[Numpy-discussion] 1.7 blockers

18 views
Skip to first unread message

Charles R Harris

unread,
Mar 24, 2012, 5:13:53 PM3/24/12
to numpy-discussion
Hi All,

There several problems with numpy master that need to be fixed before a release can be considered.
  1. Datetime on windows with mingw.
  2. Bus error on SPARC, ticket #2076.
  3. NA and real/complex views of complex arrays.

Number 1 has been proved to be particularly difficult, any help or suggestions for that would be much appreciated. The current work has been going in pull request 214.

This isn't to say that there aren't a ton of other things that need fixing or that we can skip out on the current stack of pull requests, but I think it is impossible to consider a release while those three problems are outstanding.

Chuck

Ralf Gommers

unread,
Mar 25, 2012, 12:33:49 PM3/25/12
to Discussion of Numerical Python
There's one more ticket that hasn't been looked at AFAIK and that has been keeping the buildbots (except the Linux one) on red: http://projects.scipy.org/numpy/ticket/1755 (floating point errors).

The other tickets have been looked at and either have PRs already or should be fixable with not too much effort.

Ralf

Charles R Harris

unread,
Mar 25, 2012, 12:48:39 PM3/25/12
to Discussion of Numerical Python
On Sun, Mar 25, 2012 at 10:33 AM, Ralf Gommers <ralf.g...@googlemail.com> wrote:


On Sat, Mar 24, 2012 at 10:13 PM, Charles R Harris <charles...@gmail.com> wrote:
Hi All,

There several problems with numpy master that need to be fixed before a release can be considered.
  1. Datetime on windows with mingw.
  2. Bus error on SPARC, ticket #2076.
  3. NA and real/complex views of complex arrays.

Number 1 has been proved to be particularly difficult, any help or suggestions for that would be much appreciated. The current work has been going in pull request 214.

This isn't to say that there aren't a ton of other things that need fixing or that we can skip out on the current stack of pull requests, but I think it is impossible to consider a release while those three problems are outstanding.

There's one more ticket that hasn't been looked at AFAIK and that has been keeping the buildbots (except the Linux one) on red: http://projects.scipy.org/numpy/ticket/1755 (floating point errors).


I don't know what to do about that one.There may be some compiler flags that would help.

The other tickets have been looked at and either have PRs already or should be fixable with not too much effort.


Chuck

Ralf Gommers

unread,
Mar 25, 2012, 1:07:32 PM3/25/12
to Discussion of Numerical Python

Why do you think this isn't a problem with the implementation? I haven't tried to look at this in detail, but Python itself does manage to produce a warning:
>>> 2.0**2048
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
OverflowError: (34, 'Result too large')
>>> import numpy as np
>>> np.power(2.0, 2048)
inf
>>> np.geterr()
{'over': 'warn', 'divide': 'warn', 'invalid': 'warn', 'under': 'ignore'}

Ralf

Ralf Gommers

unread,
Mar 25, 2012, 5:14:24 PM3/25/12
to Discussion of Numerical Python
On Sat, Mar 24, 2012 at 10:13 PM, Charles R Harris <charles...@gmail.com> wrote:
Why do you consider (2) a blocker? Not saying it's not important, but there are eight other open tickets with segfaults. Some are more esoteric than other, but I don't see why for example #1713 and #1808 are less important than this one.

#1522 provides a patch that fixes a segfault by the way, could use a review.

Ralf

 

Reply all
Reply to author
Forward
0 new messages