Instead of in PaintHelper::StartBufferedPaint(). If the buffer size is
set to match the size of a sub-widget (for which the paint was
triggered), then client using the buffer as a persistent vdev are unable
to paint at arbitrary locations.

For example, if we painted the ruler, then the blinking cursor won't be
able to paint to the SwEditWin area.

(cherry picked from commits 161f7533 and
43ac95ab)

Change-Id: Iba07070baafb5b802fc6da200696611afd2010d7
Reviewed-on: https://gerrit.libreoffice.org/17490Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>

Change-Id: Ie2d1cb7312de6f14a5c6de81eefd7a00be6f75c0
(cherry picked from commit 396643d4)
Reviewed-on: https://gerrit.libreoffice.org/17421Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: David Tardon <dtardon@redhat.com>

This was causing all these "libpython3.3m.so: file not recognized: File
truncated" failures.
(cherry picked from commit 57dd4d01)

RepositoryExternal.mk: for some reason Mac wants a GeneratedPackage
(cherry picked from commit d9162d24)

Change-Id: I5d983fee8d5cd313fbd0d6ece800fa8b80d81b35
Reviewed-on: https://gerrit.libreoffice.org/17398Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: David Ostrovsky <david@ostrovsky.org>
Reviewed-by: David Tardon <dtardon@redhat.com>

With scientific format, use the trick of thousand separator to detect
Engineering notation.

Change-Id: I7acdfb8bec646f9d6c1a17c334b60fa9708c8111
Reviewed-on: https://gerrit.libreoffice.org/17406Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>
(cherry picked from commit f4bdd798)
Reviewed-on: https://gerrit.libreoffice.org/17440

This partially brings back the behavior before
10a3db37.

Change-Id: I5b372ab56105c05dda6ecb9aa1eed1c6a0c72ea8
Reviewed-on: https://gerrit.libreoffice.org/17420Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>

Instead:

- create it early in Window::ImplInit(), except for the default window
  (that one is a chicken-and-egg problem, as creating a VirtualDevice
  invokes ImplGetDefaultWindow(), but creating the default window then
  would create a VirtualDevice)
- only erase the painted area of the buffer, don't create it
- use a separate bool in the frame state to track if we're in the middle
  of a (possibly recursive) buffered paint

And with that, rename CreateBuffer() to StartBufferedPaint(), as it does
not create the buffer anymore.

Change-Id: Ib33e8afa36977aa809f0ea2158a369f288cba8c7
(cherry picked from commit ea5e83ab)
Reviewed-on: https://gerrit.libreoffice.org/17445Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>

With this, vcl::Cursor does not paint on the vcl::Window directly, and
in case the output should be something other than pRenderContext, it
should be a one-liner change to do that.

(cherry picked from commit 79c9598d)

Conflicts:
	vcl/source/window/cursor.cxx

Change-Id: I57833f6f8589286260d96f412f1fcc681e6c5c8c
Reviewed-on: https://gerrit.libreoffice.org/17433Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>

Change-Id: Ibb7500527d0ff142a6ba89a20894258cb63eeed5
(cherry picked from commit 2d9db406)

Change-Id: Ia3931f3cac13894ccd51772312110367194e19c6
(cherry picked from commit 8552e928)

by reducing the width of the graphical element

Change-Id: Ie100629e2602f8f9a3f0e4e1ba353178f219c9a0
(cherry picked from commit 303b7758)

Revert the API change from e41c33b3
and use the existing css::util::XUpdatable instead.

Change-Id: I3eba4c7def98c8765a970d54a7fe84a320d4313e
Reviewed-on: https://gerrit.libreoffice.org/17382Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
(cherry picked from commit 41d1b01a)

With this, code that wants to take a persistent render context (e.g.
vcl::Cursor) will be able to do so. This commit just moves the buffer,
though: it's still created / deleted by PaintHelper.

Having it in ImplFrameData means that we'll have one buffer / one
system window: i.e. toplevel window and its sub-widgets share a buffer.

(cherry picked from commit a33e0379)

Conflicts:
	vcl/source/window/paint.cxx

Change-Id: Ic37220a2a483a7389aa04cd4313b6fc61a5408bf
Reviewed-on: https://gerrit.libreoffice.org/17414Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>

(cherry picked from commit 86d2cdda)

Change-Id: I0a876e9afa9a90f312d68611a8adf5a962da784d
Reviewed-on: https://gerrit.libreoffice.org/17413Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Miklos Vajna <vmiklos@collabora.co.uk>

With this, vcl::Cursor will be able to paint to a vcl::RenderContext,
too; not just directly, which is needed for double-buffering.

(cherry picked from commit 09bd5846)

Conflicts:
	include/vcl/window.hxx
	vcl/source/window/window2.cxx

Change-Id: I868f6cd9b08afe59611ef266911a894a26cacedc
Reviewed-on: https://gerrit.libreoffice.org/17401Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>

With this, the print preview window no longer performs direct paint.

(cherry picked from commit 46e7991b)

Conflicts:
	sw/inc/pagepreviewlayout.hxx

Change-Id: I53f05d8a3b723bc131f21a485e0ffec71484670a
Reviewed-on: https://gerrit.libreoffice.org/17394Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>

JunitTest_forms_unoapi_3 just crashed in DropTarget::disposing()
with m_pDropTarget pointing to garbage.

Change-Id: Icb5ad15a3d4b857b45553543d404e14abbac4374
(cherry picked from commit 372d4166)
Reviewed-on: https://gerrit.libreoffice.org/17388Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Miklos Vajna <vmiklos@collabora.co.uk>

With this, when the user select an other page in the print preview
window, the selection rectangle is no longer painted directly.

Change-Id: I85f1cb9d09c8545222ed49ff0acf819540d10920
(cherry picked from commit da24bd36)
Reviewed-on: https://gerrit.libreoffice.org/17381Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Jan Holesovsky <kendy@collabora.com>

With this, the print preview window does not paint the selection
rectangle on opening directly anymore.

(cherry picked from commit 0c114151)

Change-Id: Ie4d24dfc4d64c74b22dcd1476778f4eac6f4cd13
Reviewed-on: https://gerrit.libreoffice.org/17363Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Jan Holesovsky <kendy@collabora.com>

Change-Id: I6b7379323a86242b42a8a5137807c535bf1d7c46
(cherry picked from commit 6be72cc2)
Reviewed-on: https://gerrit.libreoffice.org/17374Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>

This method is used to get the sheet's name and to display it
in the tabs (sheet selector)
I've also added an unit test for it and uncommented a similar one

Change-Id: Ia866815c594a873812c71a6c86e303c869e1f093
Reviewed-on: https://gerrit.libreoffice.org/17294Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Jan Holesovsky <kendy@collabora.com>
(cherry picked from commit 50fe21c7)
Reviewed-on: https://gerrit.libreoffice.org/17368

Change-Id: I24868acec66a72abbb52b8026ed3a092dbd97632
Reviewed-on: https://gerrit.libreoffice.org/17347Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
(cherry picked from commit 11f24bc2)
Signed-off-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

Change-Id: I8e4983a91d4f97a2a20fbeed89d4e0f186c35fad
(cherry picked from commit f29ac1d9)
Reviewed-on: https://gerrit.libreoffice.org/17373Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

here we appear to check if the preceding optional entries are of certain
types. Presumably if they are optional then its legal for them not to
exist at all, just that if they are there they should be in the right
order.

Change-Id: Id147b91d7d375d7985d901cda112c7757dc3fb19
Reviewed-on: https://gerrit.libreoffice.org/17353Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>

Change-Id: I81a784ab506d0d3ae266d7341b8ff7a163dcd8d2
(cherry picked from commit 442b788b)
Reviewed-on: https://gerrit.libreoffice.org/17327Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>

Commit 0208ead7 (DOCX import: improve
btLr table cell support, 2013-02-22) made any table row that has at
least one btLr cell fixed height. This causes problems in case a table
has a minimal height with lots of content, where the fixed height gives
wrong layout, but the minimal height is correct.

Fix the problem by only making the row fixed height if <w:cantSplit/> is
set (as seen in the old bugdoc), and revert to setting the height type
to minimal in any other case.

Change-Id: Ibaf91f542e64e5caa7904df97eb6eb52618e0023
(cherry picked from commit 233a634a)
Reviewed-on: https://gerrit.libreoffice.org/17338Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Mork service wasn't registred on windows as it was generally
excluded on this platform not matter what architecture was used.

Pofile discovery was broken, as XP_WIN wasn't define and the
code fall back to use UNIX directories for profile discovery
on windows that obviously cannot work.

Change-Id: I823378a1a094a2172ba2cb3bf9bdacedb27b36a2
Reviewed-on: https://gerrit.libreoffice.org/17350Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>

Change-Id: I8bc9d5bd49366097bdabea0c234e688b15bdea42

Change-Id: I0d97689ec299dfb4e6400eee40d85b5abea1ab11
(cherry picked from commit fa8089e1)

Avoids crashing with empty context stacks.

Change-Id: I0ee8b457fdbb19b55f5c15876b7253680cde6e23
(cherry picked from commit a61fd02c)
Reviewed-on: https://gerrit.libreoffice.org/17334Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

The problem was that when inserting a post-it comment, the Paint() is
triggered for SwEditWin, which then shares a RenderContext with its
children, and at the end paints the buffer to the vcl::Window of
SwEditWin. When painting this buffer, the contents of SwAnnotationWin
was missing, as the parent SwEditWin was not allowed to paint on that
area. No explicit clipping is set at the OutputDevice level, but
vcl::Window::ImplClipChildren() still calculates a clip region based on
the child windows for the underlying SalGraphics, and that prevented
SwEditWin to paint its buffer fully.

Fix the problem by explicitly disabling this clipping for
SwAnnotationWin in its ctor in the rendercontext case.

(cherry picked from commit dcf94ab5)

Change-Id: I5eb47c20070057a79297d91242624fc2057e2dfb
Reviewed-on: https://gerrit.libreoffice.org/17329Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Commit c1f8437d (fdo#83465 RTF import:
handle font of numbering, 2014-09-25), changed the "get the me character
style of the current numbering's current level" member function to be
successfull even in case we're inside a DOCX run, not when we're inside
a DOCX paragraph, but outside runs.

While this is necessary for RTF, the side effect of this was that
unwanted run properties started to affect the above mentioned character
style in case of DOCX.

Fix the problem by enabling the "in paragraph and run" looking for RTF
only.

Change-Id: I610bfce6cec15b918fe547402360f5a894401f7e
(cherry picked from commit fc7c1a07)
Reviewed-on: https://gerrit.libreoffice.org/17323Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

bah, didn't mean to push directly without review

This reverts commit c19e079b.

which fills the toplevel window, rather than directly to the toplevel window.

It makes no difference for X, but for wayland the window decorations are part
of the toplevel window, dropping down a level means we don't draw out menu bar
under the window decoration space

Change-Id: Icec400efacd16b5d901107c13b6fa90c59cad0e6
(cherry picked from commit 298c089d)

about-dialog & windows installer imgs
use non TDF-tagline and add Development build variant for splashscreen
DS_Store for Mac dnd installer not adjusted yet

Change-Id: I4ff449d9564214a80f88c752e54064b599e9948b
(cherry picked from commit 8cfdd81b)

as opposed to implicitly keeping whatever unrelated string was there before.

Change-Id: Ifaf1b41e951e97f209ecb617b32ec4f7522b1d08
Reviewed-on: https://gerrit.libreoffice.org/17297Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

In particular when changing row.

1) Teach DbCellControl about "SelectedItems" as known value property.

2) Fix DbListBox::updateFromModel to actually use the SelectedItems
   it reads from the model, as opposed to throwing it away.

Change-Id: I7074c13b3d271bf2362aa059378aa857682a040b
Reviewed-on: https://gerrit.libreoffice.org/17298Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

...if you call ResetPoolDefaultItem() first. Crash found by Varun Dhall.

Change-Id: I409484c172fb5843270aee2425844076a008b4df
(cherry picked from commit bcb1f816)
Reviewed-on: https://gerrit.libreoffice.org/17320Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>

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.

(cherry picked from commit 16f9cf57)

filter: filter_utils test needs tools library (from Michael Stahl)

(cherry picked from commit 75d0f726)

Missing boost_headers (from Stephan Bergmann)

(cherry picked from commit bd8e8db2)

Change-Id: I0de6e4598f365fda1cccd1d3466429e52cf4067f
Reviewed-on: https://gerrit.libreoffice.org/17052Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>

... instead of using an arbitray first data record's string as column
name. We don't support header-less tables properly yet, so don't pretend
to.

Squashed 4 commits into one:

add optional ContainsHeader property

great we never had this :-(

(cherry picked from commit 6dddd1aa)

handle ContainsHeader property at database range

(cherry picked from commit ed497c01)

clear HasHeader at database range if table headerRowCount is 0

(cherry picked from commit 22c9977d)

TableRef: generate error for header-less column references, tdf#91278 related

... instead of using an arbitray first data record's string as column
name. We don't support header-less tables properly yet, so don't pretend
to.

(cherry picked from commit d7794792)

Change-Id: Id79b08ff9dfe42228ed7d6b27ad0c8cc29b1bfb0
f0d54ab1dee6c861b973dc490f6c4a1e11260546
1cf5b0ac3884320f39d439b6eecf0b39cdf6bc49
a42619ec800291b6617a61c8a89a2d54ef231cec
Reviewed-on: https://gerrit.libreoffice.org/17321Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Change-Id: I4c95f52f0b22ab574f608b93c172e0398e81974b
(cherry picked from commit 57ec66e2)
Reviewed-on: https://gerrit.libreoffice.org/17316Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>
Tested-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>