Merge pull request #129 from loli/HOTFIXES_release

Hotfixes release

Author: loli

CHANGES.txt

@@ -1,3 +1,9 @@
+v0.5.0, 2024-04-XX -- Addressed all depreciation warnings and incompatabilities
+                      Updated documentation
+                      Updated and fixed tests
+                      Added github workflows as system
+                      Introduced formatting rules and pre-commit to enforce them
+                      Removed dockerfile creation files
 v0.4.0, 2018-02-XX -- Switched to Python 3: finally compatible with modern development environements
                       Switched to simple itk for image loading/saving: read dicom series, more formats, less dependencies, cleaner code, easier to maintain
                       Documentation: installation instructions for Windows and OsX

doc/source/installation/conda.rst

@@ -2,15 +2,10 @@
 Installing MedPy with Conda
 ===========================
 
-**MedPy** does not come with an official conda package.
-But you can nevertheless install it into a conda environement using *pip* after resolving the dependencies.
+**MedPy** can be installed from `conda forge <https://anaconda.org/conda-forge/medpy>`:
 
 .. code-block:: bash
 
-	conda install -c simpleitk simpleitk
-	python3 -m pip install medpy
+	conda install conda-forge::medpy
 
-Note that the graph-cut package won't compile in the conda environement due to unmet dependencies.
-
-For conda-purists: The friendly folks from `bioconda <https://bioconda.github.io/>`_ wrapped the previous (0.3.0) version of **MedPy**
-into their distribution system (see https://anaconda.org/bioconda/medpy).
+Note that the graph-cut package might not compile in the conda environement due to unmet dependencies.

doc/source/installation/graphcutsupport.rst

@@ -18,3 +18,34 @@ These dependencies can be found in the repositories of all major distribution. F
 	sudo apt-get install libboost-python-dev build-essential
 
 Then install **MedPy** the usual way.
+
+Troubleshooting
+---------------
+
+If you experience an error like `ModuleNotFoundError: No module named 'medpy.graphcut.maxflow'`, this usually means
+that the `graphcut`` module has not been compiled successfully. To check the error log, try re-installing **MedPy** with:
+
+.. code-block:: bash
+
+	pip install medpy --no-cache-dir --force-reinstall -v
+
+In the logs, you might see the following warning:
+
+.. code-block:: bash
+
+	2021-06-30T11:07:32,684   ***************************************************************************
+	2021-06-30T11:07:32,685   WARNING: The medpy.graphcut.maxflow external C++ package could not be compiled, all graphcut functionality will be disabled. You might be missing Boost.Python or some build essentials like g++.
+	2021-06-30T11:07:32,685   Failure information, if any, is above.
+	2021-06-30T11:07:32,685   I'm retrying the build without the graphcut C++ module now.
+	2021-06-30T11:07:32,685   ***************************************************************************
+
+The error should be detailed in the lines just above.
+
+Usually, it is a problem with the linking of the `(lib)boost_python3`` lib.
+There are some inconsistent naming conventions around, rendering the file undiscoverable to **MedPy**.
+
+On Ubuntu, you should be able to locate your *libboost_python3x.so* under `/usr/lib/x86_64-linux-gnu/`.
+If your shared library file is named differently than **MedPy** expects, you might have to create a softlink like, e.g.:
+
+.. code-block:: bash
+	sudo ln -s libboost_python38.so libboost_python3.so