[Bioperl-l] test::more template
n.haigh at sheffield.ac.uk
Fri Oct 20 06:54:30 EDT 2006
If there are known bugs in a particular version of software, what is the
best approach for dealing with tests that would fail due to this bug?
Simply skip those tests that would be affected by the bug, or to fail if
the affected version is detected and report the reason so the user is
informed? Or simply bump the minimum version to one above the affected
For example, t/Clustalw has a test for at least version 1.8. It then has
some profile alignment tests that are only run if version > 1.82 is
installed. It states that versions 1.81 and 1.82 are affected by a
profile alignment bug - which i assume would make the tests fail.
More information about the Bioperl-l