Age | Commit message (Collapse) | Author |
|
Change-Id: I8018d9b5fa01d1720c0392dc5fdc4a0656f25a35
|
|
Change-Id: I6c145e984c885c7e06caa1c27bfb354ea49ad9ce
|
|
Change-Id: Ic9338867a3bf9315f20fa02d7fdb1adda2db285a
|
|
Change-Id: Ice06e639213aeb6f7f23cbf4634947dd25613db1
|
|
...passing -session= into soffice was no longer supported.
The way I tested this with Gnome 3.4 (Fedora 17) is with:
* Run gnome-session-properties to select "Options - Automatically remember
running applications when logging out"
* Create ~/.local/share/applications/lodev.desktop:
[Desktop Entry]
Exec=.../core/solver/unxlngx6/installation/opt/program/soffice
Icon=libreoffice-startcenter
Name=LOdev
Type=Application
* Run "startx /usr/bin/gnome-session -- /usr/bin/Xephyr :1"
Change-Id: Id8a6e431e4c355caafa6c7111a435ebc0d35f78a
|
|
All of those defines (should) not be used at any point in the code.
False positives may exist, because some macro names are getting
generated by some template/macro foo. This is the version which
compiled fine for me.
Also there are many commented out lines removed.
Change-Id: I6394024682e4ab3691eb72707a9363d41ba31df8
|
|
...(i.e., after logging out of a session with soffice still running, but only
with the backing window still open). The key issue was that bSessionDataExists
would always be true if /org.openoffice.Office.Recovery/RecoveryInfo/SessionData
is true, even if /org.openoffice.Office.Recovery/RecoveryList is empty. Always
calling doRestore (unless bExistsRecoveryData, but even if !bExistsSessionData)
is there to ensure that /org.openoffice.Office.Recovery/RecoveryInfo/SessionData
is eventually reset (whether or not that is really necessary).
Change-Id: I9c3fff23f8019069f36e1dd5ded634ea1917bbd4
|
|
...as the signal can potentially be delivered to the to-be-joined thread. (It
is unclear to me whether the SalMainPipeExchangeSingal_impl handler is really
needed anyway, as there are various other places that should ensure that
DisableOfficeIPCThread is called during shutdown. At least in theory, no longer
joining here can cause the thread to incur crashes etc. when it it still running
during shutdown.)
Change-Id: I677a2e31e1a58d2b8d91634c2eb5274d63ac15f4
|
|
Change-Id: Ic2e712f4447b733b79d980e178d9d6d9d8bf0e40
|
|
Change-Id: Id437c428f07cd084b5f54151a806e111d557acaa
|
|
These RID definitions aren't used at any other point. So remove
them. Unused START and END definitions were commented out for
documentation purpose.
All definitions in basctl/source/basicide/tbxctl.hrc were unused,
so I removed this file.
Change-Id: I4713188a12b9da912b2a91b85ea4e54ca1aeb994
|
|
More than two lines are removed for readability.
Change-Id: Ibff6cf68d7c512e240a54065b54a225bb23a782b
|
|
OOo after crash"
This reverts commit f8e1d2438199c4a76ceeb36352678c415e25272c, which erroneously
disabled automatic restart on Windows completely and has now been addressed in a
better way with 150bf2c95c193b30e050ac1c54d64b5485af088d "bnc#524250: Show
backing component after cancelled recovery."
|
|
...instead of an empty Writer document, which is non-intuitive when the user had
worked on something other than a Writer document before the crash.
f8e1d2438199c4a76ceeb36352678c415e25272c "win32-dont-attempt-restart.diff: Don't
attempt to restart OOo after crash" had previously attempted to fix this with a
sledgehammer approach of not restarting at all after a crash on Windows, under
the wrong assumption that there would be no restart after crash at all on Linux
either (which was apparently rather a bug caused by the newly introduced
oosplash at that time).
Change-Id: If65d2e21640d56262c486d887c4a29137cabcd0a
|
|
Update code to use factory method URLTransformer::create
Change-Id: I3fd2e838497bcfd8fc949615c0e7d60a6ea47118
Signed-off-by: Stephan Bergmann <sbergman@redhat.com>, added some tweaks.
|
|
Update calls to factories to use new SimpleFileAccess::create method
Change-Id: Ie5b0696fe2228a9033b19969245a53c21a61aa14
Signed-off-by: Stephan Bergmann <sbergman@redhat.com>, added some tweaks.
|
|
Update calls to factories to use new ::create methods
Change-Id: I01d4417820f52718836c92faf3c2fae0dc96b30d
Signed-off-by: Stephan Bergmann <sbergman@redhat.com>, added some tweaks.
|
|
|
|
|
|
|
|
Change-Id: Ia9fe54a5198811d648bb32a2ebb3ab7f9cde77c7
|
|
Change-Id: Ic00cdfce4172af0a2f0aa1aa33ef5e386d407976
|
|
When LIBO_HEADLESS is set we are not supposed to run libo without
--headless; doing so will throw an UNO exception. So better use
a sensible default.
|
|
Remove hack in sofficemain we really don't want and don't install
a lib we are not building.
|
|
Change-Id: I572a8a6dc9da4f63b7c937748b3013bab60bb6bb
|
|
|
|
See the TODO comment in sfx2/inc/sfx2/docfile.hxx about a temporary SfxMedium
ctor overload to find not yet adapted uses of another overload, to be removed
again in due time.
Change-Id: Ie22c33c32f8870ce6ebf6d500abc7a4e33d97183
|
|
|
|
|
|
Change-Id: Ib60a2a4f0d28a53d997731eb34b118cc9b9f822d
|
|
|
|
Change-Id: If02be864f958de96ae1c54fe575b1828410390b1
|
|
Change-Id: I5a254459a491b9547530d8e312260dceed21f25c
|
|
Change-Id: I6f468005c8d8d99d9251a9c4fe4629b98bc4aa5e
|
|
Change-Id: I5a097beaaeb30214ed6071bba1297e551291922b
|
|
Variables should have module name as prefix to prevent collisions.
|
|
Okay, I give up. This obviously still does not work on Windows, but it
does on MinGW when I try to simulate it.
Change-Id: I9f2d7114df0498d5cc3a71431aacc7e49a5a78fd
|
|
Change-Id: I57ed1971414789e6efbe76e69bb3fabb8b51e22d
|
|
...which happens when installing non-bundle extensions (like a plain .xcu
file), where the path computed for the description.xml contains the path to
the non-bundle extension file itself as a prefix, so access results in
E_NOTDIR.
Change-Id: Ic0f4a34b46d99d1817ceea599bb8b53dbd65bc49
|
|
Change-Id: Idc7b869a0a3755df7fb10105ce07dca3d076dd43
|
|
* Support for dead arguments has been removed.
** Especially, dead UCB_CONFIGURATION_KEY1/2_xxx values have also been removed,
and code initializing UCB has been simplified accordingly.
* For the sake of simplicity, behaviour has been changed slightly. For example,
passing both --calc --writer now always lets --writer win (whereas in the past
the one mentioned first, i.e., --calc in this example, would win).
Change-Id: Id40a19b9ef96db0d439cadd96ba491bf640d28fb
|
|
Change-Id: I72b41e4826b7e93b5c8aeacbf6d9d52e3780f090
|
|
Change-Id: I756c0a19bea7b1cc0e290d9f382a04d655819bfb
|
|
this removes dmake completely out of the build for migrated modules
build.pl now assumes modules to be gbuild, unless there is a
prj/dmake file
Change-Id: I674a036b182ee13c5ec093e83cb3d38133112d3b
|
|
|
|
|
|
|
|
|
|
|
|
...and some further clean up of needlessly complex class CommandLineArgs.
|