feat: add multimodal search and granular performance diagnostics

tools/media-eval/download.py

@@ -0,0 +1,32 @@
+"""Reacquire the locked licensed corpus. Updated bytes are never accepted."""
+import argparse
+import hashlib
+import json
+from pathlib import Path
+import urllib.request
+
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--root", type=Path, default=Path("fixtures/media"))
+    args = parser.parse_args()
+    root = args.root.resolve()
+    fixture_path = root / "fixture.json"
+    lock = json.loads((root / "labels.lock.json").read_text())
+    assert hashlib.sha256(fixture_path.read_bytes()).hexdigest() == lock["fixture_sha256"]
+    for asset in json.loads(fixture_path.read_text(encoding="utf-8"))["assets"]:
+        destination = (root / asset["path"]).resolve()
+        assert destination.is_relative_to(root)
+        if destination.is_file() and hashlib.sha256(destination.read_bytes()).hexdigest() == asset["sha256"]:
+            continue
+        request = urllib.request.Request(asset["source_url"], headers={"User-Agent": "SearchiMediaEvaluation/1.0"})
+        with urllib.request.urlopen(request, timeout=60) as response:
+            data = response.read(asset["bytes"] + 1)
+        assert len(data) == asset["bytes"] and hashlib.sha256(data).hexdigest() == asset["sha256"], f"Publisher changed {asset['id']}; keep frozen corpus, investigate a new version"
+        destination.parent.mkdir(parents=True, exist_ok=True)
+        destination.write_bytes(data)
+        print("Verified", asset["id"])
+
+
+if __name__ == "__main__":
+    main()