website: | http://offlineimap.org |
---|
Avoid lockups for IMAP synchronizations running with the "-1" command-line switch (X-Ryl669 <boite.pour.spam@gmail.com>)
Dump stacktrace for all threads on SIGQUIT: ease debugging of threading and other issues
SIGHUP is now handled as the termination notification rather than the signal to reread the configuration (Dmitrijs Ledkovs)
Honor the timezone of emails (Tobias Thierer)
Allow mbnames output to be sorted by a custom sort key by specifying a 'sort_keyfunc' function in the [mbnames] section of the config.
Support SASL PLAIN authentication method. (Andreas Mack)
Support transport-only tunnels that requre full IMAP authentication. (Steve Purcell)
Make the list of authentication mechanisms to be configurable. (Andreas Mack)
Allow to set message access and modification timestamps based on the "Date" header of the message itself. (Cyril Russo)
"peritem" format string for [mbnames] got new expansion key "localfolders" that corresponds to the same parameter of the local repository for the account being processed.
[regression] pass folder names to the foldersort function, revert the documented behaviour
Fix handling of zero-sized IMAP data items (GitHub#15).
Updated bundled imaplib2 to 2.35: - fix for Gmail sending a BYE response after reading >100 messages
in a session;
Updated bundled imaplib2 to 2.36: it includes support for SSL version override that was integrated into our code before, no other changes.
Fixed parsing of quoted strings in IMAP responses: strings like "\" were treated as having " as the escaped quote, rather than treating it as the quoted escaped backslash (GitHub#53).
Execute pre/post-sync hooks during synchronizations toggled by IMAP IDLE message processing. (maxgerer@gmail.com)
Catch unsuccessful local mail uploads when IMAP server responds with "NO" status; that resulted in a loss of such local messages. (Adam Spiers)
Don't create folders if readonly is enabled.
Learn to deal with readonly folders to properly detect this condition and act accordingly. One example is Gmail's "Chats" folder that is read-only, but contains logs of the quick chats. (E. Ryabinkin)
Fix str.format() calls for Python 2.6 (D. Logie)
Remove APPENDUID hack, previously introduced to fix Gmail, no longer necessary, it might have been breaking things. (J. Wiegley)
Improve regex that could lead to 'NoneType' object has no attribute 'group' (D. Franke)
Improved error throwing on repository misconfiguration
Smallish bug fixes that deserve to be put out.
Blinkenlights UI 6.5.0 regression fixes only.
This is a CRITICAL bug fix release for everyone who is on the 6.4.x series. Please upgrade to avoid potential data loss! The version has been bumped to 6.5.0, please let everyone know that the 6.4.x series is problematic.
This is a bugfix release, fixing regressions occurring in or since 6.4.0.
This is the first stable release to support the forward-compatible per-account locks and remote folder creation that has been introduced in the 6.3.5 series.
Idle feature and SQLite backend leave the experimental stage! ,-)
Here we are. A nice release since v6.3.3, I think.
There is nothing exciting in this release. This is somewhat expected due to the late merge on -rc3.
Here is a surprising release. :-)
As expected we have a lot bug fixes in this round (see git log for details), including a fix for a bug we had for ages (details below) which is a very good news.
What makes this cycle so unusual is that I merged a feature to support StartTLS automatically (thanks Sebastian!). Another very good news.
We usually don't do much changes so late in a cycle. Now, things are highly calming down and I hope a lot of people will test this release. Next one could be the stable!
This was a very active rc1 and we could expect a lot of new fixes for the next release.
The most important fix is about a bug that could lead to data loss. Find more information about his bug here:
http://permalink.gmane.org/gmane.mail.imap.offlineimap.general/3803
The IDLE support is merged as experimental feature.
Welcome to the v6.3.4 pre-release cycle. Your favorite IMAP tool wins 2 new features which were asked for a long time: * an experimental SQL-based backend for the local cache; * one-way synchronization cabability.
Logic synchronization is reviewed and simplified (from 4 to 3 passes) giving improved performance.
Lot of work was done to give OfflineIMAP a better code base. Raised errors can now rely on a new error system and should become the default in the coming releases.
As usual, we ask our users to test this release as much as possible, especially the SQL backend. Have fun!
Make this last candidate cycle short. It looks like we don't need more tests as most issues were raised and solved in the second round. Also, we have huge work to merge big and expected features into OfflineIMAP.
Thanks to all contributors, again. With such a contribution rate, we can release stable faster. I hope it will be confirmed in the longer run!
It's more than a week since the previous release. Most of the issues raised were discussed and fixed since last release. I think we can be glad and confident for the future while the project live his merry life.
We are now at the third week of the -rc1 cycle. I think it's welcome to begin the -rc2 cycle. Things are highly calming down in the code even if we had much more feedbacks than usual. Keep going your effort!
(my apologies if I forget somebody) ...and all active developers, of course!
The imaplib2 migration looks to go the right way to be definetly released but still needs more tests. So, here we go...
Here is time to begin the tests cycle. If feature topics are sent, I may merge or delay them until the next stable release.
Main change comes from the migration from imaplib to imaplib2. It's internal code changes and doesn't impact users. UIDPLUS and subjectAltName for SSL are also great improvements.
This release includes a hang fix due to infinite loop. Users seeing OfflineIMAP hang and consuming a lot of CPU are asked to update.
That beeing said, this is still an early release candidate you should use for non-critical data only!
First of all I'm really happy to announce our new official website. Most of the work started from the impulse of Philippe LeCavalier with the help of Sebastian Spaeth and other contributors. Thanks to everybody.
In this release, we are still touched by the "SSL3 write pending" but I think time was long enough to try to fix it. We have our first entry in the "KNOWN BUG" section of the manual about that. I'm afraid it could impact a lot of users if some distribution package any SSL library not having underlying (still obscure) requirements. Distribution maintainers should be care of it. I hope this release will help us to have more reports.
This release will also be the root of our long maintenance support.
Other bugs were fixed.
We are still touched by the "SSL3 write pending" bug it would be really nice to fix before releasing the coming stable. In the worse case, we'll have to add the first entry in the "KNOWN BUG" section of the manual. I'm afraid it could impact a lot of users if some distribution package any SSL library not having underlying (still obscure) requirements.
The best news with this release are the Curse UI fixed and the better reports on errors.
In this release I won't merge any patch not fixing a bug or a security issue.
More feedbacks on the main issue would be appreciated.
We are beginning a new tests cycle. At this stage, I expect most people will try to intensively stuck OfflineIMAP. :-)
We are beginning a tests cycle. If feature topics are sent, I may merge or delay them until the next stable release.
Yes, I know I've just annouced the v6.3.0 in the same week. As said, it was not really a true release for the software. This last release includes fixes and improvements it might be nice to update to.
Thanks to every body who helped to make this release with patches and tips through the mailing list. This is clearly a release they own.
This release is more "administrative" than anything else and mainly marks the change of the maintainer. New workflow and policy for developers come in. BTW, I don't think I'll maintain debian/changelog. At least, not in the debian way.
Most users and maintainers may rather want to skip this release.