René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"[ENH]: contour/tricontour: return values along contour","articleBody":"### Problem\r\n\r\n(A follow-up on https://github.com/matplotlib/matplotlib/issues/20548.)\r\n\r\nThe purpose of contour and tricontour is to draw contour lines along given levels. The algorithm tries to close the contour loops, but sometimes that's not possible or useful. Example:\r\n```python\r\nimport numpy as np\r\nimport matplotlib.pyplot as plt\r\n\r\nxmin, xmax = -3.0, 3.0\r\nymin, ymax = -3.0, 3.0\r\nnx, ny = 500, 500\r\n\r\ndef f(z):\r\n    return np.angle(z)\r\n\r\nx = np.linspace(xmin, xmax, nx)\r\ny = np.linspace(ymin, ymax, ny)\r\nX, Y = np.meshgrid(x, y)\r\nfZ = f(X + 1j * Y)\r\n\r\nc = plt.contour(X, Y, fZ, levels=[0.7853981633974483], colors=\"k\")\r\nplt.show()\r\n```\r\n\r\n![screenshot](https://user-images.githubusercontent.com/181628/136400404-756af36f-1441-420b-9708-1983bec5b31b.png)\r\n\r\nThe left \"arm\" of the contour is wrong. Let's verify this by computing the values:\r\n```python\r\nfor allseg in c.allsegs:\r\n    for seg in allseg:\r\n        x, y = seg.T\r\n        z = x + 1j * y\r\n        print(f(z))\r\n```\r\n```python\r\n[3.14109133 3.14108931 3.14108728 3.14108523 3.14108316 3.14108107\r\n# ...\r\n 3.13344468 3.13287655 3.13222326 3.1314641  3.13057109 3.1295054\r\n 3.12821158 3.12660761 3.12456683 3.12188272 3.1181943  3.11280868\r\n 3.10420637 3.08829091 3.04901618 2.8198421  0.78539816 0.78539816\r\n 0.78539816 0.78539816 0.78539816 0.78539816 0.78539816 0.78539816\r\n 0.78539816 0.78539816 0.78539816 0.78539816 0.78539816 0.78539816\r\n# ...\r\n.78539816 0.78539816 0.78539816 0.78539816 0.78539816 0.78539816\r\n 0.78539816 0.78539816]\r\n```\r\nNote that the specified level is `0.7853981633974483`, so the first half is completely off. [A previous comment](https://github.com/matplotlib/matplotlib/issues/20548#issuecomment-871232335) suggests that the contouring algorithm is right, though.\r\n\r\nTo fix this, I manually set the corresponding segments to `np.nan` which prevents plotting. However, this does _not_ work for tricontour as I don't have a function `f()` that I can evaluate at the segments. The contouring algorithm itself must compute the values somewhere, so my suggestion would be to expose those values to the user.","author":{"url":"https://github.com/nschloe","@type":"Person","name":"nschloe"},"datePublished":"2021-10-07T14:13:15.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":7},"url":"https://github.com/21309/matplotlib/issues/21309"}