OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

tag message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Proposals for updating the "Anatomy of a TA" section (V0.6):


Proposals for updating the "Anatomy of a TA"  section (V0.6):
 
----------------------------------------------------------------------------------------------------------
 
1- In the parts of a Test Assertion as currently stated in Anatomy V0.6,
 
replace:
 "Test Outcome" and its defined values (pass/fail)
with:
 "Indicator" (or Test Indicator) and its defined values (true/false).
 
The Indicator is a boolean expression, the evaluation of which is supposed to take place during the execution of a test case derived from this test assertion. An evaluation to "true" is an indication that the item under test is adhering (conforming) to the specification requirement addressed by the TA. An evaluation to "false" is an indication that the item under test is NOT adhering to the specification requirement. It could be that the evaluation is not possible, but that is not to be considered in the TA definition: that is a test case issue.
 
NOTES:
- the Indicator is just this: an indicator, not a proof. (in any case, a test is never a proof that the IUT will always behave consistently with the test case outcome).
- A derived test case still has the possibility to interpret this true/false outcome in a particular way to produce a test case outcome (pass/fail/undetermined). For example, if the TA only address some subcases of the specification requirement, an Indicator="true" can be interpreted as "pass" by a derived test case. But an Indicator="false" may just produce "undetermined", as other subcases for this spec requirement may be addressed by another TA. Or, a  the [test] Indicator in a TA may focus on revealing failures do adhere to a spec requirement, and does not intend to indicate adhesion. In that case, a derived test case may interpret "false" as "fail" but will not infer anything from "true".
 
Rationale: some of us expressed concerned of possible confusion with test metadata and the derived test case(s).
 
 
 
----------------------------------------------------------------------------------------------------------
2- Replace Item Under Test (IUT) with Test Assertion Target (or TAT...).
 
Rationale: "under test" suggests an operation more relevant to test case execution. It could also be confused with a larger entity such as the target of a full conformance clause (or conformance target), which could be named Implementation Under Test (or IUT too...).
 
 
 
----------------------------------------------------------------------------------------------------------
3- Use "Qualifier" (or as second choice, "Precondition") to name the optional part of a TA  that qualifies a TA target as applicable for the test Indicator.
 
- The qualifier is another boolean expression, that must evaluate to "true" for the Indicator outcome (true or false) to have any meaning w/r to adherence to the spec requirement.
- The qualifier may refer to one or more other TA, with the semantics that the test target must then evaluate to "true" by the Indicator of these referred TAs, in order to qualify.
 
 
 
Jacques


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]