Move python documentation things from opm-simulators to here

Get dune.module from opm-simulators as well so we can get the version
Trigger workflow on all changes, since this repo contains only documentation

.github/workflows/python_sphinx_docs.yml

@@ -0,0 +1,79 @@
+name: Build Python Sphinx Docs and push to gh-pages
+
+on:
+    push:
+        branches:
+          - '**'
+    pull_request_target:
+        types: closed
+        branches: main
+    repository_dispatch:
+      types: [docstrings_common_updated, docstrings_simulators_updated]
+permissions:
+    contents: write
+jobs:
+    build:
+      runs-on: ubuntu-latest
+      steps:
+        - name: Checkout source code
+          uses: actions/checkout@v4
+          with:
+            fetch-depth: 0   # Fetch all history for all tags and branches
+        - name: Get docstrings_common.json from master branch of opm-common
+          run: |
+            curl -L -o python/docstrings_common.json https://raw.githubusercontent.com/OPM/opm-common/master/python/docstrings_common.json
+        - name: Get docstrings_common.json from master branch of opm-simulators
+          run: |
+            curl -L -o python/docstrings_simulators.json https://raw.githubusercontent.com/OPM/opm-simulators/master/python/docstrings_simulators.json
+        - name: Get dune.module from master branch of opm-simulators
+          run: |
+            curl -L -o dune.module https://raw.githubusercontent.com/OPM/opm-simulators/master/dune.module
+        - name: Set up Python
+          uses: actions/setup-python@v5
+          with:
+            python-version: "3.11"
+        - name: Install poetry
+          uses: OPM/actions-poetry@master
+        - name: Install python dependencies
+          run: |
+            cd python/sphinx_docs
+            poetry install
+        - name: Build documentation
+          run: |
+            cd python
+            mkdir gh-pages
+            touch gh-pages/.nojekyll
+            cd sphinx_docs
+            # Currently we build only docs for the HEAD of the master/main branch
+            # Later we can add release tags to the list to get the docs for the releases
+            # For example: -b "main, release/2024.04/final" will build the docs for
+            # the main branch and the release/2024.04/final tag
+            # For the releases, we create snapshots of the docstrings_common.json and docstrings_simulators.json
+            # and take the ones tracked by git, only for the main, we take the current ones we fetched
+            # in steps 2 and 3 of this workflow
+
+            branch_name=${{ github.ref }}
+            if [[ "$branch_name" == refs/heads/* ]]; then
+              branch_name=${branch_name#refs/heads/}
+            fi
+            if [ "$branch_name" == "main" ]; then
+              poetry run sphinx-versioned -m main -b "main" --force --git-root ../../
+            else
+              poetry run sphinx-versioned -m main -b "main, ${branch_name}" --force --git-root ../../
+            fi
+        - name: Copy documentation to gh-pages
+          run: |
+            cp -r python/sphinx_docs/docs/_build/* python/gh-pages
+
+        - name: Deploy documentation for PR merge to main or push to main
+          if: github.ref == 'refs/heads/main'
+          uses: OPM/github-pages-deploy-action@releases/v4
+          with:
+            branch: gh-pages
+            folder: python/gh-pages
+        - name: Deploy documentation for other branches (on forks)
+          if: github.event_name == 'push' && github.ref != 'refs/heads/main'
+          uses: OPM/github-pages-deploy-action@releases/v4
+          with:
+            branch: gh-pages-${{ github.ref }}
+            folder: python/gh-pages

.gitignore

@@ -0,0 +1,14 @@
+# Files necessary to build the documentation locally, but these are tracked by other github repositories
+dune.module
+python/docstrings_common.json
+python/docstrings_simulators.json
+
+# Python sphinx build
+python/sphinx_docs/docs/_build*
+
+# Python cache directories
+**/__pycache__/
+
+# Editor autosave files
+*~
+*.swp

python/sphinx_docs/Makefile

@@ -0,0 +1,7 @@
+.PHONY: docs
+
+# Build the documentation locally for the current branch
+# NOTE: You need to commit your changes before running this command
+docs:
+	@CURRENT_BRANCH=$(shell git rev-parse --abbrev-ref HEAD)
+	sphinx-versioned -m $$CURRENT_BRANCH -b $$CURRENT_BRANCH --git-root ../../

python/sphinx_docs/README.md

@@ -0,0 +1,23 @@
+# Python scripts for building opm-simulators sphinx documentation
+
+## Installation of the python scripts
+- Requires python3 >= 3.10
+
+### Using poetry
+For development it is recommended to use poetry:
+
+- Install [poetry](https://python-poetry.org/docs/)
+- Then run:
+```
+$ poetry install
+$ poetry shell
+```
+
+### Installation into virtual environment
+If you do not plan to change the code, you can do a regular installation into a VENV:
+
+```
+$ python -m venv .venv
+$ source .venv/bin/activate
+$ pip install .
+```

python/sphinx_docs/docs/common.rst

@@ -0,0 +1,4 @@
+OPM Common Python Documentation
+===============================
+
+.. opm_common_docstrings::

python/sphinx_docs/docs/conf.py

@@ -0,0 +1,56 @@
+# Configuration file for the Sphinx documentation builder.
+
+project = "OPM Python Documentation"
+copyright = "2024 Equinor ASA"
+author = "Håkon Hægland"
+
+# Function to extract release version from dune.module file
+def extract_opm_simulators_release():
+    version_file_path = '../../../dune.module'
+    with open(version_file_path, 'r') as file:
+        for line in file:
+            if line.startswith('Version:'):
+                version_string = line.split(':')[1].strip()
+                return version_string
+    return "unknown"  # Fallback version
+
+release = extract_opm_simulators_release()
+
+# -- General configuration ---------------------------------------------------
+import os
+import sys
+
+# For regular Python packages, the path to the package is usually added to sys.path
+# here such that autodoc can find the modules. However, our Python module
+#  opm.simulators.BlackOilSimulator is not generated yet. Since it is a pybind11
+# extension module, it needs to be compiled as part of the opm-simulators build
+# process (which requires building opm-common first and so on). The full compilation
+# of opm-simulators requres time and storage resources, so we do not want to
+# do this as part of the documentation build. Instead, we will use a sphinx extension
+# (Python script) to generate documentation from a JSON input file. (This JSON file
+# is also is also used to generate a C++ header file with docstrings. The header file
+# is generated when opm-simulators is built and is then included
+# in the pybind11 binding code for the BlackOilSimulator class. In this way, the
+# opm.simulators.BlackOilSimulator Python module will also have access to the docstrings.)
+sys.path.insert(0, os.path.abspath("../src"))
+# Our sphinx extension that will use the docstrings.json file to generate documentation
+extensions = ["opm_simulators_docs.sphinx_ext_docstrings"]
+# Path to docstrings.json
+opm_simulators_docstrings_path = os.path.abspath('../../docstrings_simulators.json')
+opm_common_docstrings_path = os.path.abspath('../../docstrings_common.json')
+
+templates_path = ["_templates"]
+exclude_patterns = ["_build", "Thumbs.db", ".DS_Store"]
+
+
+# -- Options for HTML output -------------------------------------------------
+# See: https://www.sphinx-doc.org/en/master/usage/configuration.html#options-for-html-output
+html_theme = "sphinx_rtd_theme"
+html_static_path = ["_static"]
+html_context = {
+    "display_github": True,
+    "github_user": "OPM",
+    "github_repo": "opm-simulators",
+    "github_version": "master",
+    "conf_py_path": "/python/docs/",
+}

python/sphinx_docs/docs/embedded.rst

@@ -0,0 +1,54 @@
+OPM Embedded Python Documentation
+=================================
+
+The PYACTION keyword is a flow specific keyword which allows for executing embedded Python
+code in the SCHEDULE section. The embedded Python code will then be executed at the end of each successful timestep.
+
+The PYACTION keyword is inspired
+by the ACTIONX keyword, but instead of a .DATA formatted condition you
+are allowed to implement the condition with a general Python script. The
+ACTIONX keywords are very clearly separated in a condition part and an
+action part in the form of a list of keywords which are effectively injected in
+the SCHEDULE section when the condition evaluates to true.
+This is not so for PYACTION where there is one Python script in which both
+conditions can be evaluated and changes applied.
+
+In order to enable the PYACTION keyword:
+
+1. OPM flow must be compiled with the cmake switches -DOPM ENABLE EMBEDDED PYTHON=ON and -DOPM ENABLE PYTHON=ON, the default is to build with these switches set to OFF.
+
+2. The keyword PYACTION must be added to the SCHEDULE section:
+
+.. code-block:: python
+
+	<PYACTION\_NAME>  <SINGLE/UNLIMITED> /
+	<pythonscript> / -- path to the python script, relative to the location of the DATA-file
+
+3. You need to provide the Python script.
+
+
+To interact with the simulator in the embedded Python code, you can access four variables from the simulator:
+
+.. code-block:: python
+
+	# Python module opm_embedded
+	import opm_embedded
+	# The current EclipseState
+	ecl_state = opm_embedded.current_ecl_state
+	# The current Schedule
+	schedule = opm_embedded.current_schedule
+	# The current SummaryState
+	summary_state = opm_embedded.current_summary_state
+	# The current report step
+	report_step = opm_embedded.current_report_step 
+
+
+- current_ecl_state: An instance of the `EclipseState <common.html#opm.io.ecl_state.EclipseState>`_ class - this is a representation of all static properties in the model, ranging from porosity to relperm tables. The content of the ecl state is immutable - you are not allowed to change the static properties at runtime.
+
+- current_schedule: An instance of the `Schedule <common.html#opm.io.schedule.Schedule>`_ class - this is a representation of all the content from the SCHEDULE section, notably all well and group information and the timestepping.
+
+- current_report_step: This is an integer for the report step we are currently working on. Observe that the PYACTION is called for every simulator timestep, i.e. it will typically be called multiple times with the same value for the report step argument.
+
+- current_summary_state: An instance of the `SummaryState <common.html#opm.io.sim.SummaryState>`_ class, this is where the current summary results of the simulator are stored. The SummaryState class has methods to get hold of well, group and general variables.
+
+See also: PYACTION in the `reference manual <https://opm-project.org/?page_id=955>`_ for more information.

python/sphinx_docs/docs/index.rst

@@ -0,0 +1,22 @@
+Welcome to the Python documentation for OPM Flow!
+=================================================
+
+There are two Python APIs within OPM Flow:
+- running flow from Python code using the Python bindings
+- running a Python script embedded in a simulation
+
+.. toctree::
+    :maxdepth: 1
+    :caption: Contents:
+
+    introduction
+    common
+    simulators
+    embedded
+
+Indices and tables
+==================
+
+* :ref:`genindex`
+* :ref:`modindex`
+* :ref:`search`

python/sphinx_docs/docs/introduction.rst

@@ -0,0 +1,7 @@
+Introduction
+============
+
+Documentation for the OPM Python interfaces.
+
+#TODO: expand on the introduction, add information about installation and requirements for example. Some example code would also be nice.
+

python/sphinx_docs/docs/simulators.rst

@@ -0,0 +1,4 @@
+OPM Simulators Python Documentation
+===================================
+
+.. opm_simulators_docstrings::