René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"[Bug]: plt.clabel raises exception at very low DPI: `ValueError: 'codes' must be a 1D list or array with the same length of 'vertices'. Your vertices have shape (2, 2) but your codes have shape (1,)`","articleBody":"### Bug summary\n\nThis worked up to and including Matplotlib 3.7.3 and broke in Matplotlib 3.8.0.\r\n\r\nData files for sample code below:\r\n[x.txt](https://github.com/matplotlib/matplotlib/files/12782405/x.txt)\r\n[y.txt](https://github.com/matplotlib/matplotlib/files/12782406/y.txt)\r\n[z.txt](https://github.com/matplotlib/matplotlib/files/12782407/z.txt)\r\n\n\n### Code for reproduction\n\n```python\nimport matplotlib\r\nmatplotlib.rcParams['figure.dpi'] = matplotlib.rcParams['savefig.dpi'] = 4\r\n\r\nfrom matplotlib import pyplot as plt\r\nimport numpy as np\r\n\r\n# See data files attached above: x.txt, y.txt, z.txt\r\nx = np.loadtxt('x.txt')\r\ny = np.loadtxt('y.txt')\r\nz = np.loadtxt('z.txt')\r\n\r\nax = plt.axes()\r\ncontourset = ax.contour(x, y, z, levels=[90.0], linewidths=0.5)\r\nax.clabel(contourset)\r\nplt.savefig('test.pdf')\n```\n\n\n### Actual outcome\n\n```\r\nTraceback (most recent call last):\r\n  File \"/private/tmp/test.py\", line 13, in \u003cmodule\u003e\r\n    ax.clabel(contourset)\r\n  File \"/Users/lpsinger/Library/Python/3.11/lib/python/site-packages/matplotlib/axes/_axes.py\", line 6550, in clabel\r\n    return CS.clabel(levels, **kwargs)\r\n           ^^^^^^^^^^^^^^^^^^^^^^^^^^^\r\n  File \"/Users/lpsinger/Library/Python/3.11/lib/python/site-packages/matplotlib/contour.py\", line 222, in clabel\r\n    self.labels(inline, inline_spacing)\r\n  File \"/Users/lpsinger/Library/Python/3.11/lib/python/site-packages/matplotlib/contour.py\", line 622, in labels\r\n    rotation, path = self._split_path_and_get_label_rotation(\r\n                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\r\n  File \"/Users/lpsinger/Library/Python/3.11/lib/python/site-packages/matplotlib/contour.py\", line 436, in _split_path_and_get_label_rotation\r\n    return angle, Path(xys, codes)\r\n                  ^^^^^^^^^^^^^^^^\r\n  File \"/Users/lpsinger/Library/Python/3.11/lib/python/site-packages/matplotlib/path.py\", line 135, in __init__\r\n    raise ValueError(\"'codes' must be a 1D list or array with the \"\r\nValueError: 'codes' must be a 1D list or array with the same length of 'vertices'. Your vertices have shape (2, 2) but your codes have shape (1,)\r\n```\n\n### Expected outcome\n\n[test.pdf](https://github.com/matplotlib/matplotlib/files/12782396/test.pdf)\r\n\n\n### Additional information\n\nI discovered this regression in the CI pipeline from one of my own projects, in a test that sets DPI to an extremely low value for performance purposes. https://git.ligo.org/lscsoft/ligo.skymap/-/jobs/2961694\n\n### Operating system\n\nmacOS\n\n### Matplotlib Version\n\n3.8.0\n\n### Matplotlib Backend\n\nMacOSX\n\n### Python version\n\n3.11.5\n\n### Jupyter version\n\nN/A\n\n### Installation\n\npip","author":{"url":"https://github.com/lpsinger","@type":"Person","name":"lpsinger"},"datePublished":"2023-10-02T12:06:47.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":5},"url":"https://github.com/26971/matplotlib/issues/26971"}