René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"[ENH]: Mixed shading in pcolor/pcolormesh","articleBody":"### Problem\n\nCurrently [pcolor](https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.pcolor.html)/[pcolormesh](https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.pcolormesh.html) support `x` and `y` coordinate grids that have either the same size as the 2D array to plot (`shading='nearest'`), or one larger than the array (`shading='flat'`). However, it is not possible to plot an array with a coordinate grid with the same size along one dimension and one larger along the other, i.e using one shading method along `x` and the other along `y`.\n\nAn example of data in this shape I occasionally deal with are spectra defined by the edges of the frequency bins,  given at a number of points in time which I would like to plot as a spectrogram. Currently, it is not possible to make such a plot without adjusting the coordinate grid before the call to pcolor/pcolormesh .\n\n\n### Proposed solution\n\nI would propose to allow specifying the shading style individually for both dimensions and to determine the shading style individually for both dimensions as well when using `shading='auto'`. To do so, the [grid interpolation](https://github.com/matplotlib/matplotlib/blob/dd8d78b8dce60b6c8db86132892577a0b9dbe469/lib/matplotlib/axes/_axes.py#L5998) would have to be adjusted to do the grid interpolation along both dimensions separately.\n\nTo illustrate the proposed feature, an extended [pcolormesh grids and shading example](https://matplotlib.org/stable/gallery/images_contours_and_fields/pcolormesh_grids.html#auto-shading) would look like this:\n\n\u003cimg width=\"640\" height=\"960\" alt=\"Image\" src=\"https://github.com/user-attachments/assets/020c0390-f835-4f55-b477-6d7c19bb612f\" /\u003e","author":{"url":"https://github.com/Bzero","@type":"Person","name":"Bzero"},"datePublished":"2026-05-02T13:26:37.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":5},"url":"https://github.com/31607/matplotlib/issues/31607"}