Delaying flushing until the dispatch level returns to zero causes
scrolling via the scrollwheel or mouse drag to appear laggy and
jerky.
Change-Id: Ib70e7766435baa765dd0d1d704ba2fac87f7fccc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158853
Tested-by: Jenkins
Reviewed-by: Patrick Luby <plubius@neooffice.org>
I can't explain why inverting using Skia causes this bug on
macOS but not other platforms. My guess is that Skia on macOS
is sharing some data when different SkiaSalBitmap instances
are created from the same OutputDevice. So, mark this
SkiaSalBitmap instance's image as immutable so that successive
inversions are done with buffered bitmap data instead of Skia.
Change-Id: I8acf90561c48edba14a5f43d16f375f15f25820c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158880
Reviewed-by: Patrick Luby <plubius@neooffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Jenkins
<https://lists.freedesktop.org/archives/libreoffice/2023-November/091151.html>
"CppunitTest_stoc_uriproc failed on Windows" reports that
translateToExternal("file:///abc/%feef") produces an empty string (indicating
failure) instead of "file:///abc/%FEef" (as expected in
stoc/test/uriproc/test_uriproc.cxx) when osl_getThreadTextEncoding() is Shift
JIS.
This was due to how the call to rtl::Uri::encode in
Translator::translateToExternal (in
stoc/source/uriproc/ExternalUriReferenceTranslator.cxx) behaved: It internally
interpreted its input "%FE" as the single-byte Shift JIS character 0xFE. Which
gets mapped to U+2122 as an extension (see "APPLE additions over SJIS, we
convert this like Apple, because I think, this gives better result, then [sic]
we take a replacement char" in sal/textenc/tcvtjp6.tab) in readUcs4, but which
in turn doesn't get mapped back to any Shift JIS character in writeEscapeChar.
Translator::translateToExternal is the only user of
rtl_UriEncodeStrictKeepEscapes, as introduced by
6ff5d3341d "INTEGRATION: CWS c07v013_SRC680
(1.4.40); FILE MERGED: 2007/06/21 13:00:56 sb 1.4.40.1: #b6550116# Made
XExternalUriReferenceTranslator.translateToExternal more robust when the input
URL contains spurious non--UTF-8 octets like %FE (which are now copied verbatim,
instead of signalling error)."
To make the claim true that such "spurious non--UTF-8 octets like %FE" are
always "copied verbatim", regardless of text encoding being used, repurpose
rtl_UriEncodeStrictKeepEscapes to always treat any escape sequences that are
present as (potentially broken) UTF-8.
Change-Id: I0fa0b14d3e3d44e4b5514e1b73c84c407a947ce9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158888
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
which calls into the kernel to get elapsed time, instead of using the
VDSO-based CLOCK_MONOTONIC (or at least, the C++ equivalent of that)
, which is much faster
Change-Id: I99d958b0ab64b75360db4e0c8a951c37af2505bc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158809
Tested-by: Jenkins
Tested-by: Caolán McNamara <caolan.mcnamara@collabora.com>
Reviewed-by: Caolán McNamara <caolan.mcnamara@collabora.com>
SetTextString will emit one, so skip doing that if we are going to be
followed by another emission that will override the earlier one.
Change-Id: I82d804d57cd9647631bd338a3a0c4dd11989b9fc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158892
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolan.mcnamara@collabora.com>
There seems to be no reason why SE child OBJR dictionaries are separate
objects, they could just be inline.
Change-Id: I9e4fcdf56d9b7454325f9d3ef7ba55bdc090f948
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158838
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
Tested-by: Jenkins
commit 004512d87a
notebookbar: allow to create multiple instances for online
Changed SfxNotebookBar::StateMethod so in LOK case it always
enter to the code block which is meant to be used for
creating or reloading notebookbar. Change that so we check
if notebookbar was initialized for given view already.
Change-Id: Icfa1e572769043509479b7b98d055120ac3d1f74
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158872
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Reviewed-by: Szymon Kłos <szymon.klos@collabora.com>
...which is used by
<95a4c0c835>
"[clang] Reland parenthesized aggregate init patches" for the
__cpp_aggregate_paren_init feature implemented since Clang 16, and which had
caused bogus warnings in patch set 1 of
<https://gerrit.libreoffice.org/c/core/+/158838/1> "tdf#157028 vcl: PDF export:
inline OBJR dictionaries"
Change-Id: Id4118ce8d53902388ca3a80ad03f12b59e3a35e6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158842
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
No longer report a red wavy underline for misspelled
words in Writer.
The underline isn't really a character/text attribute, since
it's not part of the actual formatting of the text, but is
just shown visually.
Each platform a11y API has a better way to explicitly
report misspelling.
commit ec2e02cfa4
Author: Michael Weghorn <m.weghorn@posteo.de>
Date: Tue Oct 17 14:36:58 2023 +0200
tdf#157696 a11y: Report spelling error via IA2 "invalid:spelling" attr
and
commit 2bf88c172c
Author: Michael Weghorn <m.weghorn@posteo.de>
Date: Wed Oct 18 13:44:42 2023 +0200
tdf#157696 wina11y: Switch from custom to IA2 text attributes
added the implementation for Windows and
the Qt-based VCL plugins on Linux.
On macOS (`NSAccessibilityMarkedMisspelledTextAttribute`/
`NSAccessibilityMisspelledTextAttribute`) and for gtk3/ATK,
the corresponding attributes were already used before.
Therefore, reporting an underline as a workaround is no
longer needed.
Orca on Linux, NVDA on Windows (with [1] in place)
and CrossOver on macOS all make use of the specific
character attribute to announce spelling errors, not
the underline. If any assistive technology should stop
reporting spelling errors due to this change, it should
be fixed to handle the proper text attribute as well.
After switching to text attributes according to the
IAccessible2 specification on Windows (i.e. with,
the above-mentioned commits and the NVDA commit
to switch to IAccessible2 text attributes for LO [1]),
NVDA was starting to announce not only that a word is
misspelled, but an underline in addition, which is
no more the case now that the underlining is no more
reported.
[1] de4e8814f3
Change-Id: Ifa29e0080a4a039c5a119dd0debe5e924fa546ed
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158847
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Handling for `NSAccessibilityMisspelledTextAttribute` and the newer
`NSAccessibilityMarkedMisspelledTextAttribute` [1] has been added in
commit db0fa92637
Author: Boris Dušek <me@dusek.me>
Date: Sun Aug 11 16:37:48 2013 +0200
fdo#67980 - VoiceOver does not inform about misspelled text
and VoiceOver announces misspelled Writer text as such, so drop the
obsolete TODO.
[1] https://developer.apple.com/documentation/appkit/nsaccessibilitymarkedmisspelledtextattribute
Change-Id: Id8ae1ee660e87c06ac7f559b5cdf1b31e22aef90
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158840
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
vs-ide-integration takes tens of minutes on my system; and most
of the time, it didn't output anything to the terminal. Every
target handled by gbuildtojson takes ~2 s; with more than 870
targets, it makes sense to show the progress to avoid making an
impression of a hang.
The "GBJ" abbreviation (for "gbuildtojson") was not used in the
output before.
Change-Id: I71ab3ae59d1034f47140aad0dab6ed170140acb3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158846
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
No Functional Change intended.
Just some slight optimizing that leads to some nice flattening.
Change-Id: I46d20a56821f218973c728f51575b104e6020500
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158854
Tested-by: Jenkins
Reviewed-by: Justin Luth <jluth@mail.com>
Character properties dialog when the Highlighting tab Color page is
open and the color has not been changed
This is noticeable when character highlighting direct formatting is
already at paragraph level.
The expected result of opening the properties dialog, and immediately
pressing OK without changing anything (or alternatively, after
changing something there and then pressing Reset then OK), is no
changes in applied properties at all.
Inspiration for rework of the background tab page. With this patch
XATTR_FILL items need not be included in the InAttrs set, for
example, as part of this patch, the XATTR_FILLSTYLE, XATTR_FILLCOLOR
range is removed from the sw_CharDialog core set.
Change-Id: Ic2de53a29579c33820fc381d354a4afebe048a5b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158100
Tested-by: Jenkins
Reviewed-by: Jim Raykowski <raykowj@gmail.com>
Previous tracked changes of cycle case weren't removed, if
the selection doesn't end in a tracked change, i.e. sentences
with punctuation, or selections end in a capitalized word etc.
Fix also lost selection of multiple selected nodes.
Follow up to commit 4cf8de89ba
"tdf#157667 sw track changes: fix cycle case on multiple words",
commit dc748d7dbd
"tdf#141198 sw: fix cycle case with change tracking" and
commit 79435eb55e
"tdf#157988 sw track changes: fix cycle case on a selected word".
Change-Id: I24476c359200a25ca9bcd467c33b0ebeb1afb037
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158804
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
Shave some time off sorting field content entries for documents with
large number of fields by eliminating a second sort and the loop used
to determine if a second sort is done because a field is found in a
frame. Also corrects ordering of fields in nested frames which prior
to this patch are placed at the beginning of the field content
entries.
Change-Id: I5fbd67b6fb0ac1ef49a5df7721bb209d4cba4013
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158606
Tested-by: Jenkins
Reviewed-by: Jim Raykowski <raykowj@gmail.com>
when we drop them from our own font list, that way we don't get
suggestions for glyph fallback which we can't satisfy
Also prefer our application fonts to system fonts when they
share the same version number where they should be the same, but
more than once we have a system-side font with oddities.
Change-Id: I90bd3311e0f37bacd60d20e1c1a6769b551b8b76
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158841
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolan.mcnamara@collabora.com>
`PushButton::Click does not` toggle a `PushButton`
that is a toggle button. Therefore, call
`PushButton::Check` and `PushButton::Toggle` instead
when the acessible "press" action is performed on
a toggle button, which makes this work as expected.
The same is already done in the UITest code, see
`ButtonUIObject::execute`.
The originally rerported issue in tdf#112661 and
tdf#112662 was that there was no action available
for the "Templates" and "Recent Documents" toggle
buttons in the start center via the NSAccessibility
API on macOS at all. By now, the "press" action was
available, but performing the action (e.g. using
the Ctrl+CapsLock+Space keyboard shortcut for
VoiceOver) didn't have any effect.
The same was true when performing the action via
Accerciser using the AT-SPI Action interface when
using the qt6 VCL plugin on Linux.
With this change in place, toggling between
showing the templates and the recently used
documents in the start center works using that
action, just as it does when clicking on one
of the toggle buttons in the UI using the mouse.
For gtk3, which is using native GtkToggleButtons,
this was already working without this change in place.
Change-Id: Ie3f02ec914239e0718ca1bfb4ba701f0831bb16a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158807
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
* Update translations from branch 'master'
to 846a915406627d0663080fd22b26aa6a2f12c48e
- update translations for master
and force-fix errors using pocheck
Change-Id: I1ac45a9287f6a2b5f00788a065fd85e3da4548f4
its not used as a SVBT32, only as a little buffer
Change-Id: I7a6d364c68ae31520cbb7432da832d108a9daeff
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158787
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolan.mcnamara@collabora.com>
There must be one Link SE per Link Annotation, so ensure that a new one
is created for a new line.
(regression from commit 4c5283a3a1)
Change-Id: I2585d9e22a435d7716f48fec89a78149c129f71d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158775
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
This is a SwBidiPortion containing a nested SwLineLayout containing
SwLinePortions for the link or text, and only the SwBidiPortion had a
tag helper, so add one for the inner portion too. (Apparently this
never worked before.)
Change-Id: I803191dafe101a42e4dbe1c0167411c2f1f76a2f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/158773
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
