Updating docs

docs/conf.py

@@ -3,11 +3,13 @@
 author = "Tom de Geus"
 
 extensions = [
+    "sphinx_tabs.tabs",
+    "sphinx.ext.autosummary",
     "sphinx.ext.mathjax",
     "sphinx.ext.todo",
-    "sphinx_tabs.tabs",
 ]
 
+autodoc_type_aliases = {"Iterable": "Iterable", "ArrayLike": "ArrayLike", "DTypeLike": "DTypeLike"}
 templates_path = ["_templates"]
 exclude_patterns = ["_build", "Thumbs.db", ".DS_Store"]
 html_theme = "furo"

docs/python_install.rst

@@ -11,37 +11,30 @@ The quickest (but not the most efficient!) is to use *conda* to install *GooseEY
 
 .. warning::
 
-    This package does not benefit from *xsimd* optimisation, as it is not compiled on your hardware. Therefore compiling by hand is recommended.
+    This package does not benefit from *xsimd* optimisation, as it is not compiled on your hardware. Therefore compiling by hand can be advantageous.
 
 From source
 ^^^^^^^^^^^
 
-Start by installing the dependencies, for example using *conda*::
+Start by downloading the package::
 
-    conda install -c conda-forge pyxtensor xsimd
-
-Note that *xsimd* is optional, but recommended.
-
-.. note::
-
-    You can also use::
+    git checkout https://github.com/tdegeus/GooseEYE.git
+    cd GooseEYE
 
-        python -m pip install pyxtensor pybind11
+Then, install the dependencies, for example using *conda*::
 
-    for use without *conda*. Note that you install *xsimd* yourself in such a way that Python can find it in order to use it.
+    conda env update --file environment.yaml
 
-Then, download the package::
+Finally, install the package using::
 
-    git checkout https://github.com/tdegeus/GooseEYE.git
-    cd GooseEYE
+    python -m pip install . -v --no-build-isolation --no-deps
 
-Install the package using::
+If you want to use hardware optimisations, you can use::
 
-    python -m pip install .
+    SKBUILD_CONFIGURE_OPTIONS="-DUSE_SIMD=1" python -m pip install . -v --no-build-isolation --no-deps
 
 .. note::
 
-    The following will give more readable output::
-
-        python setup.py build
-        python setup.py install
+    Since you are compiling a C++ extension, with C++ libraries as dependencies, you cannot rely solely on *pip* to install the dependencies.
+    As observed, *conda* does contain all the dependencies.
+    Alternatively, you can use whatever non-supported method you prefer.

docs/python_usage.rst

@@ -2,9 +2,11 @@
  Functions & Usage
 ###################
 
-The Python interface is a simple wrapper around the C++ classes and
-functions. In general the Python and C++ codes are almost identical,
-with as only differences:
+Syntax
+======
+
+The Python interface is mostly wrapper around the C++ classes and functions.
+In general the Python and C++ codes are almost identical, with as only differences:
 
 -  The C++ ``xt::xarray`` are ``numpy.ndarray`` in Python.
 -  The syntax ``::`` in Python in simply a ``.``.
@@ -20,3 +22,19 @@ In Python is
 .. code:: python
 
    ensemble = GooseEYE.Ensemble([51, 51])
+
+Python only functions
+=====================
+
+Overview
+--------
+
+.. autosummary::
+
+   GooseEYE.Structure
+
+Details
+-------
+
+.. automodule:: GooseEYE
+    :members:

docs/theory_heightheight.rst

@@ -14,6 +14,14 @@ The structure factor is defined as:
     S(\vec{q}) = \langle \hat{u}(\vec{q}) \hat{u}(\vec{q}) \rangle
 
 where :math:`\vec{q}` is the angular frequency.
+It can be used to compute the roughness exponent :math:`\zeta`:
+
+.. math::
+
+    S(|q|) \sim |q|^{-(d + 2 \zeta)}
+
+where :math:`d` is the dimension of the image.
+An example is shown below.
 
 .. image:: examples/structure.svg
     :width: 700px

python/GooseEYE/__init__.py

@@ -11,6 +11,7 @@
 def ClusterLabeller(shape, periodic=True, **kwargs):
     """
     Allocate a cluster labeller.
+
     :param shape: The shape of the image.
     :param periodic: Whether the image is periodic.
     :return: A cluster labeller.