Merge lp:~jameinel/bzr/2.3-gzip-py2.7 into lp:bzr

[John A Meinel]

Stop using bzrlib.tuned_gzip.GzipFile.

This removes our use of tuned_gzip.GzipFile. It seems to be broken in python-2.7. We also only use it for Knit format repositories. And we introduced packs back in 2007. So I'm not too concerned about performance tuning 3-year-old deprecated formats.

[Vincent Ladeuil (vila)]

Great !

Not being concerned about performance doesn't imply forcing people under pdb though, so may be we can do without the pdb.set_trace() ;)

23 - return MultiParent.from_patch(zip_file.read())
24 + import pdb; pdb.set_trace()
25 + content = zip_file.read()
26 + return MultiParent.from_patch(content)

Apart from that: approve

review: Needs Fixing

[John A Meinel (jameinel)]

sent to pqm by email

[John A Meinel (jameinel)]

sent to pqm by email

[John A Meinel (jameinel)]

sent to pqm by email

=== modified file 'bzrlib/builtins.py'
--- bzrlib/builtins.py	2011-01-11 16:21:00 +0000
+++ bzrlib/builtins.py	2011-01-11 20:49:28 +0000
@@ -1,4 +1,4 @@
-# Copyright (C) 2005-2010 Canonical Ltd
+# Copyright (C) 2005-2011 Canonical Ltd
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by

=== modified file 'bzrlib/knit.py'
--- bzrlib/knit.py	2010-06-04 03:09:35 +0000
+++ bzrlib/knit.py	2011-01-11 20:49:28 +0000
@@ -1,4 +1,4 @@
-# Copyright (C) 2006-2010 Canonical Ltd
+# Copyright (C) 2006-2011 Canonical Ltd
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -54,6 +54,7 @@
 
 from cStringIO import StringIO
 from itertools import izip
+import gzip
 import operator
 import os
 import sys
@@ -1879,7 +1880,7 @@
         :return: the header and the decompressor stream.
                  as (stream, header_record)
         """
-        df = tuned_gzip.GzipFile(mode='rb', fileobj=StringIO(raw_data))
+        df = gzip.GzipFile(mode='rb', fileobj=StringIO(raw_data))
         try:
             # Current serialise
             rec = self._check_header(key, df.readline())
@@ -1894,7 +1895,7 @@
         # 4168 calls in 2880 217 internal
         # 4168 calls to _parse_record_header in 2121
         # 4168 calls to readlines in 330
-        df = tuned_gzip.GzipFile(mode='rb', fileobj=StringIO(data))
+        df = gzip.GzipFile(mode='rb', fileobj=StringIO(data))
         try:
             record_contents = df.readlines()
         except Exception, e:

=== modified file 'bzrlib/multiparent.py'
--- bzrlib/multiparent.py	2010-08-11 01:31:11 +0000
+++ bzrlib/multiparent.py	2011-01-11 20:49:28 +0000
@@ -1,4 +1,4 @@
-# Copyright (C) 2007-2010 Canonical Ltd
+# Copyright (C) 2007-2011 Canonical Ltd
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -30,7 +30,7 @@
     )
 from bzrlib import bencode
 """)
-from bzrlib.tuned_gzip import GzipFile
+from gzip import GzipFile
 
 
 def topo_iter_keys(vf, keys=None):
@@ -564,7 +564,8 @@
         zip_file = GzipFile(None, mode='rb', fileobj=sio)
         try:
             file_version_id = zip_file.readline()
-            return MultiParent.from_patch(zip_file.read())
+            content = zip_file.read()
+            return MultiParent.from_patch(content)
         finally:
             zip_file.close()
 

=== modified file 'bzrlib/repofmt/weaverepo.py'
--- bzrlib/repofmt/weaverepo.py	2010-11-20 21:41:05 +0000
+++ bzrlib/repofmt/weaverepo.py	2011-01-11 20:49:28 +0000
@@ -1,4 +1,4 @@
-# Copyright (C) 2007-2010 Canonical Ltd
+# Copyright (C) 2007-2011 Canonical Ltd
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -20,6 +20,7 @@
 ghosts.
 """
 
+import gzip
 import os
 from cStringIO import StringIO
 import urllib
@@ -40,6 +41,7 @@
     lockdir,
     osutils,
     trace,
+    tuned_gzip,
     urlutils,
     versionedfile,
     weave,
@@ -56,7 +58,6 @@
     RepositoryFormat,
     )
 from bzrlib.store.text import TextStore
-from bzrlib.tuned_gzip import GzipFile, bytes_to_gzip
 from bzrlib.versionedfile import (
     AbsentContentFactory,
     FulltextContentFactory,
@@ -589,7 +590,7 @@
             raise ValueError('bad idea to put / in %r' % (key,))
         text = ''.join(lines)
         if self._compressed:
-            text = bytes_to_gzip(text)
+            text = tuned_gzip.bytes_to_gzip(text)
         path = self._map(key)
         self._transport.put_bytes_non_atomic(path, text, create_parent_dir=True)
 
@@ -637,7 +638,7 @@
             else:
                 return None
         if compressed:
-            text = GzipFile(mode='rb', fileobj=StringIO(text)).read()
+            text = gzip.GzipFile(mode='rb', fileobj=StringIO(text)).read()
         return text
 
     def _map(self, key):

=== modified file 'bzrlib/repository.py'
--- bzrlib/repository.py	2011-01-11 03:28:39 +0000
+++ bzrlib/repository.py	2011-01-11 20:49:28 +0000
@@ -1,4 +1,4 @@
-# Copyright (C) 2005-2010 Canonical Ltd
+# Copyright (C) 2005-2011 Canonical Ltd
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by

=== modified file 'bzrlib/tests/per_versionedfile.py'
--- bzrlib/tests/per_versionedfile.py	2010-12-24 22:40:27 +0000
+++ bzrlib/tests/per_versionedfile.py	2011-01-11 20:49:28 +0000
@@ -1,4 +1,4 @@
-# Copyright (C) 2006-2010 Canonical Ltd
+# Copyright (C) 2006-2011 Canonical Ltd
 #
 # Authors:
 #   Johan Rydberg <[e-mail]>
@@ -21,6 +21,7 @@
 # TODO: might be nice to create a versionedfile with some type of corruption
 # considered typical and check that it can be detected/corrected.
 
+from gzip import GzipFile
 from itertools import chain, izip
 from StringIO import StringIO
 
@@ -51,7 +52,6 @@
     )
 from bzrlib.tests.http_utils import TestCaseWithWebserver
 from bzrlib.transport.memory import MemoryTransport
-from bzrlib.tuned_gzip import GzipFile
 import bzrlib.versionedfile as versionedfile
 from bzrlib.versionedfile import (
     ConstantMapper,

=== modified file 'bzrlib/tests/test_knit.py'
--- bzrlib/tests/test_knit.py	2010-06-20 11:18:38 +0000
+++ bzrlib/tests/test_knit.py	2011-01-11 20:49:28 +0000
@@ -1,4 +1,4 @@
-# Copyright (C) 2006-2010 Canonical Ltd
+# Copyright (C) 2006-2011 Canonical Ltd
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -17,6 +17,7 @@
 """Tests for Knit data structure"""
 
 from cStringIO import StringIO
+import gzip
 import sys
 
 from bzrlib import (
@@ -27,7 +28,6 @@
     pack,
     tests,
     transport,
-    tuned_gzip,
     )
 from bzrlib.errors import (
     KnitHeaderError,
@@ -692,7 +692,7 @@
 
     def create_gz_content(self, text):
         sio = StringIO()
-        gz_file = tuned_gzip.GzipFile(mode='wb', fileobj=sio)
+        gz_file = gzip.GzipFile(mode='wb', fileobj=sio)
         gz_file.write(text)
         gz_file.close()
         return sio.getvalue()

=== modified file 'bzrlib/tests/test_tuned_gzip.py'
--- bzrlib/tests/test_tuned_gzip.py	2010-09-25 20:08:01 +0000
+++ bzrlib/tests/test_tuned_gzip.py	2011-01-11 20:49:28 +0000
@@ -1,4 +1,4 @@
-# Copyright (C) 2006, 2009, 2010 Canonical Ltd
+# Copyright (C) 2006, 2009, 2010, 2011 Canonical Ltd
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -21,11 +21,14 @@
 # do not use bzrlib test cases here - this should be suitable for sending
 # upstream.
 from cStringIO import StringIO
-from unittest import TestCase
 import zlib
 
 
-from bzrlib import tuned_gzip
+from bzrlib import (
+    symbol_versioning,
+    tuned_gzip,
+    tests,
+    )
 
 
 class FakeDecompress(object):
@@ -42,7 +45,7 @@
         return ''
 
 
-class TestFakeDecompress(TestCase):
+class TestFakeDecompress(tests.TestCase):
     """We use a fake decompressor to test GzipFile.
 
     This class tests the behaviours we want from it.
@@ -67,14 +70,16 @@
         self.assertEqual('1234567', decompress.unused_data)
 
 
-class TestGzip(TestCase):
+class TestGzip(tests.TestCase):
 
     def test__read_short_remainder(self):
         # a _read call at the end of a compressed hunk should
         # read more bytes if there is less than 8 bytes (the
         # gzip trailer) unread.
         stream = StringIO('\0\0\0\0\0\0\0\0')
-        myfile = tuned_gzip.GzipFile(fileobj=stream)
+        myfile = self.applyDeprecated(
+            symbol_versioning.deprecated_in((2, 3, 0)),
+            tuned_gzip.GzipFile, fileobj=stream)
         # disable the _new_member check, we are microtesting.
         myfile._new_member = False
         myfile.crc = zlib.crc32('')
@@ -89,21 +94,25 @@
     def test_negative_crc(self):
         """Content with a negative crc should not break when written"""
         sio = StringIO()
-        gfile = tuned_gzip.GzipFile(mode="w", fileobj=sio)
+        gfile = self.applyDeprecated(
+            symbol_versioning.deprecated_in((2, 3, 0)),
+            tuned_gzip.GzipFile, mode="w", fileobj=sio)
         gfile.write("\xFF")
         gfile.close()
         self.assertEqual(gfile.crc & 0xFFFFFFFFL, 0xFF000000L)
         self.assertEqual(sio.getvalue()[-8:-4], "\x00\x00\x00\xFF")
 
 
-class TestToGzip(TestCase):
+class TestToGzip(tests.TestCase):
 
     def assertToGzip(self, chunks):
         bytes = ''.join(chunks)
         gzfromchunks = tuned_gzip.chunks_to_gzip(chunks)
         gzfrombytes = tuned_gzip.bytes_to_gzip(bytes)
         self.assertEqual(gzfrombytes, gzfromchunks)
-        decoded = tuned_gzip.GzipFile(fileobj=StringIO(gzfromchunks)).read()
+        decoded = self.applyDeprecated(
+            symbol_versioning.deprecated_in((2, 3, 0)),
+            tuned_gzip.GzipFile, fileobj=StringIO(gzfromchunks)).read()
         self.assertEqual(bytes, decoded)
 
     def test_single_chunk(self):

=== modified file 'bzrlib/tuned_gzip.py'
--- bzrlib/tuned_gzip.py	2010-09-25 20:08:01 +0000
+++ bzrlib/tuned_gzip.py	2011-01-11 20:49:28 +0000
@@ -1,4 +1,4 @@
-# Copyright (C) 2006-2010 Canonical Ltd
+# Copyright (C) 2006-2011 Canonical Ltd
 # Written by Robert Collins <[e-mail]>
 #
 # This program is free software; you can redistribute it and/or modify
@@ -27,7 +27,7 @@
 import zlib
 
 # we want a \n preserved, break on \n only splitlines.
-import bzrlib
+from bzrlib import symbol_versioning
 
 __all__ = ["GzipFile", "bytes_to_gzip"]
 
@@ -118,6 +118,13 @@
     Yes, its only 1.6 seconds, but they add up.
     """
 
+    def __init__(self, *args, **kwargs):
+        symbol_versioning.warn(
+            symbol_versioning.deprecated_in((2, 3, 0))
+            % 'bzrlib.tuned_gzip.GzipFile',
+            DeprecationWarning, stacklevel=2)
+        gzip.GzipFile.__init__(self, *args, **kwargs)
+
     def _add_read_data(self, data):
         # 4169 calls in 183
         # temp var for len(data) and switch to +='s.

=== modified file 'doc/en/release-notes/bzr-2.3.txt'
--- doc/en/release-notes/bzr-2.3.txt	2011-01-11 20:10:31 +0000
+++ doc/en/release-notes/bzr-2.3.txt	2011-01-11 20:49:28 +0000
@@ -66,6 +66,11 @@
   trying to set the tags in the master branch. This had been broken by the
   bug fix for bug #603395.  (John Arbash Meinel, #701212)
 
+* Stop using ``bzrlib.tuned_gzip.GzipFile``. It is incompatible with
+  python-2.7 and was only used for Knit format repositories, which haven't
+  been recommended since 2007. The file itself will be removed in the next
+  release. (John Arbash Meinel)
+
 Documentation
 *************
 
@@ -88,6 +93,10 @@
   already opened that repository).  Implementations of these APIs will
   need to be updated to accept these arguments.  (Andrew Bennetts)
 
+* ``bzrlib.tuned_gzip.GzipFile`` is now deprecated and will be removed in
+  the bzr-2.4 series. Code that was using it can just use the python
+  stdlib ``gzip.GzipFile``. (John Arbash Meinel)
+
 Internals
 *********