|
Introduction
While it is interesting to show that the chronologies produced by
can
show a kind of internal consistency, it is also important to be able
to prove that those calculations are accurate. To that end, a series
of unit tests have been designed that perform tests against the
specific conditions that
might encounter so as to demonstrate that the results that
produces
will be consistent and produce the proper results. This procudure of
software testing is known in the computer industry as unit testing
and is an important technique for computational validation and data
integrity.
Methodology
To show computation validation a test suite database with several computational
scenerios has been devised. That database is run through the Groundhog system
to produce test values. Then the values are compared against the correct
answers and those tests are given a PASS or FAIL result accordingly.
If all the tests receive a PASS standing, then there is reasonable
assurance that the core program will perform as expected.
Data:
Please note, the test suite examines whether
is able to detect both consistent and inconsistent results. Thus an
"Inconsistent" as an overall validation result is expected and is
not a failure of Groundhog to function but is a reflection of test
scenerios that are both valid and invalid. With unit testing, some
tests produce errors intentionally as a means to demonstrate data
integrety. The following link will show the live test suite database.
Test Suite Data
Validation Test Results:
To actually see if
actually produced the correct result, a separate unit test script is
run that checks the results in the test suite database against the
expected correct results. This script that runs live and is
not a copy of cached results. Click on the link to run the unit
test script:
Validation Results
|
|