Browse Prior Art Database

Utilising existing code coverage metrics and execution patterns to select well-driven and reliable code for software reuse.

IP.com Disclosure Number: IPCOM000235840D
Publication Date: 2014-Mar-26
Document File: 3 page(s) / 86K

Publishing Venue

The IP.com Prior Art Database

Abstract

Utilising existing code coverage metrics and execution patterns to select well-driven and reliable code for software reuse.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 54% of the total text.

Page 01 of 3

Utilising existing code coverage metrics and execution patterns to select well-driven and reliable code for software reuse.

Software development relies in part upon the identification of existing function and its reuse within newly developed software. This is particularly true when developing a small piece of code as part of servicing an existing product.

    However, it does not reflect the relative usage of code. For example, code may be shown from the mainline path of mission critical performance intensive logic,

which has been very heavily tested and executed and so is well proven in terms of

design and quality. Code may also be shown that resides in recovery routines or peripheral code paths that are very rarely if ever executed. There is no way to base a decision on the quality of code in terms of knowledge of its usage and exploitation

when using code search tools to return hits on pieces of function.

    Taking ideas for code reuse from existing logic can result in errors or performance implications if the software being re-used was based upon existing code which had not been heavily executed. This can lead to problems raised against products.

    The lack of a solution to this problem can result in newly written code incorporating unproven code from elsewhere within the product. The problem to be solved is that there is no gauge of the trustworthiness of an isolated code fragment when considering it for software reuse.

    This publication describes taking information about the relative usage and execution coverage of functionality within software programs and making that information available to cross-reference technology. This would allow a programmer to base their decisions about re-use, on the basis of an accurate knowledge of the exploitation and execution of existing code.

    This idea would make use of information pertaining to the relative usage of code within software modules.

    The logic would parse program code and assign a relative usage metric to the program logic. Mainline logic would be more heavily executed than recovery routines or error handling fragments and this information would be captured accordingly. Base usage could therefore be generated by identifying areas of code

which will always be used less frequently, like recovery routines.

    This information could be augmented by technology such as code coverage measurements, test coverage measurements and information about the execution paths within a product captured from customer problem determination material. Using this various data the invention would provide a relative usage correlation of code functionality within a product, based upon the relative measurements and the actual usage information.

    In addition, code which has a dependency upon other code or compiler directives could be flagged as associated when used as input to the invention correlation logic. For example, in 390 assembler, the need for a TRT instruction t...