fix: some datasets with unknown feature names could not be opened

CHANGELOG

@@ -1,4 +1,6 @@
 0.33.2
+ - fix: some datasets with unknown feature names could not be
+   opened (AssertionError regression in 0.33.1)
  - fix: workaround for sporadic JSONDecodeError when accessing DCOR
  - ref: cleanup cli.py
  - ref: cleanup util.py and deprecate `hash_class` argument in

dclab/rtdc_dataset/fmt_hdf5.py

@@ -62,14 +62,7 @@ def __init__(self, h5):
             self._features.remove("trace")
 
     def __contains__(self, key):
-        if self._is_defective_feature(key):
-            contained = False
-        else:
-            if key in self._features and dfn.feature_exists(key):
-                contained = True
-            else:
-                contained = False
-        return contained
+        return key in self.keys()
 
     def __getitem__(self, key):
         # user-level checking is done in core.py
@@ -104,10 +97,16 @@ def _is_defective_feature(self, key):
         return defective
 
     def keys(self):
+        """Returns list of valid features
+
+        Checks for
+        - defective features
+        - existing feature names
+        """
         features = []
         for key in self._features:
             # check for defective features
-            if not self._is_defective_feature(key):
+            if dfn.feature_exists(key) and not self._is_defective_feature(key):
                 features.append(key)
         return features
 

tests/test_rtdc_fmt_hdf5.py

@@ -127,6 +127,17 @@ def test_no_suffix():
     assert(len(ds) == 8)
 
 
+def test_open_with_invalid_feature_names():
+    """Loading an .rtdc file that has wrong feature name"""
+    path = str(retrieve_data("rtdc_data_hdf5_mask_contour.zip"))
+    # add wrong feature name right at the top of the list
+    with h5py.File(path, "r+") as h5:
+        h5["events"]["a0"] = h5["events"]["deform"]
+    # see if we can open the file without any error
+    ds = new_dataset(path)
+    assert len(ds) == 8
+
+
 @pytest.mark.filterwarnings(
     'ignore::dclab.rtdc_dataset.config.UnknownConfigurationKeyWarning')
 def test_trace():