[Python-Dev] Breaking Test Cases on Purpose (original) (raw)

Greg Stein gstein@lyra.org
Thu, 3 Aug 2000 11:39:08 -0700

On Thu, Aug 03, 2000 at 12:54:55PM -0500, Guido van Rossum wrote:

> Suppose I'm fixing a bug in the library. I want peer review for my fix, > but I need none for my new "would have caught" test cases. Is it > considered alright to check-in right away the test case, breaking the test > suite, and to upload a patch to SF to fix it? Or should the patch include > the new test cases? > > The XP answer would be "hey, you have to checkin the breaking test case > right away", and I'm inclined to agree. > > I really want to break the standard library, just because I'm a sadist -- > but seriously, we need tests that break more often, so bugs will be easier > to fix.

In theory I'm with you. In practice, each time the test suite breaks, we get worried mail from people who aren't following the list closely, did a checkout, and suddenly find that the test suite breaks. That just adds noise to the list. So I'm against it.

Tell those people to chill out for a few days and not be so jumpy. You're talking about behavior that can easily be remedied.

It is a simple statement about the CVS repository: "CVS builds but may not pass the test suite in certain cases" rather than "CVS is perfect"

Cheers, -g

-- Greg Stein, http://www.lyra.org/