Change-Id: I4346b6d79b46bccb5b79e27744c3cf80aa88fc9a
Reviewed-on: https://gerrit.libreoffice.org/26344Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>

See
http://crashreport.libreoffice.org/stats/crash_details/caa7efba-fbe8-4c58-966d-c4bc356b5a63

Change-Id: Idc8b3a2e482a5423080735f46f50d415835f747d
Reviewed-on: https://gerrit.libreoffice.org/26341Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

http://crashreport.libreoffice.org/stats/crash_details/d7bc6c8a-e8c2-4c1f-8291-ccab84ea892c

Change-Id: I76f9614bc54988c001c9b252707e4b5d4c588112
Reviewed-on: https://gerrit.libreoffice.org/26339Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Jenkins <ci@libreoffice.org>

so just disable the controls when in master view

Change-Id: I9551cc5c0d123e58555d248b78859c1f77de6690
Reviewed-on: https://gerrit.libreoffice.org/26276Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>

Commit c48df7ad added a isDisposed()
call in SwMailDispatcherListener_Impl::idle() and i can't see why that
would not be necessary in the other functions that access
m_pSendMailDialog if it is necessary in idle() - it's not obvious *why*
it would be necessary in idle() in the first place but obviously i don't
understand when SwSendMailDialog::dispose() is invoked (which is what
deletes m_pStatus, dereferencing which the backtrace crashes).

There is code in SwSendMailDialog::dispose() to handle the case that the
xMailDispatcher thread is still running, and stop it without joining,
which would obviously trigger this crash if it were ever executed.

(Maybe i should actually try what happens at runtime, but these
mail-merge dialogs are all scarily confusing.)

Change-Id: I550f6107b064b0c97f3d33bed5bd3830fa2e86f4

A new static member getPreferredConcurrency added to
comphelper::ThreadPool to return a configurable max
number of threads.

By default the new function returns the hardware_concurrency
value provided by std::thread. When MAX_CONCURRENCY envar is
defined, the return value is limited to whatever is set there.

Three call-sites that used std::thread::hardware_concurrency
have been replaced with getPreferredConcurrency.

Unittests added to cover the functionality of the new member.

Unittests are capped to 4 threads.

Change-Id: I3332e393a88a5ed436316fa712ed920a4b37f4af
Reviewed-on: https://gerrit.libreoffice.org/26254Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>

just to keep them together and to drop the need for a static

Change-Id: I55c784f9fc075f4542542ee2ec649eff32d708ec

By removing unnecessary variables/strings/function calls,
and making proper changes in the related .ui files.

Change-Id: I52e3f2a16519aee13b06f0efc6beb3385bcb0f1c
Reviewed-on: https://gerrit.libreoffice.org/26066Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>

drop hint of previous window to get mouse wheel event
when that window is disposed in order to drop any
references to it immediately that happens to avoid
anything else lingering too late

move the VclPtr into ImplSVData alongside the rest
of the things like this so we can remove the over-the-top
DeleteOnDeinit which itself replaced a relatively harmless
static Window*

Change-Id: I1e172071b711b6e4ded9a813ee3de730d3dfdf38

... so an already existing previous error is kept.

Change-Id: Ie731f17e93afa512542cc3ba2378031e978f9462

See crash reports at
http://crashreport.libreoffice.org/stats/signature/do_msvcr_magic+0x7
and documentation at
https://msdn.microsoft.com/en-us/library/zs6wbdhx.aspx

Change-Id: Ia9166d3b9fa10b87585821504e39cdfecbd22eda
Reviewed-on: https://gerrit.libreoffice.org/26317Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

This can be reproduced by switching to master slide sidebar in impress
, use the mouse to wheel up and down and then go straight to clicking
the window manager close button

The dispose doesn't ultimately drop its reference to the document
so keeps it open if the scroll-wheel previous event widget vclptr
keeps the widget around

Change-Id: Ic90bcadf2489094358737be7f522129bd555a1ba

Change-Id: Idedb9192938a01ecfda3dd93e69c16a896801fd7

Change-Id: I1e88e9fa6361c6f1f2aebebc101d44bc8e974283

... as GetDouble() returns 0.0 for any string and we don't want to
change that, most relevant places already check for numeric/text
beforehand.

Change-Id: Ifbc04e892f6f504040026042faa38674ced880fb

... to fire desired UNO commands from a dialog. Helpful when you
need to check some arbitrary UNO commands not supported via UI
yet.

Change-Id: I55df75ef235f5eb6922c50619610caf1c88241fb

Not all buttons are toggle buttons.

Change-Id: I6cddc619f3f760e447ea3ffea07f776e03cdaab9

This fixes CppunitTest_sw_tiledrendering in the LOK_VIEW_CALLBACK=1
case.

Change-Id: Ib79e57908c3edda0154341baba8279ede4281a42
Reviewed-on: https://gerrit.libreoffice.org/26313Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>

This was triggered by

This appears to be triggered by 08cf2fd0

commit 08cf2fd0
Author: Thorsten Behrens <Thorsten.Behrens@CIB.de>
Date:   Fri May 20 16:48:00 2016 +0200

which changed the order that things registered through
registerDispatchProviderInterceptor are used by, so swap the order of
registerDispatchProviderInterceptor calls here to sync with that

Change-Id: I047e4c7f6cb488c646df717e22c8ac91864c3938

For empty hyperlinks the EndURL() is called immediately after StartURL()
Due to the way the various buffers are written, the group closing braces
are written before the groups are opened, which is rather invalid.

Using the m_aRun buffer instead of m_aStyles appears to fix the problem.

(regression from b8907bf3)

Change-Id: I6910e1afa0ee262ae0496cf1d3aa83ae3e537ad0

... if it is missing, similar to what SfxObjectShell::DoLoad() does.

Change-Id: I5b0ae1f892355a5e9786d590c821656b58d29cf2

Parsed the same test files using XFastParser and built string, we
asssert that both the strings built from XParser and XFastParser
are identical.

Change-Id: Ie1e0dc6b676ca32dd10d75c3cb272b5c8b795ab7

Menu Sequence:
   Open Template...
   Save as Template...
   ----------------------
   Manage Templates

Change-Id: I0d0c8ca15d8dc38662e7f84a4f598933c13dd512
Reviewed-on: https://gerrit.libreoffice.org/26179Reviewed-by: Akshay Deep <akshaydeepiitr@gmail.com>
Tested-by: Akshay Deep <akshaydeepiitr@gmail.com>

These caushed an assertion failure when opening a Calc or Impress
document in gtktiledviewer.

Change-Id: If9cf1ef6c5a9d8e1b0d578b20dd3f513989b669b
Reviewed-on: https://gerrit.libreoffice.org/26311Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>

Change-Id: Ifd5df74063bcaf2f3a198b090ed2126449d7b8f8

Change-Id: I19470c7eeb961665cb5248861b030479a0796a24
Reviewed-on: https://gerrit.libreoffice.org/26290Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
Tested-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>

Change-Id: I9a2ed653bcb6afe803b781fc686c02f817d84e75
Reviewed-on: https://gerrit.libreoffice.org/26303Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>

bffvalidator (Microsoft Office Binary File Format Validator) basic
support in configure for later usage in export tests. It should be
installed from
https://www.microsoft.com/en-us/download/details.aspx?id=26794
and enabled in configure with new switch
--with-bffvalidator=<optional path to validator>

Change-Id: I6faa08d3277f0328b3dfa310222a11d115e5aa15
Reviewed-on: https://gerrit.libreoffice.org/25870Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
Tested-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>

Triggered by:

LOK_VIEW_CALLBACK=1 make -sr CppunitTest_libreofficekit_tiledrendering

The problem was that on shutdown SfxViewFrame::Current() returns 0 and
then SfxLokHelper::getView() should not be called (as documented by an
assert() in it).

Change-Id: If0475ca591408d5b041c1cbde5b553d0fb8dd056
Reviewed-on: https://gerrit.libreoffice.org/26300Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>

Change-Id: Iebddffbdc66397b1f2937ed7320f7800c8bab4cc
Reviewed-on: https://gerrit.libreoffice.org/26301Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>

Change-Id: I6a79214d24452dc2813d22d60f97eead220a7c5b

Change-Id: Ia321d2f03bf77eba87b1135799b3cb0409f23ebf
Reviewed-on: https://gerrit.libreoffice.org/26293Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>

... introduced with Excel 2010.

Change-Id: Ide47fea0e8bc109dc47f591c33ec1d1a13c08d72

Unused ever since it was introduced in 5d7ce6db
"INTEGRATION: CWS swmailmerge", and access to any SwMiscConfig members has
always been done indirectly via SW_MOD()->GetModuleConfig()->..., so looks like
aMiscConfig is unnecessary (and Oliver agrees with me on that).

Change-Id: I98b1962a43ba969cdadf1e42fdb1d3522ae4bc35

Change-Id: If4d7b51de338c026b3308a733c8f4a41478158b2

Change-Id: Id08a999764bfeeff3bb3409b46f48b278009ccbe
Reviewed-on: https://gerrit.libreoffice.org/26238Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

Replaces the defensive programming band-aid of
5c1a1d1c with a real fix.

The problem is that lcl_NonCopyCount() has some special case code
to ignore the first node in the target document, which erroneously is
executed for every bookmark, which results in the 2 bookmarks in the
bugdoc being created with nDelCount 1 and 2 so they land on the same
node, which is not allowed for cross-reference marks.

Extract the adjustment into a separate function that is called once.

(regression from 689962fe)

Change-Id: Ie14c650f7fdb259c13cb9048226da30971d2ab3c

Change-Id: I04938ceba4e4740535caee396ca8e13ca50a003c
Reviewed-on: https://gerrit.libreoffice.org/26288Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>

if the a11y option "Use automatic font color for screen display" is active.

I guess we could go to town and have a button to zoom to that option or to
directly disable it, but I've never encountered this problem before so surely
it isn't that common that we need to immediately do that.

Change-Id: Id19cac56591829f5644a8ac66c3d873c77e152f8

Change-Id: Ie4c9f67d07d81c0724c206c3a309dd5ec68fb963