[Numpy-discussion] Removing datetime support for 1.4.x series ?
Charles R Harris
Wed Feb 3 00:30:38 CST 2010
On Tue, Feb 2, 2010 at 10:45 PM, Travis Oliphant <email@example.com>wrote:
> On Feb 2, 2010, at 8:53 PM, David Cournapeau wrote:
> > Travis Oliphant wrote:
> >> I think we just signal the breakage in 1.4.1 and move forward. The
> >> datetime is useful as a place-holder for data. Math on date-time
> >> arrays
> >> just doesn't work yet. I don't think removing it is the right
> >> approach. It would be better to spend the time on fleshing out the
> >> ufuncs and conversion functions for date-time support.
> > Just so that there is no confusion: it is only about removing it for
> > 1.4.x, not about removing datetime altogether. It seems that
> > datetime in
> > 1.4.x has few users, whereas breaking ABI is a nuisance for many more
> > people. In particular, people who update NumPy 1.4.0 cannot use
> > scipy or
> > matplotlib unless they build it by themselves as well - we are talking
> > about thousand of people at least assuming sourceforge numbers are
> > accurate.
> > More fundamentally though, what is your opinion about ABI ? Am I right
> > to understand you don't consider is as significant ?
> I consider ABI a very significant think. We should be very accurate
> about when a re-compile is required. I just don't believe that we
> should be promising ABI compatibility at .X releases. I never had
> that intention. I don't remember when it crept in to the ethos.
About 1.2 after the discussion at SciPy. The general consensus was that
breaking the ABI was a very bad thing, not to be taken lightly. We are
currently bumping the .X number about twice a year, which is too frequent to
allow changes at each iteration. IMHO. I would think changes to the ABI
would be more a two/three year sort of thing and only under the pressure of
necessity. At some point we need to do a major refactoring to hide the
structures and make it easier to add types, but I don't see that in the near
future. I don't think we should add any more types to the current code after
datetime goes in, it's just too big a hassle the way things are now. I'm
thinking numpy types should basically interface to the c-types, and new
types should subclass or build new classes on top of that. That keeps things
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the NumPy-Discussion