Change-Id: Ib32b299806851fd2df1fbe8a824c1fe49a4efd99

Change-Id: I9da47d7add4fadf34fd4e3c9aa6f1e65cf100b20

The problem is that the hints are iterated in order of start position,
and if there is a hyperlink inside a ruby then its end position will be
reached after the end position of the ruby due to the sort order, which
leads to overlapping XML elements.

So for any given position, first export the end positions, then the
start positions.

Change-Id: I2db28d7a36e4e34fbd394b3a69fe4549ee905250
Example: fdo82849-1.docx

If a hyperlink ends on the same position as a ruby starts, the hyperlink is
not closed before the ruby so its end tag is written inside the ruby
element.  Example: fdo82849-1.odt

Change-Id: I4531327ee23a99e366c2de5d9b08ddabb1be69c7

Change-Id: I33c51176b4497f1fcfe5a05d07741e8b74e7243b

Change-Id: Iae673ac8ed8e5817efefd665dbf5f061e80f94bc

Change-Id: I5089ce8e13d26dcc773f63b3a6f0ef17474444e0

Change-Id: I112e85d999a160b9b6393e924e981231d17f439d

They were removed in commit 18f41dfa by
mistake.

Change-Id: I0ed17471cd9d6ef19d401210ed530dab93c47865

I.e. the copy part of copy&paste. Only the Writer bits for now.

Change-Id: Ia003e76e3b234735f472cdef125514f9771d8640

Change-Id: I8a8d78b291c4b1fb6f0ef5d97cfae7c6a6ceaf64

Change-Id: I8ea68819a87abecb2a6561fbd5b695902dd4af63

Follow-up to commit 27384cdf:
avoid SwIndexReg::Update() correcting the index of CrossRefBookmark away
from 0.

Not sure if it wouldn't be better to remove the requirement that
CrossRefBookmark must have index 0.  Or perhaps remove the SwIndex from
their position completely, like AT_PARA frame anchors.  But since the
whole bookmark stuff stupidly uses inheritance the base class has
established a requirement that a mark has a position.  But then again
the CrossRefBookmark already cannot have a second position, in contrast
to all other marks.

Change-Id: If42cc48724de98151b02469850805f5c77e84c6b

was broken with 70f56bc2 "java: reduce scope,
make member classes private"

Change-Id: I84a3fafbfdddf6bb04ab7787d344bf50ffd22994

This may reduce some degree of dependency on boost.
Done by running a script like:

git grep -l '#include  *.boost/scoped_array.hpp.' \
 | xargs sed -i -e 's@#include  *.boost/scoped_array.hpp.@#include <memory>@'
git grep -l '\(boost::\)\?scoped_array<\([^<>]*\)>' \
 | xargs sed -i -e 's/\(boost::\)\?scoped_array<\([^<>]*\)>/std::unique_ptr<\2[]>/'

... and then killing duplicate or unnecessary includes,
while changing manually
m_xOutlineStylesCandidates in xmloff/source/text/txtimp.cxx,
extensions/source/ole/unoconversionutilities.hxx, and
extensions/source/ole/oleobjw.cxx.

Change-Id: I3955ed3ad99b94499a7bd0e6e3a09078771f9bfd
Reviewed-on: https://gerrit.libreoffice.org/16289Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

...setting a new StepHelp value sometimes did not change the old value at all,
for whatever reason.

Change-Id: Id859091679bff96c1a3337bf32da42e7054cc166

Change-Id: I0c6390b54d0e9bb1cad7c30ff24ddb109d88e03b

Change-Id: I786b9590f8179a1f6c6593b7ff1dec286adda192

Many of our patches are now upstream.

Change-Id: I6c59973eea491148e92d7fa1a0b3681a9fcd2f2d

At least OutputDevice::acquire/release use a plain unguarded int and ++, --, so
apparently rely on the SolarMutex being locked whenever they are called.  Fixed
those places that caused "make check" to fail for me when temporarily adding
DBG_TESTSOLARMUTEX() to OutputDevice::acquire/release.  (A recurring pattern is
that a class fails to ensure the SolarMutex is locked around the destruction of
non-null VclPtr members.)

Change-Id: I77cba6f3908f2de1b516ce28f1c3c43b3f57a9c5

Change-Id: I8346872a3e5bc7443524e5039f66eea0d3d381cd
Reviewed-on: https://gerrit.libreoffice.org/16335Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Change-Id: Iaeaa568c24d6814bb696d040bef02470bd46e8e1
Reviewed-on: https://gerrit.libreoffice.org/16334Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Change-Id: I19edaf928ab4d99b5d3c399b59e26d515c2265b0
Reviewed-on: https://gerrit.libreoffice.org/16333Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Change-Id: I2571fcdde1689f1a168a5c5d6bd20785ce018927
Reviewed-on: https://gerrit.libreoffice.org/16332Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Change-Id: I792749951f11e99504c708bf27bec56c2d98019e
Reviewed-on: https://gerrit.libreoffice.org/16331Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Change-Id: Ifa0ffd9d3f40d6a47efe56cfe960aa33f06fa547
Reviewed-on: https://gerrit.libreoffice.org/16330Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Change-Id: Ief0118d0845009bd8fd9cb1c7d51b76c09cc23b5
Reviewed-on: https://gerrit.libreoffice.org/16329Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Change-Id: I69ec748819b1012790bb47eaf161623507ff2efe
Reviewed-on: https://gerrit.libreoffice.org/16328Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

We are now switching from ahrd coding the blacklist in the source code
to using a xml file. In a second step that xml file will be fetched from
a TDF server.

Change-Id: I1c67d366c77b2f4185bedaa552fa655cc8ddb047
Reviewed-on: https://gerrit.libreoffice.org/16327Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Change-Id: I1a74a5f1877bfa72818ff0a38788eafe77eea581

Change-Id: I748493de799902f5acf905c7ce919e3ebbf1a800

Change-Id: I63dcc98667ebfc94cc16407d283e65e1514d89fb

Change-Id: I81fd45276183b911e4560ca785221894ea2ebc88

this allows getGeneratedValues to work

Change-Id: Ia87e87afa8cdb01f1d39c84bc7d7143c101d8891

Change-Id: I9d91503e6beb08f11368fc6b80fefb61f60d4486

If it's AT_CHAR anchor it needs to have a valid nContent, if not the
SwFormatAnchor::SetAnchor() will clear nContent.

Change-Id: I876b32110eb9573c1e0807ce02417632a0fcd507

Apparently it may happen that some fly does get a different anchor node
than before, which means the RemoveAnchoredFly() will be called twice.

Prevent the problem differently by clearing the anchor in
_SaveFlyInRange().

(regression from a624cf52)

Change-Id: Id796d2e230de065113bcf1e6164d6b8a01ab933c

Change-Id: I0c8a360a264fe960303431877856b5f2f2d731fc

as things stand right now in writer if the table doesn't fit it
doesn't fit and don't try

Change-Id: I90cea4c8566460f614d9a1143a638597d5ab8ce6

Using "issue" instead of "run", but not consistently, is silly.

There is no instdir/program/soffice on OS X.

Change-Id: Ic94a969b268981981842101ed9c530a8f4211a15