From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-0.8 required=5.0 tests=BAYES_00,INVALID_DATE autolearn=no autolearn_force=no version=3.4.4 Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!usc!zaphod.mps.ohio-state.edu!van-bc!ubc-cs!uw-beaver!uw-june!jon From: jon@cs.washington.edu (Jon Jacky) Newsgroups: comp.lang.ada Subject: Re: ACVC test coverage Summary: Early versions of ACVC did *not* cover much of the language Keywords: acvc ACVC coverage Message-ID: <11746@june.cs.washington.edu> Date: 8 May 90 15:55:28 GMT References: <135307@sun.Eng.Sun.COM> Organization: U of Washington, Computer Science, Seattle List-Id: > (Vladimir G. Ivanovic asks....) > > Does anyone know the coverage of the ACVC suite? I would expect (and > hope) that it would test 100% of the syntactic rules (i.e. the > grammar) and a high percentage of the semantic rules. It may be less than that. The limitations in an early version of the validation suite are candidly discussed in a very informative paper by one of the people who built it: John B. Goodenough. Ada compiler validation: an example of software testing theory and practice. In: A.N. Haberman and U. Montinari (Eds.), System Development and Ada, Proceedings of the CRAI Workshop on Software Factories and Ada, Capri Italy, May 26 - 30, 1986. Lecture Notes in Computer Science No. 275. New York, Springer-Verlag 1986, pps. 195 - 232 Goodenough wrote: "The test suite today (Version 1.8) represents 2400 programs. ... The test suite is not yet complete. We have identified approximately 1400 additional tests that need to be written just to cover all aspects of the language adequately." Things may have improved since Goodenough's paper was written, but I've seen nothing more recent to compare it with. Jonathan Jacky jon@gaffer.rad.washington.edu Radiation Oncology RC-08 voice: (206)-548-4117 University of Washington FAX: (206)-548-6218 Seattle, Washington 98195 USA