[Numpy-discussion] NA masks in the next numpy release?
Charles R Harris
Sun Oct 23 16:42:42 CDT 2011
On Sun, Oct 23, 2011 at 3:28 PM, Benjamin Root <email@example.com> wrote:
> On Sunday, October 23, 2011, Nathaniel Smith <firstname.lastname@example.org> wrote:
> > On Sun, Oct 23, 2011 at 12:53 PM, Charles R Harris
> > <email@example.com> wrote:
> >> On Sun, Oct 23, 2011 at 12:54 PM, Matthew Brett <
> >> wrote:
> >>> I think this email might be a plea to the numpy steering group, and to
> >>> Travis in particular, to see if we can use a discussion of this series
> >>> of events to decide on a good way to proceed in future.
> >> Oh come, people had plenty to say, you and Nathaniel in particular.
> >> pointed to the pull request, anyone who was interested could comment on
> > Ah, this helps answer my initial question -- I can see how you might
> > have thought things were more resolved if you thought that we were
> > aware of the pull request and chose not to participate. That's a
> > reasonable source of confusion.
> > But I (and presumably others) were unaware of the pull request,
> > because it turns out that actually Mark did *not* point to the pull
> > request, at least in email to either me or numpy-discussion. As far as
> > I can tell, the first time that pull request has ever been mentioned
> > on the list is in Pauli's email today. (I did worry I might have
> > missed it, so I just double-checked the archives for August 18-August
> > 27, which is the time period the pull request was open, and couldn't
> > find anything there.)
> > (Also, for the record, I'd ask that next time you want to make sure
> > that there has been sufficient discussion on a controversial feature
> > that has "strong and reasonable opposition", you make more of an
> > effort to make sure that the relevant stakeholders are aware...?)
> >> Benjamin Root did so, for instance. The fact things didn't go the way
> >> wanted doesn't indicate insufficient discussion. And you are certainly
> >> welcome to put together an alternative and put up a pull request.
> > In the interests of not turning this into a game of procedural
> > brinksmanship, can we agree that the point of pull requests and such
> > is to make sure that code which ends up in numpy releases generally
> > matches what the community wants? Obviously the community has not
> > reached a consensus on this code and API, so I'll prepare a pull
> > request to temporarily revert the change, and we can work from there.
> > -- Nathaniel
> The discussion started on mark's branches, which was referred to several
> times in emails (that's how I started). When it reached a particular level
> of maturity, a pull request was made and additional work went into it. The
> initial discussion happened for quite a while.
> Plus, my understanding is that it isnt the full Nep, but the core parts
> (but I haven't checked in a while).
In its current state, it is a working implementation that can be used to
explore the API. Bit patterns are missing and the masks are handled at the
iterator level rather than in the low level ufunc loops, so it isn't
particularly fast. IIRC, Mark was careful to leave some hooks for further
development and also set things up so that in the future masks could be
adapted to allow different mask values with different interpretations.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the NumPy-Discussion