Add the local-only DuckDuckGo reference collector

tools/test_refcollect.py

@@ -0,0 +1,51 @@
+"""Tests for the local-only reference collector (no network)."""
+
+import json
+import unittest
+from pathlib import Path
+from tempfile import TemporaryDirectory
+from unittest import mock
+
+import refcollect as rc
+
+
+class SlugTests(unittest.TestCase):
+    def test_slug_folds_and_strips(self):
+        self.assertEqual(rc._slug("El Capitan Entertainment Centre"),
+                         "el-capitan-entertainment-centre")
+        self.assertEqual(rc._slug("  !!!  "), "block")
+
+
+class ManifestTests(unittest.TestCase):
+    def test_manifest_rows_are_marked_local_only_and_never_licenced(self):
+        rows = [{"title": "t", "image": "https://x/i.jpg", "thumbnail": "",
+                 "source": "https://x/", "width": 640, "height": 480}]
+        out = rc._manifest_rows(rows)
+        self.assertTrue(out[0]["local_only"])
+        self.assertEqual(out[0]["licence"], "unchecked-local")
+
+
+class SheetTests(unittest.TestCase):
+    def test_sheet_requires_a_manifest(self):
+        with TemporaryDirectory() as tmp:
+            with mock.patch.object(rc, "RESEARCH_DIR", Path(tmp)):
+                with self.assertRaises(SystemExit):
+                    rc.cmd_sheet(mock.Mock(slug="nope"))
+
+    def test_sheet_composes_from_downloaded_frames(self):
+        from PIL import Image
+        with TemporaryDirectory() as tmp:
+            root = Path(tmp)
+            slug_dir = root / "block"
+            slug_dir.mkdir()
+            for i in range(3):
+                Image.new("RGB", (640, 400), (i * 60, 0, 0)).save(slug_dir / f"{i:02d}.jpg")
+            (slug_dir / "manifest.json").write_text(json.dumps([
+                {"file": "00.jpg"}, {"file": "01.jpg"}, {"file": "02.jpg"}]))
+            with mock.patch.object(rc, "RESEARCH_DIR", root):
+                rc.cmd_sheet(mock.Mock(slug="block"))
+            self.assertTrue((slug_dir / "sheet.jpg").is_file())
+
+
+if __name__ == "__main__":
+    unittest.main()