Change-Id: Ieaf349ee43737304f51362bfe01db636e3eaf629

Change-Id: I90f8e6c4de383b05345b8df9118000167bd20d2f
Reviewed-on: https://gerrit.libreoffice.org/14735Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Julien Nabet <serval2412@yahoo.fr>

Change-Id: I9e9a00acf9503980f0c6c7cd0a20378e5ca7390d

we need that method soon in another place

Change-Id: Ic5b20acd666bd4d5183be7acc9d9ebe2c0a810f7

Change-Id: I74da9d277d21e01ffb316e6885470a1a3cbe0cf2

Change-Id: Ie77a988956927475ea384c36664ca7c3dd00b2a7

Change-Id: I3a1aaafe30f13f2eb388f7db3d78ecafaf0a4a2d

Change-Id: Ia2a50922fdfe07ff452bcc78b973d5d4943a128c

Change-Id: Idecead242c71c89cddb166cce36fa41f912298d0

This reverts commit 104a1e64.

A number sequence has to ignore all string and empty elements by
definition, see tdf#89872 for an example regression, further affected
are all functions that use ScInterpreter::GetSortArray() which in turn
uses ScInterpreter::GetNumberSequenceArray()

Change-Id: Iba139ede7bd72e23c0b7a28a8a4ff38ea816725a

Change-Id: I65f0ec335d4d0008af002857f48f4a6765d44dfd

Change-Id: Ie91b3fc3090906b321f7b0d3073de129627c0407

Change-Id: Ia525130e8c378b784b7ee8439153daa1f2a43283

... note that the LibreOfficeKit bits aren't built for OS X. I don't
recall if there is some fundamental reason why not. Was it so that for
OS X, like for iOS, a "virtual device" should be created with a
CGContext and not a pixmap buffer?

Change-Id: I6e1fedb89e7fe409baf2ea1ac7f3b65b0f1c776b

Change-Id: Id71236acb322e57f3e5af406874de4924c7524b9

gcc 4.6 reports: sorry, unimplemented: non-static data member
initializers

Moves the initialization into the constructors initializer list.

Change-Id: Id25ac042aa3207135edda2485d3c5fe8f98652af

Change-Id: I500008b32e5af07702b76afb901a3ec270453462
Reviewed-on: https://gerrit.libreoffice.org/14770Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Andras Timar <andras.timar@collabora.com>

There is no standard for this so some ATs will expect "level" and
others "heading-level"; we will expose both.

Change-Id: Ia935f7efdea11dc9be66fd9a54e1df0432c74662

1) One more place in desktop that missed the LOK_TEST check.

2) One testcase currently fails on baseline, disabled for now, need to
check later.

Change-Id: I8c8bcf2cec1618e39e8453120755dffd3feb79ce

Change-Id: Ied315d27cd0b5dbe640a2a3c87f5dc6c3dc11bf0

Change-Id: Ic662d71face370457c122946c8dbcb56d6f2bfe7

Change-Id: I105f8811f810b393cac8518b42d77f8bd4d0c763

Change-Id: Id668dc509dc3fa02e9a248716050c16a8cf91419

Do all overlay changes and state tracking in InvalidationHandler
where we get invalidation and selection / cursor messages from
LibreOfficeKit.

Change-Id: I87ccdefedc01e8bd54666b694a9a429a1fc6af76

Change-Id: Ic4ceaadcf251f7257001f8043161ce15898d7239

Change-Id: I5ad3c0f853efe1cf314f3c1afd77cecfdfc59134

Change-Id: Iaebf3e5a2f0c11efa7110b7062ecfc3ae38e0e64

Change-Id: Ifaa6c0de7d9b91706985667b72195bcfc2d610cd

One has to click & press enter to be able to type into the cell.  The typing
is visible only in the top left tile, but even in the other tiles, the text
gets there - when you manage to invalidate everything, the text appears.

Change-Id: I7c9c0a52949a514eb3de7a7fe64f11597377d39f

In principle, we should avoid messing up with zoom, and instead just work with
the MapMode, but we are not there yet - lots of places are hardcoded to work
in pixels.

Change-Id: I572b0d54fbfc72494c44ef95e7fda5e655f83fde

Change-Id: Ibd7e9e398fe24ec2b3553c8488b46b65de316da6

Change-Id: Ic05fbba4f431d9ccbe09377ec02e8eb84e3d3b17

From now on, the code for the tiled rendering is shared with the desktop
rendering, modulo few isTiledRendering() calls.

Drawing of the shapes & charts needs fixing, it does not honor the
output device settings.

Change-Id: I74cdb4e09da59aa71f31b18130829de28a93fab4

The drawing part is planned to be shared with the tiled rendering, while the
setup part has to be different.

Change-Id: I9101111d44f4602cdb92916ff3889b52bf10a8bf

Change-Id: Iaf7c31db65ecc4945f9d19e9031632443fa3823f

Can't access the SdrModel from sdr::overlay::OverlayObjectList
otherwise.

Change-Id: I0530985fb79659bd9de95d944b06edfb883f4403

Change-Id: Ie28cd768519fbdc305f98e1d764d05bd209951ca

This should fix the problem that LOK_CALLBACK_CURSOR_VISIBLE is emitted
2 times during every mouse click (hide, then show), while the motivation
behind this callback is to just hide the blinking cursor for image
selection, i.e. it's never emitted during normal text editing.

Change-Id: Id2a2b1102589f8151f640af3fcb50b646d261275