1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
|
Behavior Changes
----------------
onselect argument to selector widgets made optional
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The *onselect* argument to `.EllipseSelector`, `.LassoSelector`, `.PolygonSelector`, and
`.RectangleSelector` is no longer required.
``NavigationToolbar2.save_figure`` now returns filepath of saved figure
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
``NavigationToolbar2.save_figure`` function may return the filename of the saved figure.
If a backend implements this functionality it should return `None`
in the case where no figure is actually saved (because the user closed the dialog without saving).
If the backend does not or can not implement this functionality (currently the Gtk4 backends
and webagg backends do not) this method will return ``NavigationToolbar2.UNKNOWN_SAVED_STATUS``.
SVG output: improved reproducibility
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Some SVG-format plots `produced different output on each render <https://github.com/matplotlib/matplotlib/issues/27831>`__, even with a static ``svg.hashsalt`` value configured.
The problem was a non-deterministic ID-generation scheme for clip paths; the fix introduces a repeatable, monotonically increasing integer ID scheme as a replacement.
Provided that plots add clip paths themselves in deterministic order, this enables repeatable (a.k.a. reproducible, deterministic) SVG output.
ft2font classes are now final
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The ft2font classes `.ft2font.FT2Font`, and `.ft2font.FT2Image` are now final
and can no longer be subclassed.
``InsetIndicator`` artist
~~~~~~~~~~~~~~~~~~~~~~~~~
`~.Axes.indicate_inset` and `~.Axes.indicate_inset_zoom` now return an instance
of `~matplotlib.inset.InsetIndicator`. Use the
`~matplotlib.inset.InsetIndicator.rectangle` and
`~matplotlib.inset.InsetIndicator.connectors` properties of this artist to
access the objects that were previously returned directly.
``imshow`` *interpolation_stage* default changed to 'auto'
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The *interpolation_stage* parameter of `~.Axes.imshow` has a new default
value 'auto'. For images that are up-sampled less than a factor of
three or down-sampled, image interpolation will occur in 'rgba' space. For images
that are up-sampled by a factor of 3 or more, then image interpolation occurs
in 'data' space.
The previous default was 'data', so down-sampled images may change subtly with
the new default. However, the new default also avoids floating point artifacts
at sharp boundaries in a colormap when down-sampling.
The previous behavior can achieved by setting the *interpolation_stage* parameter
or :rc:`image.interpolation_stage` to 'data'.
imshow default *interpolation* changed to 'auto'
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The *interpolation* parameter of `~.Axes.imshow` has a new default
value 'auto', changed from 'antialiased', for consistency with *interpolation_stage*
and because the interpolation is only anti-aliasing during down-sampling. Passing
'antialiased' still works, and behaves exactly the same as 'auto', but is discouraged.
dark_background and fivethirtyeight styles no longer set ``savefig.facecolor`` and ``savefig.edgecolor``
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
When using these styles, :rc:`savefig.facecolor` and :rc:`savefig.edgecolor`
now inherit the global default value of "auto", which means that the actual
figure colors will be used. Previously, these rcParams were set to the same
values as :rc:`figure.facecolor` and :rc:`figure.edgecolor`, i.e. a saved
figure would always use the theme colors even if the user manually overrode
them; this is no longer the case.
This change should have no impact for users that do not manually set the figure
face and edge colors.
Add zorder option in QuiverKey
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
``zorder`` can be used as a keyword argument to `.QuiverKey`. Previously,
that parameter did not have any effect because the zorder was hard coded.
Subfigures
~~~~~~~~~~
`.Figure.subfigures` are now added in row-major order to be consistent with
`.Figure.subplots`. The return value of `~.Figure.subfigures` is not changed,
but the order of ``fig.subfigs`` is.
(Sub)Figure.get_figure
~~~~~~~~~~~~~~~~~~~~~~
...in future will by default return the direct parent figure, which may be a SubFigure.
This will make the default behavior consistent with the
`~matplotlib.artist.Artist.get_figure` method of other artists. To control the
behavior, use the newly introduced *root* parameter.
``transforms.AffineDeltaTransform`` updates correctly on axis limit changes
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Before this change, transform sub-graphs with ``AffineDeltaTransform`` did not update correctly.
This PR ensures that changes to the child transform are passed through correctly.
The offset string associated with ConciseDateFormatter will now invert when the axis is inverted
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Previously, when the axis was inverted, the offset string associated with ConciseDateFormatter would not change,
so the offset string indicated the axis was oriented in the wrong direction. Now, when the axis is inverted, the offset
string is oriented correctly.
``suptitle`` in compressed layout
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Compressed layout now automatically positions the `~.Figure.suptitle` just
above the top row of axes. To keep this title in its previous position,
either pass ``in_layout=False`` or explicitly set ``y=0.98`` in the
`~.Figure.suptitle` call.
|
