Age | Commit message (Collapse) | Author |
|
Just convert some loops to range based and fix strange loop logic
at 216 line.
Upd: rename *iterators* as Mike Kaganski adviced
Change-Id: Idf33b65f9c4466aa49c8f3bc13a12ad072dce7b6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137929
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
Tested-by: Jenkins
|
|
Change-Id: I57f924522a6fd93950f56c44ca65aa5131fb4aa8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137989
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I31413a6dafe1ddf1345ac6bccb96fc88ed2993c2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137985
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
restrict this to a very short length to reset ofz Timeout to force
ofz to report a new issue when restored later
Change-Id: Ibbdbb60f16ddca8367355d4c855e58aa16ea5f4a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137944
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I551112fb097a6ac2b442cd37d1a16bd2b34ecc54
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137932
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
This reverts commit 499ecbf3a36990c29dc7e1fb9b0ecb1d297c2848 because
it causes the dialog not to start with the default selected macro
auto-expanded like it used to. Possibly the other fix for tdf#150291
solves the original problem.
Change-Id: Ic91410de073fe87d4edebde18b60ffcc1ad324a3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137923
Tested-by: Jenkins
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
+ Nvidia on Windows just for device id "0x1401" in order to not being too restrictive.
Change-Id: Ie2cb2c6c6bb50ac6080721230ca3e451e886bb07
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137768
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Change-Id: I035bd25509db608d2273949cecd438d412e5a4c5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137816
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
rather than DontKnow, Drag assumptions seem to fit best
Change-Id: Iae92b29d33e69410c67b43fabb933018f1277abf
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137809
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Since `QAccessibleTextInterface::offsetAtPoint`
uses screen coordinates, but
`XAccessibleText::getIndexAtPoint` takes
local coordinates (i.e. position within the object
itself), convert that first.
Tested as follows:
1) start new document in Writer
2) paste this text into the first paragraph:
"Hello world, this is some test text. Lorem ipsum."
3) start Accerciser
4) select the paragraph in Accerciser's tree view of the
LO a11y hierarchy.
5) use Accerciser's IPython console to check the results
returned by calling the corresponding AT-SPI methods:
In [69]: acc.queryText().getOffsetAtPoint(acc.get_position(pyatspi.component.XY_SCREEN).x + 200, acc.get_position(pyatspi.component.XY_SCREEN).y + 5, pyatspi.component.XY_SCREEN)
Out[69]: 20
In [70]: acc.queryText().getOffsetAtPoint(acc.get_position(pyatspi.component.XY_SCREEN).x + 250, acc.get_position(pyatspi.component.XY_SCREEN).y + 5, pyatspi.component.XY_SCREEN)
Out[70]: 27
In [71]: acc.queryText().getOffsetAtPoint(acc.get_position(pyatspi.component.XY_SCREEN).x + 300, acc.get_position(pyatspi.component.XY_SCREEN).y + 5, pyatspi.component.XY_SCREEN)
Out[71]: 37
With this change and the Qt changes mentioned in
commit 35be93f83ac866ef18f0e06853c9818cd1d1bd56
Date: Wed Aug 3 09:18:46 2022 +0200
qt a11y: Implement QtAccessibleWidget::characterRect
, the results with the qt6 VCL plugin are the same as with
the gtk3 one.
Change-Id: Ie75fa9b7a64c413251c771f0379d9f3a3b200c30
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137795
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Implement these methods inherited from
`QAccessibleTableCellInterface` by retrieving all
row/column headers from the table (= cell's parent object) and
only returning those for the row/column that the cell itself is
in.
Tested with Accerciser and the qt6 VCL plugin as follows:
1) start LO Writer, "Table" -> "Insert Table"
2) select to create table with 2 rows, 2 columns
3) make sure "Heading" is checked, "Heading rows": 1
4) "Insert"
5) in the first row, type "First heading" into first column, "Second heading" into second column
5) start Accerciser
7) select the table element in Accerciser's treeview of the a11y
hierarchy
8) type these in Accerciser's IPython console:
In [10]: acc.queryTable().getColumnHeader(0).name
Out[10]: 'A1'
In [11]: acc.queryTable().getColumnHeader(0).get_child_at_index(0).queryText().getText(0, -1)
Out[11]: 'First heading'
In [12]: acc.queryTable().getColumnHeader(1).name
Out[12]: 'B1'
In [13]: acc.queryTable().getColumnHeader(1).get_child_at_index(0).queryText().getText(0, -1)
Out[13]: 'Second heading'
Change-Id: I69b9bd10bfe4076de9e4a05fe4aff97d1bfa4118
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137794
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Change-Id: I728456d27211916010c0174fa71b0125c7530707
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137793
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
There are specialised roles for these 2 button types
in Qt as well.
Change-Id: I67f49f0c4afc5f20a8c42fa2388f08ea9d950b91
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137785
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Qt doesn't have an extra role for this, so
use the dialog role, since `AccessibleRole::FILE_CHOOSER`
is for a "specialized dialog that displays the files in the
directory and lets the user select a file, browse a different
directory, or specify a filename."
Change-Id: Ib17b573dcd149148e72e1a47922dccf6f8fb67d5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137784
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Change-Id: I9208dccf15da918f8a30f21efb49986c69b68c5a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137783
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
The
/* Ignore window objects for sub-menus, combo- and list boxes,
* which are exposed as children of their parents.
*/
comment had been added in
commit 4d382636b0b1c555af2b98f2f41b4776fd7b5ffb
Date: Fri Sep 21 10:20:09 2018 +0200
qt5: Add basic a11y support
where the code was still actually doing that, but since
commit 7384e53bbc5163534ae898b7295b4efe51c8acde
Date: Fri Sep 28 17:00:12 2018 +0200
qt5 a11y: Use UNO Api instead of vcl::Window methods
that is no longer the case, so drop the obsolete comment.
Change-Id: Ic7739049f283dbe27f881414e87ada7616ca424f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137782
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Change-Id: I89b8a2cb778f91610b6a8e8efc0c4e6ed96a7644
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137778
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Quoting the Qt doc [1]:
> Returns the window associated with the underlying object.
> For instance, QAccessibleWidget reimplements this and returns
> the windowHandle() of the QWidget.
>
> It is used on some platforms to be able to notify the AT client
> about state changes. The backend will traverse up all ancestors
> until it finds a window. (This means that at least one interface
> among the ancestors should return a valid QWindow
> pointer).
Check if the associated QObject is a QWidget and if so,
get the associated window, otherwise walk up the a11y tree.
Note however that this change alone is not yet sufficient
for a window to actually be returned for any arbitrary a11y
object deeper down the hierarchy. This is because
walking up the a11y hierarchy currently results in new
Qt a11y objects being created for the parents instead of
using existing ones, and the newly created ones lack
the association to the widgets.
(This works in a WIP branch that remembers/caches
a11y objects, but that needs some additional work before
it can be merged.)
[1] https://doc.qt.io/qt-5/qaccessibleinterface.html#window
Change-Id: Iba05f7bd3fba30fb2ca741179abbda2d60bee980
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137753
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
The a11y hierarchy of the whole application
includes a11y objects that have no LO-internal
a11y objects associated with them, but are handled
solely by the Qt library, like the Qt application
at the root of the a11y hierarchy.
Therefore, fall back to using the QObject hierarchy
when looking for the parent of an a11y object.
This e.g. makes the scenario of starting at the top-level
application object, moving the hierarchy
down two levels, then up again work, which now
and ends up at the application object again.
With the application a11y object ("soffice.bin")
selected in the Accerciser tree view of the a11y hierarchy,
this can be achieved with the following commands
in Accerciser's IPython console:
In [4]: acc.name
Out[5]: 'soffice.bin'
In [5]: acc.get_child_at_index(0).get_child_at_index(0).get_parent().get_parent().name
Out[6]: 'soffice.bin'
It previously failed like this, which is one of the reasons
why restoring bookmarks in Accerciser didn't work when using
the qt6 VCL plugin:
In [1]: acc.get_child_at_index(0).get_child_at_index(0).get_parent().get_parent().name
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
/usr/lib/python3/dist-packages/pyatspi/__init__.py in <module>
----> 1 acc.get_child_at_index(00).get_child_at_index(0)0.(get_parent).(get_parent).nameAttributeError
In [1]: acc.get_child_at_index(0).get_child_at_index(0)
Out[2]: <Atspi.Accessible object at 0x7fd3b8d11140 (AtspiAccessible at 0x56252b30bd40)>
In [2]: acc.get_child_at_index(0).get_child_at_index(0).name
Out[3]: 'VCL ImplGetDefaultWindow (Type = 362)'
In [3]: acc.get_child_at_index(0).get_child_at_index(0).get_parent()
In [4]: acc.get_child_at_index(0).get_child_at_index(0).get_parent() == None
Out[5]: True
Change-Id: Iba197a2e943671591fa0ee5641a7b330eae8822a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137747
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Change-Id: I47f71371246f6322f41b782697d4824517375c76
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137760
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I42072354fd6f901035d1419f9f9402eac2eaaaac
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137751
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
with the motivation that for gtk4 theming for vcl scrollbar isn't
available, so it looks super yuck but this should also support the
long-press scrolling thing under gtk3 as a side effect.
Change-Id: Iada5a66087d5aa982f9213d7ec0a05f6177f4e66
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137750
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I1c343507d7db4b0805c20eabb68bf58a4432d33f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137746
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
typically a full ScrolledWindow is probably preferable, but allow
individual scrollbars to be used
Change-Id: Iab290b387421d7d97d3604e31f528769d822155a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137709
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
With this and the suggested changes ([1], [2], [3], [4])
for 4 Qt issues I ran into ([5], [6], [7], [8]),
the Accerciser scenario described in
commit 7b312771d7eb33f7410167e36efdaeca6f540b1c
Date: Thu Jul 14 08:35:53 2022 +0200
tdf#149952 gtk3 a11y: Return pos relative to window when requested
works for the qt6 VCL plugin as well.
(The complete GNOME Magnifier scenario from tdf#149952
still doesn't work with qt6, probably because a11y
events are not sent (reliably) so far.)
[1] https://codereview.qt-project.org/c/qt/qtbase/+/422251/1
[2] https://codereview.qt-project.org/c/qt/qtbase/+/422333/1
[3] https://codereview.qt-project.org/c/qt/qtbase/+/424510/2
[4] https://codereview.qt-project.org/c/qt/qtbase/+/424731/1
[5] https://bugreports.qt.io/browse/QTBUG-105031
[6] https://bugreports.qt.io/browse/QTBUG-105042
[7] https://bugreports.qt.io/browse/QTBUG-105281
[8] https://bugreports.qt.io/browse/QTBUG-105313
Change-Id: Iaccc89bf62c4617bee06f34ffebbd309a62212e9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137735
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
The bugdoc has a barcode which looks good in Writer, but it's quite
blurry in the PDF export result.
This is a problem since commit dd4a67084853a030bf4b9f1f85d728620e0604a5
(vcl: avoid downscale && upscale in DrawTransformedBitmapEx(),
2019-10-08), where the motivation was to not downscale a bitmap in case
it has a larger amount of pixels but a smaller logic size. This went
wrong here and resulted in a blurry bitmap for a not so small image.
Fix the problem by mostly reverting the above commit, because it's no
longer necessary: 68549e00d5e23aa22bc974a8151d93cd948444b3 (vcl,
BitmapEx transformed draw: special-case simple rotations, 2019-10-10)
already made sure that the rotation case doesn't use scaling from the
transform.
testTdf128630 has been adapted to pass again (after manually verifying
that the bugdoc export result is still OK), now the bad case there
simply produces a smaller bitmap.
Change-Id: Ib28881c129a0cf037a96eecd065e5cadede97051
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137737
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
Update PDF export test that depends on a mapping the font no longer has,
and use different characters that meet the test criteria.
Change-Id: I0f89297a52ffcef3513e0a788ec3bab645cc354b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137661
Tested-by: Jenkins
Reviewed-by: Heiko Tietze <heiko.tietze@documentfoundation.org>
|
|
Also update test expectation due to changed glyph order.
Change-Id: I8f6d1a057077cb6de5e5d7fce4bec29929edb8b9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137663
Tested-by: Jenkins
Reviewed-by: Heiko Tietze <heiko.tietze@documentfoundation.org>
|
|
It passed "make check" on my laptop
Change-Id: I2d231c3a8ec40b63026a4b81256b98ee4be0ae1a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137660
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Crashreport signature:
SIG Fatal signal received: SIGSEGV code: 1 for address: 0x0
program/libmergedlo.so
vcl::ImportPDFUnloaded(rtl::OUString const&, std::vector<vcl::PDFGraphicResult, std::allocator<vcl::PDFGraphicResult> >&)
vcl/source/filter/ipdf/pdfread.cxx:256
program/../program/libsdlo.so
SdPdfFilter::Import()
sd/source/filter/pdf/sdpdffilter.cxx:55
program/../program/libsdlo.so
sd::DrawDocShell::ConvertFrom(SfxMedium&)
sd/source/ui/docshell/docshel4.cxx:500 (discriminator 1)
program/libmergedlo.so
SfxObjectShell::DoLoad(SfxMedium*)
sfx2/source/doc/objstor.cxx:768
Change-Id: I2fc635de696643af6a80a29aa5af4fa128f8776c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137673
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
Change-Id: I215eb41bc2e7f442b0dc37393460bd2e53fed8c1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137646
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Use the new HarfBuzz glyph flag HB_GLYPH_FLAG_SAFE_TO_INSERT_TATWEEL if
available to prevent kashida insertion where is would interrupt shaping
(between contextual glyphs). Previously we only prevented it when there
is a ligature across the insertion position, with this change fonts that
use contextual alternate instead of ligatures will be handled better as
well.
Change-Id: Ibb42a310c1e7dbcb225a1ba3acac82154b4edb3a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137649
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
The constant PLACEABLE_SIGNATURE in graphicfilter2.cxx had wrong
endianness, which would cause the WMF file detection to fail for
placeable WMFs. Changed endianness since WMF files are in little endian.
Also added test using placeable and non-placeable WMFs.
To run the test:
make CPPUNIT_TEST_NAME="testDetectWMF" -sr CppunitTest_vcl_graphic_test
Change-Id: Ia3c61bae5ab27375636f21bc902f773804cf9273
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137634
Tested-by: Jenkins
Reviewed-by: Hossein <hossein@libreoffice.org>
|
|
* Fixed a heading problem
* Marked variables, functions and identifiers as fixed width
* Added several related links
Change-Id: If5e56c1e2a474fcfa52601ccda7911bf9e80c913
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137636
Tested-by: Hossein <hossein@libreoffice.org>
Reviewed-by: Hossein <hossein@libreoffice.org>
|
|
we only use AtkObject::accessible-role for "static", but there doesn't
seem to be a good match to anything I can see in gtk4. convert it
to accessible-role of "alert" for lack of a better idea.
Change-Id: Ic0795fed006b827914f1319150a915187188a4db
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137617
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Documents signed using an invalid (e.g., not having a trusted root)
certificate give signatures that obviously can't pass validity tests.
Change-Id: Id4b097516e06c548ea42cad65d76bbd8a6853cc4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137620
Tested-by: Mike Kaganski <mike.kaganski@collabora.com>
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
Change-Id: Ie580cd0b42ca1c928bfea20a4a9f1494021047d8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137560
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Opentype defines only "frac" and not parameters values like 1 or 2 for "frac"
See:
https://en.wikipedia.org/wiki/List_of_typographic_features
https://docs.microsoft.com/en-us/typography/opentype/spec/features_fj
So remove STR_FONT_FEATURE_ID_FRAC_PARAM_0/1/2
+ remove the extra ":" for STR_FONT_FEATURE_ID_FRAC
I gave a try with "Alegraya" font, it worked.
I also give a try with "Linux Biolinum G", there are still "None" (value "0"), "Diagonal fractions" (value "1") and "Nut fractions" (value "2").
It seems they're on the font itself like entry "'frac' Diagonal and nut Fractions".
So it shouldn't bring regression.
Change-Id: I2f46a3f4a53dc498b764dbeb1c1266589cc8d8c7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137574
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
the alignment won't match when using GtkPopover to implement these,
there's an extra border+margin between the launcher and the contents
Change-Id: Idb78d903874fee5a28e87c89951d9b3f9063c547
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137554
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I8a338a92823bd4e4c1f8123d52711d0925357ee6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/76430
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
Change-Id: I9f5de7ed1bdcd39e5d6e580a2edbd5c0dd2027e0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137278
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
(regression from commit fe38553aef2121f358fb58e450ec69314aad851e)
Change-Id: I58e1afb228dfd87b7d642ccf1b58277960f01e14
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137561
Tested-by: Jenkins
Tested-by: Julien Nabet <serval2412@yahoo.fr>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Always suspicious that some un-expected dialog / state can cause
strange behavior in a client. An initial cut at an API to make it
easier to unwind such problems by exposing the toolkit state.
Change-Id: If8f17943fa4837df4f9ca659a111dcdce5c23244
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137504
Tested-by: Jenkins
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
|
|
still a bit rough, but functional at least
Change-Id: I4b5f896a0d895ed20da3534173c20241a7fe7ea1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137510
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I9c71dd3cad1e45d0a01ab187c652711c19fe3b89
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/135444
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
the size request for the overlay comes from its main child, the drawing
area in this case
Change-Id: If814aaa1d1fc38b331b18f5e9685c405d4f0d1b3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137509
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
this was the clipboard dropdown matches the save one
Change-Id: I423aeed097cd22d48aa9efcaf95e0e8d6fd36d98
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137501
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: Ia646d6f98804ffbec1177d5964550708588ce88f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137500
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
keep any value the user explicitly set, but update one that was taken
from lpoptions if that changed
Change-Id: I02ed4ff575cb7b2580b777bb0948de2ff7211621
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137484
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Printer and driver name have a limited length (64 chars for
the printer name, 32 for the driver name) in the
`ImplOldJobSetupData` struct that is part of what gets
serialized into the "PrinterSetup" config item in
settings.xml when saving printer settings to the document
(s. `WriteJobSetup`). Longer printer and driver names
get truncated, and only the truncated name got restored.
At least for Windows, this scenario would result in the
doc printer settings not getting restored properly,
because `ImplTestSalJobSetup` (in vcl/win/gdi/salprn.cxx)
among others checks that both printer and driver name
match, which isn't the case when one version is truncated.
ODF spec, version 1.3, section 3.1.3.5 says this on settings.xml:
> 3.1.3.5 <office:document-settings>
> The <office:document-settings> root element contains
> implementation-dependent settings. The file within a
> package for the <office:document-settings> element is
> settings.xml.
Since the settings are implementation-dependent,
changing the exact format and content of
the data in "PrinterSetup" is unproblematic,
as long as compatibility with older application versions
is maintained.
Luckily, there is already a way to save variable-length
key-value pairs with `JOBSET_FILE605_SYSTEM`,
added in
commit 4bd178f6ee60f641b2a6c7fcc9f6e4b6dd01df27
Date: Fri Sep 22 11:55:00 2000 +0000
new feature in JobSetup: additional parameters
Make use of this to save the full printer and
driver name using the "PRINTER_NAME" and "DRIVER_NAME"
keys and restore the values from those if present.
For compatibility with older versions, the (potentially
truncated) values are still written and also used as
fallback on document open.
Change-Id: I55bed58023fa7a078525add79b13f2310aa9a9dd
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137454
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|