2011-01-19 20:13:38 +01:00
|
|
|
=========
|
|
|
|
ChangeLog
|
|
|
|
=========
|
|
|
|
|
|
|
|
Users should ignore this content: **it is draft**.
|
|
|
|
|
|
|
|
Contributors should add entries here in the following section, on top of the
|
|
|
|
others.
|
|
|
|
|
|
|
|
`WIP (coming releases)`
|
|
|
|
=======================
|
|
|
|
|
|
|
|
New Features
|
|
|
|
------------
|
|
|
|
|
|
|
|
Changes
|
|
|
|
-------
|
|
|
|
|
2011-03-09 08:44:41 +01:00
|
|
|
* TTYUI ouput improved.
|
|
|
|
|
2011-01-19 20:13:38 +01:00
|
|
|
Bug Fixes
|
|
|
|
---------
|
|
|
|
|
fix hang because of infinite loop reading EOF
Read() should return empty string when EOF happen, instead of looping
forever. This is the right semantics of read(), and a wrapped version
should not change it.
If you read the read(2) system call manpage, it tells you that when EOF
is seen, return value is 0; it does not say
``loop forever when EOF happen''.
After the EOF detection is patched you can see the
following exception:
WARNING: ERROR attempting to copy message 344 for account Gmail:Traceback (most recent call last):
File "/usr/lib/pymodules/python2.6/offlineimap/folder/Base.py", line 282, in copymessageto
message = self.getmessage(uid)
File "/usr/lib/pymodules/python2.6/offlineimap/folder/IMAP.py", line 216, in getmessage
initialresult = imapobj.uid('fetch', '%d' % uid, '(BODY.PEEK[])')
File "/usr/lib/python2.6/imaplib.py", line 753, in uid
typ, dat = self._simple_command(name, command, *args)
File "/usr/lib/python2.6/imaplib.py", line 1060, in _simple_command
return self._command_complete(name, self._command(name, *args))
File "/usr/lib/python2.6/imaplib.py", line 890, in _command_complete
raise self.abort('command: %s => %s' % (name, val))
abort: command: UID => socket error: EOF
Signed-off-by: Bao Haojun <baohaojun@gmail.com>
Signed-off-by: Nicolas Sebrecht <nicolas.s-dev@laposte.net>
2011-03-03 10:48:26 +01:00
|
|
|
* Fix hang because of infinite loop reading EOF.
|
2011-03-01 15:08:38 +01:00
|
|
|
* Fix regression (UIBase is no more).
|
|
|
|
|
2011-03-03 13:44:39 +01:00
|
|
|
* Make profiling mode really enforce single-threading
|
2011-01-19 20:13:38 +01:00
|
|
|
|
|
|
|
Pending for the next major release
|
|
|
|
==================================
|
|
|
|
|
|
|
|
* UIs get shorter and nicer names. (API changing)
|
|
|
|
|
|
|
|
|
|
|
|
Stalled
|
|
|
|
=======
|
|
|
|
|
|
|
|
* Learn Sqlite support.
|
|
|
|
Stalled: it would need to learn the ability to choose between the current
|
|
|
|
format and SQL to help testing the long term.
|