feat: drop duplicated images at exact same location

Author: Gigaszi

mapswipe_workers/mapswipe_workers/utils/process_mapillary.py

@@ -227,7 +227,7 @@ def get_image_metadata(
         raise ValueError(
             "No Mapillary Features in the AoI or no Features match the filter criteria."
         )
-
+    downloaded_metadata = downloaded_metadata.drop_duplicates(subset=["geometry"])
     if sampling_threshold is not None:
         downloaded_metadata = spatial_sampling(downloaded_metadata, sampling_threshold)
 

mapswipe_workers/tests/unittests/test_process_mapillary.py

@@ -314,11 +314,28 @@ def test_get_image_metadata_empty_response(self, mock_coordinate_download):
 
     @patch("mapswipe_workers.utils.process_mapillary.coordinate_download")
     def test_get_image_metadata_size_restriction(self, mock_coordinate_download):
-        mock_coordinate_download.return_value = pd.DataFrame({"ID": range(1, 100002)})
+        mock_coordinate_download.return_value = pd.DataFrame(
+            {"geometry": range(1, 100002)}
+        )
 
         with self.assertRaises(ValueError):
             get_image_metadata(self.fixture_data)
 
+    @patch("mapswipe_workers.utils.process_mapillary.coordinate_download")
+    def test_get_image_metadata_drop_duplicates(self, mock_coordinate_download):
+        test_df = pd.DataFrame(
+            {
+                "id": [1, 2, 2, 3, 4, 4, 5],
+                "geometry": ["a", "b", "b", "c", "d", "d", "e"],
+            }
+        )
+        mock_coordinate_download.return_value = test_df
+        return_dict = get_image_metadata(self.fixture_data)
+
+        return_df = pd.DataFrame(return_dict)
+
+        self.assertNotEqual(len(return_df), len(test_df))
+
 
 if __name__ == "__main__":
     unittest.main()