further clean-up

Author: jorisvandenbossche

pandas/io/parquet.py

@@ -98,9 +98,7 @@ def write(self, df, path, compression='snappy',
               coerce_timestamps='ms', **kwargs):
         self.validate_dataframe(df)
         if self._pyarrow_lt_070:
-            self._validate_write_lt_070(
-                df, path, compression, coerce_timestamps, **kwargs
-            )
+            self._validate_write_lt_070(df)
         path, _, _ = get_filepath_or_buffer(path)
 
         if self._pyarrow_lt_060:
@@ -116,48 +114,46 @@ def write(self, df, path, compression='snappy',
 
     def read(self, path, columns=None, **kwargs):
         path, _, _ = get_filepath_or_buffer(path)
-        parquet_file = self.api.parquet.ParquetFile(path)
         if self._pyarrow_lt_070:
-            return self._read_lt_070(path, parquet_file, columns, **kwargs)
+            return self.api.parquet.read_pandas(path, columns=columns,
+                                                **kwargs).to_pandas()
         kwargs['use_pandas_metadata'] = True
-        return parquet_file.read(columns=columns, **kwargs).to_pandas()
+        return self.api.parquet.read_table(path, columns=columns,
+                                           **kwargs).to_pandas()
 
-    def _validate_write_lt_070(self, df, path, compression='snappy',
-                               coerce_timestamps='ms', **kwargs):
+    def _validate_write_lt_070(self, df):
         # Compatibility shim for pyarrow < 0.7.0
         # TODO: Remove in pandas 0.22.0
         from pandas.core.indexes.multi import MultiIndex
         if isinstance(df.index, MultiIndex):
             msg = (
-                "Mulit-index DataFrames are only supported "
+                "Multi-index DataFrames are only supported "
                 "with pyarrow >= 0.7.0"
             )
             raise ValueError(msg)
         # Validate index
         if not isinstance(df.index, Int64Index):
             msg = (
-                "parquet does not support serializing {} for the index;"
-                "you can .reset_index() to make the index into column(s)"
+                "pyarrow < 0.7.0 does not support serializing {} for the "
+                "index; you can .reset_index() to make the index into "
+                "column(s), or install the latest version of pyarrow or "
+                "fastparquet."
             )
             raise ValueError(msg.format(type(df.index)))
         if not df.index.equals(RangeIndex(len(df))):
             raise ValueError(
-                "parquet does not support serializing a non-default index "
-                "for the index; you can .reset_index() to make the index "
-                "into column(s)"
+                "pyarrow < 0.7.0 does not support serializing a non-default "
+                "index; you can .reset_index() to make the index into "
+                "column(s), or install the latest version of pyarrow or "
+                "fastparquet."
             )
         if df.index.name is not None:
             raise ValueError(
-                "parquet does not serialize index meta-data "
-                "on a default index"
+                "pyarrow < 0.7.0 does not serialize indexes with a name; you "
+                "can set the index.name to None or install the latest version "
+                "of pyarrow or fastparquet."
             )
 
-    def _read_lt_070(self, path, parquet_file, columns, **kwargs):
-        # Compatibility shim for pyarrow < 0.7.0
-        # TODO: Remove in pandas 0.22.0
-        kwargs['columns'] = columns
-        return self.api.parquet.read_pandas(path, **kwargs).to_pandas()
-
 
 class FastParquetImpl(BaseImpl):
 

pandas/tests/io/test_parquet.py

@@ -6,7 +6,6 @@
 from warnings import catch_warnings
 
 import numpy as np
-from numpy.random import randn
 import pandas as pd
 from pandas.compat import PY3, is_platform_windows
 from pandas.io.parquet import (to_parquet, read_parquet, get_engine,
@@ -324,7 +323,7 @@ def test_write_index(self, engine):
         # index with meta-data
         df.index = [0, 1, 2]
         df.index.name = 'foo'
-        self.check_round_trip( df, engine, write_kwargs={'compression': None})
+        self.check_round_trip(df, engine, write_kwargs={'compression': None})
 
     def test_write_multiindex(self, pa_ge_070):
         # Not suppoprted in fastparquet as of 0.1.3 or older pyarrow version
@@ -345,11 +344,11 @@ def test_multiindex_with_columns(self, pa_ge_070):
 
         engine = pa_ge_070
         dates = pd.date_range('01-Jan-2018', '01-Dec-2018', freq='MS')
-        df = pd.DataFrame(randn(2 * len(dates), 3), columns=list('ABC'))
+        df = pd.DataFrame(np.random.randn(2 * len(dates), 3),
+                          columns=list('ABC'))
         index1 = pd.MultiIndex.from_product(
             [['Level1', 'Level2'], dates],
-            names=['level', 'date']
-        )
+            names=['level', 'date'])
         index2 = index1.copy(names=None)
         for index in [index1, index2]:
             df.index = index