[Bioperl-l] bioperl-db: rewriting tests

[Hilmar Lapp]

I'm revamping most of the tests in bioperl-db because of various hard-coded primary key assumptions and foreign key violations (I thought Pg does enforce FKs).

comment.t may serve as an example. It first adds the parent(s) it needs, does its actual tests within a catch-block, and finally cleans up after itself.

In case of a failure within the catch block the test count goes wrong though (the number of tests finally reported as failed will be correct, but their numbers won't). The fix would be to set the test counter to the expected number after the catch block. Maybe someone has a good idea about how to do this painlessly.

	-hilmar
-- 
-------------------------------------------------------------
Hilmar Lapp                            email: lapp at gnf.org
GNF, San Diego, Ca. 92121              phone: +1-858-812-1757
-------------------------------------------------------------