f0305ec0 "Fix creation/removal of GPG socketdir"
spawns additional gpgconf processes which need the same LD_LIBRARY_PATH fix as
d15f042a "CppunitTest_xmlsecurity_signing failed
in sanitizer builds"

Change-Id: I1da9e9e459c5a584f092966923f8818f115322dc
Reviewed-on: https://gerrit.libreoffice.org/63420
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

Change-Id: I79ed336058ffa9e182d3332acfeda8d6526c6b43
Reviewed-on: https://gerrit.libreoffice.org/63432Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
Tested-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>

Change-Id: I3cdce43e0ba9b17f9bf993ebcad5f64f0834ceaf
Reviewed-on: https://gerrit.libreoffice.org/63421
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Change-Id: Ide10d0186431e745c3bb17d321cf46724e801c8f
Reviewed-on: https://gerrit.libreoffice.org/63404
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

LD_LIBRARY_PATH=.../master/instdir/program \
 .../master/workdir/LinkTarget/Executable/unoidl-read --published \
 .../libreoffice-6-1-0/instdir/program/types.rdb \
 >.../master/udkapi/type_reference/udkapi.idl

* no changes

LD_LIBRARY_PATH=.../master/instdir/program \
 .../master/workdir/LinkTarget/Executable/unoidl-read --published \
 .../libreoffice-6-1-0/instdir/program/types.rdb \
 .../libreoffice-6-1-0/instdir/program/types/offapi.rdb \
 >.../master/offapi/type_reference/offapi.idl

* all the interim modifications to offapi/type_reference/offapi.idl were done in
  error and had already been reverted again (i.e., `git diff
  76f30450..5b0ceb86^
  -- offapi/type_reference/offapi.idl` is empty)

Change-Id: Ibac6d00f9b6fa3fa4d4ac491af66ea4938b36b87
Reviewed-on: https://gerrit.libreoffice.org/63424
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

b96180cb "uitest: set en_US.UTF8 for the
LibreOffice instance" had made UITest use the en_US.UTF-8 locale (passed from
LIBO_LANG to LC_ALL when starting soffice in uitest/libreoffice/connection.py),
for unstated reasons.

The mail sub-thread starting at
<https://lists.freedesktop.org/archives/libreoffice/2018-October/081318.html>
"Re: master build problems with en_US.utf8 locale" now argues that support for
an en_US.UTF-8 locale in the OS should not be necessary when building LO.  While
absence of OS-support for en_US.UTF-8 apparently doesn't break the UITests (see
<https://lists.freedesktop.org/archives/libreoffice/2018-November/081375.html>
"Re: master build problems with en_US.utf8 locale"), it feels better to make
these tests not use the en_US.UTF-8 locale at all.  At least for me, the tests
ran fine when using the C locale instead.

Change-Id: I23eb2ce540bb40a7b7d13c2a396e313966f03f6e
Reviewed-on: https://gerrit.libreoffice.org/63360
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

Change-Id: If69cee340e48d42b1be56ea6012e6686aa203f53
Reviewed-on: https://gerrit.libreoffice.org/63423
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: I48b8295fc75e40f5d58f99fc2809c28de48771d5
Reviewed-on: https://gerrit.libreoffice.org/63417Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins

This reverts commit fa2167fe ...
and fixes the merge conflicts.

Change-Id: I57fa6b9e7d35e0fc28d941f2968854a52b17b995

Change-Id: Ib0713df614e63c20965979e7e34f7481fe8ed838

Change-Id: I6b2df2aee96ab32123d3fa23fd9a2480b3ed3c72

Can't tell which chapter/reference fields will be affected by a redline,
so we have to either update all of them, or rely on the
IDocumentSettingAccess::getFieldUpdateFlags() which lets the SwIdle
update the fields, at some later time...

Change-Id: I3b9cfd17bd227a18f64dac193f0cc912768810c1

Repainting is not enough, values require up-to-date outline visibility
etc.

Change-Id: I85d99fc65a071279c2d4656a52ff82ed3b2db7d8

Change-Id: I386bce6b929e38077c01595de618b89fccdaf5fe

Change-Id: Iea24b3823b648066e331dc5e9789c79662727a12

With the old Hide implementation this actually would take the last
set-field in the redline nodes array section to set the value at the
start of the document, because there was no check for that; surely that
was wrong...

Change-Id: I63fb89ccf0067f1fe375226d6d04cafb6485c5d6

... and SwAuthorityFieldType, which needs 2 lists of the fields because
each citation may occur multiple times, so their order might change on
deletions.

This is for when there is a bibliography in the document, and it is set
to IsSequence(), i.e.,
insert Bibliography -> Number entries -> Sort by document position...

Change-Id: Iba2c31267aeebacb779713567eeb0aa6659afefb

This is quite tricky & somewhat ugly due to the SwValueField base class
& we need to duplicate its member too :(

It would be possible to filter the deleted fields in
SwDocUpdateField::MakeFieldList(), but the problem with that is that
the instance is long-lived & might be used with different layout
settings, so it's probably better to filter in the 5 or so client
functions that iterate over the GetSortList().

Note that this will also filter other fields including database fields,
hopefully that shouldn't cause problems.

Change-Id: I0cdbc7757fa529598b5dbceec0bd2c4a619be05a

Has been like that since CVS import...

Change-Id: Iefa49407de90bace5eea2f86170ba8cd29b1cfe7

Change-Id: Ifebbaa6ef21ac518f62a726dadaceb97fd5e446b

Change-Id: Ied1e89279e3071e0c6c1b414bb23044e6439258c

It's only used in one function, just use a local map.

Change-Id: I3a79876acf8b31c1a8a14e9f65ada2153e7c773a

Change-Id: I3b68a46f68fb4070d26b247a6701b510cc209d9b

While at it, rename that to ExpandImpl() so it's unique.

Change-Id: I217deb72b4dfdca777e28e18ff56e49863e66640

If the bCached=true, it shouldn't matter what the layout is, because the
field won't be expanded properly anyway.

Except for the SwInputField which disables caching, but that one often
has a different code path anyway.

For most fields it doesn't matter anyway, e.g. database fields.

Change-Id: I628195f43c5d26feba94af0a832386791c072ba1

Change-Id: I206d7345525a4de0f9ece6bac575f82ff1f96df8

Change-Id: I98530004f2d75b10fc4cd3241b06d513ed57675b

Thanks to [loplugin:nullptr] and [loplugin:implicitboolconversion]
for finding one of the call sites; 3 current C++ compilers would
otherwise have implicitly converted the parameters into nonsense
without warning.

Change-Id: I3cf6697b37616570fd56fd32da27752983a66f4a

The expansion of footnotes (and soon text fields) depends on the
layout's mode.

In SwXFlatParagraphIterator, just use whatever is SwDoc's "current"
layout, should be good enough...

Change-Id: Ica3b9eae56b14910c435c0f9c3f8ab6dbcda105f

... apparently only used by ToX code; rename uniquely & add layout
parameter.

Change-Id: I0f94787fecd913912f122a654f183ac59b16e30a

... and init it in SwGetRefField::UpdateField().

Change-Id: I69af00678e84214d4a122d8b2d940fcdda5f4ccf

Now it can store both the result for Show as well as for Hide mode.

Change-Id: I15a14b76a96ef77683cd63039b1a8f5e1e94e4c9

Add layout parameter to find the correct outline node.

Change-Id: I15477dff18c3e5e34b291a92878d938d04497436

Change-Id: I180806bfe40150398fb45f57c7d93e429464f2cb

Change-Id: I2d992719c86ec7d0203173c6f9008cf4845b3855

GetUndoComment() just returns "$1" and then
SwUndoRedlineDelete::GetRewriter() just replaces all of that;
seems to work...

Change-Id: Iaaa563b8dc6848506a5f8e186248f4f6fb9c80e9

Change-Id: Iffb09b9145b72a1ac8710e41a0a7d7ad3abe5c33

Reuse some formerly static functions in docnum.cxx

Change-Id: I8d6bcca28a4f8a6e197a1c302acdf22ce6de01ff

SwCursorShell::LeftRight() and SetCursor().

Change-Id: I71cd382f06a4050c4676864ac50fb268cd099226

Change-Id: I4576a464b37e786178a3cb31a0be922f1b55db88