kmail. As suggested by Ingo Kloecker and Nicolas Goutte.
Added missing catalogue "libkpgp" to knode.
svn path=/branches/KDE/3.5/kdepim/; revision=492217
Additionally, checking for config updates is sped up by only checking the application of new config updates instead of checking all config updates.
FEATURE: 79938
GUI:
svn path=/trunk/kdepim/; revision=372098
- jump to first new (old default)
- jump to first new or unread (old checkbox jump to unread)
- jump to last selected (new)
svn path=/trunk/kdepim/; revision=338393
either DontSendOnCheck, SendOnManualChecks or SendOnAllChecks
- use dropdown in the gui instead of checkbox
- port to KConfigXT
- add kmail-3.3-mish.sh update script which atm only updates the one key,
but can be used to collect key/value renames as we move the remaining
keys to KConfigXT and canonicalize (is that a word?) them
- hook in update script
CCMAIL: 79190-done@bugs.kde.org
svn path=/trunk/kdepim/; revision=320818
ae2-kconf_update:
This includes the kconf_update entries in kmstartup.cpp, kmail.upd (the scripts are
already committed, but not yet installed, this patch does the latter).
(this patch differs from the one posted to kmail-devel in that I added Ingo's
request for a separate update script for the enc/sign split)
ae2-configuredialog-security-page-rework:
This patch contains the configure dialog's Security Page rework.
This patch adds GUI config for:
- specifying whether or not to import attached certificates automatically.
This is currently only implemented for S/MIME (see later patches), but is
dead easy to implement for OpenPGP, too.
This sounds like more than it actually is, since gpgsm imports certs in
signatures anyway, to be able to check the sigs. This is just the way S/MIME
works. We still make it configurable, mainly b/c project management couldn't
establish a position on whether or not it should happen automatically :)
- Remove the GUI for enabling/disabling DSNs.
We never generated standards-conformant DSNs anyway; in addition to that,
this is an MTA-level feature. We simply lack the envelope data to create
them. Since we now have MDNs, which _are_ MUA-level, I'm on a quest to phase
out the KMail-pseudo-DSN support "bouncing". This is one part of it.
Why is this here? We needed the space occupied by that groupbox otherwise :)
- Add an option to suppress any MDN sending in case of encrypted messages.
We always sent an MDN, regardless of whether or not the decryption actually
succeeded, apparently some Win SW doesn't do it this way, which makes them
an Oracle. We never were one, but this is an option for the paranoid, of
which we have a few (HTML sidebar, attachment icons in the message list,
etc)
- Reorganise the Security Page:
(General,PGP,CryptPlug(Crypto Backends),individual CryptPlug config)
->(Reading,Composing,S/MIME Validation,Crypto Backends,
individual Crypto Backend configs)
svn path=/trunk/kdepim/; revision=316512
(better than <foo.h>, especially when systems could have an identity.h somewhere)
A kconfig_update script moves the identities from kmailrc to emailidentities
svn path=/trunk/kdepim/; revision=311347
- Make sure an idle cursor is used when the message box is displayed. This
wasn't the case if Kontact was started.
- Don't hardcode the app name, i.e. "KMail".
- Show different messages depending on whether the app that created the
lock file is the same or a different app than the app that's started
and on whether the instances are running on the same or on different
machines. Also don't mention the PID which would just unnecessarily
confuse normal users.
- Put meaningful text on the buttons instead of "Yes" and "No". This is
strongly recommended by the KDE GUI style guide.
- Finally, instead of just resetting the lock file and telling the user
to restart the app, simply start the app if the user tells us so.
svn path=/trunk/kdepim/; revision=304611
- drop QTimer::singleShot() call to newInstance() slot - singleShot() was never used and there is no "newInstance()" slot available anyway.
CCMAIL: 60753-done@bugs.kde.org
svn path=/trunk/kdepim/; revision=304357
#define kernel KMKernel::self()
to
#define kmkernel KMKernel::self()
because 'kernel' was a much to general term. We really shouldn't repeat the mistakes of the X developers.
I noticed this problem when I played around with KImageEffects. kimageeffects.h contains 'kernel' as parameter of some methods and so the compilation had to fail. We won't need KImageEffects in the near future, but at least we are now prepared and a clash with another 'kernel' can't happen anymore.
svn path=/trunk/kdepim/; revision=252621
has redraw problems with smart MTV mode in separate reader, though)
- Extract config profile loading from the Appearance page into a new
dialog: Better user experience and less special code in appearance
page to make sure ProfileTab::apply() is called first...
- Merge the groupware and folder config pages into a misc page
(the folder tag should probably be split into "folder" and
"behaviour")
- various small cleanups:
o s/QString()/QString::null/
o unused #includes in .h files
o added const'ness to tmps
o removed some tmps
o etc
svn path=/trunk/kdepim/; revision=242473
Aaron J. Seigo
Will Entriken
Volker Krause
Reinhold Kainhofer
Ingo Klcker
Chusslove Illich
Andreas Gungl
Laurent Montel
Stephan Johach
Luk Tinkl
Allen Winter
Carsten Burghardt
Till Adam
Marc Mutz
Michael Brade
Stephan Kulow
David Faure
Helge Deller