Russ Allbery wrote: > Raphael Geissert writes: > >> I'm a bit hesitant about how correct is to move the lab setup test to >> a t/scripts test; since it would be more appropriate as a shell script >> because it is basically just a set of external commands being run. > > It won't be if we really start testing something rather than just being > sure that lintian returns a non-zero exit status, like being sure that > the right files are created and that directories in the lab are removed > properly.
Maybe it is time to move that part of the code into a module? it would be easier to test it that way (not that I dislike black-box testing, but in this case it would be better to test it from the ground). > Plus, writing it as a Perl script means that we can use the > Test::More functions to check things and report status properly with all > the nice infrastructure that Test::More provides. > Sure. By the way, although I guess you have already read about TDD, I have found Kent Beck's "Test-driven development - by example" a very nice reading; it changed my POV. Cheers, -- Raphael Geissert - Debian Maintainer www.debian.org - get.debian.net -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
