Merge pull request #125 from FAST-HEP/BK_fix_explode_second_chunk

Fix explode function for second and subsequent chunks

CHANGELOG.md

@@ -5,6 +5,9 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ## [Unreleased]
+## [0.18.1] - 2020-06-17
+### Fixed
+- Fix binned_dataframe.explode for object-level and non-initial data chunks, PR #125
 
 ## [0.18.0] - 2020-06-17
 ### Added

fast_carpenter/summary/binned_dataframe.py

@@ -293,7 +293,7 @@ def explode(df):
     # get the list columns
     lst_cols = [col for col, dtype in df.dtypes.items() if is_object_dtype(dtype)]
     # Be more specific about which objects are ok
-    lst_cols = [col for col in lst_cols if isinstance(df[col][0], _explodable_types)]
+    lst_cols = [col for col in lst_cols if isinstance(df[col].iloc[0], _explodable_types)]
     if not lst_cols:
         return df
 

fast_carpenter/version.py

@@ -12,5 +12,5 @@ def split_version(version):
     return tuple(result)
 
 
-__version__ = '0.18.0'
+__version__ = '0.18.1'
 version_info = split_version(__version__) # noqa

setup.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.18.0
+current_version = 0.18.1
 commit = True
 tag = False
 

tests/summary/test_binned_dataframe.py

@@ -261,6 +261,11 @@ def test_explode():
     exploded = bdf.explode(pd.DataFrame(columns=["one", "two", "3"]))
     assert exploded.empty is True
 
+    df.index = np.arange(len(df)) + 100
+    exploded = bdf.explode(df)
+    assert len(exploded) == 1 + 8 + 3
+    assert np.array_equal(exploded.list, [0, 0, 1, 2, 3, 4, 5, 6, 7, 0, 1, 2])
+
 
 def test_densify_dataframe_integers():
     index = [("one", 1), ("one", 3), ("two", 2), ("three", 1), ("three", 2)]