The nAlphaTo parameter was actually transparency. Change the
implementation and the call sites to actually use alpha.
Also remove one of the calls in Graphic::colorChange, because if the
BitmapEx has no alpha channel, the call was going to do nothing anyway.
Change-Id: I0bf27835b62596ac7c497c8606ceba04fcf859a1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145205
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
...as there were no calls to its OAccessibleImplementationAccess::getUnoTunnelId
so no code apparently made use of that XUnoTunnel
Change-Id: I35d392999f1c28e81f7ff4bf395f63642eb17bce
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145250
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
There was LOK API to insert a new section with provided HTML content and
to query it, but there was no LOK API to update such created section
with new names/contents.
This is needed in case Zotero wants to store citations with refmarks, in
which case it wants to store the bibliography with sections.
Introduce a .uno:UpdateSections UNO command that can do this: the
sections will be renamed if necessary & the content will be updated. The
content update is reasonably straightforward, because the section always
contains at least one empty paragraph, so there is no danger in simply
deleting the old content before inserting the new one.
This is similar to babba47239 (sw,
UpdateFields: add new TypeName, NamePrefix and Fields parameters,
2023-01-04), but that was for refmarks / citations, this is for sections
/ bibliography.
Change-Id: Idde6d5ed1b0f25f40df502bb6b7e7ca590ef9151
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145249
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
If cell width is not given or zero lets not confuse it with too
narrow cell. Lets keep zero to allow domain mapper define some
width.
This patch does not solves autofit issue completely, but finally
tables with autofit are visible somehow and do not contain cells
with minimal almost invisible width (41).
Change-Id: I640228ed117f1d65dde9753a00b8403dda45dd40
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145148
Tested-by: Jenkins
Reviewed-by: Vasily Melenchuk <vasily.melenchuk@cib.de>
The problem was that the LOK API client wanted to change the name of the
bookmark, but nothing happened, because we had a sanity check that made
sure the provided vs existing bookmark names match.
The intention was that the bookmark name is just an ID (you can store
additional properties in per-doc custom properties, those can change),
so we can require that bookmarks never get renamed when using this
command.
Fix the problem by dropping this requirement and rather use renameMark()
to adjust the in-doc bookmark name if needed.
This also handles undo/redo for us.
Change-Id: I7a426c865ddf24f13741c0369fb3991da2e881b2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145235
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
There was no LOK API to get a list of all sections where the name
matches a certain prefix.
This is useful in case the API client wants to know what previously
inserted sections were deleted by the user as part of deleting text
content.
Add a new getCommandValues(".uno:Sections") that returns the names of
matching sections. Do not return the section text, assuming that would
be updated by the API client anyway.
In practice this is needed by Zotero in case it wants to model its
bibliography items with sections.
Change-Id: If4f02d2a27f2328020934b319d30561aeaaf6612
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145207
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
It's created by SwFrameDlg and this is one of the tabpages which are
only there for images, not for frames.
Change-Id: I1f3740e683413a2dad2bb124c843b170b45c90e4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145242
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
This reverts the change of A.I.a.i. -> A.I.a.i.1.
made by commit dd8ed1fdbc.
While it wasn't necessarily a bad change, it no longer has
any justification, since the complementary change in the
other outline definition was modified to become MLA compliant.
Rather than make any change at all, just revert back to the
original definition, and leave the level definition to the
normal 4 levels of numbering before the bullets start.
Change-Id: I88d792a6dd20bd519755349905975cf9028f9fa6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145236
Tested-by: Jenkins
Reviewed-by: Justin Luth <jluth@mail.com>
Change 1. a) -> 1) a) 1) A)
This is intended as a clean, modern style of outline
that has good visual spacing in LibreOffice.
This differentiates it a bit more from the first option,
and extends the numbered level to the typical depth of 4,
while still keeping a logical suffix pattern
and maintaining the general "theme" of the previous definition.
Yes, it is a bit odd to have a capital A following
a smaller case a. However, this is at the fourth
level, which isn't reached very often,
and I wanted to keep a choice that focuses on a small a.
Surprisingly, with the default 12pt/Liberation font, a
10) still fits in the allotted space before the tabstop,
so in most cases this outline will have proper visual spacing
all the way up to 99).
There are plenty of choices that start with 1 followed by a dot,
so I used this opportunity to create the only list
that starts with a ")" suffix. Google docs for example has a
1) a) i) choice, but for modern lists I don't think
we want to use roman numerals.
Change-Id: I34725d6aa113831eaeaa88c295b3b3f1294f8998
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145155
Tested-by: Jenkins
Reviewed-by: Justin Luth <jluth@mail.com>
As the function name says, it's the Excel compatibility name and
not the Calc UI visible name and thus does not have the _ADD or
_EXCEL2003 suffixes. Theoretically the names may even be
completely different.
Change-Id: I5f52dee77009123ebc3762defe08c937a3a59a18
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145246
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins
If you don't read the commit message or the code, it wasn't really
obvious that FN_UPDATE_TEXT_FORMFIELD works based on the current cursor.
Change-Id: Icd8bc856e5d50b995fa92559c873a50b48b98ba8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145189
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
...after 21e0d8162a "loplugin:unusedfields" had
dropped the base class
OAccessibleImplementationAccess::implGetForeignControlledParent
Change-Id: I753aad41baaa0ac1d1f7eb0a6084993df5a58d5f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145197
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Change 1.(a).i.A. -> A.1.a.1.
This is intended as a clean, modern style of outline
that has good visual spacing in LibreOffice.
It provides a simple alternative to the second style,
reversing the order while still keeping a logical suffix pattern.
This design will easily allow large/long lists
to fit in the allotted space before the tabstop.
There was no attempt to match the theme of the
previous definition, since it neither looked
good in LO's default spacing, nor was it consistent/logical.
Change-Id: If7de1302110a3c97922bd2eeba7a917444110b27
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145156
Tested-by: Jenkins
Reviewed-by: Justin Luth <jluth@mail.com>
Reviewed-by: Vernon, Stuart Foote <vsfoote@libreoffice.org>
Changed I.A.i.a) -> I.A.1.a.i.
The overwhelming result of a google search for
"official/correct outline list format"
indicates that the third level should be a 1, not an i.
MLA Handbook (according to wikipedia) uses the order
I.A.1.a.i. (a) (1) i)
Chicago Mode of Style is usually interpreted as
I.A.1.a) or I.A.1.a. with varying lower sublevels/suffixes.
Wikipedia also suggest the logical order of
I.A.1.a.i.I) A) 1) a) i) (I) (A) (1) (a) (i)
Change-Id: I7fcc8a1ee727d6cddca4bf0c6a8461538769cf5d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145147
Tested-by: Justin Luth <jluth@mail.com>
Reviewed-by: Justin Luth <jluth@mail.com>
When Zotero uses refmarks to represent citations, then a section is used
for the bibliography. While it was possible to insert refmarks with a
given name & content, inserting a section with a provided HTML content
was not possible.
A section (and not a refmark) is needed for the bibliography, as a
refmark is a text attribute, so it must start and end in the same
paragraph.
Fix this by adding a new, optional Content parameter to
.uno:InsertSection, which allows providing multi-paragraph, formatted
text.
This is somewhat similar to what commit
fa82e151d8 (sw, .uno:InsertBookmark: add a
new BookmarkText parameter and accept HTML there, 2022-12-06) did for
the bookmark case.
Change-Id: I0ccbc8a765fed4e019783d0b743c1cd2d0596093
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145188
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
* Update translations from branch 'master'
to fd1416ef15091d8b08ea7dbad77148e68e4d89f9
- update translations for 7.5.0 rc2 / master
and force-fix errors using pocheck
Change-Id: Ia0f91e00129cb802164fd4323b3072cf8a76cb5c
This was using the order used by GetItemSurrogates() which can be hard to predict
So it used to update field in uncertain ways
Now it updates field in order of the fields in file
Change-Id: I7fc39b476cae6ed6a7fb7186803d4205832d7a73
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145167
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Andras Timar <andras.timar@collabora.com>
(cherry picked from commit 83f5340538d8aa8c469de3fee6012e83f8b56317)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145129
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Table shift based on cell lect margin should be applied
for DOCX (earlier compat options) but not actual for RTF.
Maybe earlier RTF version did behave this way, but nowadays
this behavior does not match MS Word 365.
Change-Id: Icdc4fa6298167fe5f263c85164d7c4c4176be25f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145088
Tested-by: Jenkins
Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de>
Switched the fixed and variable/current field
See also I18136feebed225f7645b6869d3600ee78f945bfb and
I56f42b4ce104dbb96d8fda1bc71af727fa46d536
Change-Id: Id885c09410dc9aecc9570c8d3b7750e9d69e9c9c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145170
Tested-by: Jenkins
Reviewed-by: Jim Raykowski <raykowj@gmail.com>
This patch makes the cursor position to be the same after undo redo as
when outline folding is not active.
Change-Id: I9e1f827455afb2675f3085c1560b1e2a246f4524
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/144873
Tested-by: Jenkins
Reviewed-by: Jim Raykowski <raykowj@gmail.com>
we already know the IsAlpha is true from the outer if condition
Change-Id: I627a9b952cb63154baf773b6162b422de4f253bc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/145157
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
