chore: refine seed cache workflow details

Co-authored-by: gkorland <753206+gkorland@users.noreply.github.com>

Author: Copilot

.github/workflows/playwright.yml

@@ -53,7 +53,7 @@ jobs:
     - name: Cache seeded repository clones
       uses: actions/cache@v4
       with:
-        path: repositories
+        path: ${{ github.workspace }}/repositories
         key: ${{ runner.os }}-seed-repos-${{ hashFiles('e2e/seed_test_data.py') }}-${{ steps.seed-repo-shas.outputs.graphrag_sdk }}-${{ steps.seed-repo-shas.outputs.flask }}
         restore-keys: |
           ${{ runner.os }}-seed-repos-${{ hashFiles('e2e/seed_test_data.py') }}-

e2e/seed_test_data.py

@@ -32,11 +32,14 @@
 
 
 def repo_name_from_url(url: str) -> str:
-    return urlparse(url).path.rstrip("/").split("/")[-1].removesuffix(".git")
+    parsed_path = urlparse(url).path.rstrip("/")
+    repo_name = parsed_path.split("/")[-1]
+    return repo_name.removesuffix(".git")
 
 
-def clone_repository(url: str, path: Path) -> Path:
+def fresh_clone_repository(url: str, path: Path) -> Path:
     if path.exists():
+        # Replace any existing directory before creating a fresh shallow clone.
         shutil.rmtree(path)
 
     path.parent.mkdir(parents=True, exist_ok=True)
@@ -57,7 +60,7 @@ def load_project(url: str) -> Project:
         logger.info("Using cached repository clone at %s", repo_path)
     else:
         logger.info("Cloning repository into cache at %s", repo_path)
-        clone_repository(url, repo_path)
+        fresh_clone_repository(url, repo_path)
 
     return Project.from_local_repository(repo_path)
 

tests/test_seed_test_data.py

@@ -3,7 +3,7 @@
 import e2e.seed_test_data as seed_test_data
 
 
-def test_load_project_uses_cached_clone(monkeypatch, tmp_path):
+def test_load_project_uses_existing_cached_clone(monkeypatch, tmp_path):
     repo_path = tmp_path / "GraphRAG-SDK"
     (repo_path / ".git").mkdir(parents=True)
 
@@ -42,7 +42,7 @@ def fake_clone(url: str, path: Path) -> Path:
 
     monkeypatch.setattr(seed_test_data, "REPOSITORIES_DIR", tmp_path)
     monkeypatch.setattr(seed_test_data, "Project", FakeProject)
-    monkeypatch.setattr(seed_test_data, "clone_repository", fake_clone)
+    monkeypatch.setattr(seed_test_data, "fresh_clone_repository", fake_clone)
 
     project = seed_test_data.load_project("https://github.com/pallets/flask")