Added deduplication for mount point list (#569)

Added deduplication for mount point list.

Fixes #544

src/databricks/labs/ucx/hive_metastore/mounts.py

@@ -19,14 +19,25 @@ def __init__(self, backend: SqlBackend, ws: WorkspaceClient, inventory_database:
         super().__init__(backend, "hive_metastore", inventory_database, "mounts", Mount)
         self._dbutils = ws.dbutils
 
+    def _deduplicate_mounts(self, mounts: list) -> list:
+        seen = set()
+        deduplicated_mounts = []
+
+        for obj in mounts:
+            obj_tuple = (obj.name, obj.source)
+            if obj_tuple not in seen:
+                seen.add(obj_tuple)
+                deduplicated_mounts.append(obj)
+        return deduplicated_mounts
+
     def inventorize_mounts(self):
         self._append_records(self._list_mounts())
 
     def _list_mounts(self):
         mounts = []
         for mount_point, source, _ in self._dbutils.fs.mounts():
             mounts.append(Mount(mount_point, source))
-        return mounts
+        return self._deduplicate_mounts(mounts)
 
     def snapshot(self) -> list[Mount]:
         return self._snapshot(self._try_fetch, self._list_mounts)

tests/unit/hive_metastore/test_mounts.py

@@ -11,6 +11,24 @@ def test_list_mounts_should_return_a_list_of_mount_without_encryption_type():
     client.dbutils.fs.mounts.return_value = [
         MountInfo("mp_1", "path_1", "info_1"),
         MountInfo("mp_2", "path_2", "info_2"),
+        MountInfo("mp_3", "path_3", "info_3"),
+    ]
+
+    backend = MockBackend()
+    instance = Mounts(backend, client, "test")
+
+    instance.inventorize_mounts()
+
+    expected = [Mount("mp_1", "path_1"), Mount("mp_2", "path_2"), Mount("mp_3", "path_3")]
+    assert expected == backend.rows_written_for("hive_metastore.test.mounts", "append")
+
+
+def test_list_mounts_should_return_a_deduped_list_of_mount_without_encryption_type():
+    client = MagicMock()
+    client.dbutils.fs.mounts.return_value = [
+        MountInfo("mp_1", "path_1", "info_1"),
+        MountInfo("mp_2", "path_2", "info_2"),
+        MountInfo("mp_2", "path_2", "info_2"),
     ]
 
     backend = MockBackend()