fix: correctly handle dataset IDs and names in DBExtract

Author: paulmueller

CHANGELOG

@@ -1,4 +1,5 @@
 0.17.4
+ - fix: correctly handle dataset IDs and names in `DBExtract`
  - ref: decorate PyQtSlots
 0.17.3
  - fix: drag'n'drop to DCscope broken since 0.17.0 (#88)

dcoraid/dbmodel/extract.py

@@ -38,32 +38,28 @@ def __contains__(self, item):
             The dataset dictionary or the name or the id of the dataset
         """
         if isinstance(item, dict):
-            id_name = item["id"]
+            name_or_id = item["id"]
         else:
-            id_name = item
-        if len(id_name) == 36:  # minor optimization
-            # we probably have a UUID
-            return id_name in self.registry_id or id_name in self.registry
-        else:
-            return id_name in self.registry
+            name_or_id = item
+        return name_or_id in self.registry_id or name_or_id in self.registry
 
-    def __getitem__(self, idx_or_name):
-        if isinstance(idx_or_name, numbers.Integral):
-            return self.datasets[idx_or_name]
+    def __getitem__(self, idx_or_name_or_id):
+        if isinstance(idx_or_name_or_id, numbers.Integral):
+            return self.datasets[idx_or_name_or_id]
         else:
-            return self.get_dataset_dict(idx_or_name)
+            return self.get_dataset_dict(idx_or_name_or_id)
 
     def __len__(self):
         return len(self.datasets)
 
     def __iter__(self):
         return iter(self.datasets)
 
-    def add_datasets(self, datasets):
+    def add_datasets(self, datasets: list[dict]):
         for dd in datasets:
-            name = dd["name"]
-            if name not in self.registry:  # datasets must only be added once
-                self.registry[name] = dd
+            if dd not in self:  # dataset must only be added once
+                self.registry_id[dd["id"]] = dd
+                self.registry[dd["name"]] = dd
                 self.datasets.append(dd)
 
     @property
@@ -97,5 +93,10 @@ def collections(self):
                 coll_list, key=lambda x: x.get("title") or x["name"])
         return self._collections
 
-    def get_dataset_dict(self, dataset_name):
-        return self.registry[dataset_name]
+    def get_dataset_dict(self, name_or_id):
+        if name_or_id in self.registry:
+            return self.registry[name_or_id]
+        elif name_or_id in self.registry_id:
+            return self.registry_id[name_or_id]
+        else:
+            raise KeyError(f"Could not find name or id '{name_or_id}")

tests/test_dbmodel_extract.py

@@ -0,0 +1,38 @@
+from dcoraid.dbmodel import DBExtract
+
+
+def test_dbmodel_add_same():
+    de = DBExtract()
+    de.add_datasets([{"name": "peter", "id": "hans"}])
+    assert len(de) == 1
+    de.add_datasets([{"name": "peter", "id": "hans"}])
+    assert len(de) == 1
+
+
+def test_dbmodel_contains():
+    de = DBExtract()
+    de.add_datasets([{"name": "peter", "id": "hans"}])
+    assert "peter" in de
+    assert "hans" in de
+    assert {"name": "peter", "id": "hans"} in de
+
+
+def test_dbmodel_get_dataset_dict():
+    de = DBExtract()
+    de.add_datasets([{"name": "peter", "id": "hans"}])
+
+    assert de.get_dataset_dict("peter") == {"name": "peter", "id": "hans"}
+    assert de.get_dataset_dict("hans") == {"name": "peter", "id": "hans"}
+
+
+def test_dbmodel_get_item():
+    de = DBExtract()
+    de.add_datasets([{"name": "peter", "id": "hans"}])
+    de.add_datasets([{"name": "flink", "id": "flank"}])
+
+    assert de["flink"] == {"name": "flink", "id": "flank"}
+    assert de["flank"] == {"name": "flink", "id": "flank"}
+    assert de["hans"] == {"name": "peter", "id": "hans"}
+    assert de["peter"] == {"name": "peter", "id": "hans"}
+    assert de[0] == {"name": "peter", "id": "hans"}
+    assert de[1] == {"name": "flink", "id": "flank"}