(issue 563): bypass cartesian_product_df() and only call index_to_labels() when we know that the DataFrame read from an HDF5 file is a saved LArray object

Author: alixdamman

larray/inout/hdf.py

@@ -74,7 +74,7 @@ def read_hdf(filepath_or_buffer, key, fill_value=nan, na=nan, sort_rows=False, s
         _meta = attrs.metadata if 'metadata' in attrs else None
         if _type == 'Array':
             res = df_aslarray(pd_obj, sort_rows=sort_rows, sort_columns=sort_columns, fill_value=fill_value,
-                              parse_header=False)
+                              parse_header=False, dense=True)
             if _meta is not None:
                 res.meta = _meta
         elif _type == 'Axis':

larray/inout/pandas.py

@@ -211,6 +211,7 @@ def from_frame(df, sort_rows=False, sort_columns=False, parse_header=False, unfo
     a1   b0   4   5
     a1   b1   6   7
     """
+    dense = kwargs.pop('dense', False)
     axes_names = [decode(name, 'utf8') for name in df.index.names]
 
     # handle 2 or more dimensions with the last axis name given using \
@@ -223,8 +224,11 @@ def from_frame(df, sort_rows=False, sort_columns=False, parse_header=False, unfo
     else:
         axes_names += [df.columns.name]
 
-    df, axes_labels = cartesian_product_df(df, sort_rows=sort_rows, sort_columns=sort_columns,
-                                           fill_value=fill_value, **kwargs)
+    if dense:
+        axes_labels = index_to_labels(df.index, sort=sort_rows)
+    else:
+        df, axes_labels = cartesian_product_df(df, sort_rows=sort_rows, sort_columns=sort_columns,
+                                               fill_value=fill_value, **kwargs)
 
     # Pandas treats column labels as column names (strings) so we need to convert them to values
     last_axis_labels = [parse(cell) for cell in df.columns.values] if parse_header else list(df.columns.values)