to ensure that the commit

Find actual data area inside the main-range...
24b1827309e7fe4d77108da5732f7f4eddad3ae6

does not change results by its optimizations.

Change-Id: I07113359cac9c653f692e84a7b79b06621142aaf
Reviewed-on: https://gerrit.libreoffice.org/64658
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>

Change-Id: I08a4494ca249dbbb444643052bcff1740e089e26
Reviewed-on: https://gerrit.libreoffice.org/64785
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>

...(according to
<http://www.open-std.org/jtc1/sc22/wg21/docs/cwg_defects.html#1579> it is fixed
in C++14), but for safety, leave the configure.ac check in for some longer.

Change-Id: Ibd2f0cac228117e35ac299e2fe74207394c900cd
Reviewed-on: https://gerrit.libreoffice.org/64773
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

...but for safety, leave the configure.ac check in for some longer.

Also remove now-redundant SAL_INLINE_VARIABLE again (which was
LIBO_INTERNAL_ONLY).

Change-Id: Id049e0cb84b4f97f5859f1b16b867b39b448dec0
Reviewed-on: https://gerrit.libreoffice.org/64772
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

...but for safety, leave the configure.ac check in for some longer.

Also, save removing now-redundant SAL_FALLTHROUGH for a follow-up commit.

Change-Id: I9bf42d237aea4c09459f28275568cf340e588607
Reviewed-on: https://gerrit.libreoffice.org/64770
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

...but for safety, leave the configure.ac check in for some longer.

o3tl::array_view::max_size (include/o3tl/array_view.hxx) and
o3tl::basic_string_view::max_size (include/o3tl/string_view.hxx) started to
produce loplugin:staticmethods warnings, which I silenced by /not/ making the
functions static.  Those classes are meant to be temporary drop-in replacements
for standard classes (std::span slated for C++20, prev. std::array_view; and
std::basic_string_view, resp.), so should have the same behavior as their
standard counterparts (and making the functions static would likely cause
loplugin:staticaccess warnings at call sites).

Change-Id: If21674dbf02886f453ca447544e37b184df5a25e
Reviewed-on: https://gerrit.libreoffice.org/64768
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

so for a quiet life arrange it like that again. https://xkcd.com/1172/

Change-Id: I8d1f3a914f8792b8c91ccd1b98d1fbd2a25a4213
Reviewed-on: https://gerrit.libreoffice.org/64777Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

* Update helpcontent2 from branch 'master'
  - Update links for missing Colibre icons
    
    Change-Id: I827b755259ac7dc8de9a706dc111fd021be22306
    Reviewed-on: https://gerrit.libreoffice.org/64783
    Tested-by: Jenkins
Reviewed-by: Olivier Hallot <olivier.hallot@libreoffice.org>
    

Change-Id: Idc4dc73e61b828f989e6c8b46a6fc1fa43253ae6
Reviewed-on: https://gerrit.libreoffice.org/64740
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>

Change-Id: I797b13c3ba76869e67c12832402798ab2b65596a
Reviewed-on: https://gerrit.libreoffice.org/64776
Tested-by: Jenkins
Reviewed-by: Zdenek Crhonek <zcrhonek@gmail.com>

Change-Id: Ia82e6e5c4d6c3e78ba5fbc320d299acf3800a72e
Reviewed-on: https://gerrit.libreoffice.org/64766Reviewed-by: andreas_kainz <kainz.a@gmail.com>
Tested-by: andreas_kainz <kainz.a@gmail.com>
Tested-by: Jenkins
Reviewed-by: Xisco Faulí <xiscofauli@libreoffice.org>

Change-Id: I118039f4e47818a43da9302ce3550ea800522675
Reviewed-on: https://gerrit.libreoffice.org/64765Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Jenkins

...but for safety, leave the configure.ac check in for some longer.

Change-Id: Ife94cdfd56696edb113e32d84f563dd805e757e5
Reviewed-on: https://gerrit.libreoffice.org/64769
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

Follow-up of commit 74816dd6

Change-Id: I246df80586c8fc60efc68b5deeb865cbfcd0da13
Reviewed-on: https://gerrit.libreoffice.org/64762
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>

and trim all ranges to match this actual data area. Don't do
this optimization for COUNTIFS where there is no main-range.
This optimization is also turned off if any of the parameter
ranges are not double-refs.

Benefits in cases like -
=SUMIFS(A:A, B:B, ">=20", C:C, "<=10") and the is data only
in say A1:A10000 and rest are empty. Range trimming in this
case saves lot of execution time and memory (for vConditions
and vRefArrayConditions).

Change-Id: I6b4ad91e23f89dc48603b98000bcef4dbdb03112
Reviewed-on: https://gerrit.libreoffice.org/64657
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>

...now that /Zc:__cplusplus is available in Visual Studio 2017 version 15.7
(see <https://blogs.msdn.microsoft.com/vcblog/2018/04/09/
msvc-now-correctly-reports-__cplusplus/>).

Some external projects might run into issues when picking up /Zc:__cplusplus
directly with $(CXXFLAGS_CXX11) or indirectly via $(gb_CXXFLAGS) now, but that
appears not to be the case.

Some obsolete MSVC-specific __cplusplus checks can be removed now.  (The ones in
external/libebook/libebook-msvc.patch.1 pick up /Zc:__cplusplus via
$(gb_CXXFLAGS) in external/libebook/ExternalProject_libebook.mk.)

Change-Id: Idc6849a0000ea424522f30f61caba112fae25d40
Reviewed-on: https://gerrit.libreoffice.org/64755
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

Change-Id: I924034417bcb19a6e535f8dc427c63ff11356b23

This reverts commit 23339ec7, which is no longer
needed after TODO.

...fixing <https://github.com/flathub/org.libreoffice.LibreOffice/issues/54>
"Build against org.freedesktop.Platform 18.08"

Change-Id: I70448277ce326075435932c75a837bd7be3d4508

...which are no longer used by the LO build since
0ce173b5 "tdf#103080 Dont package Open Sans and
PT_Serif fonts"

Change-Id: I9ece7a85fe0003711eb23f332e617b780e6701cd

Change-Id: I8797ccc165fb0a6060c6232babc299114411130a
Reviewed-on: https://gerrit.libreoffice.org/64707
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Now this list is used also in rules for capitalizing first letter of
every sentence, and for correcting accidental use of Caps Lock key.

Change-Id: I4fbdbccbae32743d5e9a6757344361eefd88e167
Reviewed-on: https://gerrit.libreoffice.org/64759
Tested-by: Jenkins
Reviewed-by: Heiko Tietze <tietze.heiko@gmail.com>
Reviewed-by: Sophie Gautier <gautier.sophie@gmail.com>
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>

Respecting the axis direction in case of
normal/stacked/percent stacked Bar chart and
the legend names will be in the right order.

Change-Id: If782393a33e48dae32f919d137e1d1148a85b0b0
Reviewed-on: https://gerrit.libreoffice.org/64632
Tested-by: Jenkins
Reviewed-by: Bartosz Kosiorek <gang65@poczta.onet.pl>

...which is the latest 9.2.x currently listed at
<https://www.postgresql.org/ftp/source/>.  9.2.1 doesn't build against
OpenSSL 1.1 which dropped SSL_library_init (cf. <https://wiki.openssl.org/
index.php/Library_Initialization#libssl_Initialization>), and 9.2.24 apparently
has that covered.  (Ran into this when trying to upgrade the LibreOffice flatpak
build to org.freedesktop.Sdk//18.08, which has OpenSSL 1.1.)

On Windows, the new tarball as-is fails with

> ..\..\port\chklocale.c(214): error C2037: left of 'lc_codepage' specifies undefined struct/union '__crt_locale_data'

because at least in Windows Kits/10/Include/10.0.17763.0/ucrt/corecrt.h
(included from Windows Kits/10/Include/10.0.17763.0/ucrt/locale.h), the relevant
definitions are now

> typedef struct __crt_locale_data_public
> {
>       unsigned short const* _locale_pctype;
>     _Field_range_(1, 2) int _locale_mb_cur_max;
>                unsigned int _locale_lc_codepage;
> } __crt_locale_data_public;
>
> typedef struct __crt_locale_pointers
> {
>     struct __crt_locale_data*    locinfo;
>     struct __crt_multibyte_data* mbcinfo;
> } __crt_locale_pointers;
>
> typedef __crt_locale_pointers* _locale_t;

which presumably has changed from a past state where that lc_codepage member was
directly publicly accessible.

<https://dev-www.libreoffice.org/src/postgresql-9.2.24.tar.bz2> is a copy of
<https://ftp.postgresql.org/pub/source/v9.2.24/postgresql-9.2.24.tar.bz2>;
`sha256sum postgresql-9.2.24.tar.bz2` reports the same
a754c02f7051c2f21e52f8669a421b50485afcde9a581674d6106326b189d126 as recorded in
<https://ftp.postgresql.org/pub/source/v9.2.24/postgresql-9.2.24.tar.bz2.sha256>

Change-Id: I196dd93aa03471042efba57ea639e1bb6655de98
Reviewed-on: https://gerrit.libreoffice.org/64730
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

If the button has this property, it means it is located in the action
area, not that it is a stock button per se.

Change-Id: Ia57ad24d8cd4ff1d7b74d7ef82da0f352fe3deb3

Change-Id: I75ea7b2d407d3acfd3094f17509c89240c48f50a

Change-Id: Ia7d0a573c1902fba3fc1e69f425bbf2dc93a1c82

The difference between Impress and Writer is the used
unit. We can assume that SID_ATTR_TRANSFORM_ROT_X and
SID_ATTR_TRANSFORM_ROT_Y contains the values in twips.

Change-Id: I44ba9532348200742b2c7c369b89eb2545295684

Dialogs routed to the LOK client have limited real
estate on the screen because dialogs are rendered
on a canvas. When the listbox drop-down bleeds outside
the dialog area, on the desktop it's no problem, but
in the browser, that requires a larger canvas and
that brings a host of problems.

Here we limit the maximum length of listbox/combobox
drop-down lists to avoid this situation.

This would ideally be configured via ListBoxMaximumLineCount
alas it doesn't work.

Change-Id: I55fff28dbfc59dec36714e221f76cf4160ba1505

Change-Id: I74668ca8079d8d56b8c3f2c4513e2a2ec95872c6

Change-Id: I134f86dc0c49ddf26e7a3cfe790c49ab200de9e9

Change-Id: Ib30be297d87fa40c03974e0a97f243812dc488f3

Change-Id: I182aa395ce0f4e4185c9c85a866dec89499842f9

Change-Id: I6d0ce089be051667e965aff45a63f6b1a8342952

Change-Id: Ib22ba6956afedf1eea055e0ac3a78c53b4ee5861

Unlike C++, C doesn't allow reference-to-pointer types,
and we do have C code that wouldn't compile with ref-to-ptr.
Had to change to ptr-to-ptr, which is the proper way of
having output arrays.

For the same reason, we cannot use new/delete, rather we
must use malloc/free.

Another (lesser) issue was that we used the renderShapeSelection
API to echo back an array we give it as prefix. This made
the API unecessarily complex (in undocumented ways) and
forced the implementation to both worry about user-data
and managing the input memory. This logic is best moved
to the client and the API simply returns the output data.

Speaking of returning data, the API now returns the size
of the array it allocated and wrote to, so the client
can do a simple check on the return value directly.

Change-Id: Ida216c10d5b37efd1e0861e26b72cabb25c568e6

Change-Id: Ia09f6ab40c20c17230e8b544987a5d11b7359f2f

Change-Id: Ic1b30a89ae513b3b3ec71853830ef7600801dee0

Change-Id: I6d9e110f48b0f4c6226503d01e10d099cfcb130c

It works for Impress only now.

Change-Id: I95e3e37ae7df49b567108f6d6467038b715e886d