What’s New in Revision 1.1.1

These are the changes in SpectroChemPy-1.1.1. See Release notes for a full changelog, including other versions of SpectroChemPy.

Bug Fixes

  • output is now honored by the plotting API instead of being silently ignored, fixing a longstanding defect reported by a user in #1719. dataset.plot(), plot_multiple(), multiplot(), and the composite plotters (plot_score(), plot_scree(), plot_compare(), plot_merit(), plot_baseline(), and plot_parity()) write the finished figure to the requested file, accepting both str and pathlib.Path. The whole figure is saved once the plot is complete - legend, colorbars, and multi-panel layouts included - and before the display step, so the file is on disk even when show=True keeps a window open. The file format follows the extension, the existing savefig.* preferences drive resolution, background, and bounding box, and a missing parent directory is reported as an OSError instead of being created silently. When axes are supplied explicitly, the saved figure is the one that owns those axes, even when it is not the active Matplotlib figure. multiplot() also performs a single display step for the whole grid instead of one per panel, and plot_multiple() given a single dataset now delegates to dataset.plot(), as documented, instead of iterating the values of the dataset and dropping its show, clear, and output arguments (PR #1717).

    For example, a pen plot can be saved without displaying it:

    dataset.plot_pen(output="spectrum.png", show=False)

  • All twelve public geometry shortcuts now follow that same finalization path: plot_scatter(), plot_pen(), plot_scatter_pen(), plot_bar(), plot_lines(), plot_contour(), plot_contourf(), plot_stack(), plot_map(), plot_image(), plot_surface(), and plot_waterfall(). Both bound dataset methods and top-level functions save the completed figure before any requested display, honor show exactly once, and return the same Matplotlib axes as the equivalent dataset.plot(method=...) call. Plugin-owned plotting APIs, including the IRIS methods, are outside the scope of this core fix (PR #1720).

Developer

  • Conda main uploads are now restricted to final core release tags. Release candidates remain downloadable CI artifacts and are not published to Conda (PR #1721).

  • Version metadata commits for independently released Carroucell 0.1.10, IRIS 0.1.10, NMR 0.1.13, PerkinElmer 0.1.6, and Tensor 0.1.7 are present in the monorepo history since core 1.1.0. The core 1.1.1 release does not republish these plugins.