So don't pass a QByteArray to DwHeaders::FindField(), pass a char* instead.
This might fix a crash.
Panagiotis, please test and let us know if this helps fix the crash you are experiencing.
CCBUG: 193891
svn path=/trunk/KDE/kdepim/; revision=972261
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/kdepim/enterprise4/kdepim
........
r971085 | tmcguire | 2009-05-21 18:38:12 +0200 (Thu, 21 May 2009) | 7 lines
Don't encode the X-UID header in the charset of the message, otherwise UTF-16 messages
will cause duplicating messages on disconnected IMAP.
This is a hack, I have the feeling that the core problem is something else.
I guess it is not allowed to encode the header fields in anything non-ascii in
the first place, but decodeRFC2047String() does exactly that...
........
svn path=/trunk/KDE/kdepim/; revision=971091
will cause duplicating messages on disconnected IMAP.
This is a hack, I have the feeling that the core problem is something else.
I guess it is not allowed to encode the header fields in anything non-ascii in
the first place, but decodeRFC2047String() does exactly that...
svn path=/branches/kdepim/enterprise4/kdepim/; revision=971085
removes the Signature Blocks (SB) from prefix+"-- (end of line)" until a line that
* does not starts with prefix or
* starts with prefix+(any substring of prefix)
where prefix can only have any number of ">" and spaces.
svn path=/trunk/KDE/kdepim/; revision=946070
Tested and seems to still work.
--
Merged revisions 940772,940775,940777,940816 via svnmerge from
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/KDE/4.2/kdepim
........
r940772 | tmcguire | 2009-03-18 10:32:23 +0100 (Wed, 18 Mar 2009) | 20 lines
Backport r940441 by tmcguire from trunk to the 4.2 branch:
Don't produce garbage when forwarding a mail with a filter.
For this, I had to remove the code duplication in KMFilterActionForward::process()
and use KMMessage::createForward() instead.
This uncovered a bug in createForward (content type boundary was not saved) and a
small bug in the template parser.
Also, I added a KMMessage::dump() function to help with debugging mimelib
related troubles.
Not perfect yet, the way the template parser copies the message needs to be
improved, right now, when forwarding a signed message, the signed text gets
replaced with the template text, resulting in an invalid signature ...
CCBUG: 174942
........
r940775 | tmcguire | 2009-03-18 10:34:02 +0100 (Wed, 18 Mar 2009) | 11 lines
Backport r940516 by tmcguire from trunk to the 4.2 branch:
When replacing the text of a message after processing the template, don't try
to just replace the first text/plain part, this will cause trouble with signatures
and possibly many other situations.
Instead, delete the content of the old message, and set the processed
template text as the first body part.
If the old message had attachments, re-add them to the new message.
........
r940777 | tmcguire | 2009-03-18 10:36:48 +0100 (Wed, 18 Mar 2009) | 9 lines
Backport r940547 by tmcguire from trunk to the 4.2 branch:
Make valgrind happy.
This didn't find the cause of the problem I was searching for, though
(inline forwarding runs out of memory because allocating insanley huge
amounts of memory)
........
r940816 | tmcguire | 2009-03-18 11:52:19 +0100 (Wed, 18 Mar 2009) | 8 lines
Backport r940565 by tmcguire from trunk to the 4.2 branch:
Don't run out of memory when forwarding a mail with more than one
attachment.
The mimelib behavior here is very irritating...
........
svn path=/branches/kdepim/enterprise4/kdepim/; revision=940873
Make valgrind happy.
This didn't find the cause of the problem I was searching for, though
(inline forwarding runs out of memory because allocating insanley huge
amounts of memory)
svn path=/branches/KDE/4.2/kdepim/; revision=940777
Don't produce garbage when forwarding a mail with a filter.
For this, I had to remove the code duplication in KMFilterActionForward::process()
and use KMMessage::createForward() instead.
This uncovered a bug in createForward (content type boundary was not saved) and a
small bug in the template parser.
Also, I added a KMMessage::dump() function to help with debugging mimelib
related troubles.
Not perfect yet, the way the template parser copies the message needs to be
improved, right now, when forwarding a signed message, the signed text gets
replaced with the template text, resulting in an invalid signature ...
CCBUG: 174942
svn path=/branches/KDE/4.2/kdepim/; revision=940772
the template parser creates a new message anyway, so just copy the
original message and let the parser do the work.
Exception is text/calendar mails.
svn path=/trunk/KDE/kdepim/; revision=940552
This didn't find the cause of the problem I was searching for, though
(inline forwarding runs out of memory because allocating insanley huge
amounts of memory)
svn path=/trunk/KDE/kdepim/; revision=940547
For this, I had to remove the code duplication in KMFilterActionForward::process()
and use KMMessage::createForward() instead.
This uncovered a bug in createForward (content type boundary was not saved) and a
small bug in the template parser.
Also, I added a KMMessage::dump() function to help with debugging mimelib
related troubles.
Not perfect yet, the way the template parser copies the message needs to be
improved, right now, when forwarding a signed message, the signed text gets
replaced with the template text, resulting in an invalid signature ...
BUG: 174942
svn path=/trunk/KDE/kdepim/; revision=940441
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/KDE/4.1/kdepim
........
r904951 | tmcguire | 2009-01-03 16:15:59 +0100 (Sat, 03 Jan 2009) | 6 lines
Backport r901167 by tmcguire from trunk to the 4.1 branch:
Also detect messages with X-Priority == 2 as urgent.
Patch by Stefan Geuns <bitrain at gmail dot com>, thanks!
........
svn path=/branches/kdepim/enterprise4/kdepim/; revision=905012
Merged revisions 899972,900344 via svnmerge from
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/KDE/4.1/kdepim
........
r899972 | tmcguire | 2008-12-21 23:53:09 +0100 (Sun, 21 Dec 2008) | 16 lines
Backport r899776 by tmcguire from trunk to the 4.1 branch:
Fix inline-forwarding of messages with attachments.
The template parser copied the original message and then set the whole body to the template text.
This did not work for multipart messages, since setting the body there has no effect (and would
discard the attachment if it had any effect).
Therefore, we simply place the template text in the first text part we find in case of multipart
messages. This probably still breaks with HTML messages with attachments or with other complicated
nested structures. Oh well.
CCBUG: 178128
CCBUG: 146921
kolab/issue 3219
........
r900344 | tmcguire | 2008-12-22 17:20:28 +0100 (Mon, 22 Dec 2008) | 10 lines
Partial backport of 846650 by tmcguire:
- get rid of setBody(), it was not necessary anymore and caused the cursor
to be at the wrong place
=> the effect was that this broke inline forwarding of multipart/mixed messages,
since it set the composer's text to the complete body of the message, which included
MIME headers and the attachment. Now, we rely on the objectreeparser in the composer
to correctly get the text from the message.
........
svn path=/branches/kdepim/enterprise/kdepim/; revision=900587
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/KDE/4.1/kdepim
........
r899972 | tmcguire | 2008-12-21 23:53:09 +0100 (Sun, 21 Dec 2008) | 16 lines
Backport r899776 by tmcguire from trunk to the 4.1 branch:
Fix inline-forwarding of messages with attachments.
The template parser copied the original message and then set the whole body to the template text.
This did not work for multipart messages, since setting the body there has no effect (and would
discard the attachment if it had any effect).
Therefore, we simply place the template text in the first text part we find in case of multipart
messages. This probably still breaks with HTML messages with attachments or with other complicated
nested structures. Oh well.
CCBUG: 178128
CCBUG: 146921
kolab/issue 3219
........
r900344 | tmcguire | 2008-12-22 17:20:28 +0100 (Mon, 22 Dec 2008) | 10 lines
Partial backport of 846650 by tmcguire:
- get rid of setBody(), it was not necessary anymore and caused the cursor
to be at the wrong place
=> the effect was that this broke inline forwarding of multipart/mixed messages,
since it set the composer's text to the complete body of the message, which included
MIME headers and the attachment. Now, we rely on the objectreeparser in the composer
to correctly get the text from the message.
........
svn path=/branches/kdepim/enterprise4/kdepim/; revision=900353
The template parser copied the original message and then set the whole body to the template text.
This did not work for multipart messages, since setting the body there has no effect (and would
discard the attachment if it had any effect).
Therefore, we simply place the template text in the first text part we find in case of multipart
messages. This probably still breaks with HTML messages with attachments or with other complicated
nested structures. Oh well.
BUG: 178128
BUG: 146921
svn path=/trunk/KDE/kdepim/; revision=899776
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/KDE/4.1/kdepim
........
r890868 | tmcguire | 2008-11-30 16:08:44 +0100 (Sun, 30 Nov 2008) | 6 lines
Backport r884034 by winterz from trunk to the 4.1 branch:
fix crash when template quote indicator does not end with an underscore.
CCBUG: 174268
........
svn path=/branches/kdepim/enterprise4/kdepim/; revision=890941
- Fix quote prefix handling:
- Also highlight custom prefixes correctly.
- Make sure the prefix is saved in a header field so
that highlighting also works when re-opening a draft
Now a quote prefix consisting of the sender's initals is
highlighted correctly again.
svn path=/trunk/KDE/kdepim/; revision=859109
Sill some deprecated warnigns, I guess I'll fix them myself now.
--
Merged revisions 843622,850460,850469 via svnmerge from
svn+ssh://tmcguire@svn.kde.org/home/kde/branches/kdepim/enterprise4/kdepim
................
r843622 | ervin | 2008-08-07 15:09:51 +0200 (Thu, 07 Aug 2008) | 143 lines
New Enterprise header style, and show attachements as part of the
header.
Closes kolab/merge74 aka "Extended attachment view".
On a personal note I hope it'll make it upstream, those two smallish
features are simply gorgeous IMHO.
[SNIP]
................
r850460 | ervin | 2008-08-21 17:12:23 +0200 (Thu, 21 Aug 2008) | 10 lines
Some fixes as requested by Thomas:
1) Enterprise header's mail picture doesn't overlap subject anymore;
2) In fancy header mode the attachments button is now left aligned;
3) The list of attachment doesn't scroll anymore with the enterprise
header;
4) Enterprise header now plays well with dark color schemes and
encapsulated messages.
Fixing deprecated calls and coding style will follow.
................
r850469 | ervin | 2008-08-21 17:43:46 +0200 (Thu, 21 Aug 2008) | 3 lines
Addresses the style and indentation issues found by Thomas
(from revision 843622).
................
svn path=/trunk/KDE/kdepim/; revision=850566
encoded from the svn revision number. then include this file
only where absolutely necessary.
should reduce the number of recompiles that are triggered
simply by an svn revision.
svn path=/trunk/KDE/kdepim/; revision=837480
-use K_GLOBAL_STATIC for static data
-an attempt to fix possible crashes and "null messages appearing in folders like outbox"
Details:
M kmail/kmfolderindex_sqlite.cpp
M kmail/kmfolderindex.h
* executes "DELETE FROM messages WHERE id=.." for every for messages
with 0 serial number, especially for the outbox
* removed commented-out old impl.
M kmail/kmmessage.h
use K_GLOBAL_STATIC
M kmail/kmfoldermaildir.cpp
* minor change
M kmail/jobscheduler.cpp
M kmail/jobscheduler.h
* ~JobScheduler() use qDeleteAll() (and clear for sanity)
M kmail/networkaccount.cpp
* use K_GLOBAL_STATIC
* use value() to avoid inserting 0 values
* NetworkAccount::resetConnectionList(): use remove() instead of inserting 0
M kmail/kmmessage.cpp
* use K_GLOBAL_STATIC
M kmail/kmcommands.cpp
* KMMoveCommand::execute(): skip mesages with serial number == 0
-this avoids crashes when user tries to clean up a folder with "no subject" items
M kmail/kmmsgbase.cpp
* #ifdef code related to using_mmap when using sqlite mode
M kmail/networkaccount.h
* use QPointer<KIO::Slave> mSlave and bits for boolean members
svn path=/trunk/KDE/kdepim/; revision=809620
- Use QHostInfo instead of KNetwork
- Q3Accel->KAction
- many QAction->KAction, so the default shortcut appears correct
again in the configure shortcuts dialog
The port of Q3Accel to KAction has the side effect that the shortcut
to scroll the message up/down can now be configured.
CCBUG:96301
svn path=/trunk/KDE/kdepim/; revision=803951
the Kleo jobs don't support synchronous execution anymore in trunk a
helper class to emulate that using QEventLoop has been added.
This enables crypto support in the KMail reader again when building
without KLEO_BUILD_OLD_MAINWINDOW.
Merged revisions 731701-731715,731717-731719,731721-731734,731736-731867,731869-731870,731872-732022,732024-732421,732423-732439,732441-732793,732795-733047,733050-733088,733090-733100,733102-733148,733150-733158,733160-733463,733465 via svnmerge from
https://vkrause@svn.kde.org/home/kde/branches/kdepim/enterprise/kdepim
................
r731701 | vkrause | 2007-11-01 17:25:25 +0100 (Thu, 01 Nov 2007) | 80 lines
Merged revisions 702839,702841-702888,702890-702952,702954-703209,703211-703454,703456-703461,703463-703760,703762-703855,703857-703886,703888-703946,703948-703961,703963-704170,704172-704178,704180-704185,704187-704195,704197-704304,704306-705719,705721-706112,706114-706138,706140-706144,706146-708323,708325-708727,708729-709044,709046-709055,709057-709060,709062-712432,712434-713381,713383-713393,713395-713409,713411-714722,714724,714726-716314,716316-716323,716325-716364,716366-716466,716468-716478,716480-716901,716903-716904,716906-716925,716927-717219,717221-717248,717250-719540,719542-719544,719546-719570,719572-720093,720095-725398,725400-730723,730725-731690 via svnmerge from
https://vkrause@svn.kde.org/home/kde/branches/work/~vkrause/enterprise
........
r725440 | vkrause | 2007-10-15 14:10:48 +0200 (Mon, 15 Oct 2007) | 2 lines
Port okDecryptMIME() to Kleo::DecryptVerifyJob.
........
r726612 | vkrause | 2007-10-18 11:26:10 +0200 (Thu, 18 Oct 2007) | 2 lines
Backport some tests to ensure that everything works here as well.
........
r726622 | vkrause | 2007-10-18 11:58:40 +0200 (Thu, 18 Oct 2007) | 3 lines
Partial port of CryptPlug::checkMessageSignature(), still contains some
regressions.
........
r727046 | vkrause | 2007-10-19 16:29:41 +0200 (Fri, 19 Oct 2007) | 2 lines
Get rid of CryptPlugWrapper::SigStatusFlags and CryptPlug::SignatureMetaData.
........
r727053 | vkrause | 2007-10-19 16:44:06 +0200 (Fri, 19 Oct 2007) | 2 lines
Test case for opaque verify.
........
r728096 | vkrause | 2007-10-22 13:03:35 +0200 (Mon, 22 Oct 2007) | 2 lines
fix QByteArray/QCString usage
........
r728101 | vkrause | 2007-10-22 13:32:05 +0200 (Mon, 22 Oct 2007) | 3 lines
Yet another QByteArray vs. QCString fix, signatures are working again
now.
........
r728113 | vkrause | 2007-10-22 14:44:11 +0200 (Mon, 22 Oct 2007) | 2 lines
CryptPlugWrapper -> Kleo::CryptoBackend::Protcol
........
r728479 | vkrause | 2007-10-23 14:21:44 +0200 (Tue, 23 Oct 2007) | 2 lines
CryptPlugFactory -> Kleo::CryptoBackendFactory
........
r728487 | vkrause | 2007-10-23 14:58:33 +0200 (Tue, 23 Oct 2007) | 5 lines
- fix certificate link
- initStatus() is no longer needed
- use Kleo::CryptoBackend::Protocol::displayName() instead of the old
local implementation
........
r728867 | vkrause | 2007-10-24 16:25:12 +0200 (Wed, 24 Oct 2007) | 2 lines
Port CryptPlugWrapper::hasFeature() calls.
........
r728877 | vkrause | 2007-10-24 16:49:29 +0200 (Wed, 24 Oct 2007) | 3 lines
The concept of active plugins doesn't exit anymore in the public API,
not active backends are simply not available.
........
r729223 | vkrause | 2007-10-25 12:59:13 +0200 (Thu, 25 Oct 2007) | 3 lines
SigStatus_NUMERICAL_CODE is a CryptPlug state and doesn't seem to exist
in gpgme++.
........
r729270 | vkrause | 2007-10-25 15:24:58 +0200 (Thu, 25 Oct 2007) | 2 lines
Set isGoodSignature correctly.
........
r731317 | vkrause | 2007-10-31 12:31:52 +0100 (Wed, 31 Oct 2007) | 2 lines
Remove the remaining cryptplug usage, only dead code anyway.
........
r731387 | vkrause | 2007-10-31 16:39:17 +0100 (Wed, 31 Oct 2007) | 2 lines
Add DN parsing and reordering code from cryptplug.
........
r731690 | vkrause | 2007-11-01 16:41:21 +0100 (Thu, 01 Nov 2007) | 2 lines
disable debug code
........
................
r733465 | vkrause | 2007-11-06 12:53:02 +0100 (Tue, 06 Nov 2007) | 2 lines
Kleo provides that function already, no need to duplicate it here.
................
svn path=/trunk/KDE/kdepim/; revision=803023
Allen Winter
Thomas McGuire
Laurent Montel
Jaime Torres Amate
Nikolai Kosjar
Edwin Schepers
Volker Krause
Bertjan Broeksema
Kevin Ottens
Jarosaw Staniek