Merged revisions 993251,993269-993270,993305,993311,993326,993338,993446,993450,993461-993464,993472,993474,993477,993480,993495,993501,993515,993518,993520,993522,993531,993542,993545,993550,993557,993561,993573,993582,993591,993599,993629,993641,993738,993741-993742,993746,993846,993878,993906,993963-993964,993970,994012,994017,994063,994093,994129,994223,994262,994339,994382,994441,994459,994532 via svnmerge from
https://vkrause@svn.kde.org/home/kde/trunk/KDE/kdepim
........
r993251 | trueg | 2009-07-08 12:46:00 +0200 (Wed, 08 Jul 2009) | 3 lines
Nepomuk rcgen cleanup first step (or rather fourth or something considering tokoe's amazing work before):
- disabling templates. we do not use them anyway and they will be deprecated soon.
........
r993269 | skelly | 2009-07-08 13:14:13 +0200 (Wed, 08 Jul 2009) | 1 line
Make sure tests includes are available.
........
r993270 | skelly | 2009-07-08 13:14:17 +0200 (Wed, 08 Jul 2009) | 2 lines
Break the main window into a tabbed interface for multiple proxy model
tests.
........
r993305 | lueck | 2009-07-08 14:52:34 +0200 (Wed, 08 Jul 2009) | 1 line
typo fix
........
r993311 | winterz | 2009-07-08 15:08:11 +0200 (Wed, 08 Jul 2009) | 9 lines
Refinement of import/export of kaddressbook from/to GMX.
Patch entirely from Urs Joss. Thanks!
Joss, please close the review board request for this as closed.
Remember: we need to also put this into kcontactmanager.
http://reviewboard.kde.org/r/941
CCMAIL: tschenturs@gmx.ch
........
r993326 | djarvie | 2009-07-08 15:42:42 +0200 (Wed, 08 Jul 2009) | 1 line
Move methods out of base class
........
r993338 | vkrause | 2009-07-08 16:00:53 +0200 (Wed, 08 Jul 2009) | 4 lines
Birthday resources can have different settings, so it does not make
sense to only allow one instance. Should help with the migration problem
David is seeing.
........
r993446 | skelly | 2009-07-08 19:47:31 +0200 (Wed, 08 Jul 2009) | 6 lines
Add another configuration use for the selection proxy model.
This is getting cumbersome, so I will soon change the API and refactor
the internals.
There are also some remaining nagging bugs.
........
r993450 | skelly | 2009-07-08 19:50:58 +0200 (Wed, 08 Jul 2009) | 1 line
Enable akonadi_next to be built standalone.
........
r993461 | skelly | 2009-07-08 20:12:46 +0200 (Wed, 08 Jul 2009) | 1 line
Start some support for AmazingCompletion(tm)
........
r993462 | skelly | 2009-07-08 20:12:50 +0200 (Wed, 08 Jul 2009) | 4 lines
Create a sane implementation od AbstractProxyModelMatch.
This should be in QAbstractProxyModel. I'll patch it there later and
send a merge request upstream.
........
r993463 | skelly | 2009-07-08 20:12:53 +0200 (Wed, 08 Jul 2009) | 2 lines
Implement match for the ContactsModel. Will allow matching Amazingly
eventually.
........
r993464 | skelly | 2009-07-08 20:12:57 +0200 (Wed, 08 Jul 2009) | 1 line
Reimplement match.
........
r993472 | skelly | 2009-07-08 20:25:04 +0200 (Wed, 08 Jul 2009) | 1 line
Better match implementation.
........
r993474 | tmcguire | 2009-07-08 20:33:31 +0200 (Wed, 08 Jul 2009) | 4 lines
Prevent an infinite recursion when entering a new folder with at least one message.
Curiously, this is very similar to kolab/issue3727
........
r993477 | skelly | 2009-07-08 20:39:54 +0200 (Wed, 08 Jul 2009) | 1 line
Really fix matching this time.
........
r993480 | mleupold | 2009-07-08 20:51:54 +0200 (Wed, 08 Jul 2009) | 1 line
Fix/hide various krazy issues including most qclasses, QLatin1String and kdebug.
........
r993495 | bbroeksema | 2009-07-08 21:39:29 +0200 (Wed, 08 Jul 2009) | 2 lines
Krazy fixes: Use const references instead of const values in Q_FOREACH.
........
r993501 | mlaurent | 2009-07-08 21:44:23 +0200 (Wed, 08 Jul 2009) | 2 lines
compile++
........
r993515 | tstaerk | 2009-07-08 22:18:36 +0200 (Wed, 08 Jul 2009) | 1 line
karm->ktimetracker && dcop->dbus
........
r993518 | tstaerk | 2009-07-08 22:30:55 +0200 (Wed, 08 Jul 2009) | 1 line
karm->ktimetracker && dcop->dbus
........
r993520 | cgiboudeaux | 2009-07-08 22:41:32 +0200 (Wed, 08 Jul 2009) | 1 line
SVN_SILENT Cleanup
........
r993522 | cgiboudeaux | 2009-07-08 22:42:49 +0200 (Wed, 08 Jul 2009) | 1 line
Cleanup
........
r993531 | tstaerk | 2009-07-08 22:53:34 +0200 (Wed, 08 Jul 2009) | 1 line
What you never wanted to know about ktimetracker, but were forced to learn anyway--
........
r993542 | bbroeksema | 2009-07-08 23:06:33 +0200 (Wed, 08 Jul 2009) | 2 lines
Krazy fixes: null string assign -> clear()
........
r993545 | bbroeksema | 2009-07-08 23:19:38 +0200 (Wed, 08 Jul 2009) | 2 lines
SVN_SILENT: fix two small spelling errors in comments.
........
r993550 | cgiboudeaux | 2009-07-08 23:29:37 +0200 (Wed, 08 Jul 2009) | 1 line
Partially revert of r967753. Only kresources and plugins must be built after the optional applications.
........
r993557 | bbroeksema | 2009-07-08 23:36:51 +0200 (Wed, 08 Jul 2009) | 2 lines
Fix krazy issues: QLatin1String
........
r993561 | cgiboudeaux | 2009-07-08 23:39:10 +0200 (Wed, 08 Jul 2009) | 1 line
SVN_SILENT Cleanup
........
r993573 | bbroeksema | 2009-07-09 00:02:53 +0200 (Thu, 09 Jul 2009) | 2 lines
Krazy fixes: proper include directives.
........
r993582 | bbroeksema | 2009-07-09 00:25:27 +0200 (Thu, 09 Jul 2009) | 2 lines
Krazy fixes: proper include directives.
........
r993591 | bbroeksema | 2009-07-09 01:02:21 +0200 (Thu, 09 Jul 2009) | 2 lines
Krazy fixes: proper include directives.
........
r993599 | winterz | 2009-07-09 02:32:58 +0200 (Thu, 09 Jul 2009) | 3 lines
compile++
add cstdlib header for exit()
........
r993629 | bbroeksema | 2009-07-09 09:04:53 +0200 (Thu, 09 Jul 2009) | 2 lines
Revert part of r993495 because Protocol is an enum. Let krazy ignore these lines.
........
r993641 | bbroeksema | 2009-07-09 10:42:09 +0200 (Thu, 09 Jul 2009) | 4 lines
Reduce reloads of resources based on the SingleFileResource by using hashes of
the file. This commit prevents reloads of resources based on singlefileresource
when the resource is writing changes to the file itself.
........
r993738 | pokrzywka | 2009-07-09 14:26:52 +0200 (Thu, 09 Jul 2009) | 1 line
nicer fix for the --templates bug in nepomuk_rcgen: only add the option if there are templates specified. Not sure if that will ever be needed again though, but it doesn't hurt for now.
........
r993741 | djarvie | 2009-07-09 14:35:19 +0200 (Thu, 09 Jul 2009) | 1 line
Fix compile warning
........
r993742 | winterz | 2009-07-09 14:46:42 +0200 (Thu, 09 Jul 2009) | 6 lines
Go crazy with node validity checks to try and fix all the crash reports that I can't reproduce.
Might fix
CCBUG: 198961
MERGE: e4,4.3
........
r993746 | pokrzywka | 2009-07-09 14:51:40 +0200 (Thu, 09 Jul 2009) | 3 lines
fixed the build problems with the nepomuk agents on windows. The generated files now handle static linking.
CCMAIL:till@kdab.net
........
r993846 | winterz | 2009-07-09 17:11:23 +0200 (Thu, 09 Jul 2009) | 7 lines
Urs' cleanups
Add Urs to copyright.
I'll copy this into kcontactmanager.
MERGE: e4
........
r993878 | scripty | 2009-07-09 17:42:19 +0200 (Thu, 09 Jul 2009) | 1 line
SVN_SILENT made messages (.desktop file)
........
r993906 | jtamate | 2009-07-09 18:58:08 +0200 (Thu, 09 Jul 2009) | 3 lines
BUG: 192897
Added "at the bottom of the Security tab in this dialog" to the message.
........
r993963 | bbroeksema | 2009-07-09 22:20:13 +0200 (Thu, 09 Jul 2009) | 2 lines
Fix krazy issues: Proper include directives.
........
r993964 | bbroeksema | 2009-07-09 22:24:44 +0200 (Thu, 09 Jul 2009) | 2 lines
Krazy fix: single-char QString operations
........
r993970 | bbroeksema | 2009-07-09 22:29:56 +0200 (Thu, 09 Jul 2009) | 2 lines
Krazy fix: C++ ctors that should be declared 'explicit'
........
r994012 | bbroeksema | 2009-07-10 00:01:00 +0200 (Fri, 10 Jul 2009) | 2 lines
Krazy fixes: validity of i18n calls.
........
r994017 | bbroeksema | 2009-07-10 00:07:32 +0200 (Fri, 10 Jul 2009) | 3 lines
Krazy fix: [foreach[ This is not a call to keys of a map or set, but to a
std::vector. Ignore this one.
........
r994063 | winterz | 2009-07-10 04:16:56 +0200 (Fri, 10 Jul 2009) | 4 lines
fix spacing between the pixmap and the urllabel
BUG: 199598
MERGE: e4,4.3
........
r994093 | mlaurent | 2009-07-10 09:10:16 +0200 (Fri, 10 Jul 2009) | 2 lines
Apply patch
........
r994129 | trueg | 2009-07-10 11:15:40 +0200 (Fri, 10 Jul 2009) | 4 lines
Moved NepomukAddOntologyClasses to kdelibs. No need for a local copy anymore.
Also by default rcgen will not export classes anymore. Thus, there is no need
for the define anymore.
........
r994223 | vkrause | 2009-07-10 13:59:23 +0200 (Fri, 10 Jul 2009) | 3 lines
No need to try to migrate the not yet/no longer supported file formats
of the KABC file resource in every run.
........
r994262 | djarvie | 2009-07-10 15:36:32 +0200 (Fri, 10 Jul 2009) | 4 lines
Refactor to remove dependencies in KAEvent, KARecurrence and AlarmText on KAlarm
preferences, in preparation for Akonadi migration.
All KAEvent data independent of preferences is not in KAEventData.
........
r994339 | ervin | 2009-07-10 16:35:40 +0200 (Fri, 10 Jul 2009) | 2 lines
Let's allow people to type in the KDateEdit. Okay? :-)
........
r994382 | djarvie | 2009-07-10 16:51:41 +0200 (Fri, 10 Jul 2009) | 1 line
Fix alarms not being triggered if reminders occur AFTER the main alarm
........
r994441 | scripty | 2009-07-10 17:19:10 +0200 (Fri, 10 Jul 2009) | 1 line
SVN_SILENT made messages (.desktop file)
........
r994459 | cgiboudeaux | 2009-07-10 17:37:43 +0200 (Fri, 10 Jul 2009) | 1 line
SVN_SILENT Remove the extra comma
........
r994532 | vkrause | 2009-07-10 19:36:12 +0200 (Fri, 10 Jul 2009) | 4 lines
Use the recently introduced server-side collection content mimetype
filtering. This fixes Kolab collections not showing up in eg.
KContactManager.
........
svn path=/branches/work/akonadi-ports/kdepim/; revision=994567
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/kdepim/enterprise4/kdepim
................
r968236 | tmcguire | 2009-05-15 14:27:38 +0200 (Fri, 15 May 2009) | 16 lines
Merged revisions 967378 via svnmerge from
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/kdepim/enterprise/kdepim
........
r967378 | tmcguire | 2009-05-13 12:56:22 +0200 (Wed, 13 May 2009) | 9 lines
Add an option to also check the subfolders when checking a single folder,
and use this when troubleshooting the cache.
That way, we get all the folders on the server back, instead of only the
immediate subfolders of the folder we are checking, thus preventing
bogus error messages like in kolab/issue2972.
kolab/issue2972
........
................
svn path=/trunk/KDE/kdepim/; revision=970738
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/kdepim/enterprise/kdepim
........
r967378 | tmcguire | 2009-05-13 12:56:22 +0200 (Wed, 13 May 2009) | 9 lines
Add an option to also check the subfolders when checking a single folder,
and use this when troubleshooting the cache.
That way, we get all the folders on the server back, instead of only the
immediate subfolders of the folder we are checking, thus preventing
bogus error messages like in kolab/issue2972.
kolab/issue2972
........
svn path=/branches/kdepim/enterprise4/kdepim/; revision=968236
and use this when troubleshooting the cache.
That way, we get all the folders on the server back, instead of only the
immediate subfolders of the folder we are checking, thus preventing
bogus error messages like in kolab/issue2972.
kolab/issue2972
svn path=/branches/kdepim/enterprise/kdepim/; revision=967378
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/KDE/4.2/kdepim
........
r908525 | kloecker | 2009-01-09 22:29:44 +0100 (Fri, 09 Jan 2009) | 4 lines
Merged revision 908518 from trunk
Fix crash by accessing a KMMsgBase object that was replaced by a KMMessage object. This is fixed by using getMsgBase() instead of getMsg() if we previously had a KMMsgBase object.
CCBUG: 179765
........
svn path=/branches/kdepim/enterprise4/kdepim/; revision=914256
Fix crash by accessing a KMMsgBase object that was replaced by a KMMessage object. This is fixed by using getMsgBase() instead of getMsg() if we previously had a KMMsgBase object.
CCBUG: 179765
svn path=/branches/KDE/4.2/kdepim/; revision=908525
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/KDE/4.1/kdepim
........
r904948 | tmcguire | 2009-01-03 16:14:10 +0100 (Sat, 03 Jan 2009) | 9 lines
Backport r900449 by tmcguire from trunk to the 4.1 branch:
Fix a crash when syncing flags from a server while we have a search folder
that has a search rule which needs the complete body, because this unget()s
the message behind our back right after changing the flags.
CCBUG: 106030
........
svn path=/branches/kdepim/enterprise4/kdepim/; revision=905010
that has a search rule which needs the complete body, because this unget()s
the message behind our back right after changing the flags.
BUG: 106030
svn path=/trunk/KDE/kdepim/; revision=900449
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/KDE/4.1/kdepim
........
r883868 | tmcguire | 2008-11-13 19:37:27 +0100 (Thu, 13 Nov 2008) | 11 lines
Backport r879276 by tmcguire from trunk to the 4.1 branch:
When a message is already in the metadata map (either by UID or by MD5), don't
filter it, as it (hopefully) means we have already seen and therefore filtered
the message.
This fixes filters for online IMAP running multiple times for the same message,
and might even fix the problem of duplicate messages after filtering.
CCBUG: 95064
........
svn path=/branches/kdepim/enterprise4/kdepim/; revision=884010
When a message is already in the metadata map (either by UID or by MD5), don't
filter it, as it (hopefully) means we have already seen and therefore filtered
the message.
This fixes filters for online IMAP running multiple times for the same message,
and might even fix the problem of duplicate messages after filtering.
CCBUG: 95064
svn path=/branches/KDE/4.1/kdepim/; revision=883868
Let's hope I didn't screw this up, otherwise we'll see interesting filtering bugs.
Now the only Qt3 things left are the manage sieve script dialog (Tom, feeling like porting this?) and the
subscription dialog.
Also, remove duplicate code in deleteMessage: simply call the version for one message for each message in
the list.
svn path=/trunk/KDE/kdepim/; revision=881691
This brings a reworked folder tree and and a new message list, which is much more powerful than the old one.
New features include grouping, multiline items and the ability to customize just about everything of the look of the message list.
Also, we have a tabbed interface for opening more than one folder at once now.
An important aspect of the new message list is that drawing speed is now fast again (there was a regression in Q3ListView that
made huge folders a real pain to navigate, reading commit folders was not a nice thing to do).
There are probably more things which I forgot to list here.
Also, this is big step forward in getting rid of Qt3Support in KMail, now just some tiny bits are left.
Many thanks to SoC student Szymon Stefanek for his awesome work on his project! I can't stress enough how great this
achivment is. He even added very good API documentation to his classes :)
This requires the lastest version of qt-copy, otherwise you'll get mysterious crashes in Qt painting code from time to time.
We hope that this patch will be added to the next Qt 4.4.x version.
And finally, this merge also brings many regressions, which is not unnatural for such a big change.
In the remaining time before the KDE 4.2 release, we'll have to work on those to get it polished and regression-free.
Help here is always needed, and is a good opportunity to start working on KMail, since many regressions are low hanging fruit.
It also means the message list and the folder tree need a bit of testing.
If you find bugs, report them under the "new message list" and "new folder tree" component of the bug tracker, thanks.
I also plan to setup a wiki page with known regressions.
CCMAIL: kde-pim@kde.org
GUI:
CCBUG: 117808
BUG: 163469
FEATURE: 18170
FEATURE: 32400
FEATURE: 42107
FEATURE: 45526
FEATURE: 81272
FEATURE: 90142
FEATURE: 107450
(and probably many others)
svn path=/trunk/KDE/kdepim/; revision=881321
filter it, as it (hopefully) means we have already seen and therefore filtered
the message.
This fixes filters for online IMAP running multiple times for the same message,
and might even fix the problem of duplicate messages after filtering.
CCBUG: 95064
svn path=/trunk/KDE/kdepim/; revision=879276
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/KDE/4.1/kdepim
........
r858265 | tmcguire | 2008-09-07 18:41:52 +0200 (Sun, 07 Sep 2008) | 9 lines
Backport r855882 by tmcguire from trunk to the 4.1 branch:
Don't abort() here, but silently ignore the error.
This fixes a crash when renaming an account while the root IMAP folder is selected,
but probably only hides another problem...
CCBUG: 170149.
........
svn path=/branches/kdepim/enterprise4/kdepim/; revision=858296
Don't abort() here, but silently ignore the error.
This fixes a crash when renaming an account while the root IMAP folder is selected,
but probably only hides another problem...
CCBUG: 170149.
svn path=/branches/KDE/4.1/kdepim/; revision=858265
This fixes a crash when renaming an account while the root IMAP folder is selected,
but probably only hides another problem...
BUG: 170149.
svn path=/trunk/KDE/kdepim/; revision=855882
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/kdepim/enterprise4/kdepim
................
r854399 | vkrause | 2008-08-29 15:04:19 +0200 (Fri, 29 Aug 2008) | 13 lines
Merged revisions 850513 via svnmerge from
https://vkrause@svn.kde.org/home/kde/branches/kdepim/enterprise/kdepim
........
r850513 | vkrause | 2008-08-21 19:11:57 +0200 (Thu, 21 Aug 2008) | 6 lines
Allow deletion of message only if the IMAP ACLs allow it. The current
code assumed that deletion was not possible iff a folder is read-only,
which is too simple for IMAP.
Kolab issue 2954
........
................
svn path=/trunk/KDE/kdepim/; revision=855221
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/kdepim/enterprise4/kdepim
........
r853982 | vkrause | 2008-08-28 17:31:55 +0200 (Thu, 28 Aug 2008) | 13 lines
Forwardport SVN commit 662047 by tilladam from enterprise branch:
Apply ported version of the mail loss debugging and explicit deletions patch,
which I've been developing with the help of some adventurous users. Thanks!
This tracks all deletions that happen through user actions and adds a check
to the sync making sure that only things that were explicitely deleted
are removed during sync. If unsure, the sync now re-downloads (duplicates)
instead of removing mails, which should be safer. Also adds a lot of
conditional debugging and refactors open/close to duplicate less code.
Will has a ported version of this for 3.x, which will go into pim+ shortly.
........
svn path=/trunk/KDE/kdepim/; revision=855193
https://vkrause@svn.kde.org/home/kde/branches/kdepim/enterprise/kdepim
........
r850513 | vkrause | 2008-08-21 19:11:57 +0200 (Thu, 21 Aug 2008) | 6 lines
Allow deletion of message only if the IMAP ACLs allow it. The current
code assumed that deletion was not possible iff a folder is read-only,
which is too simple for IMAP.
Kolab issue 2954
........
svn path=/branches/kdepim/enterprise4/kdepim/; revision=854399
Apply ported version of the mail loss debugging and explicit deletions patch,
which I've been developing with the help of some adventurous users. Thanks!
This tracks all deletions that happen through user actions and adds a check
to the sync making sure that only things that were explicitely deleted
are removed during sync. If unsure, the sync now re-downloads (duplicates)
instead of removing mails, which should be safer. Also adds a lot of
conditional debugging and refactors open/close to duplicate less code.
Will has a ported version of this for 3.x, which will go into pim+ shortly.
svn path=/branches/kdepim/enterprise4/kdepim/; revision=853982
code assumed that deletion was not possible iff a folder is read-only,
which is too simple for IMAP.
Kolab issue 2954
svn path=/branches/kdepim/enterprise/kdepim/; revision=850513
https://vkrause@svn.kde.org/home/kde/branches/kdepim/enterprise/kdepim
........
r807761 | vkrause | 2008-05-14 19:00:30 +0200 (Wed, 14 May 2008) | 5 lines
Don't delete folder configs for folders that might not be in the list
here but nevertheless still exist.
Kolab issue 2667
........
r821814 | vkrause | 2008-06-18 17:21:15 +0200 (Wed, 18 Jun 2008) | 5 lines
Make sure favorite folder items get the same label independent of how
say are created.
Kolab issue 2785.
........
r823865 | tilladam | 2008-06-24 14:03:18 +0200 (Tue, 24 Jun 2008) | 7 lines
Make it possible again to accept an invitation that was sent
to a distribution list the current user is part of, by offering
all the identities in addition to the recipients of the invite
as choices for the attendee.
kolab/issue1998
........
r824289 | vkrause | 2008-06-25 17:20:44 +0200 (Wed, 25 Jun 2008) | 3 lines
Honor local subscription, fixes non-subscribed folders showing up in
the folder tree for a short time after changing local subscriptions.
........
r824806 | vkrause | 2008-06-26 18:42:06 +0200 (Thu, 26 Jun 2008) | 6 lines
The else block is supposed to belong to the outer if, so we need braces
here.
This ensures that a failed annotation job is removed correctly from the
job map and therefore avoids a crash on exit when KMail tries to cancel
the apparently still running job.
........
r830550 | vkrause | 2008-07-10 20:51:09 +0200 (Thu, 10 Jul 2008) | 23 lines
Backport SVN commit 829126 by vkrause from enterprise4 branch:
Close the folder before we unregister from the message dict.
This fixes the .index.db file being newer than the corresponding
.index.ids file, causing the latter being detected as outdated on the
next startup. In this case message wouldn't have a serial number, which
the sqlite backend poorly recovers from, causing apparent mail loss
(fixable by rebuilding the folder index). And to make it even more
confusing, this only happends for new DIMAP folders that have been
synced but not entered yet before shutting down KMail.
Kolab issue 2816
Incidentally this also fixes the mysterious non-persistence of certain
folder infos worked around in revision 830478. This happend because
groupware folders were kept open until destructions. However,
writeConfig() was called from the KMFolderMaildir dtor at which point
the virtual writeConfig() in KMFolderCachedImap wasn't available
anymore.
Kolab issue 2666
........
r834241 | vkrause | 2008-07-18 15:25:03 +0200 (Fri, 18 Jul 2008) | 7 lines
Update the highest uid even if we don't find the corresponding message
anymore or the folder is empty. This happens for example when the
message with this uid was moved away by a filter just before we got
here, which then caused mail duplication during the next sync.
Kolab issue 2801
........
svn path=/branches/KDE/3.5/kdepim/; revision=848898
only talk about "todo" or "to-do" in reference to calendar incidences.
a few config file keys still contain "todo" so that should be changed,
but will be done in another commit.
whoops! I should also check the kmail handbook.
svn path=/trunk/KDE/kdepim/; revision=840786
Some fixes for online IMAP filtering:
- In the action scheduler, don't error out when the filtered message
couldn't be moved back to the target folder. Instead, ignore the error
(but don't delete the orginal message). This fixes filtering
stopping on GMail accounts once a message was encountered that was not
moved to another folder, but stayed in the same folder (GMail prevents moving
in this case, since it thinks it is a duplicate message)
- When moving the filtered message with the action scheduler from the
temporary filter folder to the target folder, the original message wouldn't
get deleted properly.
The reason for this was that the move command thought the move failed because
a message with another serial number arrived.
Fix this by remembering the serial number (based on the message ID) when using
the action scheduler for filtering.
This fixes filters which move messages to other folders: Now the message is properly
removed from the source folder again.
- add comments and kDebug output
This does _not_ solve the following problem:
When using GMail and online IMAP, filter actions which modify the message will not have
any effect, since GMails duplicate message prevention prevents the filtered message to
be moved back to the IMAP folder.
This will not be fixed, I see no way to work around this. GMail should fix their IMAP
server instead.
Please test, I'm not sure if this is safe to backport.
CCBUG: 166150
svn path=/branches/KDE/4.1/kdepim/; revision=836580
Merged revisions 834308 via svnmerge from
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/kdepim/enterprise4/kdepim
................
r834308 | vkrause | 2008-07-18 18:10:11 +0200 (Fri, 18 Jul 2008) | 27 lines
Merged revisions 824289-825075,825077-825083,825085-825805,825807-826354,826356-827490,827492-827612,827614-827891,827893-827903,827905-827944,827946-827949,827951-828972,828974-829033,829035-830099,830101-830132,830134-830442,830444-830477,830479-830549,830551-830875,830877-830932,830934-834241 via svnmerge from
https://vkrause@svn.kde.org/home/kde/branches/kdepim/enterprise/kdepim
........
r824289 | vkrause | 2008-06-25 17:20:44 +0200 (Wed, 25 Jun 2008) | 3 lines
Honor local subscription, fixes non-subscribed folders showing up in
the folder tree for a short time after changing local subscriptions.
........
r824806 | vkrause | 2008-06-26 18:42:06 +0200 (Thu, 26 Jun 2008) | 6 lines
The else block is supposed to belong to the outer if, so we need braces
here.
This ensures that a failed annotation job is removed correctly from the
job map and therefore avoids a crash on exit when KMail tries to cancel
the apparently still running job.
........
r834241 | vkrause | 2008-07-18 15:25:03 +0200 (Fri, 18 Jul 2008) | 7 lines
Update the highest uid even if we don't find the corresponding message
anymore or the folder is empty. This happens for example when the
message with this uid was moved away by a filter just before we got
here, which then caused mail duplication during the next sync.
Kolab issue 2801
........
................
svn path=/branches/KDE/4.1/kdepim/; revision=835464
svn+ssh://tmcguire@svn.kde.org/home/kde/trunk/KDE/kdepim
........
r833475 | tmcguire | 2008-07-16 23:07:56 +0200 (Wed, 16 Jul 2008) | 32 lines
Some fixes for online IMAP filtering:
- In the action scheduler, don't error out when the filtered message
couldn't be moved back to the target folder. Instead, ignore the error
(but don't delete the orginal message). This fixes filtering
stopping on GMail accounts once a message was encountered that was not
moved to another folder, but stayed in the same folder (GMail prevents moving
in this case, since it thinks it is a duplicate message)
- When moving the filtered message with the action scheduler from the
temporary filter folder to the target folder, the original message wouldn't
get deleted properly.
The reason for this was that the move command thought the move failed because
a message with another serial number arrived.
Fix this by remembering the serial number (based on the message ID) when using
the action scheduler for filtering.
This fixes filters which move messages to other folders: Now the message is properly
removed from the source folder again.
- add comments and kDebug output
This does _not_ solve the following problem:
When using GMail and online IMAP, filter actions which modify the message will not have
any effect, since GMails duplicate message prevention prevents the filtered message to
be moved back to the IMAP folder.
This will not be fixed, I see no way to work around this. GMail should fix their IMAP
server instead.
Please test, I'm not sure if this is safe to backport.
CCBUG: 166150
........
svn path=/branches/kdepim/enterprise4/kdepim/; revision=835021
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/kdepim/enterprise4/kdepim
................
r834308 | vkrause | 2008-07-18 18:10:11 +0200 (Fri, 18 Jul 2008) | 27 lines
Merged revisions 824289-825075,825077-825083,825085-825805,825807-826354,826356-827490,827492-827612,827614-827891,827893-827903,827905-827944,827946-827949,827951-828972,828974-829033,829035-830099,830101-830132,830134-830442,830444-830477,830479-830549,830551-830875,830877-830932,830934-834241 via svnmerge from
https://vkrause@svn.kde.org/home/kde/branches/kdepim/enterprise/kdepim
........
r824289 | vkrause | 2008-06-25 17:20:44 +0200 (Wed, 25 Jun 2008) | 3 lines
Honor local subscription, fixes non-subscribed folders showing up in
the folder tree for a short time after changing local subscriptions.
........
r824806 | vkrause | 2008-06-26 18:42:06 +0200 (Thu, 26 Jun 2008) | 6 lines
The else block is supposed to belong to the outer if, so we need braces
here.
This ensures that a failed annotation job is removed correctly from the
job map and therefore avoids a crash on exit when KMail tries to cancel
the apparently still running job.
........
r834241 | vkrause | 2008-07-18 15:25:03 +0200 (Fri, 18 Jul 2008) | 7 lines
Update the highest uid even if we don't find the corresponding message
anymore or the folder is empty. This happens for example when the
message with this uid was moved away by a filter just before we got
here, which then caused mail duplication during the next sync.
Kolab issue 2801
........
................
svn path=/trunk/KDE/kdepim/; revision=834655
https://vkrause@svn.kde.org/home/kde/branches/kdepim/enterprise/kdepim
........
r824289 | vkrause | 2008-06-25 17:20:44 +0200 (Wed, 25 Jun 2008) | 3 lines
Honor local subscription, fixes non-subscribed folders showing up in
the folder tree for a short time after changing local subscriptions.
........
r824806 | vkrause | 2008-06-26 18:42:06 +0200 (Thu, 26 Jun 2008) | 6 lines
The else block is supposed to belong to the outer if, so we need braces
here.
This ensures that a failed annotation job is removed correctly from the
job map and therefore avoids a crash on exit when KMail tries to cancel
the apparently still running job.
........
r834241 | vkrause | 2008-07-18 15:25:03 +0200 (Fri, 18 Jul 2008) | 7 lines
Update the highest uid even if we don't find the corresponding message
anymore or the folder is empty. This happens for example when the
message with this uid was moved away by a filter just before we got
here, which then caused mail duplication during the next sync.
Kolab issue 2801
........
svn path=/branches/kdepim/enterprise4/kdepim/; revision=834308
- In the action scheduler, don't error out when the filtered message
couldn't be moved back to the target folder. Instead, ignore the error
(but don't delete the orginal message). This fixes filtering
stopping on GMail accounts once a message was encountered that was not
moved to another folder, but stayed in the same folder (GMail prevents moving
in this case, since it thinks it is a duplicate message)
- When moving the filtered message with the action scheduler from the
temporary filter folder to the target folder, the original message wouldn't
get deleted properly.
The reason for this was that the move command thought the move failed because
a message with another serial number arrived.
Fix this by remembering the serial number (based on the message ID) when using
the action scheduler for filtering.
This fixes filters which move messages to other folders: Now the message is properly
removed from the source folder again.
- add comments and kDebug output
This does _not_ solve the following problem:
When using GMail and online IMAP, filter actions which modify the message will not have
any effect, since GMails duplicate message prevention prevents the filtered message to
be moved back to the IMAP folder.
This will not be fixed, I see no way to work around this. GMail should fix their IMAP
server instead.
Please test, I'm not sure if this is safe to backport.
CCBUG: 166150
svn path=/trunk/KDE/kdepim/; revision=833475
Laurent Montel
Thomas McGuire
Volker Krause
Michael Leupold
Allen Winter
Ingo Klcker
Bertjan Broeksema
Till Adam