Enhance documentation and dependencies: update index.rst for improved clarity and add sphinx-design for enhanced styling

Author: CSSFrancis

docs/conf.py

@@ -33,6 +33,7 @@
     "sphinx.ext.viewcode",
     "sphinx.ext.intersphinx",
     "sphinx_gallery.gen_gallery",
+    "sphinx_design",
 ]
 
 autosummary_generate = True

docs/index.rst

@@ -1,63 +1,54 @@
-anyplotlib
-==========
+========================
+anyplotlib Documentation
+========================
 
-.. toctree::
-   :maxdepth: 2
-   :caption: Contents
-
-   getting_started
-   api/index
-   auto_examples/index
-
-Welcome to **anyplotlib** – a lightweight, interactive viewer for 1-D signals and
-2-D images, backed by `anywidget <https://anywidget.dev/>`_ and a pure-JavaScript
-canvas renderer.  The goal is to duplicate and extend the interactive plotting capabilities of Matplotlib,
-although the scope is intentionally limited in the following ways:
+anyplotlib is a lightweight, interactive plotting library for Jupyter notebooks and JupyterLab,
+backed by `anywidget <https://anywidget.dev/>`_ and a pure-JavaScript canvas renderer. It follows
+the object-oriented Matplotlib API — create a ``Figure``, call methods on ``Axes`` — while
+delivering real-time interactivity and high performance on large datasets through canvas-based
+blitting instead of SVG.
 
-1. This uses the object-oriented API of Matplotlib, not the stateful pyplot interface. This means there is
-   no ``plt.imshow`` or ``plt.plot`` — instead, you create a figure object and call methods on axes to add
-   data and customize the plot. This is a deliberate choice to avoid the pitfalls of the stateful API.
+.. grid:: 2 3 3 3
+   :gutter: 2
 
-   .. code-block:: python
+   .. grid-item-card::
+      :link: getting_started
+      :link-type: doc
 
-      import anyplotlib as apl
-      import matplotlib.pyplot as plt
+      :octicon:`rocket;2em;sd-text-info` Getting Started
+      ^^^
 
-      # matplotlib:
-      fig, axs = plt.subplots(1, 1)
-      axs.imshow(...)
+      New to anyplotlib? The getting started guide walks through installation and
+      your first interactive figure in a Jupyter notebook.
 
-      # anyplotlib equivalent:
-      fig, axs = apl.subplots(1, 1)
-      axs.imshow(...)
+   .. grid-item-card::
+      :link: api/index
+      :link-type: doc
 
-2. In matplotlib they use vector graphics (SVG) to render the plot, which is great for static images.  It's especially
-   great for making publication-quality figures.  (If you haven't tried inkscape + matplotlib SVG output,
-   it's pretty amazing.) For interactivity, it can be slow.  Anyplotlib uses a pure-JavaScript canvas renderer which is
-   much faster for interactive applications, but the quality of the output is not as good as vector graphics.  This is a
-   trade-off that we are willing to make for the sake of interactivity.
+      :octicon:`code-square;2em;sd-text-info` API Reference
+      ^^^
 
-3. Matplotlib supports a wide range of marker styles, line styles, and other plot elements.  Anyplotlib focuses on a
-   core set of features that are most commonly used in scientific plotting.  This means that some of the more
-   esoteric features of Matplotlib may not be available in Anyplotlib. In general we try to match the lower level
-   ``collections`` API of Matplotlib.
+      Full documentation of the anyplotlib API — ``Figure``, ``Axes``, plot classes,
+      markers, widgets, and callbacks — with parameter descriptions and type signatures.
 
-4. Each collection, plot, image is rendered as a single object on the canvas.  This is highly performant and more
-   importantly allows for blitting. This is one of the main reasons why the ``ipympl`` backend of Matplotlib is so slow.
+   .. grid-item-card::
+      :link: auto_examples/index
+      :link-type: doc
 
-5. Finally ``anyplotlib`` uses ``AnyWidget`` as the underlying widget framework.  This means that it can be used in any
-   environment that supports ``AnyWidget``, including Jupyter notebooks, JupyterLab, and PyCharm notebook preview.  Under
-   the hood, ``AnyWidget`` uses a pure-JavaScript implementation of the widget protocol, which allows for fast rendering
-   and interactivity.
+      :octicon:`zap;2em;sd-text-info` Examples
+      ^^^
 
-**Status**: anyplotlib v0.1.0 provides a lightweight, interactive alternative to matplotlib's
-pyplot interface for Jupyter notebooks and compatible environments. Performance is optimized for
-real-time interactivity with large datasets on canvas-based rendering.
+      Gallery of short, self-contained examples showing 1-D signals, 2-D images,
+      3-D surfaces, bar charts, interactive widgets, and more.
 
 .. toctree::
    :hidden:
+   :maxdepth: 2
+
+   getting_started
+   api/index
+   auto_examples/index
 
 * :ref:`genindex`
 * :ref:`modindex`
 * :ref:`search`
-

pyproject.toml

@@ -21,6 +21,7 @@ dependencies = [
 docs = [
     "sphinx>=8.0",
     "pydata-sphinx-theme>=0.16",
+    "sphinx-design>=0.6",
     "sphinx-gallery>=0.18",
     "pillow>=10.0",
     "matplotlib>=3.7",