Skip to content

[Bug]: Deepcopy of TextPath fails #20943

Closed
@deep-jkl

Description

@deep-jkl

Bug summary

The problem is that deepcopy of TextPath calls deepcopy of Path. In turn Path utilizes super().init for creating a new
instance. However, Path.init and TextPath.init are completely different.

Code for reproduction

from matplotlib.textpath import TextPath
TextPath((0, 0), ".").deepcopy()

Actual outcome

AttributeError                            Traceback (most recent call last)
c:\figures.py in <module>
----> 1 TextPath((0, 0), ".").deepcopy()

~\.site-packages\matplotlib\path.py in __deepcopy__(self, memo)
    284         return self.__class__(
    285             self.vertices.copy(), codes,
--> 286             _interpolation_steps=self._interpolation_steps)
    287 
    288     deepcopy = __deepcopy__

~\.site-packages\matplotlib\textpath.py in __init__(self, xy, s, size, prop, _interpolation_steps, usetex)
    394 
    395         self._cached_vertices = None
--> 396         s, ismath = Text(usetex=usetex)._preprocess_math(s)
    397         self._vertices, self._codes = text_to_path.get_text_path(
    398             prop, s, ismath=ismath)

~\.site-packages\matplotlib\text.py in _preprocess_math(self, s)
   1186             return s, True
   1187         else:
-> 1188             return s.replace(r"\$", "$"), False
   1189 
   1190     def set_fontproperties(self, fp):

AttributeError: 'list' object has no attribute 'replace'

Expected outcome

A new instance of TextPath.

Operating system

Windows 10

Matplotlib Version

3.3.4

Matplotlib Backend

module://ipykernel.pylab.backend_inline

Python version

Python 3.6.6

Jupyter version

No response

Other libraries

No response

Installation

pip

Conda channel

No response

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions