BUG: set keyword argument so zipfile actually compresses (#21144)

Author: minggli

doc/source/whatsnew/v0.23.1.txt

@@ -86,6 +86,7 @@ Indexing
 I/O
 ^^^
 
+- Bug in IO methods specifying ``compression='zip'`` which produced uncompressed zip archives (:issue:`17778`, :issue:`21144`)
 - Bug in :meth:`DataFrame.to_stata` which prevented exporting DataFrames to buffers and most file-like objects (:issue:`21041`)
 -
 

pandas/io/common.py

@@ -5,7 +5,7 @@
 import codecs
 import mmap
 from contextlib import contextmanager, closing
-from zipfile import ZipFile
+import zipfile
 
 from pandas.compat import StringIO, BytesIO, string_types, text_type
 from pandas import compat
@@ -428,7 +428,7 @@ def _get_handle(path_or_buf, mode, encoding=None, compression=None,
     return f, handles
 
 
-class BytesZipFile(ZipFile, BytesIO):
+class BytesZipFile(zipfile.ZipFile, BytesIO):
     """
     Wrapper for standard library class ZipFile and allow the returned file-like
     handle to accept byte strings via `write` method.
@@ -437,10 +437,10 @@ class BytesZipFile(ZipFile, BytesIO):
     bytes strings into a member of the archive.
     """
     # GH 17778
-    def __init__(self, file, mode='r', **kwargs):
+    def __init__(self, file, mode, compression=zipfile.ZIP_DEFLATED, **kwargs):
         if mode in ['wb', 'rb']:
             mode = mode.replace('b', '')
-        super(BytesZipFile, self).__init__(file, mode, **kwargs)
+        super(BytesZipFile, self).__init__(file, mode, compression, **kwargs)
 
     def write(self, data):
         super(BytesZipFile, self).writestr(self.filename, data)

pandas/tests/test_common.py

@@ -1,12 +1,13 @@
 # -*- coding: utf-8 -*-
 
 import pytest
+import os
 import collections
 from functools import partial
 
 import numpy as np
 
-from pandas import Series, Timestamp
+from pandas import Series, DataFrame, Timestamp
 from pandas.compat import range, lmap
 import pandas.core.common as com
 from pandas.core import ops
@@ -222,3 +223,21 @@ def test_standardize_mapping():
 
     dd = collections.defaultdict(list)
     assert isinstance(com.standardize_mapping(dd), partial)
+
+
+@pytest.mark.parametrize('obj', [
+    DataFrame(100 * [[0.123456, 0.234567, 0.567567],
+                     [12.32112, 123123.2, 321321.2]],
+              columns=['X', 'Y', 'Z']),
+    Series(100 * [0.123456, 0.234567, 0.567567], name='X')])
+@pytest.mark.parametrize('method', ['to_pickle', 'to_json', 'to_csv'])
+def test_compression_size(obj, method, compression):
+    if not compression:
+        pytest.skip("only test compression case.")
+
+    with tm.ensure_clean() as filename:
+        getattr(obj, method)(filename, compression=compression)
+        compressed = os.path.getsize(filename)
+        getattr(obj, method)(filename, compression=None)
+        uncompressed = os.path.getsize(filename)
+        assert uncompressed > compressed