bump product version to 5.3.0.3.0+

Change-Id: I35f9984728015715b9668bf94a2b84ddf1867b6a

tdf#104967 preserve isolated notes data in clipboard when closing document

Change-Id: I0e263583e27c5103c0bb90e8fe00562e46a52d98
(cherry picked from commit b8b657123cc508c906622d20669507628c93e104)
Reviewed-on: https://gerrit.libreoffice.org/33291
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
Tested-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>

tdf#104967 prevent crash when pasting notes originating from a closed document

This is only a workaround to prevent a crash, the actual note content is lost
when pasting, only a standard empty note caption will be pasted.

Change-Id: Ifaef5d30cc2024f3d03d6b2b4f56c5ccfbb67142
(cherry picked from commit cb566c056b0e8f9f73dac3cbaf497e102a247cb9)
Reviewed-on: https://gerrit.libreoffice.org/33275
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
Tested-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>

update credits

Change-Id: I678449b9e5722a1d8b7516cb934f03c48abb5fda
(cherry picked from commit 116693b67f82eba47f9ed03e7da4081849ec80ec)
(cherry picked from commit 6847c4e02eda9bcc64624dc716b88ba74f5433e9)

update emoji autocorrect files from po-files

Change-Id: Iea4e16a5c0ede574bc9e01bc0953aed9a138243d
(cherry picked from commit a7f24b6fe60d84d3ec226445e8023cf594daa72c)
(cherry picked from commit 4edc6a81057ec2424bdc304e7ee94d42eeb1623d)

Updated core
Project: translations  004b9ca41788f358d934ef5161facd3390bdaacc

update translations for 5.3.0 rc3

and force-fix errors using pocheck

Change-Id: I6b04819571adace8dd8bd2f5bfa6e37c0e4ac618
(cherry picked from commit f6f80453277b09ee87688949dce31fb350d068c0)
(cherry picked from commit 694cb5be945ceae5b8b51816fb92ee3e218a852f)

Adapt odk/examples to incompatibly changed (unpublished) API

...changed with f88f6bcfce233c336c9025d466c5c2fb0b671de3 "Resolves: tdf#35247
introduce a locale independent numeric field filter".

Makes the example at least compile again.  No idea whether any of the values
reported by getLocaleIndependentElementNames would actually need to be different
from those reported by getElementNames for this exampel.

Change-Id: I115096477b28a4ceea398cbbb473a707141d05aa
(cherry picked from commit d119e3a986a82ce94018c5ac021996ff221f3772)
Reviewed-on: https://gerrit.libreoffice.org/33592
Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Resolves: ofz#424 guard against broken dxary length

Reviewed-on: https://gerrit.libreoffice.org/33320
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
(cherry picked from commit 336764ad6b8f4d88ac579684c905c88c8c630bcd)

Change-Id: Ia2569e963edd75cd6c27399d33e73bafe8b3f073
Reviewed-on: https://gerrit.libreoffice.org/33536
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Resolves: tdf#104675 ScConditionalFormat ranges needed to start listening

Conditional format manager dialog holds a cloned list of
ScConditionalFormat and sets that at the document on OK. The ranges of a
format need to be set before ScFormatEntry/ScConditionEntry are added to
start their listeners.

Change-Id: I903553b1f6a8298838ee8753315a6ea263b29b08
(cherry picked from commit 657eea01046c7f39ee8ca4545241372177385946)
Reviewed-on: https://gerrit.libreoffice.org/33526
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

tdf#75471 Document is not focused when opening a second window

This reverts commit fa52e16b3fb1b8b051f8f64a52c126ba3cbf4d54.

Change-Id: I30fd747af1c332c361f5015641971fce45eaec01
Reviewed-on: https://gerrit.libreoffice.org/33587
Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Tested-by: Stephan Bergmann <sbergman@redhat.com>

ofz#419 same problem as wmf and emf

this is the *third* effective copy of this, see

the other two uses of LF_FACESIZE in wmf parsing

Change-Id: I475b18314784ad072fe58730c361fc5ca6e65819
Reviewed-on: https://gerrit.libreoffice.org/33256
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
(cherry picked from commit f365db4e64d4bfcab561936787759348bcd50a7a)
Reviewed-on: https://gerrit.libreoffice.org/33537
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

ofz#414 crash in DXFHatchEntity::EvaluateGroup

(cherry picked from commit 5434d51d12611eb2726ce1394eb01921d008fa89)

ofz#415 crash in DXFVector::DXFVector

(cherry picked from commit 2a9b4363ca190f1d783d540e95a031357f852858)

Change-Id: I15c8cb7aeb8c45f32357afd0ea2f550ffe11dbf7
Reviewed-on: https://gerrit.libreoffice.org/33233
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>
(cherry picked from commit a304efedd2ca20f9ad29c18c996d00aadf30cf12)
Reviewed-on: https://gerrit.libreoffice.org/33539
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

ofz: reduce scope, check stream, validate nActions

(cherry picked from commit bd78b28a1233e26c023a52df6b119d60a61d53ad)

Change-Id: Ie420ad40b0d852708828620cd26c2f746398f5a8
Reviewed-on: https://gerrit.libreoffice.org/33259
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
(cherry picked from commit e0f262b998074eb493a56ff372c35b817e856b17)
Reviewed-on: https://gerrit.libreoffice.org/33538
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

ofz#411: XPMReader::ImplGetColKey short read

Change-Id: I2e47ef6478b8349a562b294d0fbdad65c2a3b543
Reviewed-on: https://gerrit.libreoffice.org/33217
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
(cherry picked from commit caa24065bc0436cde5aa5bd5de4f76c65b1933f0)
Reviewed-on: https://gerrit.libreoffice.org/33540
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Updated core
Project: dictionaries  ee48a4b18b2260a4c6adfed7c45a8e2d3d507f1b

tdf#105396 update German dictionaries

Change-Id: I74d65e945c3ad080d95d77d846e56b4fa493c3aa
Reviewed-on: https://gerrit.libreoffice.org/33399
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

ofz#463 unable to mmap

Reviewed-on: https://gerrit.libreoffice.org/33519
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
(cherry picked from commit f6c465bc8e7583a8321f5c881cb008b980e0e3fa)

Change-Id: I509faeda019f42bbe7cdc5fc249f2ea2076bb702
Reviewed-on: https://gerrit.libreoffice.org/33541
Reviewed-by: David Tardon <dtardon@redhat.com>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

ofz#439 protect against 0 len text

Change-Id: Ib67284636235b0d43df626a4aeed67f2bc64b491
Reviewed-on: https://gerrit.libreoffice.org/33375
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
(cherry picked from commit 033b47dd2ace4d43bf0be1a6983717adf044b0c2)
Reviewed-on: https://gerrit.libreoffice.org/33535
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Resolves: tdf#105351 pass matching index key to conditional format dialog

Change-Id: I9d872140755c4ec0ae67349530b00f339057a356
(cherry picked from commit 0367acdcc59e14b0b067976a08202e8afef0ecf3)
Reviewed-on: https://gerrit.libreoffice.org/33529
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

tdf#100151: Dispose of window if DX device creation failed

Before, if the device lacked Direct3D capabilities, VclPtr
handling became compromised, and caused a crash. This was
particularly a problem in virtualized/remote desktop
connections in Windows.

Now mpWindow instance is disposed of right away if DX
device creation fails.

Change-Id: I639340380affdb31a93ce6dccd8df33a0a466df6
Reviewed-on: https://gerrit.libreoffice.org/33413
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Jenkins <ci@libreoffice.org>
(cherry picked from commit bb50474225f80b8aeea49f14ad66173462026a41)
Reviewed-on: https://gerrit.libreoffice.org/33478
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Resolves: tdf#105411 remove from entries before dispose

cause on dispose all the state change callbacks are
called, and all of the conditional formatting code assumes
that if there's an entry in the entries, that its not an
empty disposed one

Change-Id: I50d85128ba884a0719959f2d6e7ae83d6fb3431c
(cherry picked from commit 2fb220093f7178f75ebd582bbcd956c1ee7e03db)
Reviewed-on: https://gerrit.libreoffice.org/33476
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>
(cherry picked from commit cb65b1fa2ac641d0cb4c589db2f4b3a4df0d6b07)
Reviewed-on: https://gerrit.libreoffice.org/33482
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Resolves: tdf#105466 crash when changing conditional formatting condition

Change-Id: Id219a7935f83acf3924e60cb1bf02311dc5c4372
(cherry picked from commit 0b7c1dcd149cd98136e5ed3e8e0e70255d87cd16)
Reviewed-on: https://gerrit.libreoffice.org/33483
Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

python3: upgrade to release 3.5.3

- fixes some minor CVEs
- drop python-vc2013.patch.1
- drop python-3.3.3-py17797.patch.1:
  the bug was fixed in MSVC2015 runtime so not relevant
- drop python-lsan.patch.0:
  fixed upstream
- ubsan.patch.0:
  drop hunks that were fixed upstream
- python-3.5.0-tcltk.disable.patch:
  merge into msvc-disable.patch.1

Change-Id: I2aecae446539d28eaf3eb64ee67581596019335d
Reviewed-on: https://gerrit.libreoffice.org/33225
Reviewed-by: David Ostrovsky <david@ostrovsky.org>
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
(cherry picked from commit 743f9fc86f3d3b6e87bf58c0654bcdccab0ab383)
Reviewed-on: https://gerrit.libreoffice.org/33262
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
(cherry picked from commit baf09e572b377f569b00e61c10710787b1c151ba)
Reviewed-on: https://gerrit.libreoffice.org/33497
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

tdf#104050 - Add Breeze palette

Amendment to gerrit 31012; makefile adjusted

Change-Id: I89b339e49e190f4269a2950799436e94f0aee665
Reviewed-on: https://gerrit.libreoffice.org/31481
Reviewed-by: jan iversen <jani@documentfoundation.org>
Tested-by: jan iversen <jani@documentfoundation.org>
(cherry picked from commit 841ef7154ba7b287666767bfa6b48f4d0bb63c10)
Reviewed-on: https://gerrit.libreoffice.org/33280
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>
(cherry picked from commit 99c4b9effe38917eb2063d15aed0d879508c110e)
Reviewed-on: https://gerrit.libreoffice.org/33292
Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Resolves: tdf#105158 set date or time return type for DATEVALUE TIMEVALUE

... so adding/subtracting another date produces number of days instead of date,
and adding/subtracting a number produces date. But if used as the final
formula result force number type.

Change-Id: I046f5cc53d1fe8c9f6f71876787f2f19d24fe146
(cherry picked from commit b85ee27d9f8039a6442429587598426e73aeb1ba)
Reviewed-on: https://gerrit.libreoffice.org/33363
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

tdf#105212: BASIC sets Delimiter prop to integer value

LO's BASIC doesn't have first-class support for the UNO CHAR type, often uses
integer values to represent such CHAR values (cf. <https://wiki.openoffice.org/
wiki/Documentation/DevGuide/ProUNO/Basic/Mapping_of_Simple_Types>).

Since 0b07406f7147b9abbb2095d9e038b13293cb8b10 "Use C++11 char16_t for
sal_Unicode" (for non-Windows, since LO 5.1) resp.
e16fa715c43dcdf836ce8c400b6d54eae87b627d "Handle wchar_t as native C++11 type on
windows" (for Windows, since LO 5.2), C++ css::uno::Any >>=, <<=, etc. with a
sal_Unicode argument no longer silently treat the argument as sal_uInt16
instead.  That means that BASIC code putting an integer value into a UNO ANY
that shall hold a value of CHAR type may no longer work.  (Arguably, such code
only ever happened to work by coincidence.  For example, if the ANY were
processed by Java instead of C++ code, it would never have worked.)

Luckily, the use of CHAR in the UNO API is rare (for a good reason, as a single
UTF-16 code unit is hardly ever the right choice to represent "a character").
The only documented place I could find using a CHAR property is Delimiter in the
css.text.TextSortDescriptor and css.text.TextSortDescriptor2 services.  And the
only processing of such a property that I could find across the LO code base is
in the file modified here, which thus just takes a one-off special effort to
take care of this problem.

The direction from C++ to BASIC is left as-is.  The value of the Delimiter
property is now reported as a CHAR value (where in the past it was---arguably
erroneously---reported as an UNSIGNED SHORT value), and BASIC is generally
capable of handling such CHAR values it receives well.  For example, in the
BASIC code attached to tdf#105212,

>   MsgBox("Sort: " + SortDesc(1).Name  + " = " + SortDesc(1).Value)

SortDesc(1).Value will now print a (space) character instead of its numeric
value (32).  Also, any other uses of individual CHAR values in the UNO API apart
from properties appear to already be handled well enough by BASIC, as the sample
code

> Sub Main
>  tk = createunoservice("com.sun.star.awt.Toolkit")
>  dev = tk.createScreenCompatibleDevice(100, 100)
>  descs = dev.getFontDescriptors()
>  msgbox("Font """ + descs(1).Name + """")
>  font = dev.getFont(descs(1))
>  n = font.getCharWidth(97) ' 'a'
>  msgbox("Width 'a' = " + n)
>  met = font.getFontMetric()
>  msgbox("FirstChar = '" + met.FirstChar + "', LastChar = '" + met.LastChar + "'")
>  met.LastChar = 122 ' 'z'
>  msgbox("New LastChar = '" + met.LastChar + "'")
>
>  trans = createunoservice("com.sun.star.i18n.Transliteration")
>  c1 = trans.transliterateChar2Char(97) ' 'a'
>  c2 = trans.transliterateChar2Char(c1)
>  msgbox("Transliterate, '" + c1 + "' '" + c2 + "'")
> End Sub

demonstrates.

Change-Id: I2aec1ce374c024bfac61f6c832241dfeb561addc
(cherry picked from commit 1b835cdb5ef4cebeae729b1edf2a773f4a582c0f)
Reviewed-on: https://gerrit.libreoffice.org/33212
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
(cherry picked from commit 1eadd137ade5e2c4361990cce3f153e46110a1e9)
Reviewed-on: https://gerrit.libreoffice.org/33248
Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

tdf#105014: Disable related buttons when there is no menu to edit

Clicking the "Add Separator" or the "Add Submenu" button, when there is
no menu edit, causes LibreOffice to crash. So disable them properly.

Change-Id: Ibf685145a87c9e4343d83e1bb5ef6f08136c8f31
Reviewed-on: https://gerrit.libreoffice.org/32987
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>
(cherry picked from commit ce0f9962d1d1d5a4f6301be18c23b2d14efd9d6b)
Reviewed-on: https://gerrit.libreoffice.org/33247
Reviewed-by: Muhammet Kara <muhammet.kara@pardus.org.tr>
Tested-by: Muhammet Kara <muhammet.kara@pardus.org.tr>

Updated core
Project: translations  ef5aa6bddae942a0ec66fa7ff130b46fabd94729

Updated Slovenian translation

Change-Id: Idaa5cc0ba76346f676568f1b045187cbfc2cf6bb
(cherry picked from commit 969ab027d097bdf9585f875a349753d6ccfe702e)
Signed-off-by: Andras Timar <andras.timar@collabora.com>

making portable sed statements is not worth the hassle

change to git's own formatting capabilities by using the text-wrap
option with only indent settings/no rewrapping → %w(0,8,8)
and also omit the additional empty line when there only is a summary,
but no body in the commit message (remove a %n, make it %+b)
finally drop the now no-longer needed starmarker (the @)

previous differences in sed caused unnecessarily large diffs for
distro-packagers, see
https://anonscm.debian.org/git/pkg-openoffice/libreoffice-dictionaries.git/commit/?h=upstream&id=3dceac52f75030e87519e73b42babde51e471d9f
now it will change once, but then we can put all the blame on git :-)

Change-Id: Ifa719d062582c54c9c23b4ae08d4bae9a396e83c
(cherry picked from commit 562edf0f09ba4e82fb9186aa75ee88fd7f68d18f)
(cherry picked from commit 767695be0e6cce551ae661ba741694d48695def1)

bump product version to 5.3.0.2.0+

Change-Id: Ibe4223f40f99558a48c5a72995a0bd7988b5be2b

Branch libreoffice-5-3-0

This is 'libreoffice-5-3-0' - the stable branch for the 5.3.0 release.
Only very safe changes, reviewed by three people are allowed.

If you want to commit more complicated fix for the next 5.3.x release,
please use the 'libreoffice-5-3' branch.

If you want to build something cool, unstable, and risky, use master.

Change-Id: I81afd118b26b50b3d8b6a9ea944f1018030f00b1

Resolves: tdf#104950 add English (Malaysia) [en-MY] to language listbox

Change-Id: I3a6865fa61e4bd0089214f110087ce0fff194e64
(cherry picked from commit 35ed50dcd1c38bb0ec304dfc50758b96db65a8a5)
Reviewed-on: https://gerrit.libreoffice.org/33197
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
Tested-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>

sfx2: fix race in JunitTest UndoManager.checkBrokenScripts()

1) checkBrokenScripts(complex.sfx2.UndoManager)
java.lang.NullPointerException
    at complex.sfx2.UndoManager.impl_clickButton(UndoManager.java:432)
    at complex.sfx2.UndoManager.checkBrokenScripts(UndoManager.java:188)

impl_clickButton() gets the wrong accessible object and throws.

The .uno:SwitchControlDesignMode is dispatched async, so make sure
it's processesed before trying to click the button.

See UnoControl::getAccessibleContext() "in design mode, use a fallback".

This was running in "rr record --chaos".

Change-Id: Iee9f25c2fff56d32e6b579b5d44ee6df9870fd8b
(cherry picked from commit e19cdeaf7ddbcf39da47d9685d8e79e775dde7f4)
Reviewed-on: https://gerrit.libreoffice.org/33171
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>

tdf#104950 add English (Malaysia) [en-MY] locale data

en-MY is a known locale with already assigned LangID and mapping, adding just
its locale data will have it appear as 'en-MY' in the number formatter's
Language listbox instead of "English (Malaysia)", but solves the users'
problem.

Change-Id: I32a7e9985d37c962d077531bc8d15162f8e2b48c
(cherry picked from commit 4cc33214a5acf278b9549e21627ff44a01ce5f5f)
Reviewed-on: https://gerrit.libreoffice.org/33185
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>

update emoji autocorrect files from po-files

Change-Id: I9f99276e36fbb0293a8ae2dc7dcd49fb75ca3be3
(cherry picked from commit aef7612b0c75f4cec32aa2f5ec4bc1e382554429)

ofz: ensure nOldSize is > 0

Change-Id: Iaf385c9cd4fd37d56879908c00c7483762d66a0b
(cherry picked from commit dd70861fd7123c826feefa86f8af1da1f4877c97)
Reviewed-on: https://gerrit.libreoffice.org/33177
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>

tdf#104372 Update accelerator command of CTRL+M in Draw.

The UNO command of default formatting in Draw is .uno:SetDefault
instead of .uno:ResetAttributes.

Cherry-picked from fef38c79e46feaf72e31253d80a82f26b4c2bcc7.

Change-Id: I9fea2cb6b89f7869cfcd40fa6e1768952563a4f6
Reviewed-on: https://gerrit.libreoffice.org/33085
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Mark Hung <marklh9@gmail.com>
Reviewed-on: https://gerrit.libreoffice.org/33167
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>

Updated core
Project: translations  c67db291214569e8ed37beced9bc7ba416715755

update templates for 5.3.0 rc2

and force-fix errors using pocheck

(cherry picked from commit b42935352f559784853712a2612ea975265d190a)

Change-Id: Ie26cb18d50e8198010a777cc5b2f18babafc3150

tdf#97597: Ensure that each parsing thread has its own buffer.

(cherry picked from commit 4ae705d02df0ddf75b97d0e94add6994626f487e)

Change-Id: I93077f954a49b3922930e4fc86c80228be0f4dd2
Reviewed-on: https://gerrit.libreoffice.org/33069
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Kohei Yoshida <libreoffice@kohei.us>
Reviewed-on: https://gerrit.libreoffice.org/33176
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>

Resolves: tdf#101187 colossal page width wraps around if doubled

when trying to generate a reasonable upper bound clipping region

Change-Id: I53f8ea1bd6c52c88284e1f41369f83492d078b5d
(cherry picked from commit 2fd88ab1cbb4690a770ca2ca5d66157ec4906a2e)
Reviewed-on: https://gerrit.libreoffice.org/33008
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

guard against self-recursive GetBoundingBox

Change-Id: If33f0e5e326604bf731d6ad50644e3a0b24ccf51
(cherry picked from commit 7420e350906558556881afc213f942d45d3e0b60)
Reviewed-on: https://gerrit.libreoffice.org/33050
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

in extremis dump Ole10Native payload and launch system viewer on it

so embedded plain text documents/source code/etc in word documents can be viewed
by us under Linux

Change-Id: I19e19619070841fe097c70297adc2e8b96d1c581
(cherry picked from commit 75367918028dffb7a137d29644a7a6e1f7af7c9b)
Reviewed-on: https://gerrit.libreoffice.org/32992
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

tdf#105188 sd: fix shutdown crash after accessing master pages

The problem here is that the destructor of SdModule does a lot of
things, including destroying an SdXImpressDocument that is referenced
from some SdGlobalResourceContainer.

This calls SD_MOD() to get the SdModule to get some resource, but at
that point SfxApplication::GetModule() returns null, because the
sequence was changed from first deleting the SfxModules, then clearing
the pointer in ~SfxModule to null, to unique_ptr::reset(), which, at
least in libstdc++, is implemented via std::swap, so it clears the
pointer before deleting the SfxModule.

It appears rather brittle to rely on such a subtle detail, so refactor
things so that SdGlobalResourceContainer is no longer owned by SdModule
but has its own pet XTerminationListener, which means it will be
destroyed earlier, while the SdModule is still fully alive.

(regression from f7b1cd66167050afecf487e3d89ea12de74200b5)

Change-Id: I7f03f3adf431be8728ef3d65a078b536cb96f959
(cherry picked from commit 983ac87793a891855c7d25c42fe795908543716c)
Reviewed-on: https://gerrit.libreoffice.org/33033
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Always exit with error if codesign fails

That sanity check was added in
615fae2f67028f3c5c51c70c77dbaa9b9f3856d6 but we lost it at some stage.

The codesigning script has a very confusing history, in part caused by
its use for two purposes, when building for the Mac App Store and when
building for a TDF style distribution on a dmg disk image. Those who
work with the former and those who work with the latter haven't
necessarily checked that it doesn't break for the other case. Sorry.

Change-Id: I79011302f60b1f6551328c8b80e00f5d3698504c
(cherry picked from commit a6216119be2c32e9031c6bb9b7dfa07f0468d813)
Reviewed-on: https://gerrit.libreoffice.org/33071
Reviewed-by: Norbert Thiebaud <nthiebaud@gmail.com>
Tested-by: Norbert Thiebaud <nthiebaud@gmail.com>

vcl: Fix string comparison for reading text attributes

The original code seems to do the opposite.

Change-Id: Ib112bcd14338cef84c562e4a670bcb1fcc06583e
Reviewed-on: https://gerrit.libreoffice.org/33080
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Julien Nabet <serval2412@yahoo.fr>
(cherry picked from commit 99eed82939999d9a9689788a4134dd05d5c20c5a)
Reviewed-on: https://gerrit.libreoffice.org/33081

workaround for codesigning on Mac baseline - do jnilibs first

as otherwise signing fails (LibreOffice.app: code object is not signed at all
In subcomponent: path/to/foo.jnilib)

Change-Id: I3e4691f2ada408d9e76d200a3291b044753ec85c
(cherry picked from commit abdb32cbad1b85d2d47ad954aa55b468b078a05e)
Reviewed-on: https://gerrit.libreoffice.org/33072
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Norbert Thiebaud <nthiebaud@gmail.com>

CodesignRules.plist was removed in 615fae2f67028f3c5c51c70c77dbaa9b9f3856d6

Change-Id: Ibde5189ecf09ec8dfaf3223683c5480974084865
(cherry picked from commit 3e7e4df5a47b8d94ab86a8d6d0df38b1f282f980)
Reviewed-on: https://gerrit.libreoffice.org/33070
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Norbert Thiebaud <nthiebaud@gmail.com>

tdf#105017 Crash when click a "New Theme..." in Gallery on detached SideBar

Change-Id: I6a3c44a0a08cf036776976d40c985889698c08ac
Reviewed-on: https://gerrit.libreoffice.org/33041
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
(cherry picked from commit a076a062ceb02bd34a460819db61dcbfbca5c8d7)
Reviewed-on: https://gerrit.libreoffice.org/33056
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>

tdf#105294 - only add spacing for the border being set

related to bug 41542 which now allows the space-to-contents value
even if the border is not visible.

For years, LO has been resetting the spacing to zero when a line
is diabled. Setting all border distances here was a mistake that
was never caught before, and the documents created because of
this oversight will likely cause some grief to bug 41542.

Change-Id: Id2708cf20600e29e97dc6c8d8b779894ad9c38d1
Reviewed-on: https://gerrit.libreoffice.org/33055
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Justin Luth <justin_luth@sil.org>
(cherry picked from commit 9e7526044c8fa6b006b0cb791d15f2476c96ebf2)
Reviewed-on: https://gerrit.libreoffice.org/33062

tdf#105259 calculate sizes of all area tab pages on construction

When we open dialog that contains the area tab, we need to
calculate the size of area tab pages (that are triggered by
clicking on a button) and set the page area to the appropriate
combined size. Otherwise we don't account for the needed space
correctly and some page will be squished or have overlapping
controls.

This change creates all pages at dialog construction, gets
all the optimal sizes and sets the size of the page container box
to the combined (minimal) size.

Change-Id: Ie04a121810b96973f6e4502a52af675b2baacf25
Reviewed-on: https://gerrit.libreoffice.org/33040
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
Tested-by: Tomaž Vajngerl <quikee@gmail.com>
(cherry picked from commit e6d7d737522124350a17a3cfdee055f03200a274)
Reviewed-on: https://gerrit.libreoffice.org/33048
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Heiko Tietze <tietze.heiko@googlemail.com>

a11y crash on accessing disposed TabControl during close of impress

Change-Id: I3064eb136ed3a6a32acd64bfb2f6fbe31595e1cf
(cherry picked from commit ff30d20c1ae3a30717123c267d52eb2b3105a1b7)
Reviewed-on: https://gerrit.libreoffice.org/32997
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>

ofz: ensure all not explicitly set indexes are zero

Change-Id: Idbe6ceeb61d3dfe26f281349181e7a60f7e59000
(cherry picked from commit 14d9b3519adac8543b52ddb84e49eecbdd1d9eb2)
Reviewed-on: https://gerrit.libreoffice.org/33035
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

avoid infinite loops in dxf rendering

Change-Id: Id2eb5540241f0792cde35c444fc2c0c5b7778353
(cherry picked from commit 57779b5f3a49fedd952aed70ddcce22f48b98ea5)
Reviewed-on: https://gerrit.libreoffice.org/33042
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

tdf#105071 MM no emails sent when using doc as mail body

The conditional statement exits if there is data to process and
continues if there is no data to process. Thus all records are lost.

This commit reverses this behaviour so that data is processed
by the rest of the Mail merge activities.

Change-Id: I4c92cbd7ca55be7842b07ad887813e417430ba4f
Reviewed-on: https://gerrit.libreoffice.org/32994
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
Tested-by: Jan-Marek Glogowski <glogow@fbihome.de>
(cherry picked from commit 03ec9c397c676faff51041d992f6d715c475ff13)
Reviewed-on: https://gerrit.libreoffice.org/33012
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

fix interchanged parameters

first one found by cppcheck
> sc/source/filter/excel/xelink.cxx 637
> funcArgOrderDifferent
>  warning Function 'InsertExtName' argument order different:
>  declaration 'rnExtSheet, rnExtName, rUrl, rName, rArray'
>  definition 'rnExtSheet, rnExtName, rName, rUrl, rArray'

Change-Id: I6b9bdbf58a8eb271e873bf8dfc30041a72419f59
Reviewed-on: https://gerrit.libreoffice.org/32927
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Julien Nabet <serval2412@yahoo.fr>
(cherry picked from commit ebadc6017827edef2bca7ed389d1af9ebffb96e9)
Reviewed-on: https://gerrit.libreoffice.org/32989
Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

tdf#105015: Don’t ignore all printer fonts

For whatever reason we query printer output device for fonts and for
whatever reason GDI will only report fonts from the printer which
often in formats we want to ignore. This results in aborting the
application when we reject all fonts on the printer!

Just special case printer output devices and hope for the best.

I think this commit is partially blamed as it is evidenced now that
there was a good reason to only abort for window output devices, but
I’m leaving this as it is for now:

commit ea7102c93c33884a68d4dba0de7d52f8ed4f4579
Author: Chris Sherlock <chris.sherlock79@gmail.com>
Date:   Tue Mar 25 21:54:20 2014 +1100

    fdo#74702 OutputDevice::ImplInitFontList() fails if no fonts on device

    It makes no sense that ImplInitFontList() only fails for Window
    instances. I have carefully checked all the functions that use
    this function, and there are no good cases when no fonts won't cause
    problems. In fact, we have a number of functions that specifically
    rely on the fact that ImplInitFontList will populate
    OutputDevice::mpFontCollection with at least one font.

    Therefore, I'm making this abort if it can't populate the collection,
    regardless of whether it is a Window, Printer or VirtualDevice.

    I have also refactored GetDefaultDevice - I now check the default
    pOutDev parameter to see if it is NULL (the default), in which case
    it is referring to the default window, so I call on
    Application::GetDefaultDevice() instead of going straight to the
    pimpl data structure used by the Application class.

Change-Id: I3f34ba0cea9f88966ace3a5ef7fac868972ea5ac
Reviewed-on: https://gerrit.libreoffice.org/33014
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

SvxFontNameBox_Impl was not disposed

This is visible after toolbar customization, when the old
font box stays visible. Turns out that currently the toolbox
controller is supposed to own the item window, not the toolbox.

Regression of:

commit 970a66f8c919ea0524f216f40d21b3e2a8c88ccc
Author: Maxim Monastirsky <momonasmon@gmail.com>
Date:   Sun Oct 16 01:27:10 2016 +0300

    Use UNO registration for font name toolbox control

Change-Id: I29a1f06d97c9a981507ff62889e2d7fc6ac9d688
Reviewed-on: https://gerrit.libreoffice.org/32943
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Maxim Monastirsky <momonasmon@gmail.com>
(cherry picked from commit e1f5db9579602dd02a80c1d1ea1d078606542f16)
Reviewed-on: https://gerrit.libreoffice.org/32956
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

tdf#104372 Update UNO command of default formatting in Impress.

The UNO command of default formatting in Impress is .uno:SetDefault
instead of .uno:ResetAttributes, which is the UNO command of clear
direct formatting in Writer.

Change-Id: I1f167e7902c333eaa47a07431f67c76d809838c4
Reviewed-on: https://gerrit.libreoffice.org/32531
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Mark Hung <marklh9@gmail.com>
Reviewed-on: https://gerrit.libreoffice.org/33031
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

jdbc clob character stream: return 1 when returning a single odd byte

if at that point nBytesToRead is 0, it may be that it was 1, but we
have read one byte from the buffer (m_buf). So in this case, return 1,
not 0 (which would signal EOF).

Change-Id: I229e53f1c38c80f709df244a3509caccd69c8ecf
Reviewed-on: https://gerrit.libreoffice.org/33028
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Updated core
Project: translations  8717b9da50e69a92e7afb89f641631ac2a7b51bf

Updated Slovenian translation

Change-Id: I2a83d61ba3142e336af58bd6250ef342d11711c2

tdf#105151 - be more conservative about EndTracking / Tracking calls.

We really shouldn't be emitting these into widgets after they are
disposed - it serves only to confuse things. Restore behaviour to how
it was before the VclPtr change for this.

Change-Id: Ife666929f8aac0e4431a5e43970769692c11888c
Reviewed-on: https://gerrit.libreoffice.org/32785
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
(cherry picked from commit d788d37a16d382d7ff32a19e8d65e5c605ac55dd)
Reviewed-on: https://gerrit.libreoffice.org/32789
Reviewed-by: Eike Rathke <erack@redhat.com>

Resolves: tdf#105209 CHF currency symbol is CHF instead of SFr.

Change-Id: Ic216c7b901bf6da779e161f35e5143739267b066
(cherry picked from commit 28a0737f8c21e5c9f789af328348e51b6044c28c)
Reviewed-on: https://gerrit.libreoffice.org/32966
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>

tdf#105182 sd: avoid iterating too far in SdOutliner::Initialize()

Considering the valid indexes are extended by "-1" for backwards
iterators and "size()" for forward iterators, it's obvious that a
not-yet-at-the-end-in-the-other-direction iterator can be incremented
once, but not necessarily twice.

Why this code even wants to increment it twice isn't obvious to me.

Change-Id: I578c8c6202049ebe6dbed41b8276a6bfa0566bbc
(cherry picked from commit aa1ee198b5b55d0a92418eb3294c93553e8513dd)
Reviewed-on: https://gerrit.libreoffice.org/32973
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>

tdf#105199: Keep URL intact

Change-Id: I85681fcc81246414332f88dc46ce4a4b60896c97
(cherry picked from commit 6a68e364faa0d384f1e3bf397f5decaadecf9b3b)
Conflicts:
sd/source/ui/dlg/PhotoAlbumDialog.cxx
Reviewed-on: https://gerrit.libreoffice.org/32930
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

update credits

Change-Id: I7e32a57cf6941d36e0bbcfb447065c4b0d56f8c3
(cherry picked from commit 05d509482cedb7d7c4697a76e80352e430fd8c7e)

update emoji autocorrect files from po-files

(cherry picked from commit b365d56ed9862de518f91b34633faadce0cb0962)

Change-Id: Ifc78ba8ae27544e8588c18cd5ba962a9fdeb1497

Bump copyright year to 2017

Change-Id: I26ddcdd7bb9616c0599eec9101603d4be1ea4147
(cherry picked from commit e5da108ef01872f460c176b6a9173e346e1d59a4)
Reviewed-on: https://gerrit.libreoffice.org/32597
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>
Tested-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

tdf#104807 Placeholder field doesn't resize

Change-Id: I55f1ff2be091ea5e8475e22abdf62aa5c0c435be
Reviewed-on: https://gerrit.libreoffice.org/32921
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>
(cherry picked from commit acc939f6cb832bac281274210e3737c1ba06216c)
Reviewed-on: https://gerrit.libreoffice.org/32965
Tested-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

tdf#97598 xlsx import: do not apply any scenarios

Excel does not automatically apply scenarios, so neither should LO.

Scenarios appear to first be supported in 2009 and this section that
applies the "mnShown" scenario comes from
> commit 0851da4d8a0a557f1e9a31af652a530c615c2989
> CWS-TOOLING: integrate CWS dr68

mnShown should only mean the last scenario that was shown, so
mark it as active, but don't apply it.
In Excel, mnCurrent tracks and auto-selects a scenario in the dialog box,
so that is irrelevant to us.

Change-Id: I6b4a9b14733d6ab6dc2283a569f0e2484f81c24f
Reviewed-on: https://gerrit.libreoffice.org/32432
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Eike Rathke <erack@redhat.com>
(cherry picked from commit dfc6f4c1be58b088dd099f6f8bb6103bf962e144)
Reviewed-on: https://gerrit.libreoffice.org/32941
Reviewed-by: Justin Luth <justin_luth@sil.org>

Resolves: rhbz#1408500 gtk3 toolbar borders too distinct vs native gtk3 ones

Change-Id: I7ce284d3e6f999e2ac08326dc40b060bf70d7d4b
(cherry picked from commit 5088b2066651ed133dcfd93388f7ac7187844d33)
Reviewed-on: https://gerrit.libreoffice.org/32954
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Maxim Monastirsky <momonasmon@gmail.com>

tdf#105077 sw: fix wrong assert corner-case in SwRegHistory::InsertItems

There is a tricky corner-case when an item set is inserted that
contains items that are all identical to what's in the node's item set.

In this case a SwHistorySetText will be created for an overwritten hint,
but no new AUTOFMT hint will be inserted so there is no
SwHistoryResetText.

So it looks like AUTOFMT can't be checked at all here.

(regression from b44e70e3d17c79eaf78f9bfe8bd0275d7ed9325f)

Change-Id: I83a241588cef6a9ec074d4f0e39322987394ca86
(cherry picked from commit db4badfc971b9cc60809c3408f579bae04a77c34)
Reviewed-on: https://gerrit.libreoffice.org/32942
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Jenkins <ci@libreoffice.org>

Related tdf#100151: use disposeAndClear for mpShowWindow (sd)

See https://bugs.documentfoundation.org/show_bug.cgi?id=100151#c33

Change-Id: I81e483189eacee086b680367148f8dd3ca2d3493
Reviewed-on: https://gerrit.libreoffice.org/32904
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Julien Nabet <serval2412@yahoo.fr>
(cherry picked from commit bce35b8e13b0d82ba54bf3d380f448dad0ee13bb)
Reviewed-on: https://gerrit.libreoffice.org/32947

tdf#104830, need an own termination listener for lib objects

The destruction of the SwDLL object happens already through the normal
termination listener but the other termination listeners might still
depend on it. Also the outstanding events might need the SwDLL instance
to be still around.

This makes the destruction of the instance explicit and at a time when
it should be safe. We should use the same code for calc, impress, math
and base as well.

Change-Id: I50b8f30426f5a4a54e362e748fe962839abca73e
Reviewed-on: https://gerrit.libreoffice.org/32926
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Updated core
Project: translations  86d6efcc43918d0c1252b02e808f921cbc4f5a34

update translations for 5.3.0 rc2

and force-fix errors using pocheck

Change-Id: Iaa074827e5f9e841a2f013a7de4e5e902fe88824

Resolves: tdf#104711 adjust range reference constructed of named anchors

A constructed range reference of named anchors (i.e. extended during
expression compilation) forms a range listened at that needs to be
adjusted if the cell is shifted and one or both anchor parts are
individually named relative references.

Currently the resulting range (ScComplexRefData aka DoubleRef) does not
know to distinguish between whether its anchors are the result of a
literal range in a named expression or the result of individually named
anchors, so such RelName DoubleRef needs to be adjusted unconditionally
when shifting or moving the formula cell.

Change-Id: I75e2cc79ac60116671acce7641567337d5f5f7ed
(cherry picked from commit 439b2a134218b93e6ca9fa23005b89c19498f586)
Reviewed-on: https://gerrit.libreoffice.org/32923
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>

ucb: cmis::DataSupplier::queryContent() looks rather questionable

It's not immediately obvious if the caller is responsible for
checking the index validity here, but all the other sub-classes of
ResultSetDataSupplier that i looked at do check the index so do
the same here.

Change-Id: Ib0c5c38cb28282f08752cdb03202e4d6f3566693
(cherry picked from commit 84de69638362c7b575560e0da3efbc709b7ac476)
Reviewed-on: https://gerrit.libreoffice.org/32906
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: David Ostrovsky <david@ostrovsky.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

ucb: ReacquireableGuard::reacquire() is a no-op

So remove this junk and use osl::ResettableGuard instead which actually
works.

Change-Id: Ieb49fab16c94f0a2847ee5a1b95b52f2c141c674
(cherry picked from commit a5a94537d804f20a0d6472ef2e5995cee2d5b2fe)
Reviewed-on: https://gerrit.libreoffice.org/32907
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: David Ostrovsky <david@ostrovsky.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

Resolves: rhbz#1409634 crash on close storing clipboard of database table

Change-Id: I8e692314b4ef278f7881780d51bcecb60ca80c3f
(cherry picked from commit 6c2725038b627535ba97ead89dadb96e0ff69c3c)

ofz#394 xbm ParseDefine problem

Change-Id: I75fb761d23fdb3780088a9d209200429edb57a26
Reviewed-on: https://gerrit.libreoffice.org/32877
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Eike Rathke <erack@redhat.com>

tdf#103225 revert setting of a fixed size for the dialog

Setting fixed size of a dialog causes a mess with HiDPI and could
also cause problems with certain languages because the translated
strings could streach the dialog size. It is better to let the
dialog calculate its ideal size depending on the content, which
in some cases is not ideal but at least it shouldn't create a
mess.

The decision to revert this was done in design hangout on
2016-12-22.

Change-Id: I0ad1e38e1625cdbbf65b846118995bd5d3ebb47e
Reviewed-on: https://gerrit.libreoffice.org/32912
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
Tested-by: Tomaž Vajngerl <quikee@gmail.com>
(cherry picked from commit 62a94dcf0c750e335d6ce272d8a5df9813910d5a)
Reviewed-on: https://gerrit.libreoffice.org/32913
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>
Tested-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

tdf#104944 gtk3: Ignore mouse up event if the wrong frame

Change-Id: I3d0aad6d2b434866df9110d3b8493fec2d2c94fe
Reviewed-on: https://gerrit.libreoffice.org/32465
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Maxim Monastirsky <momonasmon@gmail.com>
(cherry picked from commit 23a13694a5a70098b5ecd58b99d99618111f3ea5)

tdf#105009 sw: fix a11y crash when removing drawing object

SwFrame::RemoveDrawObj() calls SwAnchoredObject::ChgAnchorFrame(0)
so the SwAnchoredDrawObj has no anchor frame later when
SwAccessibleMap::InvalidateCursorPosition() asks for the parent.

(regression from 76c549eb01dcb7b5bf28a271ce00e386f3d388ba)

Change-Id: Id55cb5fc41a4e37e863498265d1565e1621d508e
(cherry picked from commit ade1d4c36053c2a7aed959e3dd09ebc1b8430ea8)
Reviewed-on: https://gerrit.libreoffice.org/32802
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

tdf#105143 DOCX import: enable DoNotCaptureDrawObjsOnPage layout compat option

Because that's what Word does to show only part of the shape in the
bugdoc.

Conflicts:
writerfilter/source/filter/WriterFilter.cxx

Change-Id: Ic5cb84cace9237671d71eda0c64e9dadfe244cb9
Reviewed-on: https://gerrit.libreoffice.org/32782
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-on: https://gerrit.libreoffice.org/32784
Reviewed-by: Jan Holesovsky <kendy@collabora.com>

Updated core
Project: translations  c54ca62057567513359adf7d7dce38167b54eb29

Updated Slovenian translation

Change-Id: Ifbe0a2092a580deef5a4accc8ab4124789eaedae

recognize the broken "Err:xxx" written by 5.2 and earlier, tdf#105024 related

... and handle same as "#ERRxxx!" if present.

Change-Id: I1ebb31d628b080c52b450a8fe624c20e9e1188b7
(cherry picked from commit 75d963bc7bb87429f304d29138c27178880c039a)
Reviewed-on: https://gerrit.libreoffice.org/32793
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins <ci@libreoffice.org>

read single error constant formula as such, tdf#105024 related

... without creating a token so when writing again no leading '=' is prepended,
with which we can enable 5.2 to read such thing correctly, and when re-reading
in 5.3 it also doesn't lead to a "real" formula.

Change-Id: I26fbd20536436b49b781e2bbb5bba1dc6bafbb37
(cherry picked from commit 4fcbe16959c839bfacf745cfa554b234e639f794)
Reviewed-on: https://gerrit.libreoffice.org/32792
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins <ci@libreoffice.org>

check length of string as this can be called untokenized, tdf#105024 related

i.e. during import of ODFF

Change-Id: I7f5419d393f89d8a84efca7444e8dde3a3e9199f
(cherry picked from commit b36bf9f567f5b531f526dad6776c84e06203396f)
Reviewed-on: https://gerrit.libreoffice.org/32791
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins <ci@libreoffice.org>

out of bounds string access

Change-Id: Iecbe24f90cefe34a447cd52a44967a4b09b4394b
(cherry picked from commit 00f17500520bcb2634146496adc8de9f4d28eea5)
Reviewed-on: https://gerrit.libreoffice.org/32787
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins <ci@libreoffice.org>

Resolves: tdf#105061 Only one text box shown in case of 'between'

non-layout ScCondFormatList widget not responding to layout child change

Change-Id: I7edd4fc54de9e71fb86b47a9b182bfc22fa4af25
(cherry picked from commit ac561ab34ea9d26243fe3791f624c440dcf7ae6d)

tdf#105127 VML import: handle <v:shape style="flip:..." path="...">

I don't see an easy way to implement this via UNO, so use the internal
API. As to the internal API usage,
SdrEditView::MirrorMarkedObjVertical() (for UI) and
SvxMSDffManager::ImportShape() (for WW8 import) are example client code.

Change-Id: I9bf27788db32fd35d6b56e0f1a240c4b7abc5604
(cherry picked from commit ac03883cd66a2d58c17c8dac555a053586e46625)
Reviewed-on: https://gerrit.libreoffice.org/32778
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Jan Holesovsky <kendy@collabora.com>

gbuildtoide: builddir is not necessarily same as srcdir

Change-Id: Ied9f9abcc75f2edcd518ac247907f573aa21f35e
Reviewed-on: https://gerrit.libreoffice.org/32405
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
Tested-by: Jenkins <ci@libreoffice.org>

libpng: update to release 1.6.28

Change-Id: Ie22e5eb6656ef1d2e7903011ea72a1aa26d4da7c
Reviewed-on: https://gerrit.libreoffice.org/32768
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>

check "#ERRxxx!" constants for accepted error values, tdf#105024 follow-up

Change-Id: I9a11695710baa2f4e022c8e07f01b962cfabe2e7
(cherry picked from commit 6b89bcf85a911d043c9d93e843be12e6f23adedd)
Reviewed-on: https://gerrit.libreoffice.org/32763
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>

check token array code errors when testing for equality, tdf#105024 related

To not end up with a series of equal errors when importing individual error
cells and attempting to group them.

Change-Id: Idfcbb2e7077fc8799ef925c2c2e17188ac5e3b14
(cherry picked from commit 44021bd4018f4f97d9f0f6b6e1f0a256e8853537)
Reviewed-on: https://gerrit.libreoffice.org/32762
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>

Resolves: tdf#105024 generate and parse detailed "#ERRxxx!" error constants

Change-Id: I4e5d98cdbbc2f3e7746cd3892f70151376bc5808
(cherry picked from commit 857be5b2db4b2726306199bc279942f621bf55d8)
Reviewed-on: https://gerrit.libreoffice.org/32761
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins <ci@libreoffice.org>

process all outstanding events before shutdown, tdf#104969, tdf#104286, tdf#104399

Change-Id: Id09c3dfdc94c430d5dcb2aebb017f17db80f17e6
Reviewed-on: https://gerrit.libreoffice.org/32486
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Reviewed-on: https://gerrit.libreoffice.org/32667

tdf#105024 use context to generate "error only" formula string

The ScGlobal::GetErrorString() shortcut produced error constants in the
localized UI language, which was wrong anyway when storing ODFF. Use compiler
with context instead to generate the proper symbols.

However, this currently maps all not defined Err:xxx to #NULL! so to preserve
those additional work is needed.

Change-Id: I1de7621d6349d5b00a5aac1cedc9f265c78c7db1
(cherry picked from commit 896dd3962cd7e62bb2a9ebc6f97ff77936ec06a8)
Reviewed-on: https://gerrit.libreoffice.org/32752
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins <ci@libreoffice.org>

Resolves: tdf#104919 crash with empty ScNavigatorDlg member

commit 292560a957498bed70b625c992ec6bcf631dceb3
Author: Noel Grandin <noel.grandin@collabora.co.uk>
Date:   Mon Oct 17 13:23:23 2016 +0200

    loplugin:unusedmethods

detected the intended methods to do this were unused, so restore then, but hook
them up this time

Change-Id: I3b7e4a557696a91a70956487468b99fec6324a93
(cherry picked from commit f91dc0342de13e648dba5d9a93c01db23ae83533)

require firebird 3.0 for build (don't allow 2.5)

Keep firebird_integer_x64le.odb around for a future 3.x firebird that
will allow opening Firebird 2.5 databases, so that we can test this
capacity in our tests.

Change-Id: I05dbef51284bdb25132ff6cb661659430eea6a92
(cherry picked from commit 860ecb9e583627ab43097784ad98b41afd983ff6)
Reviewed-on: https://gerrit.libreoffice.org/32753
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Jenkins <ci@libreoffice.org>

handle paste special of single formula cell if IsEmptyDisplayedAsString()

Similar to CopyCellsFromClipHandler.

Change-Id: I7da8a55c3648c6f04c5f3b11459764fd42f51399
(cherry picked from commit 3afe82bd63fde41d2a88418fb64e4ff587b05436)
Reviewed-on: https://gerrit.libreoffice.org/32743
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins <ci@libreoffice.org>

handle paste special of single formula cell with error result

This apparently was never implemented, unlike in the block pasting code that
handles it fine. Formula cells with error results were always copied as is if
numeric was requested, even if formulas weren't requested.

Change-Id: Id550c4e757b6bb2c06aa0637328216383cdf3d6b
(cherry picked from commit 12ecd30476f17c6f6efde976f8e56d604eda0f1e)
Reviewed-on: https://gerrit.libreoffice.org/32742
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins <ci@libreoffice.org>