tdf#146597 sw: fix freezing with GroupShape

in case it is anchored to character and wrap is none.

Regression from 8da3586cff9cc11cf5db985d19851f21f0d42eb8
"tdf#141220 sw: fix textbox popped out of off-page shape".

Note: This is a backport-compatible fix made also for the
7.3 backport. There will be a less urgent 7.4-only fix
for its unresolved group case soon.

Change-Id: I421c85e2dd16ad2649bc299fc4733d462be8d7b7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130478
Tested-by: László Németh <nemeth@numbertext.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
(cherry picked from commit e835c2c0d3aff4f1ee5d29c9eaf36400b1aca6a7)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130456
Tested-by: Jenkins
Signed-off-by: Xisco Fauli <xiscofauli@libreoffice.org>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130715
Reviewed-by: Attila Bakos <bakos.attilakaroly@nisz.hu>

use internal SHA256Thumbprint api

Change-Id: I6a51359af58dbb79b6a0399944030dbcbe97152b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130963
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
(cherry picked from commit f30bc3ccdfb48356910d64ee4a3dda685b11de93)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131203
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

tdf#147414 sw_redlinehide: fix cursor position after AutoCorrect

Change-Id: Ia06cd4e1a74a21788e4d0ddb5f0481e8a144d863
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131147
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
(cherry picked from commit e2076b31a91d3882f3deeaa5d3d4659da0e4b17c)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131135
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>

tdf#147310 sw_redlinehide: recreate frames for whole table deleted

SwUndoDelete calls MakeFrames with end being end node of the table, but
it needs to be the following node (with a frame).

(regression from commit 723728cd358693b8f4bc9d913541aa4479f2bd48)

Change-Id: Id0974c8349be5aef9630822738eae9462bbcb4f0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131112
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
(cherry picked from commit 6f20bcb152948a24dbe40ca2e6c4ecef2bebf853)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131132
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>

tdf#136632: Use 'Save transparency' when exporting selected objects for PNG

To avoid some regression, consider that transparency should be enabled by default.
So only remove transparency if it's required explicitely.

+ change QA test since "Translucent" property isn't a bool but a sal_Int32
In officecfg/registry/schema/org/openoffice/Office/Common.xcs:
<prop oor:name="Translucent" oor:type="xs:int" oor:nillable="false">

Change-Id: I9bfb6aae8e2e8a31dee82f73a9165989129f222a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131057
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131061
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
(cherry picked from commit 43356e4d9fe022f24265747cb4b0ae2fc3cd3236)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131133

cid#1500515 Dereference after null check

Change-Id: I6483c3a036c5c56c8c6515b5b8f08720c8bd1a6b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130258
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>

Fix flaky test on Win64

Obviously the left values are unstable; they were excluded on macOS,
and they fail exactly on the same lines for me locally on Win64 build.
So just test that the values are restored after delete and undo.

Change-Id: I63dc50d223edb052a0ada200189cef94c2e0f8e5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130738
Tested-by: Mike Kaganski <mike.kaganski@collabora.com>
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
(cherry picked from commit da77aec27e3121c895a6c6c287225216489cc8bc)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131024
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
Tested-by: Jenkins

tdf#143239 sw: layout: invalidate fly position when moving off page

It's possible that flags on the fly are all valid when it is moved off page
when SwLayAction::FormatContent() returns and then it may not be positioned
again, keeping its (wrong) position on the old page and will appear to be
invisible unless you click on it.

(regression from commits c799de145f7e289f31e3669646e5bd12814e6c5e
 and eb85de8e6b61fb3fcb6c03ae0145f7fe5478bccf)

Change-Id: If07d5af7b47eb288bef71d1b9e3459197b50fdc9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130535
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
(cherry picked from commit b14bb255199e7d6db6ec9155b5d9237cb35fdba7)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130776
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>

proper columns range check (tdf#147830)

Change-Id: Ib23bc3f194fa418035e7c70a4f3e1b682f9c46b8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131181
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
(cherry picked from commit d032b81bbf2ca87114d15749d74166d0ba414972)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131195
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>

Sukapura: Update Frame Line Color icons

Change-Id: Ie6cab6f9853f4a6d338f6291b5e093efe1b0cec1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131192
Tested-by: Jenkins
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>

Sifr: tdf#147838 image for Base preview

Change-Id: I3348ae989b57477ffa741b7f4e38ec9114fef66e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131190
Tested-by: Jenkins
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>

KJ: tdf#147838 image for Base preview

Change-Id: Iee19c5a60aa0f1f7e17a446661f1db3741e1350f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131145
Tested-by: Rizal Muttaqin <rizmut@libreoffice.org>
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>

Breeze: tdf#147838 image for Base preview

Change-Id: Ib37ec53fed6d6d9ac2a39f9e81aff5564fb1d34a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131187
Tested-by: Jenkins
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>

elementary: tdf#147838 image for Base preview

Change-Id: Ie840eae9f0ee2e05d76b21f7595e9d157b61baf5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131143
Tested-by: Jenkins
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>

Colibre: tdf#137520 Increase & Decrease Font icons

Change-Id: Idea7c926304008c00b281da6c9a37ceae4e6c9c4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131141
Tested-by: Jenkins
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>

Colibre: tdf#147838 revise Base dummy preview icons

Change-Id: I72f19849f48566769a12be8a3b04308c8fcf7846
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131139
Tested-by: Jenkins
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>

Sukapura: tdf#147835: Fix paragraph direction

Change-Id: I6cfb44ed407048c5ed77783889519ccb595541b6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131137
Tested-by: Jenkins
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>

ofz#45314 limit to valid day/month indexes

Change-Id: Ibf53fa1a0c1db3046a25367fb79da3b90f7cd924
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131021
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
Tested-by: Jenkins

tdf#140077 sw_redlinehide: fix crash on SplitNode()

The problem is that CutImpl() was called with nLen=0 and so returned
without sending any hint to SwTextFrame.

Then the MergedPara gets recreated, but the problem is that the
SwParaPortion is never invalidated so it still contains the portions of
the previous merged paragraph, and painting it will crash.

But it turns out that there isn't a hint that would actually work here;
instead, generalise a condition in SwTextNode::SplitContentNode()
to simply throw away the portions.

(regression from commit eb92dc08f2abf5ed088da0d736266f213adf00de)

Change-Id: I06091ca695ea9180b32e61ddb88ce64fbf20443e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131039
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
(cherry picked from commit 6518c45dc0c2fb67500af85b97ed40466fd1d1e0)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131026
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>

ofz#45255 handle short O_hhea table

Change-Id: I649dfdb3eff868441125fbf62ff3c7d0d18d8d19
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131016
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz: Too many hatch points

Change-Id: I08cb9d09a9bb48ab31763f50bc2fa23cf723330f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131014
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz: record less than 10 bytes is invalid

Change-Id: Ie6b88efbc12b4c7fddb7459e50cba28fcbcf35fe
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131011
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

clamp and add some logging like SvmReader

LIBREOFFICE-OWMTGGWJ

Change-Id: I8f744e1ab2684a0f0995abcc3e753a684a3b970a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130982
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

sw: really prevent footnote frame against delete while it's moved

This reverts commit fe5d3fbfe63fe8b433776bd3a0508dd712b868b0
(on master only)

It turns out that checking a single column's IsMovingFootnotes() is not
enough - as frames can move both forwards and backwards the entire chain
of columns would need to be checked.

(Most callers of MoveLowerFootnotes() move forwards, but one place in
tabfrm.cxx moves from a follow to master.)

But it turns out that this is probably the wrong way in any case: most
likely the intention in FormatAnchorFrameAndItsPrevs() is to format
previous frames in the same layout environment, so if there is a section
or column inside a footnote then this upper should be formatted, while
if the footnote is inside a section or column this upper should not be
formatted; this should make calls during MoveFootnotes_() safe as it
should prevent the formatting of frames in the footnote boss moving a
footnote that is already being moved.

So tweak the fix in commit fa1bcc22921941b2cd8a0b32fe0d15655d12d607
a little to make it more general.

Also it was previously possible that for a section with columns,
both the section branch and the column branch was taken, which seems
supperfluous.

Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130544
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
(cherry picked from commit 3f569ffc9238e6bf2915e78bf21c844ca5f1270d)

add a layout test to exercise UXTSOREL

Change-Id: I57b29dd7d73c08bdbf227676c4ce9034aabf418a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130493
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
(cherry picked from commit 10d46c0498b89350645cf45e744a528d3d8f1ac1)

Change-Id: I39487640322339fe4d511e845d9c6bced2ba9dad
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130831
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
(cherry picked from commit b845927ed4e3987a578b0c8ded46ddbc7833e7ff)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130847
Reviewed-by: Caolán McNamara <caolanm@redhat.com>

ofz#45073 don't dereference null pointer on bogus font

Change-Id: I28616696833b87c300d2ee7061a15921042934b9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130613
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130936
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
Tested-by: Jenkins

add a SwFrameDeleteGuard to SwSaveFootnoteHeight

Change-Id: Ic4f0bb7c26cc021119b09682075a05d5207b48a7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130897
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
(cherry picked from commit ac605821ecf3ea263d4812ebec352314a2dbff42)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130855
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

compare authors using Thumbprint

Change-Id: I338f58eb07cbf0a3d13a7dafdaddac09252a8546
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130929
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
(cherry picked from commit 65442205b5b274ad309308162f150f8d41648f72)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130866
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz#45165 verify table offsets

Change-Id: I99649a03da28d3b896bed72999b50803ce0a4761
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130858
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz: Use-of-uninitialized-value

Change-Id: If10e8d2465ef6de62583f6547e3f68e92002f3f8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130864
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz: check hmtx offset

Change-Id: I650a37472c70771d40febe52efcb723195856421
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130856
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

Handle symlinks

LIBREOFFICE-PQ967WYM

Change-Id: I5e4df92401e009fe6b7238565d1223cfb77d10ad
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130930
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
(cherry picked from commit c1ac79906bde5c902be5cb0ae6b530da96e52169)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130861
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz: Use-of-uninitialized-value

Change-Id: I358d826a1dafe345f610d26d0d517a7357c838d2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130854
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
Tested-by: Jenkins

update credits

Change-Id: Iccc04d9704a9ee9800d2015b38e55e707ff69c41
(cherry picked from commit defa0142c92e50e7a24689bebac593db434bb778)

ofz: detect endless loop in font processing

Change-Id: I4e6c61d8be15a560f43b5d37d646e7bad9739eb7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130779
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
Tested-by: Jenkins

protect frame from triggering deleting itself

LIBREOFFICE-N4LA0OHZ

Change-Id: I0d24277665a317f047b286fe0f0878b3814ded65
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130766
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
(cherry picked from commit ee2a192923bf709d05c174848e7054cd411b205a)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130840
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

svx: the SdrObject of SvxGraphicObject can be nullptr

Crashreport:

        Graphic::GetType() const
                vcl/inc/impgraph.hxx:122
        SvxGraphicObject::getPropertyValueImpl(rtl::OUString const&, SfxItemPropertySimpleEntry const*, com::sun::star::uno::Any&)
                svx/source/unodraw/unoshap2.cxx:1466 (discriminator 1)
        SvxShape::_getPropertyValue(rtl::OUString const&)
                svx/source/unodraw/unoshape.cxx:1747
        non-virtual thunk to SvxShape::getPropertyValue(rtl::OUString const&)
                svx/source/unodraw/unoshape.cxx:1732

During SVG export of a selected shape.

Change-Id: I29b56d49b0c5418dfe8fec8491ac8667e5ade2ed
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130740
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130819
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
(cherry picked from commit c1ce7d9f2c4036ae71f8cfa71e4bee4067f4d8d2)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130849
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>

ofz: don't read past end of record

Change-Id: I9fced38faf46dce9f4cc2b96e351e7ae945d0ac1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130827
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
Tested-by: Jenkins

ofz: don't register style if hbox load failed

Change-Id: I4d9d5d76f0c2385871003720e933ed1926f66c70
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130771
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
Tested-by: Jenkins

Update git submodules

* Update translations from branch 'libreoffice-7-2'
  to 3fa6133e8446fe5eb90825616fe46d2df7444c71
  - update translations for 7.2.6 rc2

    and force-fix errors using pocheck

    Change-Id: Id35150757d9d75ecc9e7a9631827ca76d910330d

ofz: measure legal range from glyph offset, not table start

Change-Id: I04c1036e004be678b70a7df197141970aa3c8b16
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130750
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130826
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

Sukapura: tdf#147664 Revise Increase Font & Decrease Font, Default Style,

Small Caps icons

Change-Id: I1348cc7fbde63d335ba3676a4c424d76f8cf2349
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130681
Tested-by: Jenkins
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

Resolves: tdf#145580 need to use gtk_im_context_filter_keypress

for at least xim, ibus works fine. To reproduce under Fedora with gtk3
can use a keyboard layout of "US International with dead keys" with
export GDK_BACKEND=x11
export GTK_IM_MODULE=xim

and 'a in writer comment or calc header/footer dialog

Change-Id: I49425887dccc23c4fadf2bc007b6e83fc7993f7a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130802
Tested-by: Jenkins
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

ofz: glyph data must be at least 10 bytes long to be useful

Change-Id: I312c33c598013feced15c6f2dbcc66e493b703e6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130767
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
Tested-by: Jenkins

ofz#44997 Use-of-uninitialized-value

Change-Id: I3803232ddc7346967e22770ae2bdbc3378779f8b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130488
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130458
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

cid#1501377 Dereference after null check

Change-Id: I9dceb5c33f62df16b43429ba502b0d12180fa403
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130694
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
(cherry picked from commit f5f3eaa63f24093453af08ad01967ae17f108af2)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130674
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz: verify table offsets

Change-Id: I49fa886508b3bf553d4f638bfe53c37fa3e2f338
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130683
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

fail more gracefully if m_aTmpPosition is empty

LIBREOFFICE-N4LA0OHZ

Change-Id: I7f863151f753ad5605c4f1f280cfd79aa4c6bce4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130772
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz: measure maximum possible contours

Change-Id: Ie039abd835fef06514edde12b99e17360f5481a5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130774
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz#44991 keep paragraph's that failed to load until import is complete

to avoid dangling references to them

Change-Id: Ic8710f63a8cf6a6a665348c7c84bf0b43f0e2c83
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130643
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz#45081 check font length

Change-Id: Ib8cea70652ae90403db3546c07d24a517b1ec93e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130635
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz#45082 Out-of-memory

Change-Id: I274a6beeec0013f5b974017500fe31bcecabfa86
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130637
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

tdf#146463 Apply data series filter to the chart legend

Consider a data series as hidden if there is no unfiltered value, do not
consider data labels. It solves a regression introduced when working on
tdf#134059.

Change-Id: I6072bf469137d70e0aa353ffbeb0d0a1216ffa5f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/128476
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
(cherry picked from commit 7c5cfe54357e4e8d08b4696e80741097d3335dcb)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130645
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
(cherry picked from commit 73e443d32d5c275da953c29f4c493ff565426e8d)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130671

check if cast is to the right type

LIBREOFFICE-311XVJ95

Change-Id: I159f516daafad3e4088677fe2c8c6f5423b3e264
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130670
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ensure null terminator

LIBREOFFICE-WB8DT2Q9

Change-Id: I98529325bbd3ff475ba84b4991eb17240440df4b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130668
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
Tested-by: Jenkins

chart2: Break labels text and limit the size to the axis size

Previously a long label wouldn't be broken into multiple lines
and be limited to the axis size if the chart sizing was automatic.
This would cause the label to distort the whole chart and make the
chart area very narrow. With this change the label text is limited
to the axis width and gets broken into multiple lines if this is
necessary.

BarChartVeryLongLabel.odp provides a test document which includes
automatic size and fixed size chart are. We make sure the area
that the label text occupies is not larger than the chart wall
size.

Change-Id: If58bfa3e51ab68f720f22df5416ae305401bcd34
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129814
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
(cherry picked from commit ca22f1ce703c1d03fc1b1ed1f5ac1fc903b09e6f)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129898
Reviewed-by: Dennis Francis <dennis.francis@collabora.com>

chart2: crop x-axis label if it's too long on fixed size charts

If the label text is too long on a fixed size chart area, crop it
as we can't resize the chart area to accomodate for text breaking
into multiple lines.

The algorithm to determine if cropping is needed has assumed that
we need this in case the x-axis label is vertical, so there was
no check to actually make sure the text is horizontal or vertical
and in case it was horizontal, it has taken height into account
and not width, so nothing got cropped.

Change-Id: I8e73027fc51722280418c9e0be34ce487c41171e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129813
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
(cherry picked from commit 92da88582ae8d5ddbd786f9936e1b2b46eaddf2c)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129897
Reviewed-by: Dennis Francis <dennis.francis@collabora.com>

ofz#44991 keep within legal ArrowShape range

Change-Id: I6bf70f27a66ba9b355a3f7ae0c68a696a7284871
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130562
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz#44991 Use-of-uninitialized-value

Change-Id: I0e3bcce489b665bac3e470e743394935a2ba3a18
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130558
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz#44991 don't skip over terminator

Change-Id: Ibc942705a788db60c104d00916a45d595596285e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130560
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

ofz#45073 initialize names to NULL in case of 0 glyphs

Change-Id: Ie410d2a6f3d2fcda19b1925675bd7515673b07a7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130564
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

Colibre: tdf#147686 Fix arrow of Text Direction icons (large size)

Change-Id: I5e70cc6d4439ca80dd53851f4d710628aa90fd2c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130639
Tested-by: Jenkins
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

Colibre: tdf#137520 Adjust some icons size to make them balance

Change-Id: I168b00b5132742984520486535c391a7a46a902f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130615
Tested-by: Rizal Muttaqin <rizmut@libreoffice.org>
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>
(cherry picked from commit 61105658881ac0b05b2d92a3a6723a7b6a6f1d72)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130629
Tested-by: Jenkins

lastPoint might be 0xFFFF

LIBREOFFICE-KYYAZMB9

Change-Id: Ic0d95bd39a01dc1e5e0fec83dcc2c40b3f23b747
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130462
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>

uitest: blind fix for UITest_writer_tests7 failed

Traceback (most recent call last):
  File "/home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/sw/qa/uitest/writer_tests7/forms.py", line 76, in test_tdf140239
    self.assertEqual("true", get_state_as_dict(xURL)['Enabled'])
AssertionError: 'true' != 'false'
- true
+ false

Change-Id: I535f1dd5d755b5896faa55651f5a1ff6c58cec4b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125814
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
(cherry picked from commit 89f64b9f0dcdc27fdc6827b5aca42af1b38ede25)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130494
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

add to suffix denylist

add the macOS specific suffixes from
https: //support.google.com/mail/answer/6590?hl=en#zippy=%2Cmessages-that-have-attachments

Change-Id: I3a834694e3aefb757974cd00ade009874c68c8b4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130397
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
(cherry picked from commit 3e67678af2f6f8f7d25c5ad67b3828e96bc293b8)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130448

add to suffix denylist

mostly https: //support.google.com/mail/answer/6590?hl=en#zippy=%2Cmessages-that-have-attachments

but see also:
https://www.howtogeek.com/137270/50-file-extensions-that-are-potentially-dangerous-on-windows

Change-Id: Ibe3abbdcdb6f82a73d245318ef97d86789d00523
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130394
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130382
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

update credits

Change-Id: I153162daf766b49806d327b64d9a6c574f275b12
(cherry picked from commit f7b99812c088b1738f25fb57a821b90b8520d8dd)

keep paragraph's that failed to load until import is complete

LIBREOFFICE-509JU93T

Change-Id: I526edb182fed4fa023cce58e78a650a7c2046ed3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130326
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

Breeze: tdf#147603 Star icons in Extension dialog

These icons are actually for Calc's conditional formatting,
but the definition has been declared in links.txt:

cmd/sc_stars-full.png sc/res/icon-set-stars-full.png
cmd/sc_stars-empty.png sc/res/icon-set-stars-empty.png

Change-Id: Ib3bb9b9d863e368919551c1e6c082c84bab8a0bf
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130401
Tested-by: Rizal Muttaqin <rizmut@libreoffice.org>
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>
(cherry picked from commit 1b32cac107141ce8f3c04cb797d6e82dc0f54cf3)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130376
(cherry picked from commit 11841378b7c2e5f6b0c5e2493e6a94b1f48d5dd5)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130377

Colibre: tdf#147603, tdf#147150

- Star icons in Extension dialog
- "Point and Lines" Net Chart Type

Change-Id: I730228371f66169fdcd4578918da4615415dac08
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130375
Tested-by: Rizal Muttaqin <rizmut@libreoffice.org>
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>

Sifr: tdf#147603 Star icons in Extension dialog

These icons are actually for Calc's conditional formatting,
but the definition has been declared in links.txt

cmd/sc_stars-full.png sc/res/icon-set-stars-full.png
cmd/sc_stars-empty.png sc/res/icon-set-stars-empty.png

Change-Id: Idf1cfcf79ec09311fb5ee76b88d29a2b950c4fc8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130373
Tested-by: Jenkins
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>

libxml2: upgrade to release 2.9.13

Fixes CVE-2022-23308

Change-Id: I1b3bf5cf58d7d1f39c224b0d898176c95107fbf5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130241
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
(cherry picked from commit d50a7151431335d1431bccef000ae39f84bdf135)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130260
Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de>

cid#1500612 Dereference after null check

otherwise this doesn't make sense

Change-Id: I321a86cf8f9f05639b686a4c15b2e8fca2e10955
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130256
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

Better handling of msi

LIBREOFFICE-SK4E5D8N

Change-Id: I44f25a47ab6ffeb9d2b679874c8c96af1319eb2c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130317
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

Avoid unnecessary empty -Djava.class.path=

Change-Id: Idcfe7321077b60381c0273910b1faeb444ef1fd8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130242
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
(cherry picked from commit 5e8f64e50f97d39e83a3358697be14db03566878)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130265
Reviewed-by: Caolán McNamara <caolanm@redhat.com>

clear ScDocumentImport position cache if iterators are invalid

SheetDataBuffer::finalizeArrayFormula calls

ScCellRangeObj::setArrayTokens
ScDocFunc::EnterMatrix
ScDocument::InsertMatrixFormula

and InsertMatrixFormula calls the variant of ScColumn::SetFormulaCell
which doesn't take a sc::ColumnBlockPosition& param when SetFormulaCell
adds a cell to the column so any iterators belonging to ScDocumentImport
are invalid.

Change-Id: Ic2814ecbeafdeb99632d2a255ed6c1dedf7376b1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130151
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>

upgrade to expat 2.4.6

Change-Id: Ibec5c6b2a321c48725e0888135c09d9b1835d0b7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130254
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

tdf#147166 EditTextObjectImpl copy ctor doesn't exactly copy EditTextObjectImpl

and this is apparently relied on, so eliding the copy gives unexpected
results.

EditTextObjectImpl::Clone returns a copy of *this, but the
EditTextObjectImpl copy ctor explicitly does not copy the "PortionInfo"
member, so in:

commit fb8973f31f111229be5184f4e4223e963ced2c7b
Author: Caolán McNamara <caolanm@redhat.com>
Date:   Sat Oct 10 19:21:38 2020 +0100

    ofz#23492 the only user of this ctor throws away the original of the clone

    so we can take ownership of the original instead

where the copy was optimized away we want from a state where there was a
new EditTextObjectImpl with an empty PortionInfo member to one where the
PortionInfo of the EditTextObjectImpl was retained.

So explicitly clear this unwanted info.

It's very hard to make rational judgements about code if a copy behaves
differently than the orignal :-(

Change-Id: I642d60841d6bdccbf830f8a2ccdbd9f542a8aa18
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130202
Tested-by: Jenkins
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

fix Writer key event delayed compression (tdf#145963)

SwEditWin::KeyInput() was stopping the flush timer on every
invocation, which means that as long as key events kept coming
the timer would be started again and again and would not time out,
and so the input buffer would not be actually processed. It seems
that AnyInput(KEYBOARD) there actually does not normally manage
to return true ... except for Win11 where it seems that now
GetQueueStatus() detects even key-up events even though it didn't
before (I've not debugged this enough to be certain).

I think it still makes sense to have the compression in case Writer
get overloaded to help it process the pending input events, so fix
this (but if there are problems, I think an easy solution is simply
to drop it, it doesn't seem to be that necessary). Also decrease
the timeout from 200 to 20ms, as 200ms is a very visible delay.

Change-Id: I760ad9c4f58726a80bd6a9fbf93f2771f10cbdb9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130083
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130123
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

Sukapura: tdf#138400 General Number Format icons

Change-Id: I71005bb44e407bdbc48654f7399a397d03ef86a7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130249
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>
Tested-by: Rizal Muttaqin <rizmut@libreoffice.org>

Colibre: tdf#138400 General Number Format icons

Change-Id: If1ee2bb381b2b27f4f6a405cdde5dc58a35b0f73
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130247
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>
Tested-by: Jenkins

upgrade expat to 2.4.5

CVE-2022-25235
CVE-2022-25236
CVE-2022-25313
CVE-2022-25314
CVE-2022-25315

Change-Id: I1cb0449411fe938fe47ab47cead685fd04e137dd
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130156
Tested-by: Jenkins
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

Resolves: tdf#147393 SpinButtons should be can-focus

Change-Id: I092ed9c3421267ea3b2c671020e7079c3f792e17
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130095
Tested-by: Jenkins
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

fix range checking in calls like ScDocument::GetNote()

Change-Id: I5612e765b3484b0515f4a16030ee19133ae3126a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130076
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
(cherry picked from commit 957d99a539df6e21fd40370938ca5dab1613cf8c)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130105
Reviewed-by: Caolán McNamara <caolanm@redhat.com>

tdf#143817 Fix HUD deleting text contents in Writer

Change-Id: Ie28388016c475907c5d8fbe51906bc4a01d6b0d5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130091
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
(cherry picked from commit 3c94a0f7729f1638ab3b273a186c48de859cfb49)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130092
(cherry picked from commit 4886374fca8f17de78ae558033aa677bcefc0a4e)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130096
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>

Better handling of terminals

LIBREOFFICE-3MQTR80G

Change-Id: Ifa52d7f65aae473056202b46788c5870a8ce38c5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130090
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>

Refetch in case SwTextNode::InsertItem causes it to be deleted

Change-Id: Ie189ae4d37ffc33e58bb405f2521d1ae02a9997b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130087
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>

libxslt: upgrade to release 1.1.35

Fixes CVE-2021-30560

Change-Id: I334662ddc40955780321133be9aee23858e04dc1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130023
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>

for unallocated columns check default column attributes (tdf#132057)

The problem was that this was returning false for the protected
attribute just because a column was not allocated, but the default
attributes had the flag set (so if the column had been allocated
first it would have the flag set too).

Change-Id: I2ef1ef40cafb7e8fc6f7b561c0a376af63f2ad26
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129984
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
(cherry picked from commit a717029e217621482ef799731f945090c6d6be4b)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130015
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
(cherry picked from commit 5bdf7c93abd15dc422586228186e822096ce1517)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130018

bump product version to 7.2.7.0.0+

Change-Id: I30efc9f314b589161a0a752f700daabb3ea21f9d

tdf#113785 sc: IsDataFiltered must be normalized

I can't believe this hasn't caused major issues
and has survived as a bug for so long.

Due to the way IsDataFiltered is coded,
it is required that the range is normalized
in order to get any kind of meaningful result,
so lets ensure that.

Change-Id: I2ede77f738fbaeb05a0f1425a2e88e59fca08e9e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129735
Tested-by: Jenkins
Reviewed-by: Justin Luth <jluth@mail.com>
(cherry picked from commit bda200a5e9c4592bd61b7924fa171ec3265bfd24)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129761
Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130049
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>

allow building with macOS SDK 12.1

Change-Id: I1ce9901f3487bc848501ffff0794f261dbb6285a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130035
Tested-by: Jenkins
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>

configure.ac: Update kf5 include/lib check to work with KF5 >= 5.91

The 'kcoreaddons_version.h' header was moved to the 'KCoreAddons'
subdirectory in kcoreaddons commit

    commit d971ba9bb27a3e8f18a116692fdf98c1729ac244
    Author: Ahmad Samir <a.samirh78@gmail.com>
    Date:   Sat Jan 15 14:14:13 2022 +0200

        Install kcoreaddons_version.h in /usr/include/KF5/KCoreAddons/

        Instead of /usr/include/KF5/kcoreaddons_version.h.

        For more details see:
        https://invent.kde.org/frameworks/kservice/-/merge_requests/79

        GIT_SILENT

Adapt the KF5 check to check for the 'KFileWidget' header and
the 'libKF5KIOFileWidgets.so' library instead.
This is more exact anyway, since we require KIOFileWidgets,
and the previous check didn't ensure that.
(So in case only KF5CoreAddons but not KIOFileWidgets was installed,
configure would previously pass, but the build fail.)

[1] https://invent.kde.org/frameworks/kcoreaddons/-/commit/d971ba9bb27a3e8f18a116692fdf98c1729ac244

Change-Id: I7ecf7f29dce85e1a6e6b0d4f2519fa37ab04ca84
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129996
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
(cherry picked from commit 5fd5e42bf28a7910321c6b6d76257e7386839fbc)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130008
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>

Resolves: tdf#147421 Do not use OUString::replaceAll() to strip null-bytes

It reallocates and concatenates for each replacement so for
massive amounts takes ages.

Change-Id: Ibe1673fd4775c5b95833000669c1a24e718fd77c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129971
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins
(cherry picked from commit 4b0c17609c2cca326bbcc9e8488a327a4a9ea952)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129945
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

Update git submodules

* Update translations from branch 'libreoffice-7-2'
  to d37ec099e87a39897ce9f7904485b145e18065a0
  - update translations for 7.2.6 rc1

    and force-fix errors using pocheck

    Change-Id: I8f4ba182aa954c2e721788900df3e81816aeb397

tdf#113785 sc FillAutoSimple: also optimize in negative direction

By only coding the optimization for the downward autofill,
it totally broke an upward autofill.
So this patch simply fixes the LO 4.3ish regression
from commit a995462e6855061816c6529c366f20ace2b45868.

This fix means that the non-hidden cells get the auto-value
instead of being erased.

However, the hidden cells are still erased
(which is the topic of tdf#119957).

I'm not going to bother with a unit test for this
portion. If I can solve bug 119957 as well,
then the unit test for that can cover both situations.

Change-Id: If6320ccf87fa8893ca6766c265b7760cc46ed7d8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129723
Tested-by: Jenkins
Reviewed-by: Justin Luth <jluth@mail.com>
(cherry picked from commit 618084819babc839510860b74b36631749093c4c)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129753
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
(cherry picked from commit f3035cdb6a8d3ef56180640ef9d04031012250b9)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129760
Reviewed-by: Eike Rathke <erack@redhat.com>

Update language-subtag-registry to 2021-12-29

Change-Id: Ia9e1c68bd13974ec7403016197cc2d91c193f7bb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129860
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins
(cherry picked from commit 7cd3013a3704b8365e295f4260e242273d45d27e)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129841
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

Resolves: tdf#147325 "Edit in SQL View" should accept tab

Change-Id: I286281c317c30e5c189747f2d4844a0d5dd0828f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129830
Tested-by: Jenkins
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

use our own tempfile mechanism

Change-Id: Ie7853ea8bd083d1056872cfad4323ff6cede235f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129071
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
(cherry picked from commit 45c15e1112e03b8b211ed0968ed1b16d5ed4ddfe)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129758

Remove COMPILER_PLUGINS_CXX from distro-configs/Jenkins/linux_clang_dbgutil_64

It had originally been added with e754d0931ca75403647cc16100edd98e7e5ceadb
"Remove CXXFLAGS_CXX11 from Clang plugin compilation", so "if
COMPILER_PLUGINS_CXX is not set, simply default it to g++ instead of trying to
construct an acceptable CLANGCXX value from CXX (which would be Clang).  (The
problem with using Clang without CXXFLAGS_CXX11 is that Clang, unlike GCC,
typically defaults to C++03, but building compilerplugins requires C++11 at
least.  That would cause e.g. the Gerrit/Jenkins linux_clang_dbgutil_64 builds
to fail---but which also needs COMPILER_PLUGINS_CXX to be explicitly set to 'g++
-std=c++11' as GCC on those machines is still 4.8.5 defaulting to C++03."  But
that should no longer be an issue with contemporary Clang, which defaults to >=
C++11 for quite a while now.

On the other hand, when trying to update the Clang used by
<https://ci.libreoffice.org/job/gerrit_linux_clang_dbgutil/> from 5.0.2 to
12.0.1, and adding

> export COMPILER_PLUGINS_CXX="ccache $LODE_HOME"/opt_private/gcc-7.3.0/bin/g++

to
<https://git.libreoffice.org/lode/+/refs/heads/master/bin/linux_clang_dbgutil_64.env>
(where this setting arguably belongs, rather than in
distro-configs/Jenkins/linux_clang_dbgutil_64, anyway), which is needed
because that version of Clang (and thus loplugin built against it)
cannot be built with the baseline CentOS 7 GCC 4.8.5, that setting of
COMPILER_PLUGINS_CXX got overriden by the one in
distro-configs/Jenkins/linux_clang_dbgutil_64, and configure failed due to

> configure:21069: checking clang/Basic/SourceLocation.h usability
> configure:21069: ccache g++ -std=c++11 -c -I/home/tdf/sberg/lode/opt_private/clang-llvmorg-12.0.1/include -std=c++14   -fno-exceptions -fno-rtti -D_GNU_SOURCE -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS -I/home/\
> tdf/sberg/lode/packages/llvm-llvmorg-12.0.1.src/clang/include -I/home/tdf/sberg/lode/opt_private/clang-llvmorg-12.0.1/include -std=c++14   -fno-exceptions -fno-rtti -D_GNU_SOURCE -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_L\
> IMIT_MACROS -I/home/tdf/sberg/lode/packages/llvm-llvmorg-12.0.1.src/clang/include conftest.cpp >&5
> g++: error: unrecognized command line option '-std=c++14'
> g++: error: unrecognized command line option '-std=c++14'
> configure:21069: $? = 1

Change-Id: Ic33b116090f648ef645febb4fbb28ceb6a2a7cae
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129692
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
(cherry picked from commit a45f057d9d2bcd28e6b4342bbdf45fec38a43ac1)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129763
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>