Fix/hide some deprecations #7857

QuLogic · Jan 17, 2017

I found the first commit lying around on disk, so better get it merged sometime.

anntzer · Jan 17, 2017

lib/matplotlib/projections/geo.py

@@ -190,7 +190,7 @@ def set_longitude_grid(self, degrees):
        """
        Set the number of degrees between each longitude grid.
        """
-        number = (360.0 / degrees) + 1
+        number = int(360.0 / degrees) + 1


360 / degrees, no need for .0 anymore (... especially given that you're casting to int immediately after). Same below.

Also I have no knowledge about this part of the code but are you sure it's shouldn't be math.ceil(360/degrees)? (Just guessing.)

Actually, I don't think this line is correct. It should probably be number = int(np.round(360 / degrees)). Right now, if degrees = 360, then number is equal to 2, which I am pretty sure is wrong.

Builtin round should be enough, then.

Do NOT use built-in round, that introduces differences between Python 2 and 3. Use numpy's.

I think we should always use numpy stuff when we can. We depend on numpy, so there is no reason to not use it (unless I am missing something).

I don't understand; if degrees = 360, it's a nice round division, so it doesn't matter if you use ceil, round, or plain int.

I think the confusion was whether it's working correctly. I'm not sure what rounding behavior we want, but the + 1 is needed to work with the call to linspace below, since we're removing the first and last points. Would probably make more sense as:

number = np.round(360 / degrees) ticks = np.linspace(-nppi, np.pi, number, endpoint=False)[1:] self.axis.set_major_locator(FixedLocator(ticks))

I rewrote these cases in a way that I hope makes it clearer.

tacaswell · Jan 17, 2017

lib/matplotlib/testing/compare.py

@@ -211,7 +211,7 @@ def convert(filename, cache):
 verifiers = {}

 # Turning this off, because it seems to cause multiprocessing issues
-if matplotlib.checkdep_xmllint() and False:


iirc @anntzer deletes this in another PR?

I deprecated it, so during the deprecation period the "False" should indeed come first to avoid triggering a deprecation warning.

ah, fair enough.

QuLogic · Jan 21, 2017

There seems to be a bug with the spectral deprecation warning; it's not triggered after the production of the reversed version in cm.py on Python 2.7, only.

QuLogic · Jan 21, 2017

That last commit should fix the warning on 2.7; not sure if we should backport it as at the moment, users on 2.7 don't see the warning they're supposed to.

QuLogic · Jan 22, 2017

Rebased due to conflicts.

NelleV · Jan 22, 2017

I just want to make sure I understand your last comment.
If we include this patch, users won't see the spectral warning anymore, or without this patch, they don't see the deprecation warning?

QuLogic · Jan 22, 2017

Without the last commit, on Python 2.7, users don't see the deprecation warning:

$ python
Python 2.7.12 (default, Sep 29 2016, 12:52:02) 
[GCC 6.2.1 20160916 (Red Hat 6.2.1-2)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import matplotlib
>>> matplotlib.__version__
'2.0.0'
>>> import matplotlib.pyplot as plt
>>> plt.get_cmap('spectral')
<matplotlib.colors.LinearSegmentedColormap object at 0x7f61bfd5f0d0>

$ python3
Python 3.5.2 (default, Sep 14 2016, 11:28:32) 
[GCC 6.2.1 20160901 (Red Hat 6.2.1-1)] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import matplotlib
>>> matplotlib.__version__
'2.0.0'
>>> import matplotlib.pyplot as plt
>>> plt.get_cmap('spectral')
/usr/lib64/python3.5/site-packages/matplotlib/cbook.py:136: MatplotlibDeprecationWarning: The spectral and spectral_r colormap was deprecated in version 2.0. Use nipy_spectral and nipy_spectral_r instead.
  warnings.warn(message, mplDeprecation, stacklevel=1)
<matplotlib.colors.LinearSegmentedColormap object at 0x7f985bbc89e8>
>>>

QuLogic · Jan 22, 2017

Oh, and to be clear, the second commit hides the warning when running tests only. I'm only wondering about backporting the last commit and not the entire PR.

anntzer · Jan 25, 2017

lib/matplotlib/tests/test_path.py

@@ -97,7 +97,7 @@ def test_make_compound_path_empty():
 def test_xkcd():
    np.random.seed(0)

-    x = np.linspace(0, 2.0 * np.pi, 100.0)
+    x = np.linspace(0, 2.0 * np.pi, 100)


2.0 -> 2 (now that we're there...)

anntzer · Jan 28, 2017

lib/matplotlib/cm.py

@@ -83,8 +83,9 @@ def _generate_cmap(name, lutsize):

 # We silence warnings here to avoid raising the deprecation warning for
 # spectral/spectral_r when this module is imported.
-with _warnings.catch_warnings():
+with _warnings.catch_warnings(record=True):


This seems way too complicated. I would just invoke getitem using the superclass method, i.e. use dict.__getitem__(cmap_d, name) instead of cmap_d[name] at the right place. (The first form does not raise a warning).

That's a good idea and really simplifies things.

This fixes "DeprecationWarning: object of type <class 'float'> cannot be safely interpreted as an integer." raised by latest NumPy.

Do the calculation in degrees instead of trying to "optimize" the calculation in radians.

In 2.7, the hidden __warningregistry__ gets set and then the warning is never seen again, regardless of the filter setting. Instead, just never trigger the warning in the first place (thanks @anntzer).

Backport spectral/spectral_r deprecations fixes from #7857

anntzer reviewed Jan 17, 2017

View reviewed changes

tacaswell reviewed Jan 17, 2017

View reviewed changes

tacaswell added this to the 2.1 (next point release) milestone Jan 17, 2017

QuLogic force-pushed the deprecations branch from 4d0b263 to a6316f5 Compare January 20, 2017 22:59

QuLogic changed the title ~~Fix/hide some deprecations~~ [MRG] Fix/hide some deprecations Jan 20, 2017

QuLogic changed the title ~~[MRG] Fix/hide some deprecations~~ Fix/hide some deprecations Jan 21, 2017

QuLogic changed the title ~~Fix/hide some deprecations~~ [MRG] Fix/hide some deprecations Jan 21, 2017

QuLogic force-pushed the deprecations branch from 0396a6d to 88b5bc1 Compare January 22, 2017 03:14

QuLogic force-pushed the deprecations branch from 88b5bc1 to b358d10 Compare January 23, 2017 07:32

anntzer reviewed Jan 25, 2017

View reviewed changes

QuLogic force-pushed the deprecations branch from b358d10 to fe8a90e Compare January 25, 2017 22:50

QuLogic closed this Jan 28, 2017

QuLogic reopened this Jan 28, 2017

anntzer reviewed Jan 28, 2017

View reviewed changes

QuLogic added 5 commits January 28, 2017 15:34

Pass integers to np.linspace/np.logspace's count.

97e170d

This fixes "DeprecationWarning: object of type <class 'float'> cannot be safely interpreted as an integer." raised by latest NumPy.

Catch warnings from deprecated colourmaps.

ffbc9bd

TST: Really avoid calling checkdep_xmllint in tests.

dcaad03

Make GeoAxes.set_{latitude,longitude}_grid clearer.

d6149e1

Do the calculation in degrees instead of trying to "optimize" the calculation in radians.

Fix spectral/spectral_r deprecations on Python 2.7.

e214769

In 2.7, the hidden __warningregistry__ gets set and then the warning is never seen again, regardless of the filter setting. Instead, just never trigger the warning in the first place (thanks @anntzer).

QuLogic force-pushed the deprecations branch from fe8a90e to e214769 Compare January 28, 2017 20:35

anntzer changed the title ~~[MRG] Fix/hide some deprecations~~ [MRG+1] Fix/hide some deprecations Jan 28, 2017

NelleV merged commit 0ce413f into matplotlib:master Jan 29, 2017

NelleV added a commit that referenced this pull request Jan 29, 2017

Merge pull request #7982 from QuLogic/fix-spectral-deprecation

6061a43

Backport spectral/spectral_r deprecations fixes from #7857

QuLogic deleted the deprecations branch January 31, 2017 09:03

QuLogic changed the title ~~[MRG+1] Fix/hide some deprecations~~ Fix/hide some deprecations Jan 31, 2017

Search code, repositories, users, issues, pull requests...

Uh oh!

Fix/hide some deprecations #7857

Fix/hide some deprecations #7857

Uh oh!

Conversation

QuLogic commented Jan 17, 2017

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

QuLogic commented Jan 21, 2017

Uh oh!

QuLogic commented Jan 21, 2017

Uh oh!

QuLogic commented Jan 22, 2017

Uh oh!

NelleV commented Jan 22, 2017

Uh oh!

QuLogic commented Jan 22, 2017

Uh oh!

QuLogic commented Jan 22, 2017

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!