Age | Commit message (Collapse) | Author |
|
Change-Id: I9d15d8edcc6746c055336bf31e5b0c15cb971ffe
Reviewed-on: https://gerrit.libreoffice.org/17061
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
|
|
Both the document statistics- and state-manager have their
own modified flags. There is a cyclic dependency between the
the two in that updating the document's statistics also marks
the document as modified. Of course when a document is edited
the statistics modified flag is set to trigger an update.
To avoid a perpetual cycle, the statistics manager resets the
document's modified state to that before setting the new
statistics. However, this doesn't reset the statistics
modified flag, which was set when the document was modified
by setting the new statistics. Hence, the statistics thinks
there are modifications in the document when there isn't.
This patch is to make DocumentStateManager::ResetModified()
symmetrical to DocumentStateManager::SetModified() by
reseting the modified flag of the statistics manager.
The idle CPU drops to nil on unmodified documents after this.
However, for modified documents the statistics is recalculated
perpetually until the document is saved. This will need a
different patch to fix.
Change-Id: Ib5936bc1acbda34fcac6a00ec46eaa31e4363885
Reviewed-on: https://gerrit.libreoffice.org/16774
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I6d3c76c0aafe0ad77b597a2cb8331e6f8219e2d2
|
|
Change-Id: I6b9089daaa2bb7d0dfc1072296fc51401e5fc88a
|
|
Change-Id: If7c6c1c5000893fca4205162526b4b4035637c11
|
|
Change-Id: Ia8df87e2b6803ebe4ab8d3defa147d20ca8d7422
|
|
Change-Id: Iee933f042186326ef4d8cd5da0090348de7a1aae
Reviewed-on: https://gerrit.libreoffice.org/17018
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>
|
|
... of SwTableFormat to SwTableBoxFormat for the temporary SwTableBox.
Change-Id: I74118b1b028bd71561f9afe6acd1cc95991a3761
|
|
Change-Id: Ib668bb948a2cd557db7b8804f8a81800048f88bd
|
|
The simple color distance on RGB color space is not good enough.
It leads to a better result if we use the scheme of the primary
colors.
This method works well with MSO highlighting color palette, but
not neccessarily in general. In highlighting palette light and
dark variant of the same color type (e.g. blue and light blue) has
the exactly same scheme.
Change-Id: Ied08b4c388b8020326709d01d2de290afdd9d77b
|
|
In sw it was still returning a string item, although the
slot type was changed to boolean back in 2005 with
4cce2a2f08238e1e03055877aa89bc755ec06f59.
In sc the status update was simply missing.
Change-Id: I8a6dd157d1e401754b3a9a93dc83458dcdd4f089
|
|
...to avoid lots of loplugin:staticmethods warnings. Also enables DBG_ASSERT
etc. also for --enable-debug builds in addition to --enable-dbgutil builds.
Change-Id: Ib89ecd9ab8ce7abb2c64790ace248b31f9d2b64d
|
|
Change-Id: Iaf6fec2e19af2cf393bc547c3ba439010819238f
Reviewed-on: https://gerrit.libreoffice.org/16949
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
|
|
Currently many toolbar controllers are basically doing the
same thing - show some docked toolbar. So the idea here is
to have one generic controller that will handle all these
cases in a unified way. It will get the name of the toolbar
that it's supposed to show from the configuration, based on
the command it was registered for.
Right now it can handle both simple (i.e. with DROPDOWNONLY
bits) and split (i.e. with DROPDOWN bits) scenarios, where
for split kind it replaces the function (and the image) of
the main part of the button, with the last selected function,
for easy reusing. A button is considered to be a split one,
if the name of the initial default command was passed along
with the sub toolbar name, otherwise it will be a simple
button.
The core change is in framework/. Other parts are a usage
example, in form of converting the custom shapes buttons
to this new controller.
Change-Id: I087cc58c3db1889ca69a26546d4f00fe07e2a58d
Reviewed-on: https://gerrit.libreoffice.org/16967
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Maxim Monastirsky <momonasmon@gmail.com>
|
|
These implementation names of Calc, Draw, Impress, and Writer documents are the
most likely to be checked by existing third-party code, so revert the
corresponding parts of 3099c70b11c7e5b80fe4dbe3dc99171fb38c6fc2 "Fix various
XServiceInfo implementations" (and instead add fixups in CppunitTest_serives).
Change-Id: I1a760adbf4c89a1b4477d6e628abba8d2028ccfe
|
|
its not the reported crash, which has "gone away" which might
be a duplicate of tdf#90502
the switch only handled RES_TITLE_CHANGED and RES_DESCRIPTION_CHANGED so if its
anything else, e.g. OBJ_DYING, then don't attempt GetNoTextNode
Change-Id: I642beb66613481cbc7ee18647f0204a67d670a84
|
|
Change-Id: I667bb264f92024b72f230c2ddbba3887471345f2
|
|
Change-Id: Iec36c7e4f4fbc2ee2ee25d4d0c8488340ba7d8c4
Reviewed-on: https://gerrit.libreoffice.org/16968
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
|
|
and remove SvxBrushItem::SetDoneLink since the field it sets
is unused.wq
Change-Id: Ide95a295fa8004f1ddab5e560f01d36d36658a72
Reviewed-on: https://gerrit.libreoffice.org/16943
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
|
|
Change-Id: I0a166d1b58b23cac96ae27d446d6d8ed2442df8e
|
|
The most relevant signal member function appears to be connect(),
so let's create a wrapper function for that now, without the more
esoteric ordering features for now.
Move the signal member itself to a new pImpl.
The benefits are worth it: preprocessor input reduced by 2.8GB,
that's 9% of the total (excluding system headers which are not counted
because they don't generate dependencies).
Change-Id: I0aaeda51a5630a348bb12c81a83f67afbd508a14
|
|
It just includes a bunch of other boost headers; mostly we need
boost/noncopyable.hpp so include that directly.
This eliminates 831 MB(!) of boost/preprocessor/seq/fold_left.hpp
completely, which is the 2nd biggest header after ustring.hxx.
Change-Id: I3df55770adcb46e56f389af828e8ba80da2dc1f2
|
|
Change-Id: Ica62926486831f126cce54d651e230cee3360074
Reviewed-on: https://gerrit.libreoffice.org/16901
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>
|
|
Change-Id: Iedb296732c0819a1d6cdc1c59b3f1718e2cd6d38
|
|
Change-Id: Ife2351bc8f0a3ee261b525aeb4e8e254174a42a0
|
|
Change-Id: I603940c64a1580a73c4f5c59b403caecd0ac75a5
|
|
There isn't really a limit for the annotation count in the RTF/DOCX spec.
Thanks Miklos for hints on this.
Change-Id: Ib4493b518acb3cabe04312c37b5c6a9c6072cb7e
|
|
Change-Id: I1fbf5a355340e20de881a030bc75ebbfce960d4f
|
|
Change-Id: Ie84087ec0d456a37c6af3e36075d840b802aebf6
Reviewed-on: https://gerrit.libreoffice.org/16837
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
|
|
Less ugly or just differently ugly? Hard to tell...
Change-Id: I1265f07f39ebbc65acfcc30242bc7cd7d46207e7
Reviewed-on: https://gerrit.libreoffice.org/16821
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Björn Michaelsen <bjoern.michaelsen@canonical.com>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
|
|
Change-Id: I9953cd58e4e285f655f11afccc1d6ee535f85d7e
|
|
Change-Id: I3e91817c5113a391668ef0f35d6b8df4702f9715
|
|
Change-Id: I4924251babfcea9a87e7e688c904264dd80f9359
|
|
Change-Id: I020c7ce4d9aa93cb0292378a410cdb9f7788e920
Signed-off-by: Michael Stahl <mstahl@redhat.com>
|
|
Change-Id: I4e5bb81a14ee996bddc8585b5d350487f5f20d50
|
|
Change-Id: Icfff86678f6b9d85c0bc9bc77e450891d388cb2a
|
|
Change-Id: I7a75b1b9fc14d4f9d5a5e7a61dc116d62a28d237
|
|
Change-Id: Id839c650721efed3bc734da49ebd5a0074897cbe
|
|
Change-Id: Iae7355d136ded368543c3825875df54f691f5219
Reviewed-on: https://gerrit.libreoffice.org/16813
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>
|
|
Change-Id: I2908fc052d44a8bddf8ebb628f67ce346ac519c7
|
|
Change-Id: I0a35a010d3edc8caea30e16813c937f1fa4d3ac8
Reviewed-on: https://gerrit.libreoffice.org/16797
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>
|
|
Regression from commit beb4aa21d61f0d66392d596be86fb57b4b167239
(SwViewShell::ImplEndAction: avoid direct paint, 2015-06-29), the
problem was that the lcl_PaintTransparentFormControls() call performs
direct paint, so it should be called only in case we don't do an async
paint via invalidate.
As expected, this call is no longer needed for the original i#107365
bug document in case SwViewShell::Paint() will be called by the main
loop shortly.
Change-Id: Ia27c551ed838d610f972f20abbb4ae9f0e1241b6
|
|
Change-Id: Ib7ea6624ac6b112779b0e64b08805538b8d6afff
Reviewed-on: https://gerrit.libreoffice.org/16796
Reviewed-by: Samuel Mehrbrodt <s.mehrbrodt@gmail.com>
Tested-by: Samuel Mehrbrodt <s.mehrbrodt@gmail.com>
|
|
Change-Id: Ic136cce6abef44291b7236a6d709f0eee391f311
Reviewed-on: https://gerrit.libreoffice.org/16784
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
Tested-by: Noel Grandin <noelgrandin@gmail.com>
|
|
Regression from commit 1d9e837698baab795efee6cc8b0e289528d5cac6 (Use
GetBoundRect() instead of Frm() in layout., 2011-04-29), this causes
problems for double-buffered rendering and also seems to be pointless,
as just above we already added the rectangle to the repaint area.
Change-Id: I3dc67143e8332ab941a683414ee318dbef13de88
|
|
One one hand, a problem since commit
330b860205c7ba69dd6603f65324d0f89ad9cd5f (fdo#68787 DOCX import: handle
when w:separator is missing for footnotes, 2013-09-04) was that the type
attribute from <w:footnote w:type="separator"> resulted in two
ooxml:CT_FtnEdn_type tokens, ignoring too many paragraph ends for
footnotes, which resulted in missing paragraph style on footnotes.
On the other hand, fixing the first problem showed that it wasn't
correct that commit 9389cf78e304a5a99bcf1745b9388e14ac36281a (cp#1000018
RTF import: empty para at the end of footnote text got lost, 2013-11-15)
unconditionally removed the RemoveLastParagraph() call in
DomainMapper_Impl::PopFootOrEndnote(). It turns out that RTF and DOCX
have different semantics here, the footnote is always within a <p></p>
pair in DOCX, while in RTF a \par at the end of a
footnote means an empty paragraph. Fix that by conditionally restoring
the removed RemoveLastParagraph() call.
Change-Id: I33020ac761c94addfec8164a17863565e4453b07
|
|
is e.g. an SwDrawFrameFormat instead of an SwFlyFrameFormat during
CppunitTest_sw_uiwriter
Change-Id: I97511cf516f76b12e04da61ce6fd51f61ef83241
|
|
Change-Id: I9b9f198aa6da2f6734194207b4c4bdec3ee1d4c0
|
|
Change-Id: I1af1d6bc150c16a2c6b0fe788a41c8c18caee6c6
|
|
Change-Id: I1b510a6194282dfa4a9001d473127c5ebc8b44eb
Reviewed-on: https://gerrit.libreoffice.org/16731
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
|