Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

REL: Prepare for 1.14.1 release. #10630

Merged
merged 1 commit into from Feb 20, 2018

Conversation

Projects
None yet
4 participants
@charris
Copy link
Member

commented Feb 19, 2018

  • Update release notes.
  • Create 1.14.1-changelog.rst

[skip ci]

@charris charris force-pushed the charris:prepare-1.14.1 branch from e538842 to babcf6f Feb 19, 2018

@charris charris added this to the 1.14.1 release milestone Feb 19, 2018

default.
* The sort order in ``np.unique`` when ``axis=<some-number>`` will now always
be lexicographic in the subarray elements. Previously there was an
optimization the could sort the subarrays as unsigned byte strings.

This comment has been minimized.

Copy link
@Nodd

Nodd Feb 19, 2018

Contributor

Typo: replace the could by that could.

@charris charris force-pushed the charris:prepare-1.14.1 branch from babcf6f to f38f51c Feb 19, 2018

complex values,
* Problems with ``np.einsum`` with the new ``optimized=True`` default. Some
fixes for optimization have been applied and ``optimize=False`` is now the
default.

This comment has been minimized.

Copy link
@eric-wieser

eric-wieser Feb 19, 2018

Member

Perhaps "is once more the default", or something emphasizing we're back to the 1.13 behavior?

This comment has been minimized.

Copy link
@charris

charris Feb 19, 2018

Author Member

This was also in 1.13, indeed, since the axis parameter was introduced.

This comment has been minimized.

Copy link
@eric-wieser

eric-wieser Feb 20, 2018

Member

What I mean is I think we should be explicit here that 1.14.1 restores the 1.13.3 behavior

Affected users should read the the 1.14.1 Numpy User Guide for advice on how to
manage this transition, under "basics/structured arrays/accessing multiple
fields". A new method `numpy.lib.recfunctions.repack_fields` has been
introduced to help with this. All other changes to structured arrays described

This comment has been minimized.

Copy link
@eric-wieser

eric-wieser Feb 19, 2018

Member

We're choosing to leave this hidden then @ahaldane?

This comment has been minimized.

Copy link
@charris

charris Feb 19, 2018

Author Member

I copied the notes from master, probably this was only in the 1.14.x branch.

This comment has been minimized.

Copy link
@ahaldane

ahaldane Feb 20, 2018

Member

I thought this would be useful info for users, but if it's not part of the standard update-release-note I don't mind skipping it.

This comment has been minimized.

Copy link
@charris

charris Feb 20, 2018

Author Member

I've added back part of this to the list of problems fixed.

This comment has been minimized.

Copy link
@ahaldane

ahaldane Feb 20, 2018

Member

Looks good to me!

REL: Prepare for 1.14.1 release.
* Update release notes.
* Create `1.14.1-changelog.rst`

[skip ci]

@charris charris force-pushed the charris:prepare-1.14.1 branch from f38f51c to 7b49658 Feb 20, 2018

* The change in 1.14.0 that multi-field indexing of structured arrays returns a
view instead of a copy has been reverted but remains on track for NumPy 1.15.
Affected users should read the 1.14.1 Numpy User Guide section
"basics/structured arrays/accessing multiple fields" for advice on how to

This comment has been minimized.

Copy link
@eric-wieser

eric-wieser Feb 20, 2018

Member

Can we get an rst link for this, so that the sphinx release note page links to it?

This comment has been minimized.

Copy link
@ahaldane

ahaldane Feb 20, 2018

Member

Unfortunately those docs are inside a .py file: numpy/doc/structured_arrays.py, so I'm not sure that will work out well.

https://github.com/numpy/numpy/blob/maintenance/1.14.x/numpy/doc/structured_arrays.py

This comment has been minimized.

Copy link
@charris

charris Feb 20, 2018

Author Member

Let's not over design this, it's just a release note. The documentation will go up after the release.

This comment has been minimized.

Copy link
@charris

charris Feb 20, 2018

Author Member

I'd like to get rid of the *.py documentation files at some point. It is confusing to have the documentation scattered about like that.

@charris charris merged commit 7849751 into numpy:maintenance/1.14.x Feb 20, 2018

2 checks passed

continuous-integration/appveyor/pr AppVeyor build succeeded
Details
lgtm analysis: Python No alert changes
Details

@charris charris deleted the charris:prepare-1.14.1 branch Feb 20, 2018

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.