Add media retention cleanup command

tests/test_job_runner.py

@@ -1,3 +1,5 @@
+import subprocess
+import sys
 from pathlib import Path
 
 import pytest
@@ -72,6 +74,66 @@ def test_main_publishes_staged_feed_after_successful_crawl(
     assert not staged_path.exists()
 
 
+def test_main_holds_media_cleanup_lock_during_crawl(
+    monkeypatch: pytest.MonkeyPatch, tmp_path: Path
+) -> None:
+    out_dir = tmp_path / "out"
+    public_path = feed_output_path(out_dir=out_dir, feed_slug="demo")
+    staged_path = staged_feed_output_path(out_dir=out_dir, feed_slug="demo")
+    public_path.parent.mkdir(parents=True)
+    public_path.write_text("<rss>old</rss>\n", encoding="utf-8")
+    staged_path.write_text(VALID_FEED, encoding="utf-8")
+
+    def assert_media_lock_is_held(*, process, feed, spider_arguments) -> int:
+        lock_path = out_dir.resolve() / ".media-retention.lock"
+        script = """
+import fcntl
+import sys
+from pathlib import Path
+
+lock_path = Path(sys.argv[1])
+lock_path.parent.mkdir(parents=True, exist_ok=True)
+with lock_path.open("a", encoding="utf-8") as lock_file:
+    try:
+        fcntl.flock(lock_file.fileno(), fcntl.LOCK_EX | fcntl.LOCK_NB)
+    except BlockingIOError:
+        sys.exit(0)
+    else:
+        fcntl.flock(lock_file.fileno(), fcntl.LOCK_UN)
+        sys.exit(2)
+"""
+        completed = subprocess.run(
+            [sys.executable, "-c", script, str(lock_path)],
+            cwd=Path.cwd(),
+            capture_output=True,
+            check=False,
+            text=True,
+        )
+        assert completed.returncode == 0, completed.stdout + completed.stderr
+        return 0
+
+    _patch_worker_dependencies(
+        monkeypatch, exit_code=0, run_crawl=assert_media_lock_is_held
+    )
+
+    exit_code = job_runner_module.main(
+        [
+            "--job-id",
+            "1",
+            "--execution-id",
+            "2",
+            "--db-path",
+            str(tmp_path / "republisher.db"),
+            "--out-dir",
+            str(out_dir),
+            "--stats-path",
+            str(tmp_path / "stats.jsonl"),
+        ]
+    )
+
+    assert exit_code == 0
+
+
 def test_main_does_not_publish_unusable_staged_feed_after_successful_crawl(
     monkeypatch: pytest.MonkeyPatch, tmp_path: Path
 ) -> None:
@@ -137,7 +199,7 @@ def test_main_does_not_publish_staged_feed_after_failed_crawl(
 
 
 def _patch_worker_dependencies(
-    monkeypatch: pytest.MonkeyPatch, *, exit_code: int
+    monkeypatch: pytest.MonkeyPatch, *, exit_code: int, run_crawl=None
 ) -> None:
     monkeypatch.setattr(
         job_runner_module,
@@ -161,5 +223,5 @@ def _patch_worker_dependencies(
     monkeypatch.setattr(
         job_runner_module,
         "_run_crawl",
-        lambda *, process, feed, spider_arguments: exit_code,
+        run_crawl or (lambda *, process, feed, spider_arguments: exit_code),
     )