Change-Id: I29be97c1a0e783abcf24720a7c5bc6c832e78eb3

Now the default for using the SVG tiny profile is false.
Now the default for use native text decoration is true.

Fixed the following issues:

1) Text decoration (underline, strike-out) was not exported.

2) Commit 0a4e9138 (tdf#56467: improve export of formulas to SVG) had a
side effect: all exported text portions have position attributes (x,y)
even if they are on the same line (y_old == y_new): this caused that
underlined text belonging to 2 different `tspan` element was not
underlined by a continuous line but there was a gap where a `tspan`
ended and the next `tspan` started.

3) In commit (tdf#37650: further improvement of svg export)
StartTextShape method is invoked even when positioned characters are
used but the `implWriteText` method always adds a `text` element by
itself so there is no need to call StartTextShape.
The changes for non-positioned characters have been embedded inside the
`WriteTextPortion` method.

Change-Id: Iee0d6a6816dcbd2d97f2fbdf969f012de678604e
Reviewed-on: https://gerrit.libreoffice.org/19672Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Jan Holesovsky <kendy@collabora.com>

Change-Id: I2e2ebc5a12794e9641a6195f07fa4b2d1a7369a9

This is used to know the current font/back color

Change-Id: Ic524fcb5c26e1c15ec9c50b2879465152ac8ee34

Don't disable "Insert table" option in case of Ole object

Change-Id: Ibe87d35db320f4a9d5c971e3f7b25199fcacbde4
Reviewed-on: https://gerrit.libreoffice.org/18897Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>
(cherry picked from commit 2135b4a8)
Reviewed-on: https://gerrit.libreoffice.org/19067
(cherry picked from commit 53b8f63d)

But you can use SAL_OVERRIDE instead.

Change-Id: Id593aa5f734b291043dee58f0e5ac6ff8ee421ea
Reviewed-on: https://gerrit.libreoffice.org/19653Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
(cherry picked from commit fc3044ce)

Change-Id: Id7ee86412288e96fda016acaeb965a53963408e0
(cherry picked from commit b4326c93)

...which uses the logic already available in VCL's ImplImageTree to locate the
image zip files and find fallbacks for incomplete themes and for localized
images.

(cherry picked from commit 6948c546, plus
f5174c89 "...but do not erase localize
attributes from image elements," which is needed to make localized image
references work again)
Conflicts:
	vcl/osx/vclnsapp.mm

Change-Id: Ic1c15fcacb6596a27a2b051093232902202bf472
Reviewed-on: https://gerrit.libreoffice.org/19644Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>
(cherry picked from commit 74453dbe)

Accept new UBlockCode enum values.
Just make it compile, without adding resource strings and blocks.
Excerpt of 5b7e8efb

Change-Id: Ia14e4c1c85f8be3e7bae0eb6f5bec1ae33f1a0bf
Reviewed-on: https://gerrit.libreoffice.org/19632Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
(cherry picked from commit 4df5049c)

Options specified in PPD files can be grouped using the
keywords "OpenGroup" and "CloseGroup".
The keyword "InstallableOptions" is used as a group name
for a group containing options that define
optional hardware features of the printer that can
be present or not (s. section 5.4 in version 4.3 of
the PPD specification).

As they are not print job specific, it is recommended
not to show them in the print dialog.

To be able to distinguish those options, the
PPD group name was added as an attribute to the PPDKey
class.

Reviewed-on: https://gerrit.libreoffice.org/19623Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>
(cherry picked from commit f8d0dc09)

Conflicts:
	include/vcl/ppdparser.hxx

Change-Id: I4a3abf23a711ad98556c0b608a07ef0a91e77e2b
Reviewed-on: https://gerrit.libreoffice.org/19630Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
Tested-by: Jan-Marek Glogowski <glogow@fbihome.de>
(cherry picked from commit 2f03a169)

Change-Id: If58fdc2eaf1b7b2148e3db2ebba6f8c0f2e9c92b
(cherry picked from commit 32d2fb8e)
(cherry picked from commit e783fe7d)

Change-Id: Id61e37e4e9ea62c3dba224e55ddfdc71fb0bfc8c
Reviewed-on: https://gerrit.libreoffice.org/19542Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Tested-by: Stephan Bergmann <sbergman@redhat.com>
(cherry picked from commit a3a61471)

The problem is that the precompiled headers' dependency files are not
run through concat-deps, hence they directly refer to headers of
external libraries' headers, which are not targets in the build system;
therefore re-building an external library does not cause the dependent
PCH to re-build and (at least with MSVC) the object files don't depend
on headers included via PCH anyway, so we get the recent link failure
in comphelper with MSVC, which wasn't rebuilt for icu's ABI change.

To fix that just use concat-deps, which re-writes header dependencies to
UnpackedTarball target dependencies.

Change-Id: Ic7555822925aaa1ff09b29bb73801fb83923bfab
(cherry picked from commit 475cf23e)
Reviewed-on: https://gerrit.libreoffice.org/19550Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
(cherry picked from commit 942da061)

This should fix build failures due to file locking preventing concurrent
access to cli_ure/source/cliuno.snk.

Change-Id: Iefd16ed83a01523b3612844c2f777516dc4e44c6
(cherry picked from commit 01518d3a)
Reviewed-on: https://gerrit.libreoffice.org/19483Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
(cherry picked from commit d0d03f22)

Change-Id: I583fc112333df6af3898b03f632780538b5f046f
Reviewed-on: https://gerrit.libreoffice.org/19469Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
(cherry picked from commit f5cf8524)
Reviewed-on: https://gerrit.libreoffice.org/19572Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
(cherry picked from commit ce260308)

regression from

commit 8f705df1
Author: Tomaž Vajngerl <tomaz.vajngerl@collabora.com>
Date:   Fri Jul 18 13:36:13 2014 +0200

    bnc#881024 Don't world transform font size in WMF/EMF import

which seems to be a regression from

commit 1281c1f5
Author: Armin Le Grand <alg@apache.org>
Date:   Mon Dec 10 16:08:58 2012 +0000

    Resolves: #i121382# Corrected size handling including font size handling...

    for EMF/WMF imports

Change-Id: I87307c9993d3343b4db636257e2567b10a73b79d
(cherry picked from commit 4eb70290)
Reviewed-on: https://gerrit.libreoffice.org/19559Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
(cherry picked from commit 8965f2b4)

this is an intermittent crash in the real world but under event testing we can
see that the PaintView is informed that the page has been deleted, but doesn't
clear its use of the page if it so happens that there was earlier modification
events that caused the fire-modification-timer to be launched

If there happens to be no such events then the page is correctly cleared

If the timer is active, then the page is not cleared, but eventually the
timer will fire and access the deleted page and fall over and die

minimal change for backporting ease, follow ups now

(cherry picked from commit f8b97599)

Change-Id: I090ad62cc175545902574c93dec7082225d97ab7
Reviewed-on: https://gerrit.libreoffice.org/19551Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Miklos Vajna <vmiklos@collabora.co.uk>
(cherry picked from commit d6a6261f)

Somehow, accessible text can lose focus and so when
attempting to set the cursor, the string we were just
working with may no longer be available, so verify that
the cursor position is still valid before setting it.

If the text can just disappear like that, then potentially
we are still forcing the cursor into the wrong place
(since the string is different than expected).  However
we can at least make sure the program doesn't
crash/exit without saving...

Change-Id: I8dcaf4625d187cb75008bdc0af0e3a546090a018
Reviewed-on: https://gerrit.libreoffice.org/19414Reviewed-by: Justin Luth <justin_luth@sil.org>
Tested-by: Justin Luth <justin_luth@sil.org>
(cherry picked from commit d9cf4744)
Reviewed-on: https://gerrit.libreoffice.org/19418Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Miklos Vajna <vmiklos@collabora.co.uk>
(cherry picked from commit 9f008e2a)

Change-Id: I8a86447bdd227621ff7b0b3551798ee343379c74
Reviewed-on: https://gerrit.libreoffice.org/19437Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-by: Norbert Thiebaud <nthiebaud@gmail.com>
Tested-by: Norbert Thiebaud <nthiebaud@gmail.com>
(cherry picked from commit b392270d)

Change-Id: Ifa833d3adfce5dcc247a8c6e2f972c0638e5c05c

dynamic atexit destructor and const static variables reportedly
cause crash, let's avoid this.

Change-Id: I24234084f810854606dde9914bee1c98ebc2e3dd

Change-Id: Ie6de4342b0e6f76076717f1fbdb18adf453ce963

flat_logo.svg: transparency, bigger log, bigger text
logo.png: new, replacing green number 5

Change-Id: I314d28f33a148ff438dae8de4db7b08097318fe0

Change-Id: I9e79c07aa3cbb70076d9bab07294062df4be02c6

Change-Id: I71d49b20a7ce96bbbbe3d130a8e8230c7afc0351

Since the test would fail to link on Windows, due to

ifeq ($(GUIBASE),unx)
$(eval $(call gb_Library_add_exception_objects,sofficeapp,\
    desktop/source/lib/init \
))
endif

in Library_sofficeapp.mk. Given that
CppunitTest_libreofficekit_tiledrendering is marked as Linux-only as
well, be consistent and have this test as Linux-only, too.

Change-Id: I6c8884398eba5dcf8a74c9cdc96c869b6108fb7e
(cherry picked from commit ac882c30)

Change-Id: Icd6df15347c48a5e42860092f4ee664e3a3d5699
(cherry picked from commit c791bef5)

Change-Id: Ida5028969782be792b32b952d3adba0c30dd8bae
(cherry picked from commit faa316e6)

(cherry picked from commit 5b4c29b1)

Conflicts:
	sc/inc/docuno.hxx

Change-Id: I0b9de068ddf0f4ff92d8fbf003b7529516f1f80a

(cherry picked from commit b08546eb)

Conflicts:
	sd/source/ui/inc/unomodel.hxx

Change-Id: I528ac9f9f687d2940c6477b1d33264f1e523051f

(cherry picked from commit 7efbbe98)

Conflicts:
	sw/inc/unotxdoc.hxx

Change-Id: Ib59ea43700815c53cdd4be819e2e9cf35c6f89e9

Change-Id: I393ec4427674cd5d77d0e9a069ffe159d14e38b1
(cherry picked from commit e531f846)

Also, closeDoc() is called by tearDown(), so no need to call it manually
at the end of tests.

Change-Id: Ib8f61a09fa3fc0885d7ea90ce96210bff4cc9f98
(cherry picked from commit d491ae56)

(cherry picked from commit 4c6db362)

Conflicts:
	sc/inc/docuno.hxx

Change-Id: If50b4b5baea36e161675afd368fc54bdec01d9a5

(cherry picked from commit 8522948b)

Conflicts:
	sd/source/ui/inc/unomodel.hxx

Change-Id: I8bc7316d9304d9e764ee846fe3af34599bf6fc35

Change-Id: I34998229e7f5cac4c62c859861783be3c161f9bf
(cherry picked from commit 6552767a)

(cherry picked from commit 981a9748)

Conflicts:
	sw/inc/unotxdoc.hxx

Change-Id: I9d0fad3904e74b44b0b126974ace4025f7a4fc5b

Change-Id: I385f64b7d5015c9a34f34a436b0ee2ce6b3a83d3
(cherry picked from commit 98cdb563)

The Calc spell checker idle handler is currently running in the
REPAINT / MEDIUM priority class - much too high.

This races with the Calc GUI repaint for large documents, like
the one attached to tdf#94485.

So move it to the LOWER priority queue, to fix the lagging GUI.

Change-Id: I6e4b05dd1a977182ec13f7e3a05f0722f502f226
(cherry picked from commit 0db68ef9)
Reviewed-on: https://gerrit.libreoffice.org/19556Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>
(cherry picked from commit be4e0ce8)

MDETERM uses lcl_LUP_decompose() which as the name suggests does LUP decomposition.
This patch allows additive cancellations to occur in a cleaner way while doing the *only*
additive operation in LUP decomposition, that is while computing Shur complement.
This patch does not change the high level semantics of the algorithm.
Also note that this change makes improvement only in the case where matrix elements entered
by the user are *integers*. The change allows MDETERM to evaluate to exact 0.0 for
singular integer matrices.

The steps to calculate Shur complement are :

for i = k+1 to n
  aik = aik / akk;
  for j = k+1 to n
    aij = aij - akj*aik

This is now modified as :

for i = k+1 to n
  for j = k+1 to n
    aij = ( aij*akk - akj*aik ) / akk

Without this change MDETERM() for certain singular matrices used to evaluate to a tiny non zero value,
which also caused MINVERSE() to generate a wrong output.
An example of such a matrix is :
1 2 3
4 5 6
7 8 9

Change-Id: Idd4211ddceab1b758fd05bfd57f7eecd5d4fd1a0
Reviewed-on: https://gerrit.libreoffice.org/19534Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>
(cherry picked from commit e1e73a97)
Reviewed-on: https://gerrit.libreoffice.org/19578
(cherry picked from commit 26b38bab)