Change-Id: I7a1653424edd93b2de0597ce8a94aa5f4fa85622
Reviewed-on: https://gerrit.libreoffice.org/20258Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>

Change-Id: Iaf8e7a46be9c92a7cd11136cb903cc4b9ffe54f5

Change-Id: I9deb28a3a5420a43f8604a56145c8274bb4cc8d9

Change-Id: I00bde6564be6fb7c7f6d2ccb1faa7dd94e414635

Change-Id: I5c0800c353b46ea1fe359a893bdd8215f70cd809

Change-Id: I5191a48d3cadc8b9e4c1cc5f4ea99adb0071002f

Change-Id: I3a07361bb66a62533475618c9b50f184d791a799

Change-Id: I6928dec92655e4655af6c519405712892bf7d870

Previously the text was too long, ending on second page,
and table was on third page here, but is expected on second.
Change font to 'Liberation Serif' which seems to be used elsewhere,
and set line spacing to 'single' which makes the test succeed.

Change-Id: I4fc903c5d83308608e70e3e440d5277abd487052

Change-Id: Id1453e3cf1c8aef22536b4e0797cca7e654316c0

Change-Id: Iffc0c9b2215c85760856ab876fa6e04613cb4200
Reviewed-on: https://gerrit.libreoffice.org/20662Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>

Change-Id: Ibcffe9d8b85d552f53246a3ac751000673705ab3

Design choices:
Sseparator = comma except if decimal point is comma in locale setting, then semi-colon
Character set = Unicode (UFT-8)

Change-Id: I11eb88bf44cedaddfb52d3570f427d9e0d26b4cc

adds detection of css::table::BorderLineStyle::NONE
to the LineListBox and stops painting NONE style lines

Change-Id: Ib24fa25a9e3e3ae685f2000685f01cae42d1653f
Reviewed-on: https://gerrit.libreoffice.org/20639Reviewed-by: Oliver Specht <oliver.specht@cib.de>
Tested-by: Oliver Specht <oliver.specht@cib.de>

Change-Id: I040dd62737237ad7e016201a798ee42e27853556
Reviewed-on: https://gerrit.libreoffice.org/20656Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>

Project: dictionaries  38ddd6c1bd743353e76496b48e8ab6ddb295f5f8

Add Guarani thesaurus (gug-PY)

Add Guarani thesaurus to gug locale

Warning: The .dat file is not perfectly conformant to the LO
build for thesaurus;

The dat and idx file were generated by external program and were
moved directly into the extension

The file Dictionary-gug.mk has lines commented to show where to
restore the "normal" build once the dat file is fixed.

Change-Id: Ia39ec69f7289e28965cde21135f7e63fcb32ee66
Reviewed-on: https://gerrit.libreoffice.org/20275Reviewed-by: Olivier Hallot <ohallot@collabora.co.uk>
Tested-by: Olivier Hallot <ohallot@collabora.co.uk>

Replace some -ldl by $(DLOPEN_LIBS)
Add some FREEBSD tests

Change-Id: Ic30fe29b938d877694db2927775f89de15c02d87
Reviewed-on: https://gerrit.libreoffice.org/20652Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Matúš Kukan <matus@libreoffice.org>

Change-Id: If723dd72479ab8d072581e9ffbf7a1462c5fecda

Change-Id: Ib8d0e3d240e693c678d575b4791f69ca760919ec

this bug was caused by commit
2aacf6c2
"sc: boost::ptr_vector->std::vector"

Since the code in question likes passing around a pointer
to the element of the vector, convert to using
std::vector<std::unique_ptr>

Change-Id: I9e9676fe7c2dc32e23ba6708aaea1f16c1bf2ff8

Change-Id: I0103ca0da3a39e48416617aab4372cde8be0bbab
Reviewed-on: https://gerrit.libreoffice.org/20654Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Martin Hosken <martin_hosken@sil.org>

Change-Id: I5f35e0aeb6cef5920891eaa338c1d9b29e0ce554

Change-Id: I7141510fbda1158511698f73c48274008f0a41dc

Change-Id: I5a8daa1c0f2b62dbf56fa9fc1d14555f0ff613e2

Change-Id: Iada331a12232fd83b6813e6db65bd011d3746142

Change-Id: I9e9372ea9df6b959493445901b70a9ee6d3c0dac

Change-Id: Ib3acd5b853283dce2cdcbe72d7992637368fdf6b

Change-Id: Ib6f55bc0ce6c09ea523174cb630c6d7cd9808f96

Other code is out of legacy mode now, so can't legacy as a proxy.

Change-Id: Ie3807a3af680b707f2f08d058db955bc9cae6c2b
Reviewed-on: https://gerrit.libreoffice.org/20647Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>

(in an attempt to cause tinderboxes to rebuild)

Change-Id: I5cc94a988303b7b1ff85ec09c3d4f88d300f73b0

To selectively use the sticky mechanism, i.e. not in an actual move of a
range during cut&paste or drag&drop. Actually currently only in
ScTokenArray::AdjustReferenceOnShift()

Change-Id: I223e5013e4b820efb4ecb9e8ce8aa04a8bc01e27

Change-Id: I639638cb9cc387178183760137c0510802b8c6d9

...which contained the bulk of .cxx files that should logically go into
Library_pdfimport.  It was likely (only) used so that check_targets
CppunitTest_sdext_pdfimport, Executable_pdf2xml, and Executable_pdfunip could
access the library's internals without exporting them.  For the CppunitTest, use
the standard gb_CppunitTest_use_library_objects hack instead.  For the two
Executables, make that _use_library_objects hack available for Executables, too.
(It is a bit unclear whether those two Executables are really needed, they are
only referenced from the dead dmake-based
sdext/source/pdfimport/test/testdocs/makefile.mk and from vcl/README,
respectively; but just keep them alive for now.)

Change-Id: Ia2478508de216678be7a2302aba0c48f80de9d91
Reviewed-on: https://gerrit.libreoffice.org/20645Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

Change-Id: I962544005b5d408f7c044a02eefe09b87d8a81fe

When importing SdrObjCustomShape for each an Outliner and a
VirtualDevice as RefDevice are created and held. On 32bit systems this
breaks the import with many such objects (in the bugdoc around 4300). On
64bit it works, but more memory as necessary is used. To avoid this,
flush the UNO API implementations at these objects after one page is
imported.  Do not do this for each single object, that will break the
connector import which does some processing later.

Change-Id: I7296edd40f01c1a40258808d78bf39e6d4fe29bf
Reviewed-on: https://gerrit.libreoffice.org/19990Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
Tested-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>

... and fix the breakages it finds.

Change-Id: Ibc7289cc0cd7fb5648d686bd55afff9016f58b3b
Reviewed-on: https://gerrit.libreoffice.org/20638Reviewed-by: Björn Michaelsen <bjoern.michaelsen@canonical.com>
Tested-by: Jenkins <ci@libreoffice.org>

...introduced in b9d1006e "re-base on ALv2
code," probably due to a merge conflict or something like that.

Change-Id: Ib38cd24e935f951482ba99f483a419a9a7782ae8

error: assuming signed overflow does not occur when assuming that
(X + c) < X is always false [-Werror=strict-overflow]
     if ( rAddress.Row() < Row() )

Change-Id: Ia643491657680fffcaacae4a474af479b9a07e35

...to prevent errors like 5ac6e002 "Memory leak"

Change-Id: I3e20393af628849d8a387b491b75e1aacdea982a

...even if no correpsonding xPopupMenu proxy has been set up for some i (because
that sub-menu has never been opened by the user).  Otherwise, the Menu that
m_pVCLMenu->pItemList->maItemList[i]->m_pSubMenu points to will be deleted
(during a recursive xSubMenuManager->dispose call), and when later for some
sub-menu j > i that /has/ been opened by the user SetPopupMenu is called, that
can lead to

> Menu::GetPopupMenu(unsigned short) const at vcl/source/window/menu.cxx:835
> Menu::GetAccessible() at vcl/source/window/menu.cxx:1400
> OAccessibleMenuItemComponent::getAccessibleParent() at accessibility/source/standard/accessiblemenuitemcomponent.cxx:339
> non-virtual thunk to OAccessibleMenuItemComponent::getAccessibleParent() at accessibility/source/standard/accessiblemenuitemcomponent.cxx:335
> atk_object_wrapper_new(com::sun::star::uno::Reference<com::sun::star::accessibility::XAccessible> const&, AtkObject*) at vcl/unx/gtk3/a11y/../../gtk/a11y/atkwrapper.cxx:839
> atk_object_wrapper_ref(com::sun::star::uno::Reference<com::sun::star::accessibility::XAccessible> const&, bool) at vcl/unx/gtk3/a11y/../../gtk/a11y/atkwrapper.cxx:795
> AtkListener::handleChildAdded(com::sun::star::uno::Reference<com::sun::star::accessibility::XAccessibleContext> const&, com::sun::star::uno::Reference<com::sun::star::accessibility::XAccessible> const&) at vcl/unx/gtk3/a11y/../../gtk/a11y/atklistener.cxx:157
> AtkListener::notifyEvent(com::sun::star::accessibility::AccessibleEventObject const&) at vcl/unx/gtk3/a11y/../../gtk/a11y/atklistener.cxx:296
> comphelper::AccessibleEventNotifier::addEvent(unsigned int, com::sun::star::accessibility::AccessibleEventObject const&) at comphelper/source/misc/accessibleeventnotifier.cxx:281
> comphelper::OAccessibleContextHelper::NotifyAccessibleEvent(short, com::sun::star::uno::Any const&, com::sun::star::uno::Any const&) at comphelper/source/misc/accessiblecontexthelper.cxx:188
> OAccessibleMenuBaseComponent::InsertChild(int) at accessibility/source/standard/accessiblemenubasecomponent.cxx:436
> OAccessibleMenuBaseComponent::ProcessMenuEvent(VclMenuEvent const&) at accessibility/source/standard/accessiblemenubasecomponent.cxx:624
> OAccessibleMenuBaseComponent::MenuEventListener(VclMenuEvent&) at accessibility/source/standard/accessiblemenubasecomponent.cxx:572
> OAccessibleMenuBaseComponent::LinkStubMenuEventListener(void*, VclMenuEvent&) at accessibility/source/standard/accessiblemenubasecomponent.cxx:569
> Link<VclMenuEvent&, void>::Call(VclMenuEvent&) const at include/tools/link.hxx:84
> Menu::ImplCallEventListeners(unsigned long, unsigned short) at vcl/source/window/menu.cxx:355
> Menu::SetPopupMenu(unsigned short, PopupMenu*) at vcl/source/window/menu.cxx:825
> framework::MenuBarManager::RemoveListener() at framework/source/uielement/menubarmanager.cxx:556

looking at the dangling pSubMenu pointer (and even if the dangling pointer is
never dereferenced, UBSan will try to determine whether the static_cast in

  return static_cast<PopupMenu*>(pData->pSubMenu);

in Menu::GetPopupMenu, vcl/source/window/menu.cxx, is fine and cause a crash).

Change-Id: I5e5f6dba266580e3c2e7693877ea79a07d80bb74