Notebook tests (#978)

* Fix notebook test runs

* Delete old issue template

* Add notebook CI action

* Print temp directories

* Print more env

* Move printing up

* Use runner_temp

* Try using current directory

* Try TMP env

* Re-write TMP

* Wrong yml

* Fix echo

* Only export if windows

* More logging

* Move export

* Reformat env write

* Fix braces

* Switch to in-memory execution

* Downgrade action perms

* Unused import

Author: natoverse

.github/ISSUE_TEMPLATE.md

@@ -84,10 +84,6 @@ jobs:
           poetry run python -m pip install gensim
           poetry install
 
-      - name: Check Semversioner
-        run: |
-          poetry run semversioner check
-
       - name: Check
         run: |
           poetry run poe check

.github/workflows/python-ci.yml

@@ -0,0 +1,70 @@
+name: Python Notebook Tests
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+permissions:
+  contents: read
+  pull-requests: read
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+  # Only run the for the latest commit
+  cancel-in-progress: true
+
+env:
+  POETRY_VERSION: 1.8.3
+
+jobs:
+  python-ci:
+    strategy:
+      matrix:
+        python-version: ["3.11"]
+        os: [ubuntu-latest, windows-latest]
+      fail-fast: false # Continue running all jobs even if one fails
+    env:
+      DEBUG: 1
+      GRAPHRAG_API_KEY: ${{ secrets.OPENAI_NOTEBOOK_KEY }}
+      GRAPHRAG_LLM_MODEL: ${{ secrets.GRAPHRAG_LLM_MODEL }}
+      GRAPHRAG_EMBEDDING_MODEL: ${{ secrets.GRAPHRAG_EMBEDDING_MODEL }}
+
+    runs-on: ${{ matrix.os }}
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: dorny/paths-filter@v3
+        id: changes
+        with:
+          filters: |
+            python:
+              - 'graphrag/**/*'
+              - 'poetry.lock'
+              - 'pyproject.toml'
+              - '**/*.py'
+              - '**/*.toml'
+              - '**/*.ipynb'
+              - '.github/workflows/python*.yml'
+
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+
+      - name: Install Poetry
+        uses: abatilo/actions-poetry@v3.0.0
+        with:
+          poetry-version: $POETRY_VERSION
+
+      - name: Install dependencies
+        shell: bash
+        run: |
+          poetry self add setuptools wheel
+          poetry run python -m pip install gensim
+          poetry install
+
+      
+      - name: Notebook Test
+        run: |
+          poetry run poe test_notebook

.github/workflows/python-notebook-tests.yml

@@ -125,6 +125,7 @@ _test_all = "coverage run -m pytest ./tests"
 test_unit = "pytest ./tests/unit"
 test_integration = "pytest ./tests/integration"
 test_smoke = "pytest ./tests/smoke"
+test_notebook = "pytest ./tests/notebook"
 index = "python -m graphrag.index"
 query = "python -m graphrag.query"
 prompt_tune = "python -m graphrag.prompt_tune"

pyproject.toml

@@ -1,38 +1,33 @@
 # Copyright (c) 2024 Microsoft Corporation.
 # Licensed under the MIT License
 import subprocess
-import tempfile
 from pathlib import Path
 
 import nbformat
 import pytest
 
-DOCS_PATH = Path("../../docsite")
+NOTEBOOKS_PATH = Path("examples_notebooks")
 
-notebooks_list = list(DOCS_PATH.rglob("*.ipynb"))
+notebooks_list = list(NOTEBOOKS_PATH.rglob("*.ipynb"))
 
 
 def _notebook_run(filepath: Path):
     """Execute a notebook via nbconvert and collect output.
     :returns execution errors
     """
-    with tempfile.NamedTemporaryFile(suffix=".ipynb") as temp_file:
-        args = [
-            "jupyter",
-            "nbconvert",
-            "--to",
-            "notebook",
-            "--execute",
-            "-y",
-            "--no-prompt",
-            "--output",
-            temp_file.name,
-            filepath.absolute().as_posix(),
-        ]
-        subprocess.check_call(args)
-
-        temp_file.seek(0)
-        nb = nbformat.read(temp_file, nbformat.current_nbformat)
+    args = [
+        "jupyter",
+        "nbconvert",
+        "--to",
+        "notebook",
+        "--execute",
+        "-y",
+        "--no-prompt",
+        "--stdout",
+        filepath.absolute().as_posix(),
+    ]
+    notebook = subprocess.check_output(args)
+    nb = nbformat.reads(notebook, nbformat.current_nbformat)
 
     return [
         output