which uses a Wide Spectrum Language, WSL, whose semantics were specially developed for use in proof of program transformations. The re-engineered code exhibits equivalent functionality to the original but differs in the degree of data abstraction and representation. Previous transformational re-engineering work has concentrated upon control flow restructuring, which has highlighted a lack of support for data restructuring in the maintainer's tool-set. Problems have been encountered during program transformation due to the lack of support for data re-engineering. A lack of strict data semantics and manipulation capabilities has left the maintainer unable to produce optimally re-engineered solutions. It has also hindered the migration of programs into other languages because it has not been possible to convert data structures into an appropriate form in the target language. The main contribution of the thesis is the Data Re-Engineering and Abstraction Mechanism (DREAM) which allows theories about type equivalence to be represented and used in a re-engineering environment. DREAM is based around the technique of "ghosting", a way of introducing different representations of data, which provides the theoretical underpinning of the changes applied to the program. A second major contribution is the introduction of data typing into the WSL language. This allows DREAM to be integrated into the existing transformation theories within WSL. These theoretical extensions of the original work have been shown to be practically viable by implementation within a prototype transformation tool, the Maintainer's Assistant. The extended tool has been used to re-engineer heavily modified, commercial legacy code. The results of this have shown that useful re-engineering work can be performed and that DREAM integrates well with existing control flow transformations.