In order to not require moving the delete redlines out of the document
body, rely on the DeleteSel() implementation skipping over the redlines,
and also pass the shell's layout to the various SwDoc functions that
are called.

Previously, the redlines were not deleted by DeleteSel() due to
SwAutoFormat setting RedlineFlags::Ignore.

Change-Id: Ib05ecb534993368d81f66e70124617968b4c8e9a

Change-Id: I001e61305ee8416f851f834d3068bd91de8281f8

(regression from 9fcef57a)

Change-Id: Id6ea5841dd25a30519deb825c3ae0aaa7469c250

Rename also DelPrevPara() and DeleteCurNxtPara().

Also document the state transitions.

Change-Id: I0b874852a4c1aab8d5d3a55b9dfef360e313089c

This is a remnant of code that was removed with commits
2cbc57c8 CWS swqbf91 and
becf02e5 CWS swqbf89.

The level counter was apparently abused to store the load
status of SwGrfNode's image.

Since the counter is always 0, the condition is always true and
TriggerAsyncRetrieveInputStream() is always called here; let's hope
not calling it isn't going to break anything.

Change-Id: I4cef0ddab081ed2df1fc6e88230a8c19a5632d7e

... should use layout.

Change-Id: Iabb29bb49558ef1c805ede3300042b815f62136a

Missing TextFrameIndex conversion.

Change-Id: Ie7b58b74aa6bb1e2ff8199031a9f59f5c3ba68c0

... and tests.

Change-Id: I847a9441b73cdfc9bd826fad2c9075a9aa1d7d2d

Change-Id: Ib9a13eab25b02b919d472ae864213b110e7794c7

Change-Id: I1ac09d3dfd95f2a98142c4bfe51aad04926e2d38

Move SwTextNode::GetScalingOfSelectedText() to SwTextFrame & simplify.

Change-Id: I9e9f0059665fb128de6be12a15ca2b4a95ebc5ff

This is really to complicated & should be simplified, but while it seems
reasonable to assume that a frame always exists (or that nodes without
frames should be skipped), there needs to be a way to update a
SwScriptInfo if it turns out to be invalid...

Change-Id: I185733c2e1e93b0c4d2453eeb3ea48cd4b9b6779

Change-Id: I2bc29816af93ae265c0f0d11977b2d694e00ad6c

Change-Id: I97a2081a142fd1f74a5d6be534f094fdc4db45ab

Change-Id: I8a85d1fe580ce8c8a52d0ee00d36e499b5e903dc

Apply to/reset only the paragraph properties node.

Change-Id: I4dd3973a758f4afcc30f905b7351e9687d050e33

Change-Id: I5131f5a3ff261a4d86e65b4afff3eb8cdd98cebf

Skip hidden nodes, apply to properties-node.

Change-Id: I9c511e87c431bd39dc6efdcddf2cad9bbf18c592

Change-Id: Iaf6101f02bd5dbeb9a07e796932186b5e2b1a563

It's really quite annoying to have RES_BREAK applied to all nodes; so
apply the node-items only to the first-/properties-node and skip hidden
nodes.

Keep applying text/character items as before, except skip over hidden
nodes.

Change-Id: I085afd59af78caf1ca8dac4080c53bdbdb4e56a1

MergeValues would set everything to invalid.

Change-Id: I45795fe9bd113399e276ac42851cbf5a198c9694

Merge items from first-node and props-node, as they are used in the
text formatting.

Change-Id: I4cbab55bc3b0f8bc6dc5bd3577da53b3160018a3

Change-Id: I1de6b544a4868f0d30077ba2c6d195bffd2e8b00

Change-Id: I78d32031ebc42d749dbc750b7127983810bbfb1d

If there are hints, the aFormatSet.Differentiate(rSet) will cause the
pre-existing items in the parameter rSet to win, whereas without hints
the SwContentNode::GetAttr() calls rSet.Set() which clears everything.

It looks like WW8Export::WriteFootnoteBegin() was sort of relying on
this; presumably it wants the pCFormat to win over node's items but
hints on the position to win over pCFormat.  Well that's what it
currently gets anyway, so try to preserve that.

The other use in SwUnoCursorHelper::SetPropertyValues() looks like an
oversight.

Change-Id: I80e18e040413fcf26f6f77844ed731de31a464d2

... so handle them as well in SwTextFrame::SwClientNotify().

Change-Id: Ifa8bf8f44bcd08a64dc36939bad58efefbd7af55

Move XElementAccess Java tests to C++
for ScCellsObj.

Change-Id: If721582a36ef5a82fba20d4ec7e0806d1d0524d9
Reviewed-on: https://gerrit.libreoffice.org/64365
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>

Change-Id: I76b440e72a80f20d05af477a38e9695cb17cb979
Reviewed-on: https://gerrit.libreoffice.org/64354
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>

Change-Id: Ia7b49c0ae685f6a18e9d372a689643422f1f6a04
Reviewed-on: https://gerrit.libreoffice.org/64189
Tested-by: Jenkins
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>

Locking files on WebDAV shares was implemented in tdf#82744,
in particular:

commit 26e6d4b0
    tdf#82744: fix WebDAV lock/unlock behaviour - part 1
commit bc9a8ddb
    tdf#82744: fix WebDAV lock/unlock behaviour - part 2
commit b4576f3d
    tdf#82744: fix WebDAV lock/unlock behaviour - part 3
commit fed29842
    tdf#82744: fix WebDAV lock/unlock behaviour - part 4

Before that, files on WebDAV shares were accessed w/o locking since
the beginning of [OOo] time. This option allows to revert to
the old behaviour i.e. non-locking WebDAV access.

For concurrent read-write WebDAV access by multiple users it is
recommended to keep UseWebDAVFileLocking at default value (true)

Change-Id: Id92177631d4854d6c7bf412164acbd9740db9592
Reviewed-on: https://gerrit.libreoffice.org/64116
Tested-by: Jenkins
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>

The class XElementAccess was missing the macro OOO_DLLPUBLIC_TEST,
and so it wasn't possible to use these test cases.

Change-Id: I358bb840c6088ea25b60ee57b8c69f31ab71ddbb
Reviewed-on: https://gerrit.libreoffice.org/64363
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>

Change-Id: Ie85b3a61ded8c6fd7a1f8c463912062669b3719b
Reviewed-on: https://gerrit.libreoffice.org/64357
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>

The constraints explicitly said that the width should be larger than the
height, but it was the opposite as constraints were not parsed.

Unfortunately it would be too brave for globally start handling all
constraints which lack a forName, so add a switch to opt in for this,
and use that with the conn algorithm. All clients should migrate to
bRequireForName=true at some stage, though.

Change-Id: I24ae79b141c0f7a11e4d19f141759fc1dd2169b0
Reviewed-on: https://gerrit.libreoffice.org/64350Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins

Change-Id: I10ff8b59ba707d5795338ff5e9037473d31337bc
Reviewed-on: https://gerrit.libreoffice.org/64361
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>

Change-Id: I25bded0e5c1340ca61c747b0125825ed13a9150f
Reviewed-on: https://gerrit.libreoffice.org/64349
Tested-by: Jenkins
Reviewed-by: Rizal Muttaqin <riz_17_oke@yahoo.co.id>

Also a group separator character followed by a non-digit is not.

Change-Id: Id57e43fe7692706c5532fb05ad394224265c2750
Reviewed-on: https://gerrit.libreoffice.org/64358
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>

The shape was created, but we literally tried to create a "conn" type,
while that has to be resolved to the relevant arrow type based on the
context.

This means now arrows show up between the parent-child pairs (but their
size is not yet correct).

Change-Id: I82594e46579e4ef723093e1dd0ba31bfcbbec4a0
Reviewed-on: https://gerrit.libreoffice.org/64348Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins

This is a followup of e22ab5e6.

Change-Id: I7052e113173204f7cfcd40622f52c5d4e14b3aa9
Reviewed-on: https://gerrit.libreoffice.org/64139
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>

Default unset bit now does not accept/skip group separators in
numbers.

See .idl description comment for why this is incompatible and how.

This actually uncovered a "bug" (or at least unexpected) in the
Math parser that parsed "0," as one entity instead of "0" followed
by ",". As obtaining the text form appends a blank after each
entity the sw/qa/extras/rtfexport/rtfexport.cxx testMathEqarray()
testcase had to be adapted.

Change-Id: I2b605742ab3d8a9bb81ad6880faf050535dc80fa
Reviewed-on: https://gerrit.libreoffice.org/64270
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>

Move XEnumerationAccess Java tests to C++ for ScCellsObj.

Change-Id: I3ec78142db7cfa924bf5663ca44cca80b6a0e556
Reviewed-on: https://gerrit.libreoffice.org/64315
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>