Continuous integration for tests

.github/identify_collections_to_test.py

@@ -0,0 +1,51 @@
+import os
+import json
+import fnmatch
+import subprocess
+
+
+def get_changed_files():
+    return subprocess.check_output([
+        'git', 'diff',
+        '--name-only', 'origin/main...'
+    ]).decode().splitlines()
+
+
+def get_collections_from_file(file):
+    with open(file, 'rt') as f:
+        data = json.load(f)
+
+    collections = []
+    for k in data.keys():
+        if 'Collection' in data[k].keys():
+            collections += [data[k]['Collection']]
+
+    # we're only running in response to changes in data_summaries/*.json files,
+    # so if there's such a file without a collection field we don't like that
+    # and should fail the test
+    assert len(collections) > 0
+
+    return collections
+
+
+def main():
+    changed_files = get_changed_files()
+
+    collections_to_test = []
+    for file in changed_files:
+        if not fnmatch.fnmatch(file, 'data_summaries/*.json'):
+            continue
+
+        if not os.path.exists(file):  # i.e., file was deleted
+            continue
+
+        collections_to_test += get_collections_from_file(file)
+
+    if collections_to_test:
+        print("::set-output name=collections::" + '\n'.join(collections_to_test))
+    else:
+        print("::set-output name=collections::")
+
+
+if __name__ == "__main__":
+    main()

.github/workflows/ci.yml

@@ -0,0 +1,40 @@
+name: CI to test new collections
+
+on:
+  pull_request:
+    paths:
+      - 'data_summaries/**.json'  # Trip only when data summary changed/added
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v2
+
+    - uses: actions/setup-python@v2
+      with:
+        python-version: '3.11'
+
+    - name: Cache Python dependencies
+      uses: actions/cache@v2
+      id: cache
+      with:
+        path: ~/.cache/pip  # This path can vary depending on your OS and Python version
+        key: ${{ runner.os }}-pip-${{ hashFiles('**/requirements.txt') }}
+        restore-keys: |
+          ${{ runner.os }}-pip-
+
+    - name: Install dependencies
+      if: steps.cache.outputs.cache-hit != 'true'
+      run: |
+        pip install -U pip
+        pip install -r requirements.txt
+
+    - name: Identify Collections
+      run: python .github/identify_collections_to_test.py
+
+    - name: Run Tests
+      run: |
+        while IFS= read -r collection; do
+          python src/test_new_collection.py --collection "$collection"
+        done <<< "${{ steps.collection.outputs.collections }}"