Change-Id: I82ce013a42ac1cd9bb4f3842e9fba8fbc6056b3c

Change-Id: Icf46bc3c086a8221e08a4bb18e1f64355043cdbb

Change-Id: I9e271c3b7ff49d33f4f6f3f8c50b70ac5ce1541e

create SwIterator with SwLayoutFrame instead of SwTabFrame
the wrong usage worked with tools/rtti because of incomplete
type information above of SwLayoutFrame

Change-Id: I9967fa1b534fc28dc5e0cd1fa5b64179faa9b94d
Reviewed-on: https://gerrit.libreoffice.org/21394Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Oliver Specht <oliver.specht@cib.de>

Change-Id: I759fe8769daccea07c2157bfb5912df8ba285534

Change-Id: I8da6d5f60b6c65ff0da8659045da1cd7666fb1d9

Change-Id: Id4017bdfe18f8224d28a9f787865c7a880aeeed8

Change-Id: I664e3815bcdaff82a03309ae387154914471a80b

Accessor and mutator created for full stop centered flag, removed
bit field.

See commit description in 8bfccd3a ("vcl: Create accessor
and mutator for font scaling in FontMetric") for reasoning behind
patch.

Unit tests
- check to ensure that can set full stop centered flag
- check equality operator on FontMetric after setting full stop
  centered flag

Change-Id: I9cacb0fbf9ea65cfebcaebdc9f0481c0a796cbcf
Reviewed-on: https://gerrit.libreoffice.org/21413Reviewed-by: Chris Sherlock <chris.sherlock79@gmail.com>
Tested-by: Chris Sherlock <chris.sherlock79@gmail.com>

Change-Id: Ic2e7213254b60cf3c384dd7ab4696092dbbc5108
Reviewed-on: https://gerrit.libreoffice.org/21412Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Chris Sherlock <chris.sherlock79@gmail.com>

Change-Id: I1844e4fc7fa08ee5526e7b9ff76224572cd4d089

Change-Id: I7d00e1eb3bb6d2da18ea2e6ae531aeb18e27c214

This is fragile code! There are actually *two* classes that do almost
precisely the same thing, they are:

- ImplFontMetric, and
- ImplFontMetricData

They both have much in common, including their class name, and even
most of their functionality. In fact, they both have common accessor
functions. When I look at the code, it looks like OutputDevice is
actually given an ImplFontMetricData object, which it then uses to
populate an ImplFontMetric object...

Basically, I'm going to merge these classes. To do so, I'm going to
do the following:

Step 1: Implement accessor functions for ImplFontMetric and FontMetric
        (then remove the friendship of this class to OutputDevice!)
Step 2: Write a unit test for each accessor function in ImplFontMetric
Step 3: Ensure that ImplFontMetric and ImplFontMetricData use some
        sort of smart pointer (probably an intrusive_ptr like I did
        ages ago with FontCharMap)
Step 4: Merge the two classes together once their class interfaces
        are the same and I am satisfied they do the same thing
Step 5: Find all instances of inefficient usage - for instance, I can
        do away with the code that copies the ImplFontMetricData
        attributes into an ImplFontMetric object.

Change-Id: I07c1cb848774b130fa2ca60b51da53e07754dd00
Reviewed-on: https://gerrit.libreoffice.org/21399Reviewed-by: Chris Sherlock <chris.sherlock79@gmail.com>
Tested-by: Chris Sherlock <chris.sherlock79@gmail.com>

Project: translations  3b2b4114cc4585399aa5db74ce702b0382bc01d8

update translations for 5.1.0 rc2

and force-fix errors using pocheck

Change-Id: I24437fdaa13a74a85f58314771245375fec94108

Change-Id: I276ce9a356086001c208902ba94641863c2f2fdb
Reviewed-on: https://gerrit.libreoffice.org/21408Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Michael Meeks <michael.meeks@collabora.com>

It seems that images must be mentioned in some src file.
Unfortunately had to copy the save icon in place of the
modified icon in galaxy, to satisfy the resource compiler.

Change-Id: If3fa7a8a194a0289c79fa10987138dfb35eaeca2

Change-Id: Ie679259deccce87949d7bf8767af707827b4134c

... if a document contains foreign elements or attributes.

In fact since ODF 1.2 the office:version attribute is mandatory and any
document that omits it is therefore invalid, while "extended conforming"
documents are allowed to contain foreign elements and attributes.

This mysterious check was there since CVS initial import, but without
any justification.

Change-Id: Ifeafad2b7af41f06356461adb7ae65dbf7bae11d

Initially we start with a very low priority, so that the lame
bits of code that do eg. focus, and cursor rendering before the
document is visible do not cause a swap, flash. Then after we've
processed a REPAINT priority idle (hopefully our first paint) we
adjust the swap priority to highest.

Essentially a fusion of Tor's approach and mine.

Change-Id: Ib9b78a18fb9359ac8c3dee1cfeb30177c08fe162
Reviewed-on: https://gerrit.libreoffice.org/21405Reviewed-by: Tor Lillqvist <tml@collabora.com>
Tested-by: Tor Lillqvist <tml@collabora.com>

Change-Id: I01786551aef0274ae11fe673cc5b0c250e8817cc
Reviewed-on: https://gerrit.libreoffice.org/21403Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Henry Castro <hcastro@collabora.com>

Change-Id: I0b22af14c7dd684c25cfee26e8c12b95b40479ee
Reviewed-on: https://gerrit.libreoffice.org/21400Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Michael Meeks <michael.meeks@collabora.com>

It also export to different Fmt when only have the difference of
UserDefinedAttributes for the text table cell.

Reviewed-on: https://gerrit.libreoffice.org/20765Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: jan iversen <jani@documentfoundation.org>

Let's try to push it again, it's unclear what the tinderbox failures
were that caused it to be reverted.

Change-Id: I022b627926870b883eea363567a34b61c34ab353

Change-Id: I78a25118d68ab7ad1db5240910f0974ba1e8381f

Reviewed on:
	https://gerrit.libreoffice.org/21395

Change-Id: I4f96b322b0306a10d7d47d3b71db9b0716575fe4

This reverts commit 0ba75220. It breaks
CppunitTest_desktop_lib on Linux.

Change-Id: I6ee9d028813c970b9ac5b9f0574af932e73111bd
Reviewed-on: https://gerrit.libreoffice.org/21293Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

Change-Id: Id2e8e9c1a9686fd674d712498acb7374a81077d7
Reviewed-on: https://gerrit.libreoffice.org/20163Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Mark Hung <marklh9@gmail.com>
Tested-by: Mark Hung <marklh9@gmail.com>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

Call it MathML 2.0 instead.

Change-Id: Ifecd52cfcded680bbddf680f2600a42b1c7568d9
Reviewed-on: https://gerrit.libreoffice.org/21304Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

There was a use of transparency because Watermark was always defined
Now we test if text of Watermark is empty or not
Remark: it seems we don't cope not-text Watermark in pdf export
(additional cleaning+renaming)

Change-Id: I9d9c759a80c0a61e4f0a94c0fa295f2950caa3bc
Reviewed-on: https://gerrit.libreoffice.org/20990Tested-by: Jenkins <ci@libreoffice.org>
Tested-by: Jan-Marek Glogowski <glogow@fbihome.de>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

Gets rid of some flashing when using OpenGL.

Change-Id: I1273b960725bf868e7340eac3f2bad2f20e73d91
Reviewed-on: https://gerrit.libreoffice.org/21390Reviewed-by: Tor Lillqvist <tml@collabora.com>
Tested-by: Tor Lillqvist <tml@collabora.com>

Files can no longer be saved on Google Drive.
The bug was introduced in 0398e1e3.

Google Drive may have user name in the form:
name.surname@domain.com
In some places the user name should be escaped before using it.

Change-Id: Idee3e57c0ad1dcbbd517b60c387868bfc706f27e
Reviewed-on: https://gerrit.libreoffice.org/21198Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: jan iversen <jani@documentfoundation.org>
Tested-by: jan iversen <jani@documentfoundation.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

This reverts commit 791ad3e1, as discussed on
IRC:

<sberg> os_cib, ^ or was that intentional?
<os_cib> sberg: There was no reason for an exact type check. Afterwards the
 pointer is static_casted to an SdrRectObj anyway.
<sberg> os_cib, but how is that "Afterwards..." part relevant for what the if()
 should actually check?
<os_cib> sberg: The detective function puts some object types into a certain
 drawing layer (polygons, circles and rectangles). It won't find any other
 object type in that layer.

Change-Id: I41d66f7b19217d0ba7780c1a999bc1289a6d3234
Reviewed-on: https://gerrit.libreoffice.org/21387Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

Change-Id: I9d1b47fc7fc26254f331967dd58f60d88bd2112f
Reviewed-on: https://gerrit.libreoffice.org/21036Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

Offline discussion about tdf#96067 "Crash on undo row inserts" brought up the
idea to warn about cases where uses of dynamic_cast are statically knwon to
always fail.  Clang's clang::AST::CXXDynamicCastExpr::isAlwaysNull already
implements such a check, reporting true if the casted-from class is final, but
has two issues:

For one, it does not work for template code, when one of the involved types is a
template parameter type (so e.g., DestType->castAs<PointerType>() can crash).

For another, it misses the opportunity to report true if the casted-to type is
final and only derives from the casted-from type non-publicly.  My hope was that
this, after the "final" decorations in 548c4323
"Mark some classes as final," might turn up the culprit of tdf#96067 (with a
scenario similar to the failed dynamic_cast on private derivation in
63b67ab5 "Use public derivation, and remove
then-unnecessary downcasts")---but not so.

Change-Id: I962ee19820758f9c601f4a292da7f37fa9dff5ce

...namely exactly those for which fa91dd31
"5th step to remove tools/rtti.hxx" (which is assumed the culprit in tdf#96067
"Crash on undo row inserts") had replaced exact-type checks of the form

  v->Type() == TYPE(X)

with type-or-subtype-of checks of the form

  dynamic_cast<X *>(v) != nullptr

which is only sound if X is a final class.  (Except for one case already handled
differently in 791ad3e1 "Stick to exact-type
check.")

Change-Id: Ia822c55c7c0e982d42c6a97589307e577721e324

fa91dd31 "5th step to remove tools/rtti.hxx" had
changed this from an exact-type check

  pObject->Type() == TYPE(SdrRectObj)

to a type-or-subtype-of check

  dynamic_cast< const SdrRectObj* >(pObject) != nullptr

but SdrRectObject is not a final class (SdrCaptionObj, SdrCircObj, ... derive
from it).

Change-Id: Ia820ee5475fbaf7a00ab0093d949c3cbd3406f83

Change-Id: I6ce5dc2333429fa472e2d605a95385912175910e

Guess we don't actually need the unique_ptr there.

Change-Id: I9b041b084567be6d8d354558da4e6916fcf81639

Project: help  8ecafdaa4e5b954941031e384b88c34309b7b27e

<ahelp> should not be empty/self-closing element (<ahelp ../>)

f5fa46ed8626910d710fd535518ddf9e756f9356 already fixed (tried to) the
empty/self-closing ahelp tag, however <ahelp ..></ahelp> is merged to
<ahelp .. /> by "make translations", so po files ended up with the
self-closing form again. Surrounding a child/PCDATA with the tag solves
the problem.

Change-Id: I58725f6a7e5999609262abec2987a6d05445a7b2