Compat: pandas MultiIndex .labels to .codes (#103)

Closes #99

recordlinkage/base.py

@@ -282,7 +282,11 @@ def _dedup_index(self, df_a):
         # Remove all pairs not in the lower triangular part of the matrix.
         # This part can be inproved by not comparing the level values, but the
         # level itself.
-        pairs = pairs[pairs.labels[0] > pairs.labels[1]]
+        try:
+            pairs = pairs[pairs.codes[0] > pairs.codes[1]]
+        except AttributeError:
+            # backwards compat pandas <24
+            pairs = pairs[pairs.labels[0] > pairs.labels[1]] 
 
         return pairs
 

recordlinkage/utils.py

@@ -1,5 +1,6 @@
 import warnings
 from functools import wraps
+from packaging import version
 
 import numpy
 
@@ -77,6 +78,10 @@ def func_wrapper(*args, **kwargs):
 
 # Checks and conversions
 
+def is_min_pandas_version(min_version):
+    """Check if pandas version is larger or equal the version passed."""
+    return version.parse(pandas.__version__) >= version.parse(min_version)
+
 
 def is_label_dataframe(label, df):
     """check column label existance"""

tests/test_misc.py

@@ -10,6 +10,7 @@
 
 import recordlinkage as rl
 from recordlinkage import index_split
+from recordlinkage.utils import is_min_pandas_version
 
 
 def test_multiindex_split():
@@ -24,7 +25,10 @@ def test_multiindex_split():
         ptm.assert_index_equal(result_index_chunk, expected_index_chunk)
 
         assert len(result_index_chunk.levels) == 2
-        assert len(result_index_chunk.labels) == 2
+        if is_min_pandas_version("0.24.0"):
+            assert len(result_index_chunk.codes) == 2
+        else:
+            assert len(result_index_chunk.labels) == 2
 
 
 def test_options():