...by keeping the worker threads around by default instead of joining and later
re-spawning them; see 5b1df7709d "Document the
Emscripten threads issue". But lets be cautious and change the bJoin default
only for Emscripten.
Also, cap the thread pool size to MAX_CONCURRENCY=4 and increase to
-sPTHREAD_POOL_SIZE=6. In an experimental setup where the Emscripten build
starts up with the start center rather than a Writer document, that is just
enough to cover the thread-spawning needs when executing the start center
window's paint task: Four threads for the comphelper::ThreadPool (used from
within drawinglayer::convertToBitmapEx -> ... ->
BitmapBasicMorphologyFilter::filter), and one each for
configmgr::Components::WriteThread (see e8358d0a0f
"Keep around a single configmgr::Components::WriteThread instance") and
salhelper::TimerManager.
Change-Id: I5b1d0e9dc09f05fb3b679c8dc1c38ee45c61c0f8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/171004
Reviewed-by: Stephan Bergmann <stephan.bergmann@allotropia.de>
Tested-by: Jenkins
and
cid#1556287 Check of thread-shared field evades lock acquisition
(1) std::mutex can be initialised at load time, since it is a
single-word object, i.e. no need to use a static function local.
(2) std::mutex is really fast, so no need to use tricky double-checked
locking
Change-Id: Ieb2239908204baaaccfe9d1e0e2c8ed369d108fb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170833
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
It seems we must first retrieve the array of values from:
- officecfg::Office::Common::BulletsNumbering::DefaultBullets
- officecfg::Office::Common::BulletsNumbering::DefaultBulletsFonts
before retrieving the value by index.
Change-Id: I9cdf124f73c52f3027ae8e97a73e01e0efb73603
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/171003
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Change-Id: Ib0243889621663ddb9b6789ede7643d54f06041e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170988
Tested-by: Jenkins
Reviewed-by: Nagy Tibor <tibor.nagy.extern@allotropia.de>
* Update helpcontent2 from branch 'master'
to e35fb29cb092771840ff876bbb4350e7570e1627
- Related tdf#69192 better placeholder string for command line example
Change-Id: I0a151668cd213906f6a7e523eacac48da8e50b20
Reviewed-on: https://gerrit.libreoffice.org/c/help/+/171001
Reviewed-by: Olivier Hallot <olivier.hallot@libreoffice.org>
Tested-by: Jenkins
- "Certificate Path" dialog (select Mozilla profile for NSS)
- resizing broken since commit db9e1ef04d
- also increased default width to account for long filesystem paths
- "Digital Signatures" dialog
- "Select Certificate" dialog
- resizable=True is the default (Glade removes explicit defaults)
- Make long translations wrap instead of being cut.
Glade set expand=False and fill=True as defaults inside GtkBox.
Change-Id: I2d7719e83e5162d06d30e25b3e307fa23e0d605e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170694
Tested-by: Jenkins
Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de>
...instead of joining and later re-spawning one. This helps with the current
Emscripten setup (see 5b1df7709d "Document the
Emscripten threads issue"), and probably doesn't matter much either way on other
platforms (so just get it in unconditionally). (Renaming the delay_ member
variable to delayOrTerminate_, to make its overall role more clear.)
(I ran into this when trying to turn the Emscripten build from starting up with
a Writer document to starting up with the start center, and then manually
opening a new Writer document. The resulting different usage scheme happened to
exhaust our current -sPTHREAD_POOL_SIZE=4 pool quickly, so this is one of
multiple commits to address that.)
Change-Id: I1bd28604b4640d2afad982bfd82b1acee8200e26
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170993
Reviewed-by: Stephan Bergmann <stephan.bergmann@allotropia.de>
Tested-by: Jenkins
Despite that we bailed out when both old and new value set,
gcc-13 cannot see that.
Change-Id: Ie727a14bb29a4b1cc304d2ce077c62f72cece19f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170973
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
when adding a change in the document (Track changes mode), round 4
Addresses "situation 4" when a new change makes a change to an entire
previous change or to multiple previously tracked changes.
Change-Id: I02175db6c7bc1258f8e7c05835b05ff79b7d7b83
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170984
Reviewed-by: Jim Raykowski <raykowj@gmail.com>
Tested-by: Jenkins
::pData type
Casting a void* RedlineData::pDate to either a SwRedlineDataParent*
or SwRedlineDataChild* works because of the order and variable types
declared in the SwRedlineDataChild and SwRedlineDataParent classes
match until an OUString is declared in SwRedlineDataParent. That
explains why the non-patched code behaves as expected even though
testing if the entry has children to determine which cast to do is
flawed. Better is to test the iterator depth which is what this patch
does.
Change-Id: I24ca0990a4edbd50a7e3dc5d0be4c5312c240921
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170860
Tested-by: Jenkins
Reviewed-by: Jim Raykowski <raykowj@gmail.com>
for all the Text field made a global define
of the NonBreakSpaces mask
Change-Id: I7107f941ee9008415325aefe5487555d91dbf1ab
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170596
Tested-by: Gabor Kelemen <gabor.kelemen.extern@allotropia.de>
Tested-by: Jenkins
Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de>
That code had been added with 9fa4eff9be
"profilesafe: Enhancements to BackupFileHelper", but it looks dubious:
For one, I cannot find proof for the claim that MSVC would run destructors of
static objects upon _exit. Building a simple C++ test program
> #include <iostream>
> #include <stdlib.h>
> struct S { ~S() { std::cerr << "hello\n"; } };
> S s;
> void f() { _exit(1); }
> int main() { f(); }
with contemporary MSVC 17.10 and executing it shows no "hello" stderr output
(see
<https://gcc.godbolt.org/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:c%2B%2B,selection:(endColumn:13,endLineNumber:5,positionColumn:13,positionLineNumber:5,selectionStartColumn:13,selectionStartLineNumber:5,startColumn:13,startLineNumber:5),source:'%23include+%3Ciostream%3E%0A%23include+%3Cstdlib.h%3E%0Astruct+S+%7B+~S()+%7B+std::cerr+%3C%3C+%22hello%5Cn%22%3B+%7D+%7D%3B%0AS+s%3B%0Avoid+f()+%7B+_exit(1)%3B+%7D%0Aint+main()+%7B+f()%3B+%7D%0A'),l:'5',n:'1',o:'C%2B%2B+source+%231',t:'0')),k:32.19670284753087,l:'4',m:100,n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:vcpp_v19_40_VS17_10_x64,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'0',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:1,lang:c%2B%2B,libs:!(),options:'',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+x64+msvc+v19.40+VS17.10+(Editor+%231)',t:'0'),(h:output,i:(compilerName:'x86-64+clang+18.1.0',editorid:1,fontScale:14,fontUsePx:'0',j:1,wrap:'1'),l:'5',n:'0',o:'Output+of+x64+msvc+v19.40+VS17.10+(Compiler+%231)',t:'0')),k:67.80329715246913,l:'4',m:100.00000000000001,n:'0',o:'',s:1,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4>).
If there indeed were issues with MSVC and non-standard _exit, a solution might
be to use standard _Exit instead (in FatalError in desktop/source/app/app.cxx).
And for another, when bExitWasCalled was true in
configmgr::Components::~Components, it would still have called
writeThread_->join() (i.e., it would potentially still have called
writeModFile from within configmgr::Components::WriteThread::execute()), but (a)
only after waiting out its full
> delay_.wait(std::chrono::seconds(1));
(because it missed to call configmgr::Components::WriteThread::flush, which
calls delay_.set()), and (b) potentially running into a deadlock because it
called writeThread_->join() with *lock_ locked, which
configmgr::Components::WriteThread::execute wants to lock too.
(I came across this because it got into my way when trying to adapt the
lifecycle of configmgr::Components::WriteThread to the needs of the Emscripten
build.)
Change-Id: Icb4ebf00d1d316b80c29f7bd91b86614ef4ac430
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170940
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <stephan.bergmann@allotropia.de>
One place that was missed was when moving from one content control
to another (for example, using the tab key).
Change-Id: I62a3af2a5a7036aaac605d592c244a58f5b65a12
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170967
Reviewed-by: Justin Luth <jluth@mail.com>
Tested-by: Jenkins
This makes the "sentence" drawing area from the spelling
dialog (cui/uiconfig/ui/spellingdialog.ui) show up with
role panel instead of "unknown" in Accerciser when using
the qt6 VCL plugin.
This is the same as with gtk3, where the `TEXT_FRAME` role
is mapped to `ATK_ROLE_PANEL`.
Change-Id: I41133c51c64749ccd808cb675fe647a0daefac33
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170944
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
Up to now these needed to be decomposed to be
rendered and did not support animation. I already
added animation, but that was a compromize.
Done that now in the correct way, so that animation
is possible and can be directly rendered using
a FillGraphicPrimitive2D if wanted, including one
extra layer of alpha, too.
This will also enhance other primitive processors,
but of course most those that do handle that now
directly, so CairoPixelProcessor2D does that.
Note that it does not need to support animation
in any way (that is part of the primitive stack),
but just direct tiled rendering.
Change-Id: I8860098dfb3f2369e361579cf1deea7c67b662b9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170937
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@me.com>
we had an ODR violation with a constant named PROPERTY_DATASOURCE which
is also defined inside dbaccess
regression from
commit bacb92275c
"add dbu to --enable-mergelibs=more"
Change-Id: I315ba82e0bccc49e01bd6a438b7e671b971410dc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170939
Tested-by: Xisco Fauli <xiscofauli@libreoffice.org>
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
This change fixes a vertical text glyph metrics regression that was
introduced while migrating PDF export to use metrics from HarfBuzz. The
root cause was incorrectly caching CJK glyph heights in place of widths
for vertical text.
Change-Id: I8426fc902658d1c045b42e760028f26b09eeef93
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170935
Tested-by: Jenkins
Reviewed-by: Jonathan Clark <jonathan@libreoffice.org>
commit 24103bdf3f
Author: Michael Weghorn <m.weghorn@posteo.de>
Date: Thu Jul 18 08:58:10 2024 +0200
tdf#161501 icon choice ctrl: Use tab text colors
Instead of using the generic window or highlight
text color, use the specific text colors for tabs
when painting entries of the vertical tab control,
similar to how it's done for the non-vertical
tab control (see `TabControl::ImplDrawItem`).
(...)
had added code to set the font color for the different
states of a vertical tab control item, but as this
was in the `if (bSelected)` block, only the one
for selection was actually used, so the other cases
were consequently removed in
commit a4befb29f3
Author: Caolán McNamara <caolan.mcnamara@collabora.com>
Date: Sat Jul 20 19:20:13 2024 +0100
cid#1610738 Logically dead code
again.
Thanks to Chris Sherlock for pointing this out!
Reintroduce setting colors, but drop the `bSelected`
condition altogether. Drop explicitly setting a fill
color, which the implementation for the non-vertical
tab control in `TabControl::ImplDrawItem` also doesn't
do and which would e.g. result in an odd grey background
for non-selected entries when used unconditionally with
the gen VCL plugin.
I see no visual difference with or without this commit
in place on Windows or for the gen VCL plugin or the kf5 VCL
plugin with the Breeze style on Linux, but other theming/styles
could depend on according colors being set.
Change-Id: Iccf7170f2de04cead23607977ac8cf7acbc471f3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170926
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
When the mouse leaves the vertical tab bar,
unset the entry to highlight as mouse-hovered,
as the previously one no longer is when the
mouse is outside the area of the tab bar.
This fixes the issue of the previously
mouse-hovered entry still being painted
with mouse-hover feedback when slowly moving the
mouse to the right away from an entry in the
"Insert" -> "Page Style" dialog with gen or kf5
at least, as shown in attachment 195467
of tdf#161501.
Change-Id: I646cab54f5031ed6a8aa88d4a162bb3a2456eec4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170923
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
It is always set, so no need to have an
icon choice control specific window bit to
specify whether or not to highlight the
currently mouse-hovered entry. Just always
do it.
Change-Id: Ib259b6b1576e1968221c4f98661a3a93e600c93b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170922
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
While GetIsDummyParaAddedForTableInSection() is true
for the entire section, it should only be used
against GetIsFirstParagraphInSection()
where the dummy para exists.
make CppunitTest_sw_ooxmlexport19 CPPUNIT_TEST_NAME=testTdf138093B
Change-Id: I457c0aa25eb7bd086c5c15bd889d9cd89c51db95
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170921
Reviewed-by: Justin Luth <jluth@mail.com>
Tested-by: Jenkins
This fixes a 24.2 regression
from commit 5082d50d24
writerfilter: use content controls for text in block SDTs
which was made even worse
with 24.2 commit 7c4dba1def.
Initially, the starting mark was made as soon as SdtContent was seen.
After that, you could get a w:tbl, or a w:tc which could trigger
the start of a table, and potentially add a dummy paragraph
which needed to be jumped over (at least for the first SDT,
but not for subsequent SDTs).
In the second patch, the starting mark was made just before
the text was appended, so all of the cell/table creation
has already been completed, so there is no more need to
jump over any dummy paragraphs, even for the first table.
Affected unit tests were
- table-start-2-sdt.docx
- sdt-company-multipara.docx
- paragraph-sdt.docx
make CppunitTest_sw_ooxmlfieldexport CPPUNIT_TEST_NAME=testParagraphSdt
make CppunitTest_sw_ooxmlfieldexport \
CPPUNIT_TEST_NAME=testTdf158661_blockSDT
Change-Id: I4a8359cec27f88f6a2fc0464aa240b523a279e42
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170920
Tested-by: Jenkins
Reviewed-by: Justin Luth <jluth@mail.com>
* Update translations from branch 'master'
to d3e4d5213d29f57c45e2199822f2a68df9a884f6
- update translations for 24.8.0 rc2
and force-fix errors using pocheck
Change-Id: If72ef1ab34b88547d6caa85763242e8f2a231015
Instead of duplicating the same code in both,
`QtGraphics::GetResolution` and
`QtSvpGraphics::GetResolution`, create a static
helper method `QtGraphicsBase::ImplGetResolution`
in their base class and call that.
While at it, also move `QtGraphics::GetResolution`
to the source file where all of the other
`QtGraphics` methods are implemented.
Change-Id: I721459e8b65756f214fee77ac4d3cb8e500f0b57
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170930
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Include X.509 certificates under Tools->Options->UserData
among the keys for signing.
Add a new checkbox to Save file dialog, to sign with that
selected key easily.
The checkbox is disabled if there's no matching key found.
Change-Id: I9fc16790c479819cd1f35bcad040d0ebc7c4bdef
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170619
Tested-by: Jenkins
Reviewed-by: Sarper Akdemir <sarper.akdemir@allotropia.de>
Don't modify the bound rect of tab items used used e.g.
for drawing the focus/selection/mouse-hover rectangles/indicators
in order to center the content horizontally and vertically.
Making the bound rect smaller in vertical direction
in `SvxIconChoiceCtrl_Impl::Center` resulted in margins
between entries for the non-native drawing path (e.g. the
gen VCL plugin on Linux) for the text-only case
(e.g. "Format" -> "Page Style" dialog in Writer).
Making sure that the content is centered should rather
happen in the drawing routine.
Horizontally centering already happens for the icon-case
anyway, as the `PAINTFLAG_VER_CENTERED` gets passed as flag
to the `PaintItem` call for the `WB_ICON` case in
`SvxIconChoiceCtrl_Impl::PaintEntry` anyway.
And for text-only mode, which was apparently the only one
for which centering vertically in `SvxIconChoiceCtrl_Impl::Center`
had an effect, vertically drawing text was implemented in the drawing
code path in
commit 180f0c1ec8
Author: Rafael Lima <rafael.palma.lima@gmail.com>
Date: Tue Jul 16 20:52:28 2024 +0200
tdf#161501 Improve visuals of selected entries in Vertical tabs
. This commit addresses the
> For the gen VCL plugin, there's still a small gap between
> text-only items (e.g. in the "Format" -> "Page Style" dialog
> which has a different cause and will be addressed in a separate
> commit.
aspect mentioned in earlier commit
Change-Id: Iac34098ef8d0f90f1ac0844df63839a6a99b83a7
Author: Michael Weghorn <m.weghorn@posteo.de>
Date: Mon Jul 22 11:39:48 2024 +0200
tdf#161501 icon choice ctrl: Drop extra space between items
Change-Id: I2077efc8ce1e97d52cf078d63bb4e4b528e389eb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/170895
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
