ENH: Add columns argument to read_feather() (pandas-dev#24025) (pandas-dev#24034)

Author: nixphix

doc/source/whatsnew/v0.24.0.rst

@@ -25,6 +25,7 @@ New features
   dataframe's indexes from the resulting Parquet file. (:issue:`20768`)
 - :meth:`DataFrame.corr` and :meth:`Series.corr` now accept a callable for generic calculation methods of correlation, e.g. histogram intersection (:issue:`22684`)
 - :func:`DataFrame.to_string` now accepts ``decimal`` as an argument, allowing the user to specify which decimal separator should be used in the output. (:issue:`23614`)
+- :func:`DataFrame.read_feather` now accepts ``columns`` as an argument, allowing the user to specify which columns should be read. (:issue:`24025`)
 
 .. _whatsnew_0240.values_api:
 

pandas/io/feather_format.py

@@ -84,7 +84,7 @@ def to_feather(df, path):
 
 
 @deprecate_kwarg(old_arg_name='nthreads', new_arg_name='use_threads')
-def read_feather(path, use_threads=True):
+def read_feather(path, columns=None, use_threads=True):
     """
     Load a feather-format object from the file path
 
@@ -93,6 +93,10 @@ def read_feather(path, use_threads=True):
     Parameters
     ----------
     path : string file path, or file-like object
+    columns : sequence, default None
+        If not provided, all columns are read
+
+        .. versionadded 0.24.0
     nthreads : int, default 1
         Number of CPU threads to use when reading to pandas.DataFrame
 
@@ -116,6 +120,8 @@ def read_feather(path, use_threads=True):
         int_use_threads = int(use_threads)
         if int_use_threads < 1:
             int_use_threads = 1
-        return feather.read_feather(path, nthreads=int_use_threads)
+        return feather.read_feather(path, columns=columns,
+                                    nthreads=int_use_threads)
 
-    return feather.read_feather(path, use_threads=bool(use_threads))
+    return feather.read_feather(path, columns=columns,
+                                use_threads=bool(use_threads))

pandas/tests/io/test_feather.py

@@ -26,13 +26,16 @@ def check_error_on_write(self, df, exc):
             with ensure_clean() as path:
                 to_feather(df, path)
 
-    def check_round_trip(self, df, **kwargs):
+    def check_round_trip(self, df, expected=None, **kwargs):
+
+        if expected is None:
+            expected = df
 
         with ensure_clean() as path:
             to_feather(df, path)
 
             result = read_feather(path, **kwargs)
-            assert_frame_equal(result, df)
+            assert_frame_equal(result, expected)
 
     def test_error(self):
 
@@ -74,6 +77,16 @@ def test_stringify_columns(self):
         df = pd.DataFrame(np.arange(12).reshape(4, 3)).copy()
         self.check_error_on_write(df, ValueError)
 
+    def test_read_columns(self):
+        # GH 24025
+        df = pd.DataFrame({'col1': list('abc'),
+                           'col2': list(range(1, 4)),
+                           'col3': list('xyz'),
+                           'col4': list(range(4, 7))})
+        columns = ['col1', 'col3']
+        self.check_round_trip(df, expected=df[columns],
+                              columns=columns)
+
     def test_unsupported_other(self):
 
         # period