-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Fred Drake wrote:
On Fri, Aug 29, 2008 at 5:26 PM, Malthe Borch <mborch@gmail.com> wrote:
I didn't know there was a controversy, but I do remember that there was consensus that ``extras_require`` is not the most elegant solution.
Oh, there is!
If you can advise a different way to avoid pulling in ``zope.app.testing`` I'm happy to revert the change; otherwise I think we should live and let live with it since it at the very least does the job.
There's no good way to avoid dependencies like zope.app.testing; because that's part of the test environment, the tests won't show whether there are problems when it's removed. If you want to fly what you test, test dependencies can't be eliminated.
BTW, that phrase is disingeuous: as one who worked on testing one of the earliest fly-by-wire fighter jets[1] (long ago in a galaxy far far away), I can assure you that the plane did *not* have all its various test jigs attatched to it as it rolled down the runway. - Before assembly, components get tested in isolation using whatever jigs were appropriate, with testing done at each "rollup" of a new assembly. - After integration into the airframe, the set of tests possible were vastly reduced: a minimal set of tests could be done in a preflight check without external equipment, but any failure in such a test resulted in pulling the indicated part for diagnosis using dedicated equipment. [1] http://en.wikipedia.org/wiki/F-16_Falcon#Fly-by-wire Tres. - -- =================================================================== Tres Seaver +1 540-429-0999 tseaver@palladion.com Palladion Software "Excellence by Design" http://palladion.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFIuWiZ+gerLs4ltQ4RAkzBAKCY7ldytq7k9TqaDWljm2gk39DhGQCgl+c9 R+dM/gOr/v/O4f8Ww/xV61o= =Yy63 -----END PGP SIGNATURE-----