gh-132983: Introduce compression package and move _compression module (GH-133018)

* Introduces `compression` package for https://peps.python.org/pep-0784/

This commit introduces the `compression` package, specified in PEP 784
to re-export the `lzma`, `bz2`, `gzip`, and `zlib` modules. Introduction
of `compression.zstd` will be completed in a future commit once the
`_zstd` module is merged.

This commit also moves the `_compression` private module to
`compression._common._streams`.

* Re-exports existing module docstrings.

Author: emmatyping

Lib/bz2.py

@@ -10,9 +10,9 @@
 __author__ = "Nadeem Vawda <[email protected]>"
 
 from builtins import open as _builtin_open
+from compression._common import _streams
 import io
 import os
-import _compression
 
 from _bz2 import BZ2Compressor, BZ2Decompressor
 
@@ -23,7 +23,7 @@
 _MODE_WRITE    = 3
 
 
-class BZ2File(_compression.BaseStream):
+class BZ2File(_streams.BaseStream):
 
     """A file object providing transparent bzip2 (de)compression.
 
@@ -88,7 +88,7 @@ def __init__(self, filename, mode="r", *, compresslevel=9):
             raise TypeError("filename must be a str, bytes, file or PathLike object")
 
         if self._mode == _MODE_READ:
-            raw = _compression.DecompressReader(self._fp,
+            raw = _streams.DecompressReader(self._fp,
                 BZ2Decompressor, trailing_error=OSError)
             self._buffer = io.BufferedReader(raw)
         else:
@@ -248,7 +248,7 @@ def writelines(self, seq):
 
         Line separators are not added between the written byte strings.
         """
-        return _compression.BaseStream.writelines(self, seq)
+        return _streams.BaseStream.writelines(self, seq)
 
     def seek(self, offset, whence=io.SEEK_SET):
         """Change the file position.

Lib/compression/__init__.py

@@ -1,4 +1,4 @@
-"""Internal classes used by the gzip, lzma and bz2 modules"""
+"""Internal classes used by compression modules"""
 
 import io
 import sys

Lib/_compression.py renamed to Lib/compression/_common/_streams.py

@@ -0,0 +1,5 @@
+import bz2
+__doc__ = bz2.__doc__
+del bz2
+
+from bz2 import *

Lib/compression/bz2/__init__.py

@@ -0,0 +1,5 @@
+import gzip
+__doc__ = gzip.__doc__
+del gzip
+
+from gzip import *

Lib/compression/gzip/__init__.py

@@ -0,0 +1,5 @@
+import lzma
+__doc__ = lzma.__doc__
+del lzma
+
+from lzma import *

Lib/compression/lzma/__init__.py

@@ -0,0 +1,5 @@
+import zlib
+__doc__ = zlib.__doc__
+del zlib
+
+from zlib import *

Lib/compression/zlib/__init__.py

@@ -5,7 +5,6 @@
 
 # based on Andrew Kuchling's minigzip.py distributed with the zlib module
 
-import _compression
 import builtins
 import io
 import os
@@ -14,6 +13,7 @@
 import time
 import weakref
 import zlib
+from compression._common import _streams
 
 __all__ = ["BadGzipFile", "GzipFile", "open", "compress", "decompress"]
 
@@ -144,7 +144,7 @@ def writable(self):
         return True
 
 
-class GzipFile(_compression.BaseStream):
+class GzipFile(_streams.BaseStream):
     """The GzipFile class simulates most of the methods of a file object with
     the exception of the truncate() method.
 
@@ -523,7 +523,7 @@ def _read_gzip_header(fp):
     return last_mtime
 
 
-class _GzipReader(_compression.DecompressReader):
+class _GzipReader(_streams.DecompressReader):
     def __init__(self, fp):
         super().__init__(_PaddedFile(fp), zlib._ZlibDecompressor,
                          wbits=-zlib.MAX_WBITS)

Lib/gzip.py

@@ -24,9 +24,9 @@
 import builtins
 import io
 import os
+from compression._common import _streams
 from _lzma import *
 from _lzma import _encode_filter_properties, _decode_filter_properties  # noqa: F401
-import _compression
 
 
 # Value 0 no longer used
@@ -35,7 +35,7 @@
 _MODE_WRITE    = 3
 
 
-class LZMAFile(_compression.BaseStream):
+class LZMAFile(_streams.BaseStream):
 
     """A file object providing transparent LZMA (de)compression.
 
@@ -127,7 +127,7 @@ def __init__(self, filename=None, mode="r", *,
             raise TypeError("filename must be a str, bytes, file or PathLike object")
 
         if self._mode == _MODE_READ:
-            raw = _compression.DecompressReader(self._fp, LZMADecompressor,
+            raw = _streams.DecompressReader(self._fp, LZMADecompressor,
                 trailing_error=LZMAError, format=format, filters=filters)
             self._buffer = io.BufferedReader(raw)
 

Lib/lzma.py

@@ -16,7 +16,7 @@
 from test.support import import_helper
 from test.support import threading_helper
 from test.support.os_helper import unlink, FakePath
-import _compression
+from compression._common import _streams
 import sys
 
 
@@ -126,15 +126,15 @@ def testReadMultiStream(self):
     def testReadMonkeyMultiStream(self):
         # Test BZ2File.read() on a multi-stream archive where a stream
         # boundary coincides with the end of the raw read buffer.
-        buffer_size = _compression.BUFFER_SIZE
-        _compression.BUFFER_SIZE = len(self.DATA)
+        buffer_size = _streams.BUFFER_SIZE
+        _streams.BUFFER_SIZE = len(self.DATA)
         try:
             self.createTempFile(streams=5)
             with BZ2File(self.filename) as bz2f:
                 self.assertRaises(TypeError, bz2f.read, float())
                 self.assertEqual(bz2f.read(), self.TEXT * 5)
         finally:
-            _compression.BUFFER_SIZE = buffer_size
+            _streams.BUFFER_SIZE = buffer_size
 
     def testReadTrailingJunk(self):
         self.createTempFile(suffix=self.BAD_DATA)
@@ -742,7 +742,7 @@ def testOpenPathLikeFilename(self):
     def testDecompressLimited(self):
         """Decompressed data buffering should be limited"""
         bomb = bz2.compress(b'\0' * int(2e6), compresslevel=9)
-        self.assertLess(len(bomb), _compression.BUFFER_SIZE)
+        self.assertLess(len(bomb), _streams.BUFFER_SIZE)
 
         decomp = BZ2File(BytesIO(bomb))
         self.assertEqual(decomp.read(1), b'\0')

Lib/test/test_bz2.py

@@ -1,4 +1,3 @@
-import _compression
 import array
 from io import BytesIO, UnsupportedOperation, DEFAULT_BUFFER_SIZE
 import os
@@ -7,6 +6,7 @@
 import sys
 from test import support
 import unittest
+from compression._common import _streams
 
 from test.support import _4G, bigmemtest
 from test.support.import_helper import import_module
@@ -861,13 +861,13 @@ def test_read_multistream(self):
     def test_read_multistream_buffer_size_aligned(self):
         # Test the case where a stream boundary coincides with the end
         # of the raw read buffer.
-        saved_buffer_size = _compression.BUFFER_SIZE
-        _compression.BUFFER_SIZE = len(COMPRESSED_XZ)
+        saved_buffer_size = _streams.BUFFER_SIZE
+        _streams.BUFFER_SIZE = len(COMPRESSED_XZ)
         try:
             with LZMAFile(BytesIO(COMPRESSED_XZ *  5)) as f:
                 self.assertEqual(f.read(), INPUT * 5)
         finally:
-            _compression.BUFFER_SIZE = saved_buffer_size
+            _streams.BUFFER_SIZE = saved_buffer_size
 
     def test_read_trailing_junk(self):
         with LZMAFile(BytesIO(COMPRESSED_XZ + COMPRESSED_BOGUS)) as f:
@@ -1066,7 +1066,7 @@ def test_readlines(self):
     def test_decompress_limited(self):
         """Decompressed data buffering should be limited"""
         bomb = lzma.compress(b'\0' * int(2e6), preset=6)
-        self.assertLess(len(bomb), _compression.BUFFER_SIZE)
+        self.assertLess(len(bomb), _streams.BUFFER_SIZE)
 
         decomp = LZMAFile(BytesIO(bomb))
         self.assertEqual(decomp.read(1), b'\0')