Age | Commit message (Collapse) | Author | Files | Lines |
|
Change-Id: Ic07722c3e1f75d909ec8f123919a1898dfde05b0
|
|
Change-Id: Iebddffbdc66397b1f2937ed7320f7800c8bab4cc
Reviewed-on: https://gerrit.libreoffice.org/24817
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
|
|
(cherry picked from commit 958b9a7fbdd58fdce762021917155c58fbb90d18)
Change-Id: Ifbed5e534ba79d32b7188bb7fb7108338b6e124d
Reviewed-on: https://gerrit.libreoffice.org/24722
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
|
|
DrawingML: The useBgFill attribute specifies that the shape fill
should be set to that of the slide background.
Reviewed-on: https://gerrit.libreoffice.org/23039
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>
(cherry picked from commit f3d1ac75c4b7fa63022e54a9cbff46ba99535076)
Change-Id: I8b568e730f00326d51e7b604579f4ff990b26f8a
|
|
For tdf#99450, see https://bugs.documentfoundation.org/show_bug.cgi?id=99450#c6 for full details
tdf#99462 is also a consequence of this double insert.
See https://bugs.documentfoundation.org/show_bug.cgi?id=99462#c2
Change-Id: I474495457088b93e0e86ea2e504f61c383ba059d
Reviewed-on: https://gerrit.libreoffice.org/24327
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Julien Nabet <serval2412@yahoo.fr>
(cherry picked from commit 618e7622d08b20f6ea5f38144b61a187aced86af)
Reviewed-on: https://gerrit.libreoffice.org/24330
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>
(cherry picked from commit 2aa8be5936cab3fa2a0d895b414cd0dc9dc3f319)
|
|
In the tiled rendering case, the slides, no matter
if it is inserted or deleted, the part names always return
sequential names i.e. Slide 1, Slide 2, ..., Slide N.
However the client side needs to know what slides had been
deleted or inserted, so it is necessary to send the hash codes.
Change-Id: I0e9caeec660c3e42dd9f751bdce7690f9ad365a1
Reviewed-on: https://gerrit.libreoffice.org/24267
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
|
|
In case an image had text, then ShapeExport::WriteGraphicObjectShapePart()
wanted to write "only the text", but PowerPointShapeExport::WriteTextShape()
had no idea how to write an image, so at the end nothing was exported.
(cherry picked from commit fc70e4c4e192372f77511bc6ce2bc77b9c9539be)
Conflicts:
sd/qa/unit/export-tests.cxx
Change-Id: I6c1ad0b41d4c5dc260b952322fb8a59e7f175603
Reviewed-on: https://gerrit.libreoffice.org/24017
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Andras Timar <andras.timar@collabora.com>
(cherry picked from commit e8e761ed003f5e509fea8eebd47111b1ed6aa672)
|
|
Change-Id: I5dd1d37ea783a65f0c8f3ad9c3c9cabf72ae036d
|
|
Change-Id: I0968689630e10f838c075e86357eb36a9a220d0d
|
|
Otherwise the table is inserted in the top left corner.
Change-Id: Ia5d7099d4bace6f18cd7e8d3409427119240b756
|
|
The width was decent, except that on small paper (A6) it was
wider than the paper size.
The height was terrible - the minimum size possible. It was too small
for even a single line of text. That made it hard to grab the table edge
(instead grabbing a row edge) when trying to resize.
Now the height and width are limited to the page/view size,
and based on the number of rows created. One possible enhancement
would be to use the border width instead of the page width.
Change-Id: I4e106a7bc34f65886e3c80d308499242c0fdd7d8
Reviewed-on: https://gerrit.libreoffice.org/21171
Reviewed-by: Justin Luth <justin_luth@sil.org>
Tested-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
|
|
Just force the whole ValueSet to be formatted. Individual rectangles
are already being invalidated (when PREVIEW_CHANGED event comes), but
something naughty must be happening in ValueSet that those changes
don't show.
Change-Id: I4fd8af36e34c3a445d7b16b59e45256f016fd9df
Reviewed-on: https://gerrit.libreoffice.org/23115
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>
(cherry picked from commit 1b98d7a3c4f1de39226e44e5b27351d8bfbbe6c0)
Reviewed-on: https://gerrit.libreoffice.org/23116
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>
(cherry picked from commit 819166500c1b45e4241fdb47f2de6653e4495372)
Reviewed-on: https://gerrit.libreoffice.org/23253
(cherry picked from commit 71f4f7fc85fbc872cd480613e95530f122f530f8)
|
|
Change-Id: If0f93ab9414f9013aab580bafc67f77891dfd8cc
|
|
LOK_CALLBACK_PARTS_COUNT_CHANGED did not cover more complex scenarios, like
more pages deleted at the same time, etc.
Instead, we need to trigger the LOK_CALLBACK_DOCUMENT_SIZE_CHANGED, and the
client has to be smart enough to update the previews in a way that does not
consume too much resources - ie. it has to have a link between the part names
(which are unique) and the previews, and request only the missing ones.
Change-Id: I36ff5dc86f360e3c7bd2c55fae8aaa99e054b4eb
|
|
Change-Id: I0a1ed390462e068dc59e360b9b5dd4a798bac21c
Reviewed-on: https://gerrit.libreoffice.org/22746
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>
(cherry picked from commit a30f48acf6b54199de1cc726f2d3e3d996280c65)
|
|
Apparently these are allowed to be empty, which means "nothing".
(regression from db1d278dcc308c73eb5edebc20481c96e7f479d8)
(cherry picked from commit af57a81d0c28944b424649f024c28f444a1ab2d9)
Change-Id: I13f7998e0986b26c34929afd40b1b4f0fc9efdff
Reviewed-on: https://gerrit.libreoffice.org/22724
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
(cherry picked from commit 13ed8b7035a3a3128e246ad2f5b3be059c411c57)
|
|
Change-Id: I19987a73e277df5e9e5457b22214cd08ace19c4b
|
|
In the tiled rendering case, slide sorted view is not created.
This revert some portion commit 80d7c5859b9e7a834a915d7e8bbbe9bc2130108a
Change-Id: Ifc2b7535f36ca69268de3e462bdd50ade9ec3853
Reviewed-on: https://gerrit.libreoffice.org/22542
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Henry Castro <hcastro@collabora.com>
|
|
Change-Id: I3ac9b0fe4d5a91d16af559ee8e38bc7be97cc694
Reviewed-on: https://gerrit.libreoffice.org/22408
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Henry Castro <hcastro@collabora.com>
|
|
In the tiled rendering case, when a slide is deleted or inserted
the sorted slides are updated on client side.
However, when .uno:Undo and .uno:Redo actions are requested on
client side the sorted slides are required to update all
sorted slides.
So every time when .uno:InsertPage, .uno:DeletePage, .uno:Undo, .uno:Redo
actions are requested, it will notify verbose action (PageInserted, PageDeleted)
with index on client side to update the sorted slide index.
Change-Id: Iebda2aa11be13aea8fbb6d0cc50442805d7485e9
Reviewed-on: https://gerrit.libreoffice.org/22309
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Henry Castro <hcastro@collabora.com>
|
|
For some reason, maDisplayName is not initialized always. Use
GetDisplayName() that handles such a case gracefully.
Change-Id: Ib62d4f03dcd3b5571749b540e5af7f90299e2067
Reviewed-on: https://gerrit.libreoffice.org/22348
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Andras Timar <andras.timar@collabora.com>
(cherry picked from commit e95db1865c7a095cde5e9cb338394555767b5c65)
|
|
Change-Id: Ife006636295f62a9e35f71ab3b0089dabb0ba623
|
|
Change-Id: Ia3ee81ced4f74c0d029a478bd59eff44d72ef327
|
|
Conflicts:
sd/source/ui/view/Outliner.cxx
Change-Id: I07df8d0330390ac599aac364581aee1c9fd0f809
|
|
... in order to set the row's min height
Regression from 4f2c8194f485b1527fb4f4dfe23ce804937f1f9c
After this commit, the row's min height was set based only on
the cells containing text in the row, but the problem appeared
when the row didn't have any cell with text.
Change logic to check wether there's text in the cell and in the row.
Now, height in SdImportTest::testRowHeight() is 507 instead of
508 but I can't figure it out why. However, I believe there's
no harm in change the test from 508 to 507 as, visually speaking,
the difference can't be distinguish.
Change-Id: I0b3a14c34eaeaa8e77227860ca290fb79a0302ce
Reviewed-on: https://gerrit.libreoffice.org/21692
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: jan iversen <jani@documentfoundation.org>
(cherry picked from commit 13d4398820ded5914f635757865e258db2db2b57)
Reviewed-on: https://gerrit.libreoffice.org/22009
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>
(cherry picked from commit 7583d19a58f9aa0fed51c248c1773285e2cb39cf)
|
|
We immediately cancel all the dialogs that potentially come up when using
LibreOfficeKit; which means that when you tried to save a .docx to a remote
server (which triggered the 'alien format' warning), the save operation
couldn't be completed.
Change-Id: I6bb5eadac994c1f515d7a49299c21960b3491bbe
|
|
The DBG_ASSERT followed by the same test actually returning immediately
prevents the actual 'search' after a 'search all' being performed.
I hope this does not have any negative consequences; but if it does, we should
fix the root cause, instead of having this kind of defensive programming here.
Change-Id: I909533f301dda9c20cab1968de45e5fa7975e852
|
|
Revert "bnc#584721: Import subtitle block to master slides"
This reverts commit b3d50feaa87b670baf68288974005ac26ad31736.
It seems the imported subtitle block makes export corrupt the
output. Better to have a valid output.
(cherry picked from commit a9db17aabc3ac970e6e4754e28310d20138144d9)
Conflicts:
sd/qa/unit/import-tests.cxx
Change-Id: I1a3d4f03e4fa0c4431a5394495682a1b9d7677bc
Reviewed-on: https://gerrit.libreoffice.org/20921
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>
(cherry picked from commit e18c062be49042945644db438608a6e58c7dc98a)
|
|
Change-Id: Ic6e0ed45b118b27a86f8657671ae877ce494c5d4
|
|
So the flag will be reset on early return as well.
(cherry picked from commit 6372080cd396a7e8608806a35a1be68d288b8fda)
Conflicts:
sd/source/ui/view/Outliner.cxx
Change-Id: Ib086ec4a02dda291a0291a8ac40660c16de0fbf6
|
|
Given that the edit/outliner views can come and go, avoid the lifecycle
problems with just passing a pointer to the sdr model to editeng, and
then it'll always have the up to date "are we searching" information.
editeng can't depend on svx, so provide an interface class SdrModel can
implement.
(cherry picked from commit 7b5d20983dfbfb458898eeab54828ba5fef5841f)
Conflicts:
editeng/source/editeng/editview.cxx
editeng/source/editeng/impedit.cxx
include/editeng/outliner.hxx
sd/qa/unit/tiledrendering/tiledrendering.cxx
svx/source/svdraw/svdedxv.cxx
sw/inc/PostItMgr.hxx
sw/source/uibase/docvw/PostItMgr.cxx
sw/source/uibase/docvw/SidebarWin.cxx
Change-Id: I3b98011593b00ac0fab05b6b9c591dd20d94c579
|
|
But they do cause annoying flashing.
(cherry picked from commit b9565ef0a73c235cd1e14fce9031db6e9237c524)
Conflicts:
include/svx/svdmodel.hxx
svx/source/svdraw/svdmodel.cxx
Change-Id: Ic313a15429c5db98c5660a5274aa49e95dd217e5
|
|
Change-Id: I5a74dd60f61fc0fdb88f08383d579b14222aa55d
Reviewed-on: https://gerrit.libreoffice.org/21366
Reviewed-by: Marco Cecchetti <mrcekets@gmail.com>
Tested-by: Marco Cecchetti <mrcekets@gmail.com>
|
|
Support for text documents and presentations.
Added a command line option for gtktiledviewer:
--hide-page-shadow.
Reviewed on:
https://gerrit.libreoffice.org/21210
(cherry picked from commit 6b7d41094d06bbb4c248927d02318cf1b5faba0a)
Conflicts:
include/svx/svdpntv.hxx
libreofficekit/qa/gtktiledviewer/gtktiledviewer.cxx
svx/source/svdraw/svdpntv.cxx
Change-Id: I1e427693d7af40cb5731d1730ac5b7c486d45c29
|
|
Change-Id: I251d079d6362cf09d014623663cf9355a9c79a2e
(cherry picked from commit be343661b076f3f57e7eb9ca1a5a4d3529752b58)
|
|
warning: now table are skipped because current implementation is not
able to iterate through table cells and when a match occurs on the first
cell the text object iterator stops working;
Change-Id: I36ca4dabe88b0eb8fd89d3fe3dcc6951e5c03d0e
Reviewed-on: https://gerrit.libreoffice.org/21166
Reviewed-by: Jan Holesovsky <kendy@collabora.com>
Tested-by: Jan Holesovsky <kendy@collabora.com>
|
|
In the tiled rendering case the map mode is disabled,
and conversion pixel to logical is not computed.
Change-Id: I0d54aaa897f0fe0df6e8b55334053dc66b6ea925
Reviewed-on: https://gerrit.libreoffice.org/20912
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Henry Castro <hcastro@collabora.com>
|
|
When LOOL requires to insert a graphic, it uses UNO command ".uno:InsertGraphic",
and the corresponding URL file image argument.
However Impress, ignores URL file image argument, and by default
pop up a dialog, so no image is inserted in LOOL.
Change-Id: I31dd1b1536307ef2dbb6b3d49812b310560c919f
|
|
Change-Id: I4b7422eb77edd39a2c6bd427aa8c10d62b1e370d
|
|
Added a unit test for testing that text decorations are exported
correctly.
Change-Id: I2bd71974242a0007726fbdd5ef5637a9ec62fd47
Reviewed-on: https://gerrit.libreoffice.org/19870
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Andras Timar <andras.timar@collabora.com>
|
|
Change-Id: Iafeaecad612b724c4eeb85e0c01c942afb6445d8
|
|
1) Write endParaRPr so that PowerPoint display them properly.
2) Original design forbid properites with default value to be exprted, now fixed.
3) Automatic colors are written as white or black based on
whether background is dark.
Note that tdf#77881,tdf#80520,tdf#89525 depend on this.
Change-Id: I255c16f35149b738be2daf2800b1c90389f2c7cf
Reviewed-on: https://gerrit.libreoffice.org/17472
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Norbert Thiebaud <nthiebaud@gmail.com>
Reviewed-on: https://gerrit.libreoffice.org/19862
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>
Tested-by: Katarina Behrens <Katarina.Behrens@cib.de>
(cherry picked from commit 0b5e3885bc706ed02acac1c682c90171700b0cac)
|
|
(cherry picked from commit 479325dec83ea2747f3ce27ca7e817695b15e1bb)
Conflicts:
sc/inc/docuno.hxx
sd/source/ui/inc/unomodel.hxx
sw/inc/unotxdoc.hxx
Change-Id: I9a6a75457078dc6383673f4c1a2012b69b5cefdd
|
|
Change-Id: I9e48407f6f5eb5cc4c36439b72648bb8d43deb79
|
|
Reviewed-on: https://gerrit.libreoffice.org/19883
Reviewed-by: Andrzej Hunt <andrzej@ahunt.org>
Tested-by: Andrzej Hunt <andrzej@ahunt.org>
(cherry picked from commit 81b8ca683d44ba9c37f2dc8c74470a86ce70513f)
Conflicts:
sc/inc/docuno.hxx
sd/source/ui/inc/unomodel.hxx
sw/inc/unotxdoc.hxx
Change-Id: I8d1f63208baf277b0a9d15908f3ea7ff3b56bf10
|
|
This helps ensure that images are swapped in when we actually
render tiles. Previously we'd sometimes have placeholders
instead of the image, which results in either an invalidate
(+rerender of that tile) once the image is swapped in (for
normal tiles) or a permanently missing image in the preview tiles.
Change-Id: I1a16a913faf9fad20e40a5d1aad3de187038c7a2
Reviewed-on: https://gerrit.libreoffice.org/19890
Reviewed-by: Andrzej Hunt <andrzej@ahunt.org>
Tested-by: Andrzej Hunt <andrzej@ahunt.org>
|
|
Change-Id: Icdb7d1cc250640f428984b96b0a51ac950013cb2
(cherry picked from commit 0c39c456f47c14e2b7f86d1bc6857dc0e2ca6cb3)
Reviewed-on: https://gerrit.libreoffice.org/19814
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
(cherry picked from commit f318e585265fe5b863bd9927f5aad415ba060173)
|
|
the index variable m_FrameList used to reference the current frame uses
the state EMTY_FRAMELIST to mark as invalid, which is the max integer
number. Thus all usages have to be secured to not use m_FrameList
if set to EMPTY_FRAMELIST. This was missing in some places.
the max integer
number. Thus all usages have to be secured to not use m_FrameList
if set to EMPTY_FRAMELIST. This was missing in some places.
Reviewed-on: https://gerrit.libreoffice.org/19753
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>
Cherry-picked from f0cef70cd4164342b218fbee34bf57eedc22c998
Change-Id: Ie1cf4ae2cb79d821f634ec78b621a2647abc61a8
Reviewed-on: https://gerrit.libreoffice.org/19787
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
(cherry picked from commit 6b3b080f8cedc1b496022b18e477af0c7361fba3)
|
|
It's cleaner to let the apps do this themselves than exposing their
underlying vcl::Window.
(cherry picked from commit bfd79be417358822023691cf7b7b2946906100ca)
Conflicts:
sc/inc/docuno.hxx
sd/source/ui/inc/unomodel.hxx
sw/inc/unotxdoc.hxx
Change-Id: Iff2442dd325fa65a0cf3ad4aa7f918542dab1e4c
|
|
Don't disable "Insert table" option in case of Ole object
Change-Id: Ibe87d35db320f4a9d5c971e3f7b25199fcacbde4
Reviewed-on: https://gerrit.libreoffice.org/18897
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>
(cherry picked from commit 2135b4a88d755c0b79a3b93b2269f08b709220c5)
Reviewed-on: https://gerrit.libreoffice.org/19067
(cherry picked from commit 53b8f63d0a8a8d1f99e2f4bf20934cf52c2b6bcd)
|