ARROW-8307: [Python] Add memory_map= option to pyarrow.feather.read_t…

…able

Having this option makes it easier to do apples-to-apples performance tests of compressed versus uncompressed files (there may be other reasons not to memory map but either way doesn't hurt to have the option)

Closes #6832 from wesm/ARROW-8307

Authored-by: Wes McKinney <wesm+git@apache.org>
Signed-off-by: Wes McKinney <wesm+git@apache.org>

python/pyarrow/feather.py

@@ -204,7 +204,7 @@ def read_feather(source, columns=None, use_threads=True):
     return read_table(source, columns=columns).to_pandas(use_threads=True)
 
 
-def read_table(source, columns=None):
+def read_table(source, columns=None, memory_map=True):
     """
     Read a pyarrow.Table from Feather format
 
@@ -214,14 +214,16 @@ def read_table(source, columns=None):
     columns : sequence, optional
         Only read a specific set of columns. If not provided, all columns are
         read.
+    memory_map : boolean, default True
+        Use memory mapping when opening file on disk
 
     Returns
     -------
     table : pyarrow.Table
     """
     _check_pandas_version()
     reader = ext.FeatherReader()
-    reader.open(source)
+    reader.open(source, use_memory_map=memory_map)
 
     if columns is None:
         return reader.read()

python/pyarrow/tests/test_feather.py

@@ -156,7 +156,10 @@ def test_read_table(version):
     table = pa.Table.from_pandas(data)
 
     result = read_table(path)
+    assert_frame_equal(table.to_pandas(), result.to_pandas())
 
+    # Test without memory mapping
+    result = read_table(path, memory_map=False)
     assert_frame_equal(table.to_pandas(), result.to_pandas())