Change-Id: I68654a5fa4cbb7403c00b103e2be5929ccb516a2

regression from

commit 10e77ab3
Author: Thorsten Behrens <thb@documentfoundation.org>
Date:   Sun Nov 2 22:37:32 2014 +0100

    coverity#1242508: swapped arguments

rename the variables to indicate which are source and
which are dest indexes and that the aNewBitmap
argument is always, and assert that it must be, of
equivalent rotated source to the Bitmap which has
ImplConvolutionPass called on it.

Change-Id: If9715b9f29655da66c6981c0f7cab3d89c528ed7

Change-Id: I5a95556b2a9975af150aeb65d6f363f643349a49

not entirely sure when this stopped working, but our contempory code is riddled
with asserts and throws for negative sizes so rework this to explicitly mirror
the preview rather than try the no longer working technique

Change-Id: Ie90283cad2797ab1c37848469b007dc01a2c9d75

Change-Id: I91745c868310b301e86cf9b5d6607ae362e2b945

Change-Id: I399bf576fe89283f839842959f06b8fe83f13b2b

Change-Id: Ifa143e5a8f0e8254fbd324032951517d3a30c4c8

If there is a SAXParseException, OOXMLDocumentImpl::resolve() should not
ignore it, because if it occurs in a substream some end tag handlers may
not have been run and the DomainMapper may be in an inconsistent state,
so continuing to parse the outer document is probably not a good idea.

Also add some exception mangling so sfx2 can present a useful error
dialog.

Change-Id: I169ba6db25f2ae264af08a64edf76a6bf6757f85

Change-Id: I9891269a765ace15497a3cd3b1729be941bc3c1b

Change-Id: I1cd932ae520ba20eff8ef447614e91e1cc2032f8

Change-Id: I1811dead8f104a445bfee9282a60ecedeca872ed

lp502369-3.doc contains several StgEntry that are invalid and have type
0 (STG_EMPTY).  Not sure if sot Storage::FillInfoList() should filter
these out.

Change-Id: I493cbb346723a3be4f8bc93de1030e68c1216b50

Change-Id: Ie4ecb334666587d25b94d8bf2b1a9fe60d03d232

Change-Id: I53ad45fef0e17ce15a500959206b9b042e550e49

It is only called from sc/qa/unit/opencl-test.cxx anyway, and there it is
presumably intended to make OpenCL be used whenever at all possible, with no
requirements on which opcodes are used in a formula or the size of the formula
group.

Change-Id: Ib8a28dfa15794374e849aa1cdcc5288ef8aa017e

Change-Id: Ia9af98d580a694c2b8c5dae213d443d94c143091

(and its later fixup)

This reverts commits
 e8a68c1f
 e60b5899

It was a work-around for the real underlying issue, which was that the
result of dbaccess::ResultSet::isFirst() et al were clobbered by moves
made by its clones.

The BrowseBox has two different cursors:
1) One for data to edit (which is kept on the current/active row)
2) One for data to *paint*
The second is a clone of the first.

The real underlying issue is fixed by:

  commit d7c9a1d9
  Author: Lionel Elie Mamane <lionel@mamane.lu>
  Date:   Thu Jan 22 10:49:42 2015 +0100

      fdo#88475 RowSetBase: reposition cache before interrogating it

      Change-Id: I28d62673fdf10ee6507d38bb7c79c08e4b40902f

Change-Id: Ib4516eec5e47d95d53dbb5002f7258514c93f2c9

This fixes a regression introduced by "fdo#88551 no need to update content from field when activating grid cell"
that when switching from design mode to normal mode
before one moved row at least once, the current cell (inevitably of the first row)
was always displayed empty.
Before, the update was done by the activation event (which was too often), so now we need to do it once at initialisation.
Not sure if using updateFromModel instead of UpdateFromField makes a difference. Hope not. Else need to change that.

Change-Id: I878dea0f91b370a4f83c3c1a3ed185a51ac9f0b7

Change-Id: I6024352f9d7c68d8075d90b5954ec3ba662a06f0

Even when called with _bMoved==false, that is when inserting row or when asked to move to the position we are already in.

Change-Id: Icff5a18638f7e850db4678cdf82021f56f135d1f

Change-Id: Ide3f05cc66353787acb33b33340577624f25a8bc

This partially reverts:

  commit d20232a7
  Author: Lionel Elie Mamane <lionel@mamane.lu>
  Date:   Thu Dec 6 13:22:06 2012 +0100

      Don't force refresh when higher up code did not request it

      Change-Id: I0f415c96fc05c1d776d14885751aef020c42f4ae

which did not take into account that the cache is shared with the
clones.

Finish the above commit more deeply, that is
more aggressively avoid moving the cache if we don't need to:
when we merely reposition the cache in case another RowSet(Clone)
has moved it, no need to force a move and by that to force a refresh.

Change-Id: I28d62673fdf10ee6507d38bb7c79c08e4b40902f

Change-Id: I87744f86d1413973709a46a58ebc03a39bce842c

Change-Id: I126e06a0c7b442dfc4fcad7c05ff6caf07896a8f

Change-Id: Ia23865a38eff10304a1c1748b2b4f17af7b4d3ae

Change-Id: Ie7ff70660a95711bcedc96da72694e5c46717274

Change-Id: I9d798beba649a946446893063386a8e521455a48

Change-Id: Ie0097c20429104f10eb188ae0cbfcf5bc5a44e14

LibreOffice recognizes MS Office Table Of Contents, but LO files
saved as .doc didn't show the Cross-Refence heading - it was blank.
Now bookmarks using the LO naming convention are also imported
as TOC bookmarks from .doc files.

Testcase included.

Change-Id: Icae5c0de31856ccc75ede9a97c0ea046dd97325f
Reviewed-on: https://gerrit.libreoffice.org/14018Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>

The Itanium C++ ABI mandates that for a unique (complete) C++ type a single
unique symbol for the type's RTTI name is used across a process's dynamic
objects (so type equivalence can be determined via pointer comparison on the
RTTI names).

GCC nowadays deviates from that, using strcmp to determine equivalence, so it is
resilient to RTTI names being bound locally within dynamic objects (which has
performance benefits, but also makes it impossible to have unrelated types that
happen to have the same name "encapsulated" in individual dynamic objects---
whether or not that would violate the ODR would be open to interpretation of how
dynamic objects fit into the C++ Standard).

LLVM sticks to the Itanium ABI, which becomes notable in at least two places:

For one, libc++abi's __dynamic_cast uses strict checking.  It still has a
_LIBCXX_DYNAMIC_FALLBACK for now that additionally uses strcmp checking and
syslogs visibility violations.  Mac OS X uses libc++abi with
_LIBCXX_DYNAMIC_FALLBACK enabled, and running LO routinely logs dynamic_cast
errors to the Console there.

For another, RTTI-based UBSan checks unconditionally only use strict checking
(cf. isDerivedFromAtOffset in lib/ubsan/ubsan_type_hash.cc).  This causes false
positives from Clang -fsanitize=function and -fsanitize=vptr even on Linux not
using libc++abi.

Therefore, introduce SAL_DLLPUBLIC_RTTI to mark types for which RTTI needs to
have default visibility under the Itanium/LLVM semantics.  There is
unfortunately no way to mark only the (implicitly generated) RTTI symbols for
default visibility, but at least with the cases where SAL_DLLPUBLIC_RTTI is used
for now that is no real problem---any class type marked SAL_DLLPUBLIC_RTTI only
has inline (covered by -fvisibility-inlines-hidden) or undefined pure virtual
functions.  It appears that even the vtables of those classes remain hidden, at
least with Mach-O on Mac OS X.  (That also means there is no need for a
SAL_DLLPRIVATE_RTTI marker analoguous to the---also superfluous in retrospect---
CPPU_GCC_DLLPRIVATE one.)

Nevertheless, the number of exported symbols of course increases when
SAL_DLLPUBLIC_RTTI is "active."  For a full-blown --enable-dbgutil build on Mac
OS X,

  find instdir/LibreOffice.app/Contents -name \*.dylib\* -exec nm -gU {} \; \
    wc -l

increased from 125541 to 139239.  For Linux, an option might be to "activate"
SAL_DLLPUBLIC_RTTI only for __clang__ plus !ENABLE_RUNTIME_OPTIMIZATIONS.

The set of types marked SAL_DLLPUBLIC_RTTI with this patch (wholesale cppumaker-
generated UNO enum, struct, and interface types; plus some IEmbeddedHelper and
IUndoManager) is chosen so that a full "make check" on Mac OS X no longer
syslogs any dynamic_cast errors to the Console.

Change-Id: I42fa6ec01c2503ec24bcd9c0518abb112afa3235

Project: help  50920f837dbb2af263070b00891d788986dd21a4

T.INV returns one-tailed, not two-tailed values

Change-Id: Icc5b44667068543c4513410c954370895f2a3820
Reviewed-on: https://gerrit.libreoffice.org/14093Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

We don't want silly non-informative comments in our C++ code. It is even more
pointless to put them into generated OpenCL code.

Change-Id: I89afc7234807a3ec5379c5d55a2ea8d52bde7600

Change-Id: I22e3b90671f1dc4f9720758a7a5b79f56fa2ff49
Reviewed-on: https://gerrit.libreoffice.org/14098Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>

Moves the common code of Fmt based lists and simple vector lists
into two intermediate template classes, of which one provides
the SwFmtsBase interface.

Change-Id: If963c19293584654a3d17438221e3fca8bdfd4f9
Reviewed-on: https://gerrit.libreoffice.org/13448Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: I9460e458e9ac5274d68bd11b0ed77b06d664ce0e

Change-Id: Ibbc35d1a9264817f3fa8e3229ec1de5f51800e97

Change-Id: I20697955457b1dadc334cb0f29c600687ae2d181

Migration from Version 3 to 4 make this fail

Entry .uno:Open and .uno:OpenFromWriter are the same,
skip the entry .uno:OpenFrom

Change the operator, and a lot of comments
change the match in startWith for better reading
change test when old-cmd ".uno:Open"
change to the stephan solution

Change-Id: I0861c1e7d30b861acc753e3258707f3b6a338768
Signed-off-by: Stephan Bergmann <sbergman@redhat.com>

Change-Id: If218d83297bfdd39471d5502c0639248c96af9c6