why does software fail?
wrong reqs missing reqs impossible reqs faulty design faulty code bad implementation
what is the objective of testing?
discover faults and fixing them.
what is fault identification?
determining what fault caused a failure.
what is fault correction?
making changes to remove faults.
Name the seven types of faults.
What is an algorithmic fault?
algorithm does not produce desired results.
page 8.
idk
What are the different types of test (in order).
What is egoless programming?
there is not individual tie or ownership to any program.
Who performs the tests?
independent test team.
What is closed or black-box testing?
assertion testing with a wide rage of inputs.
advantage: free of internal constraints.
disadvantage: cannot test all inputs.
What is clear or white-box testing?
test cases that execute all statements or all control paths.
What factors to consider when choosing tests?
Number of possible logical paths.
Nature of the input.
Amount of computation.
Complexity of algorithms.
What is code walkthrough (code review)?
present code and documentation to review team for comments.
informal process.
What is code inspection (code review)?
more formal process than code walkthrough.
preparation and meeting times…
fault discovery rate.
page 18. take the L if its on the test.
page 19. sigh.
What is test thoroughness?
What are the 6 methods for integration testing?
what is component driver (integration testing)?
a routine that calls a particular component and passes a test case to it.
mocha “users create” test a particular portion of a unit.
what is stub (integration testing)?
special-purpose program that simulates activities of missing components.
what is bottom up testing (integration testing)?
when a system is decomposed.
test from the lowest level to the highest level.
what is top down testing (integration testing)?
when a system is decomposed.
test from the highest level to the lowest level.
what is modified top-down (integration testing)?
from the top.
each component is unit tested before merging.
after merging, test again.
what is big-bang (integration testing)?
uses stubs and drivers to test individual components.
then everything is combined and tested again.