[SciPy-dev] Add bz2 support to io.mmio [not tested] <--
Stéfan van der Walt
Thu Apr 10 14:57:18 CDT 2008
On 10/04/2008, Pearu Peterson <firstname.lastname@example.org> wrote:
> I think in this particular case that would have been an overkill. We
> are all busy.
Therefore, we should we jealous of our time and write the tests now,
instead of postponing them into the future (where I can guarantee that
a) they'd take longer to write and b) they won't be written). A test
in time saves nine, I've heard.
> When I looked what was needed to implement the feature,
> I saw that only a simple patch was required that I could do in 5
> minutes. So, I did it. Otherwise I would have postpone it and asked
> to submit a feature request.
I'm very happy you wrote the code, and I certainly don't want to
dissuade you from doing so in the future. I do think that we need a
better way of tracking missing tests than adding "[not tested]" to the
change log, though. I didn't see an open ticket on the original
feature request (maybe I missed it?), but until the test is in place,
there should be a corresponding open ticket (even if just to help me
out, next time I want to improve the test coverage).
> While I strongly agree with the importance of unittests, we should not
> be too bureaucratic too, imho, because it can crush down creativity
> which is also important. So, we need to move forward along a middle
> In this particular case, I'd say, you have been looking for a problem
> that was not there. :)
I can't really go along with any middle ground on testing, I'm afraid
(do I sound like a broken record yet?). To me that reeks of bad
software development practice. If it were up to me, there wouldn't be
a single line of untested code in NumPy.
More and more people depend on NumPy to do their work and research,
and we have a duty towards them to produce reliable, high quality
Your change consists of only 5 lines of code, but what happens when
the bz2 module changes its calling convention in Python 3k? We'd
happily ship SciPy without even knowing that the compressed matrix
market loading was completely unusable.
I sincerely hope that I am overreacting, and that there is nothing to
be concerned about, but as it stands I see a very real problem here.
> No offense taken:) Though I have been inactive for some time now, I have
> been still following the list and overall scipy/numpy development. But
> since you are using past tense, I gather I have been writeoff (I hope
> this is the right term) :)
Not at all -- but I can't very well comment on code (and tests!) that
haven't been written yet :)
More information about the Scipy-dev