This makes CppunitTest_sw_ww8export pass on Windows again.

Change-Id: I16fed4eabbe7b9ccdcc0c71361b85b0e13f2245a

Change-Id: Ic23c434770159ddea1f05ac0ef49572b387ebb54

Change-Id: Ic05027fafc9bed8d37306be9c7da63e53d1c6196

Change-Id: I8499c79358c598f71585234d7007e981ff4d88e1

Change-Id: I2a059f9332215678936d78990ee36a5ae2161712

Change-Id: I5a4fc60b30829cf8d9e8e7f142331dbb0007ac9b

Commit fe5527f1 (desktop, sysui: let
--convert-to imply --headless, 2014-12-22) changed
CommandLineArgs::ParseCommandLine_Impl(), so that --convert-to implies
--headless. However, that worked only when ./soffice.bin was launched;
when the ./soffice wrapper started oosplash, then a logo was still
shown, as oosplash has its own list of switches that disable the splash
screen.

Fix the inconsistency by adding "convert-to" to oosplash's list as well.

Change-Id: I47e30db3b658f179f10e8a447b5d7cea083fcc60

Change-Id: Ia8002d76443dee819b57779df5213707bd52f890

Change-Id: I132c046323e1c7805fc7ead6f247732c1843eb24

Change-Id: I18f2abfeee3b7e440aace632aec1296a1beab1f7

Change-Id: Ia28bac4b12f23f3e85e42e323b35d8715a6d5d02

fixes vcl/osx/a11ywrapper.mm:1037: error: ambiguous overload for
'operator<<' (operand types are 'std::basic_ostream<char>' and 'CGPoint')

Change-Id: Iabf9a1e795c9fe89ff5d6ff379aa6a18dd7dbf65

Change-Id: Id077493f9e462a00c427bc493f823fed79617c07

Change-Id: Ib3966118e5dde496f294c7ee224f4651a6192388

Change-Id: Ia9ed86b0d96cac76c2e1639065a4ac594c2966b1

Change-Id: I745695a92023e7a14ee547ca6bef5981dd329d19

Change-Id: Ie94c7502b0a82ca406b8d76e18def7bab8233a00

Change-Id: I11de468ab65aa4dec761fbcc1e9416528c332dd5

Change-Id: Idf69d375eff4bfff7115e59530b37200ee79f086

Change-Id: Id0579d323016d760ec89185026a3c0803345695b

On one hand, this reduces code duplication. OTOH, the
DocxAttributeOutput::FlyFrameGraphic() version was a stub, so this
improves DOCX export of image frames as well.

Change-Id: I7cd5e5feb8f1e63d074f82263c3d71088c0be30f

IRC exchange:

<tml> reverting 16bfddca makes the CppunitTest_sc_subsequent_filters_test not
crash for me (Linux, with the AMDAPPSDK-2.9-1 OpenCL implementation, but no
AND hardware, just an (Intel) CPU), but then it instead fails the
ScFiltersTest::testSharedFormulaXLS

<tml> the problem (in the non-reverted case) is that for some reason the
pDVR->GetArrayLength() on line 131 returns a ridiculously large value in
sc/source/core/opencl/formulagroupcl.cxx

<tml> that seems to cause the alllocated buffer to actually be of zero size,
according to valgrind (even if no OpenCL call returns failure...)

<moggi> let me have aquick look

<moggi> just comment out that line for now and I'll have a look at it during
the weekend

Change-Id: I5f9499ba1576cfa6d0945ecb19260a4b89c83e72

Change-Id: I38501b371b8c63ab2e6fb019e17416e139120908

Remove some pointless line breaks followed by a huge indentation (past the end
of the previous line).

Change-Id: I81925962c6eddc7ee3137135000fcc8e55222105

This is the "old" pre OOo2 fontwork stuff, take the document from #i14297# to
see an example of this old fontwork object and use
tools->customize->add->format->fontwork to enable the fontwork dialog

Change-Id: Ic7151937f8dccd8bfaf146f6a058fc03c49632ae

Change-Id: I15fda8ab103a7cdad18ef6cc36debf983819e6f3

Project: help  a958bc9cd64c9658a021296ebcd85a88b8e50d7f

Controller.FormOperations always returns <Null> in AOO. LibreOffice OK.

Change-Id: I3471227b97802d5f3cf5d41f5530d28163fee37d

Change-Id: I2b351530c5eba062449e02621cfca6ba34bf9c58

Change-Id: Icbd1a46dc426a0527422d55f77e67eac755bbc8b

Change-Id: I11326c2873aad5116fd70bfa31eb94e93fef3f40

Change-Id: Ieede6dec05f63ed0fa1dde376b2e89c381601cd6

Change-Id: Ie9241d0a28dc3871c9242123c9f000100e1a2fa1

"implicit conversion from 'long' to 'sal_uInt16' (aka 'unsigned short') changes
value from 9223372036854775807 to 65535," and at least the call to getHandle for
the long nHand argument of SwFltControlStack::SetAttr in
SwWW8ImplReader::Read_AtnBook makes it look like it better return long?

Change-Id: I594ac3ee60a72c0eb2e13f802371ff1e8ade7676

Change-Id: Ib5649d8979093bb147c61eebcf95a472ad899672

Change-Id: I2cbba2a2dc8d2763ddfe982890be52f9fcea4ae1

And add a minimal reproducer that shows how the old way was broken.

Change-Id: Ic2dadf9905d603b0fd0573651b235ecd5dd70e71

So that later SwWW8ImplReader can read these structures in a way similar
to bookmarks.

Change-Id: I8e5bc804832b944336701ac12fe6cb50e17c6b94

As commit b1cd83c6 (fix crash loading
ooo#93570-3.doc, 2014-07-04) notes, the way how we map WW8 CP indexes to
sw character positions is working only in the very simple cases.

However, to do that properly, we need a way to store the start position
of an annotation mark before we create it, so add a
RES_FLTR_ANNOTATIONMARK that can do so.

Change-Id: I7cb3564621803fef9a76edaac48773cab7649ee8