fix: integrity check failed for empty datasets

CHANGELOG

@@ -1,3 +1,5 @@
+0.61.5
+ - fix: integrity check failed for empty datasets
 0.61.4
  - docs: update basins section
  - docs: cleanup build process

dclab/rtdc_dataset/check.py

@@ -199,7 +199,6 @@ def __init__(self, path_or_ds):
                         level="alert",
                         category="warning"))
             self.finally_close = True
-        np.max(self.ds["index"])
 
     def __enter__(self):
         return self
@@ -319,8 +318,19 @@ def iter_count_compression(h5):
             data=data))
         return cues
 
+    def check_empty(self, **kwargs):
+        """The dataset should contain events"""
+        cues = []
+        lends = len(self.ds)
+        if lends == 0:
+            cues.append(ICue(
+                msg="The dataset does not contain any events",
+                level="alert",
+                category="feature data"))
+        return cues
+
     def check_feat_index(self, **kwargs):
-        """Up until"""
+        """The index of the dataset should be monotonous"""
         cues = []
         lends = len(self.ds)
         if "index" in self.ds:

tests/test_rtdc_check_dataset.py

@@ -175,6 +175,24 @@ def test_icue():
     assert cues[0].msg in cues[0].__repr__()
 
 
+def test_ic_empty():
+    """Identity checker should not fail for empty datasets"""
+    path = retrieve_data("fmt-hdf5_polygon_gate_2021.zip")
+    path_empty = path.with_name("empty.rtdc")
+    with RTDCWriter(path_empty) as hw, dclab.new_dataset(path) as ds:
+        hw.h5file.attrs.update(ds.h5file.attrs)
+        hw.h5file.attrs["experiment:event count"] = 0
+    with check.IntegrityChecker(path_empty) as ic:
+        assert len(ic.ds) == 0
+        cues = ic.check_empty()
+
+    assert len(cues) == 1
+    cue = cues[0]
+    assert cue.level == "alert"
+    assert cue.category == "feature data"
+    assert cue.msg == "The dataset does not contain any events"
+
+
 def test_ic_expand_section():
     ddict = example_data_dict(size=8472, keys=["area_um", "deform"])
     ds1 = new_dataset(ddict)
@@ -714,11 +732,3 @@ def test_wrong_samples_per_event():
           + "(expected 10, got 566)"
     viol, _, _ = check_dataset(h5path)
     assert msg in viol
-
-
-if __name__ == "__main__":
-    # Run all tests
-    _loc = locals()
-    for _key in list(_loc.keys()):
-        if _key.startswith("test_") and hasattr(_loc[_key], "__call__"):
-            _loc[_key]()