[Python-checkins] cpython (3.2): #9559: Append data to single-file mailbox files if messages are only added

[petri.lehtinen]

http://hg.python.org/cpython/rev/5f447a005d67
changeset:   77833:5f447a005d67
branch:      3.2
parent:      77819:d07319eeae87
user:        Petri Lehtinen <petri at digip.org>
date:        Thu Jun 28 13:48:17 2012 +0300
summary:
  #9559: Append data to single-file mailbox files if messages are only added

If messages were only added, a new file is no longer created and
renamed over the old file when flush() is called on an mbox, MMDF or
Babyl mailbox.

files:
  Lib/mailbox.py           |  18 ++++++++++++++--
  Lib/test/test_mailbox.py |  29 ++++++++++++++++++++++++++-
  Misc/NEWS                |   4 +++
  3 files changed, 46 insertions(+), 5 deletions(-)


diff --git a/Lib/mailbox.py b/Lib/mailbox.py
--- a/Lib/mailbox.py
+++ b/Lib/mailbox.py
@@ -587,16 +587,19 @@
         self._file = f
         self._toc = None
         self._next_key = 0
-        self._pending = False   # No changes require rewriting the file.
+        self._pending = False       # No changes require rewriting the file.
+        self._pending_sync = False  # No need to sync the file
         self._locked = False
-        self._file_length = None        # Used to record mailbox size
+        self._file_length = None    # Used to record mailbox size
 
     def add(self, message):
         """Add message and return assigned key."""
         self._lookup()
         self._toc[self._next_key] = self._append_message(message)
         self._next_key += 1
-        self._pending = True
+        # _append_message appends the message to the mailbox file. We
+        # don't need a full rewrite + rename, sync is enough.
+        self._pending_sync = True
         return self._next_key - 1
 
     def remove(self, key):
@@ -642,6 +645,11 @@
     def flush(self):
         """Write any pending changes to disk."""
         if not self._pending:
+            if self._pending_sync:
+                # Messages have only been added, so syncing the file
+                # is enough.
+                _sync_flush(self._file)
+                self._pending_sync = False
             return
 
         # In order to be writing anything out at all, self._toc must
@@ -695,6 +703,7 @@
         self._file = open(self._path, 'rb+')
         self._toc = new_toc
         self._pending = False
+        self._pending_sync = False
         if self._locked:
             _lock_file(self._file, dotlock=False)
 
@@ -731,6 +740,9 @@
         """Append message to mailbox and return (start, stop) offsets."""
         self._file.seek(0, 2)
         before = self._file.tell()
+        if len(self._toc) == 0:
+            # This is the first message
+            self._pre_mailbox_hook(self._file)
         try:
             self._pre_message_hook(self._file)
             offsets = self._install_message(message)
diff --git a/Lib/test/test_mailbox.py b/Lib/test/test_mailbox.py
--- a/Lib/test/test_mailbox.py
+++ b/Lib/test/test_mailbox.py
@@ -942,7 +942,32 @@
         self._box._refresh()
         self.assertTrue(refreshed())
 
-class _TestMboxMMDF(TestMailbox):
+
+class _TestSingleFile(TestMailbox):
+    '''Common tests for single-file mailboxes'''
+
+    def test_add_doesnt_rewrite(self):
+        # When only adding messages, flush() should not rewrite the
+        # mailbox file. See issue #9559.
+
+        # Inode number changes if the contents are written to another
+        # file which is then renamed over the original file. So we
+        # must check that the inode number doesn't change.
+        inode_before = os.stat(self._path).st_ino
+
+        self._box.add(self._template % 0)
+        self._box.flush()
+
+        inode_after = os.stat(self._path).st_ino
+        self.assertEqual(inode_before, inode_after)
+
+        # Make sure the message was really added
+        self._box.close()
+        self._box = self._factory(self._path)
+        self.assertEqual(len(self._box), 1)
+
+
+class _TestMboxMMDF(_TestSingleFile):
 
     def tearDown(self):
         super().tearDown()
@@ -1217,7 +1242,7 @@
         return os.path.join(self._path, '.mh_sequences.lock')
 
 
-class TestBabyl(TestMailbox, unittest.TestCase):
+class TestBabyl(_TestSingleFile, unittest.TestCase):
 
     _factory = lambda self, path, factory=None: mailbox.Babyl(path, factory)
 
diff --git a/Misc/NEWS b/Misc/NEWS
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -81,6 +81,10 @@
 Library
 -------
 
+- Issue #9559: If messages were only added, a new file is no longer
+  created and renamed over the old file when flush() is called on an
+  mbox, MMDF or Babyl mailbox.
+
 - Issue #14653: email.utils.mktime_tz() no longer relies on system
   mktime() when timezone offest is supplied.
 

-- 
Repository URL: http://hg.python.org/cpython