kathuria
Joined: 29 Aug 2007
Posts: 9
|
 Posted: Sat Sep 15, 2007 3:01 pm Post subject: What was Measured..... |
 |
|
For each stage, the following was measured:
(1) The time spent designing tests. This included time spent finding test conditions, ruling out infeasible
coverage conditions, deciding that code not worth covering (e.g., potential weak mutation faults) was
correct, and designing test cases from test conditions. The time spent actually writing the tests was
not measured, since it is dominated by extraneous factors. (Can the program be tested from the command
line, or does support code have to be written? Are the inputs easy to provide, like integers, or
do they have to be built, like linked lists?)
(2) The number of test conditions and test cases written down.
(3) The percent of coverage, of all types, achieved. This is the percent of total coverage conditions, not
just feasible ones.
(4) The number of feasible, impossible, and not worthwhile coverage conditions. |
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
