* Update helpcontent2 from branch 'master'
to 12e13425715a23ff426715e0bb7248d4aea55be9
- tdf#129097 Fix inconsistent names for Name Box/Sheet Area Box
In text/scalc/01/04070100.xhp
- Added variable tags to H1 to allow the page to be more easily linked from other pages
- Updated reference to "Sheet Area" to read "Name Box" instead
In text/scalc/02/06010000.xhp
- Added index bookmarks to make the page easier to find
- Added a hid bookmark so that pressing F1 opens this page when the Name Box is selected in Calc
- Added variable tags to H1 to allow the page to be more easily linked from other pages
- Added variable tags to name in syname_box so it can be embedded in other pages. Changes to the menu option will update the name in linked pages.
- Added a Related Topics Section
In source/text/scalc/04/01020000.xhp
- Replaced reference to "Sheet Area" with an embedded link to text/scalc/02/06010000.xhp
In source/text/scalc/main0206.xhp
- Replaced reference to "Sheet Area" with an embedded link to text/scalc/02/06010000.xhp
Change-Id: Iddd26877397f3ffcf7ee38b95c38550f1a216d87
Reviewed-on: https://gerrit.libreoffice.org/c/help/+/167947
Reviewed-by: Olivier Hallot <olivier.hallot@libreoffice.org>
Tested-by: Jenkins
For unclear reason, this shifts text on all tiles except the top left one.
I wasn't able to track where the coordinate is changed: it seems that the
pixel offset of the virtual device, its MapMode's origin, as well as the
start point passed to the DrawTextArray method, are all only changed by a
tiny amount, if at all, so it should be no more than a pixel off compared
to before commit cc3663bbae (tdf#160702:
improve text positioning, 2024-04-20). However, it is already more than
half the tile size (more than 128 pixel) offset in 75% scale case for the
second tile; and it's completely off the tile for all the rest (third+ in
a row), and for greater scales (100%+) even for second tile.
Change-Id: I64dc24bea4bab0cac90f11f2500bba0fd9bc7855
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168041
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
Tested-by: Jenkins
Found the following memory leaks using Xcode's Instruments application:
1. Posting an NSAccessibilityUIElementDestroyedNotification
notification causes [ AquaA11yWrapper isAccessibilityElement ]
to be called on the object so mark the object as disposed
before posting the destroyed notification and test for disposed
in all of the standard NSAccessibility selectors to prevent
any calls to likely disposed C++ accessibility objects.
2. In [ AquaA11yWrapper accessibilityHitTest: ],
[ AquaA11yFactory wrapperForAccessibleContext: ] already retains
the returned object so retaining it until the next call to this
selector can lead to a memory leak when dragging selected cells
in Calc to a new location. So autorelease the object so that
transient objects stay alive but not past the next clearing of
the autorelease pool.
3. [ AquaA11ySelectionWrapper selectedChildrenAttributeForElement: ] is
expected to return an autoreleased object.
4. [ AquaA11yFactory wrapperForAccessible: ] is not a getter. It
expects the caller to release the returned object.
5. CreateNSString() is not a getter. It expects the caller to
release the returned string.
Change-Id: I824740d7e3851b0c3e31e2c009860aa822c94222
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168034
Reviewed-by: Patrick Luby <guibomacdev@gmail.com>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Jenkins
The WakeupThread is an attempt to avoid needing to call gettimeofday
and/or update a visual status bar very regularly, and to have a thread
marking the passing of time to emit progress updates infrequently.
Re-factor this to have a single time-keeping thread, so it will be
easier to shutdown and re-start for LOK; and also to simplify some
of the complexity lurking here.
Change-Id: Ia95890e5d6041a029484aa3f7df13b59a0397086
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/167949
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Jenkins
Clear all `ScAccessibleCell` references that are owned by
`ScAccessibleSpreadsheet` when disposing the sheet instead
of when the sheet is deleted.
While `ScAccessibleSpreadsheet::disposing` is always called,
`ScAccessibleSpreadsheet::~ScAccessibleSpreadsheet` is not called
until all of the sheet's `ScAccessibleCell` instances have been
deleted. So make sure that `ScAccessibleSpreadsheet::disposing`
releases all `ScAccessibleCell` references.
This might possibly reduce some of the leaks mentioned in [1].
[1] https://gerrit.libreoffice.org/c/core/+/167961/comments/15a6a4e4_91a77291
Change-Id: Iaf14338945c1899d54c8e7f8a16f38a48316ac98
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/167985
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
shave 2% off time by inlining GetDoc which is quite hot
Change-Id: I20ede177e6032761598b1b977e90ff969bd2162a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168012
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Jenkins
A fixed-height cell can contain a bigger image, which is cropped by
cell boundaries. It was not possible to resize this image with a simple drag & drop, because its size changed to the cell size immediately.
Now it's possible, like MSO does.
Follow-up to commit 30de13743f
"tdf#160836 sw: resize rows at images cropped by row height".
Change-Id: I9d63e6e526135b6d8541f904c363b9186bb3cfc4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168023
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
takes the time from 22s to 20s for me
Change-Id: Ib401c03ba13f82047c8376741e3547aadf5b18df
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168011
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
In `QtPlayer::getPreferredPlayerWindowSize`, retrieve the
video resolution from the meta data and return that for the
preferred window size if available, instead of always returning
an empty rect.
For the sample presentation referenced in tdf#145735,
the size retrieved this way also matches what
`GtkPlayer::getPreferredPlayerWindowSize` returns
(720x400).
As mentioned in
commit 441d8ed9be
Author: Michael Weghorn <m.weghorn@posteo.de>
Date: Mon May 20 16:25:09 2024 +0200
tdf#145735 avmedia qt: Use QtMultimedia for Qt 6 media playback
, the behaviour without this was:
> * In non-presentation mode, the placeholder
> shown until the video gets started using the controls
> in the sidebar is just an "audio icon", not a frame
> from the actual video. (This might be related to the
> fact that `QtPlayer::createFrameGrabber` currently
> returns an empty reference.)
With this in place, the placeholder is now a different
icon, a "video icon" (but still not a frame from the actual
video).
Change-Id: I295abe6b7973fece3fcaf9a38af2602cb08610a7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168013
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Set all of the content margins of the layout that
holds the widget for video playback to 0.
Quoting [1]:
> Sets the left, top, right, and bottom margins to use around the layout.
>
> By default, QLayout uses the values provided by the style. On most platforms,
> the margin is 11 pixels in all directions.
This makes the odd margin/frame shown around the video
on video playback mentioned in
commit 441d8ed9be
Author: Michael Weghorn <m.weghorn@posteo.de>
Date: Mon May 20 16:25:09 2024 +0200
tdf#145735 avmedia qt: Use QtMultimedia for Qt 6 media playback
disappear.
[1] https://doc.qt.io/qt-6/qlayout.html#setContentsMargins
Change-Id: I29f1093caa986e6818fd575b031375ded0a572bd
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168010
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Make SlideSorterView update the current page properly when it is the
MainViewShell.
Also fixes Notes Pane not properly updating on SlideSorter
view mode.
Change-Id: If4fed4e84951cb4a1707874b288ac46e6d5c26da
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168022
Tested-by: Jenkins
Reviewed-by: Sarper Akdemir <sarper.akdemir@allotropia.de>
15:47 <vmiklos> noelgrandin: any objections to go further with your writerfilter rename and turn sw/qa/writerfilter/cppunittests/* into sw/qa/writerfilter/* ? (suggest to either do it now or not at all if we're past the libreoffice-24-8 branch-off)
15:59 <noelgrandin> vmiklos, not at all, go for it
Change-Id: Ic0eb6907463389f3be326cdf183002f1bfcb42be
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168024
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
* Special indicators such as "°" for no-break space but also
"," for zero width space are hidden if non-printable chars
are disabled
* Special indicators use the same color (NON_PRINTING_CHARACTER_COLOR)
* Hard hyphen becomes blueish with non-printable chars on
* Hard blank as well as hard hyphen are controlled by Formatting Aids >
Non-breaking spaces; and both highlighting and indicator are hidden
with non-printable chars off
Change-Id: Ifa29371587ba44d825e04368c5f5c5bf7ed94c7f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/167987
Tested-by: Jenkins
Reviewed-by: Heiko Tietze <heiko.tietze@documentfoundation.org>
The "early rejection test" removed here is a historical
artifact when UpdateForKeyPress was created.
in the current state it checked if
pOutlineViewShell->GetActualPage() was different from
itself.
In the OutlineViewShell::UpdatePreview, there's an
additional check if we have actually moved to a new page, so
we can just safely remove this.
Also fixes Notes Pane not properly updating on Outline View Mode
Change-Id: Ic9de2e7302e76260809f90360533f0bfe6e05d2e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168017
Reviewed-by: Sarper Akdemir <sarper.akdemir@allotropia.de>
Tested-by: Jenkins
Makes it so that notes pane handles some previously
incorrectly handled keys itself. e.g. PageUp & PageDown
Change-Id: I19f73fa2782900a752889544dcc79fc180ee52d4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168008
Reviewed-by: Sarper Akdemir <sarper.akdemir@allotropia.de>
Tested-by: Jenkins
Regression from commit 02c4350820 (sw keep
aspect ratio: add filter for this setting, 2021-06-07), once UI units
are set to poins (instead of cms), the image dialog for the bugdoc was
showing 5% width instead of 48%.
48% is roughtly correct, visually the image is taking half of the body
frame width. Previously the bad rel size didn't happen because we didn't
save the "keep aspect ratio" to documents, so it was off by the time the
dialog was initialized.
Fix the problem by introducing a new flag, so we can differentiate
between the user changine the width or height vs the dialog being
initialized. RelSizeClickHdl() is meant to adjust the other axis in the
user case, and this is not wanted in the init case.
A higher level fix would be to make sure once aspect ratio is kept,
that ratio is stored in documents explicitly, so we can say 50% wide
with e.g. 4:3 ratio, that would avoid all this trouble by even looking
at the calculated sizes when we want to work with percents. This storing
of the aspect ratio is not done here.
Change-Id: I901e7f6d5e6f7f1349d7beeb05985ddbf99a34a2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168015
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
See
<https://bug-attachments.documentfoundation.org/attachment.cgi?id=194324>,
Word has a longer separator line for the foot/endnote than Writer for
this bugdoc.
Writer defaults to 25% of the body frame width in the SwPageFootnoteInfo
ctor, and we don't seem to change that in the DOCX import. Word has a
static 2 inches setting, which is only reduced if it would go outside
the body frame.
Fix the problem by extending SwFootnoteContFrame::PaintLine() in the
DocumentSettingId::CONTINUOUS_ENDNOTES case to do the same.
I searched the OOXML spec and the MS implementer notes, they don't
specify this 2 inches length, but it seems static: the value doesn't
change with the page size. With this, the single-section bugdoc is now
rendered fine.
Change-Id: I3bb23680937580179b8d37c940ea14e0f80fc7f4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168009
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
Was previously changed the other way around in
commit 28c89e58d9
Date: Wed Aug 30 15:16:01 2023 +0200
use concrete type for OAccessibleMenuBaseComponent::m_aAccessibleChildren
for no apparent reason.
Change-Id: I0d0c6627134bcef8ac9732cc8a5213ab266afbf7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168005
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
... which is only used in the next line anyway.
Change-Id: I3bb038cfba3b3b31df45ca316242d54442db7220
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168004
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
As discussed in the ESC call of 2024-05-23,
CppunitTest_vcl_gtk3_a11y fails relatively often recently:
> 38 CppunitTest_vcl_gtk3_a11y gerrit_linux_clang_dbgutil
> + this week: have a new problem with CppunitTest_vcl_gtk3_a11y (Cloph)
> + failure is not new, but the frequency increased recently
> + assume: timing issue
Jenkins logs show that failures are due to running into
timeouts, example from [1]:
[...]
found run attribute: CharTopBorder=<Any: (com.sun.star.table.BorderLine2) 0x6831c50>
found run attribute: CharTopBorderComplexColor=<Any: (void)>
found run attribute: MMToPixelRatio=<Any: (float) 1.25>
matching atspi attribute is: mm-to-pixel-ratio=1.25
unknown:0:Test1::TestBody
An uncaught exception of type com.sun.star.uno.RuntimeException
- timeout from dbind at /home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/vcl/qa/cppunit/a11y/atspi2/atspiwrapper.hxx:302
Increase the timeout for method calls from 2.5 to 10 seconds
in the hope that will help.
(Issue is not reproducible locally for me.)
[1] https://ci.libreoffice.org/job/gerrit_linux_clang_dbgutil/162811/consoleFull
Change-Id: Ib0eef753cf9f574ee5b2a013a46a0a600c91eb34
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168002
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
The gtk3 VCL plugin uses a custom implementation that
has a listbox in a popup (s. vcl/uiconfig/ui/combobox.ui).
The popup window gets shown when expanding the combobox, e.g.
using Alt+Up/Alt+Down while the edit has focus.
Once the popup is closed again, focus is given to the edit
again.
That one correctly reports focus when checked via
`gtk_widget_has_focus`, but for some unknown
reason, the listbox in the popup assumes/claims
it still has global keyboard focus as well, despite it no
longer being visible and the `gtk_widget_has_focus` doc [1]
explicitly mentioning:
> Determines if the widget has the global input focus.
> See gtk_widget_is_focus() for the difference between
having the global input focus, and only having the focus
> within a toplevel.
The tree view listbox still assuming focus leads to it
emitting invalid active-descendant-changed a11y events
when the selected entry changes, which e.g. breaks focus
tracking by the Orca screen reader.
As a consequence, current Orca git main
(as of commit d70919051ef01384c473e08321a88cec52549cd2)
doesn't doesn't announce the new value of e.g. the "Font Name"
combobox in the Writer toolbar when changing it via the
Up/Down arrow keys any more once the popup has been shown.
(It works fine before showing the popup for the first time.)
To avoid that problem, manually unset the focus
using `gtk_widget_send_focus_change`.
Its doc [2] mentions
> This function is not meant to be used by applications.
, but since the focus is apparently not unset
automatically, still use it as a workaround for now.
[1] https://docs.gtk.org/gtk3/method.Widget.has_focus.html
[2] https://docs.gtk.org/gtk3/method.Widget.send_focus_change.html
Change-Id: Ifd32c2a66b4a7b9e827f5c8dc2927c7e359a42aa
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/167994
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
* Update helpcontent2 from branch 'master'
to c9ef3ec7d50ab940d9f9d2a4c255c2cd16a47048
- tdf#130838 tdf#154996 Add warning about copying/moving in record mode
In text/scalc/01/02180000.xhp
- Add statement that sheets can't be moved or copied in Record Changes mode.
In text/shared/00/edit_menu.xhp
- Updated howtoget instructions for "Track Changes" and "Record Changes", including adding Tabbed UI instructions for "Record Changes" and hiding instructions for the Track Changes toolbar behind an application switch, because Calc does not have a Track Changes toolbar.
In text/shared/01/02230000.xhp
- Added variable tags to H1 so that the page can be more easily linked from other pages
In text/shared/01/02230100.xhp
- Moved howtoget function to text/shared/00/edit_menu.xhp
- Updated embed link for howtoget section
Change-Id: I4776c0d45cd5d58a822cf00ede81ae3769ef8218
Reviewed-on: https://gerrit.libreoffice.org/c/help/+/167943
Tested-by: Jenkins
Reviewed-by: Olivier Hallot <olivier.hallot@libreoffice.org>
* Update helpcontent2 from branch 'master'
to 35c331d42654a28c9852187474b102db1c987e53
- tdf#105988 tdf#130024 Update Edit Points Bar help page
Created new file text/shared/02/eliminate_points.xhp so that more detailed information on "Eliminate Points" mode is on its own page.
- Updated the description of the "Eliminate Points" feature to more accurately describe how the feature works.
In AllLangHelp_shared.mk
- Added new page text/shared/02/eliminate_points.xhp
In source/text/shared/01/05270000.xhp
- Added section and variable tags to allow content to be easily embedded in other pages.
In source/text/shared/main0227.xhp
- Added section and variable tags to allow content to be easily embedded in other pages.
- Moved the link to "Converting 2D Objects to Curves, Polygons, and 3D Objects" to a new Related Topics section
- Replaced "Edit Points" section with an embedded link to the description in text/shared/01/05270000 (Edit Points mode) to avoid repetition
- Replaced "Edit Points" section with an embedded link to the description in text/shared/02/eliminate_points.xhp to avoid repetition.
- Created a Related Topics section
In text/simpress/02/10080000.xhp
- Added variable tags to allow the page to be easily linked from other pages
- Changed the Title and Heading 1 from "Curve" to "Curves and Polygons" to reflect current menu options.
Change-Id: Ib9038f35b97b4177e11171bfdedd10c0a4149dd2
Reviewed-on: https://gerrit.libreoffice.org/c/help/+/167939
Reviewed-by: Olivier Hallot <olivier.hallot@libreoffice.org>
Tested-by: Jenkins
See <https://bugs.documentfoundation.org/attachment.cgi?id=194306>, the
total height of the endnote separator is now correct, but the
distribution of upper space, line width, lower space is still bad, when
compared to Word.
Our model is 2 spacings and a line width, while Word seems to simply
split the amount of available space: 60% goes above the separator and
the rest goes below.
Fix the problem by breaking up the monster expression for the separator
rectangle in SwFootnoteContFrame::PaintLine(), and then the compat +
horizontal case can do the 60% spacing in the Word compat mode.
The width of the separator is still not correct.
Change-Id: Ic065c138041892225b3c1b7d4bd64d4f1c625611
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/167990
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
