Change-Id: Ib0a2d8d69f5bd43872c949d79c685f71657dc9db

Change-Id: Idc75fd0735452ffdf6a21e3597cd3fe74f651581

This method calls DocxSdrExport::writeDMLDrawing(), which may call back
WriteTextBox(), which may call WritePostponedDMLDrawing() again.  The
result is that we try to flush drawings inside a shape which were
postponed outside of it.  Luckily, StartRunProperties() asserts this, so
instaed of silent corruption, such an attempt crashes.

Fix the crash by saving the postponed drawings on the stack, and
restoring them after the shape export finished.

CppunitTest_sw_ooxmlsdrexport's testAnchorIdForWP14AndW14 is a
reproducer for this problem (when shape with text is imported as shape
with textbox).

Change-Id: Id5aeda33472655697717401c24dd54e7efabacd9

RTFSdrImport::resolve() is called for \picprop and creates an XShape
that is stored in RTFSdrImport::m_xShape and also
DomainMapper_Impl::m_aPendingShapes;
later RTFDocumentImpl::resolvePict() completely ignores that XShape
and creates a new one, which is also inserted in the document;
the first XShape is effectively leaked.

Try to avoid that by re-using the exising m_xShape in resolvePict().
Not sure if there are any problems with doing this, it's all a bit
confusing.

Change-Id: I98456242acb0766f547eb8f7d877f51d53323f3a

If the shape properties are inside \picprop destination, don't set
shapeType.

(regression from 9f1f7199)

Actually this does not fix the root cause, this is just a work-around,
the extra shape is still inserted but it's invisible now.

Change-Id: I6cf093de2a5657533f393863ed8010ae083bec16

also

coverity#706301

Change-Id: Ie6eba530c9a2f822226fbd1b1a6aa49deaf99659

also
coverity#706515
coverity#706516
coverity#706517
coverity#706528
coverity#706529

Change-Id: I5beb753d5cc8ed976b3c11319cc36415adebd1e6

Change-Id: Ib3b25a0aab2389be68fc8339f9b08a87de419b33

Regression introduced with 83f77ab0.

Interpreter errors are transported using NaN coded doubles, using simple
setNan()/isNan() to flag and ignore non-numeric values skips all error
values.

Change-Id: I0d3cb30262bc5ba7ee77e53a2bc45e56569fbc4b

ForceArray parameters weren't propagated and enforced to array arguments
on svJump tokens (FormulaJumpToken), namely IF, CHOOSE, IFERROR and
IFNA.

Change-Id: Icf9074f11b826655a52858d778d9a0122d207aa4

Change-Id: I89ea074bacf7884fe8b8471cfd208f643326a7e1

Change-Id: I1da5edfdfd15ec8205de6467bd6c88dc5596426f

Change-Id: Ie2299485b8017232b6cfcb855b6f2dbff2c5db6e

Change-Id: I7414fffb0ab0aa7c52ba58d666c26abc4367798c

Change-Id: I41d841718da2eab8a6f424df694ba8ea4e450bac

Change-Id: Ie2ec61c75d036dbad68bd43d5a2eb9834dadd735

Change-Id: Ic951a16c3d44b36ce82f05381c98e13e74b8bc35

Change-Id: Id5ce3555c487aa7abcf97855a036663c91dd28b1

Change-Id: I9e2fd1249061fcf386a8812a42450e52d37bdc5c

Change-Id: Icf80cff1f68e85bf9103aedc95ccca80e2d32681

I can see nowhere that m_pHScrollbar or m_pVScrollbar are
set to NULL, deleted yes in the dtor, but nowhere NULLed.

Change-Id: I3012be6de1117757237884deebacc9e0e29dc7a7

The TextRange property of a shape is its anchor position: if that's
adjusted, then also set the textbox's RES_ANCHOR to the new position.

Without this, e.g. shapes anchored in headers have their textboxes
anchored in the body text, CppunitTest_sw_ooxmlexport's testfdo78420 is
a reproducer.

Change-Id: I83ed09875c3f0360c581c331507ad2b9d05ffb3a

Also, let the new SwTextBoxHelper::restoreLinks() restore also the
RES_CNTNT of the *old* draw formats, not only the link between the new
draw and fly formats.

This allows properly preserving the link between draw and fly formats,
when they are in the header (and so copied in and out variously).

Change-Id: I101ff06533e2ea27abea8bed171ed69c9649ebe8

In case a shape (has a draw format) has a textbox (RES_CNTNT of the draw
format), then that's just a pointer to that content, but the draw format
doesn't own it: the matching fly format does. So ignore that content
when deleting the layout format in case of draw formats: that ensures
when both the draw and the fly format is deleted, deletion is only
performed once.

Change-Id: Idb4bb19130a6b9acd8f8d3710b9982801b416dda

see commits:
  c5c317ed

Change-Id: I3a886bde11dc3620cffeb3e2e579c7b37285d9c5
Reviewed-on: https://gerrit.libreoffice.org/9759Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: Ia3ad4032f500e95db5db520292e6c8b07b0f5c0b

Change-Id: Ia4d919d88bf22168aa770ad35fd8360b0b155efe

Change-Id: Iee4e5dd2936979a9bcb8bd5db781e51551270a7f

Change-Id: Ie3a19a5c3de0021b89eefbacb91c88e38396b8cb

It's static library since d324745f.

Change-Id: Id8474fb28d40a2373e2b0c85439cb5a96c469c85

Change-Id: I6a759eb3c5bff4dcf4603d648053e17a6d4ff837

"LC_ALL=C make CppunitTest_sw_rtfexport" was still failing after
e47a02b1 "fdo#77979: sw: RTF export: write non-
ASCII font names encoded," so for each given eTextEncoding determine a Windows
charset that can at least encode the font's name, and not only for Unicode-
related eTextEncodings.

Change-Id: If547566bb0cffc60411d8f667d76749a904f7a3f

Problem: Many classes inherit directly from std::vector
See comment#6 by kendy for more details.

Solution: Removed class SwpHstry altogether and moved its
destructor's functionality to SwHistory's destructor.
Changed type of m_SwpHstry from SwpHstry to vector<SwHistoryHint*>

Change-Id: I32a7e111fef7c5e7b83ecee6469e173b5a21a5bf
Reviewed-on: https://gerrit.libreoffice.org/9725Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
Tested-by: Noel Grandin <noelgrandin@gmail.com>

Change-Id: I0fe3ee06d8112207f3ae3e0e8e40fa74d043ce11

Instead of this hacky collada2gltf patch.

Change-Id: I9f15c4ee1c9ea1a1a64116691b12b696c1d63ac5

libgltf: Parser releases the glTFHandle, but it was not connected to
the corresponding handle by RenderScene::initScene. So when rendering
stopped during file loading because of missing file, Parser's handle
member was an uninitialized pointer so it crashed by calling release
method.

Change-Id: I80099195341766f474143014d5949703d47a6fd8

Plus fix some problems related to temp files.
e.g. kmz converter was creating temp files next
to the input file.

Change-Id: Ib3f367fe1c4ef3cb25f5ca1c3d06b2dde78c5e45

Change-Id: I8c7b7a311cbd38058540d3462b0497ebe3d28a9e

There is no an accepted mime type for glTF models yet, so
follow collada's mime type: model/vnd.collada+xml.

Change-Id: I1e23a95dbb0caea781dade819f8e08235cb5c1dc

Change-Id: I75e69ba1acafbab8362536f02f4d39a8fc2deed1