home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.software-eng
- Path: sparky!uunet!noc.near.net!mv!jlc!john
- From: john@jlc.mv.com (John Leslie)
- Subject: Re: Value of High Code Coverage Metrics in Testing
- Message-ID: <1992Dec14.200333.5592@jlc.mv.com>
- Organization: John Leslie Consulting, Milford NH
- References: <1992Dec14.072812.13689@syacus.acus.oz.au> <ssimmons.724336081@convex.convex.com>
- Date: Mon, 14 Dec 1992 20:03:33 GMT
- Lines: 27
-
- paulb@syacus.acus.oz.au (Paul Bandler) writes:
-
- > One of the metrics I have been asked to measure is the 'Branch Flow
- > Analysis' (BFA) percentage achieved during unit and system testing.
- > i.e. How much of the potential paths through the code is actually
- > excersized during these testing phases.
- >
- > 1) Do people think that this is a valuable metric?
- > 2) Is it a cost effective excersize to get engineers to achieve a particular
- > %BFA as a completion criteria?
- > 3) What is a realistic %BFA to aim for?
-
- 1) Yes; 2) Yes; 3) 100%.
-
- Totally untested code should not be let out the door. If a particular
- branch of the code cannot be tested, it deserves to be removed.
-
- OTOH, maybe you're proposing to enumerate all branches and count the
- combinations of them which have been tested. This idea is quite useless.
- In a modern object-oriented GUI system, the number of possible combinations
- exceeds the capacity of today's computers to count.
-
- There is a middle ground, wherein each branch would be tested in each
- "state" of the system. This sounds very promising to me, but I haven't yet
- seen it done in practice.
-
- John Leslie <john@jlc.mv.com>
-