Window and cocoa NSWindow types
VCL's Window type and Cocoa's NSWindow type are quite different.
Naming variables as if they were the same introduces gratuitous
complexity especially when debugging stacks where both types are used.
The names of NSView type variables have been adjusted too.
(cherry picked from commit 5e6120c21a979c05bc6b9297035e31a47b98b785)
Conflicts:
vcl/aqua/source/app/vclnsapp.mm
vcl/aqua/source/gdi/salgdi.cxx
vcl/aqua/source/gdi/salgdiutils.cxx
vcl/aqua/source/window/salframe.cxx
vcl/aqua/source/window/salobj.cxx
vcl/inc/vcl/sysdata.hxx
vcl/source/window/syschild.cxx
vcl/source/window/window.cxx
Change-Id: Ic4143dfddfa9502a529c871ac4ce0bea413b6ce9
internal link, take it as it is
(cherry picked from commit d44e9a9244558f63b3feaf923802ed32b4931283)
Conflicts:
sd/source/filter/ppt/pptin.cxx
Change-Id: Ic2541bcf7e368a4e9f2cc87353be165fc203dbac
This reverts most of commit a50a2111a31e4d9dd5e2659a146e9de8e292de75, where the
-Werror,-Wunused-variable for SwView::SetMoveType's aColor (after
3e8fe4d8e19be2ccd8f5bb898530e2f615a90321 "Remove the Navigator buttons below the
scrollbar in Writer" had removed its use) had erroneously make me think that
SwView::SetMoveType now effectively didn't do anything anymore, failing to note
the
m_nMoveType = nSet;
part that was still relevant.
Change-Id: I5e43886806533e80e238fef8d836e09e28209b6d
Change-Id: Ie2e1edee804f3e85eab08d05adecfc25ebbc77c1
Reviewed-on: https://gerrit.libreoffice.org/6934
Reviewed-by: Joren De Cuyper <jorendc@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
a9d85d62a889288b17899c8defc020da487d8b36 used
ScInputHandler::EnterHandler() to reset all sort of things related to
input EditEngine, but that is a handler for Enter and does not enter a
handler and actually attempts to finalize input, which lead to various
unwanted side effects.
Introduced ScInputHandler::InputTurnOffWinEngine() as only the input bar
window EditEngine needs to be reset in the window switching case (which
EnterHandler also does hence it did prevent that bug). The approach
could be polished with further refinement but most importantly fixes the
actual problems now.
Change-Id: I9a0bc452b49ba11a3313cafbc1e5972f41dc65c7
This was used as a tag for various classes, but the generated result
wasn't used anywhere, so let's just get rid of it.
Change-Id: I61dc51368177eddb6d04dba94063ba102f4ca1e5
This can happen when Excel's original shared formula range is across
multiple columns, in which case we split that into multiple groups (one
per column), and initialize top cell only for the left most column.
This fixes the crasher on the doc from fdo#31296.
Change-Id: I9e4def9836947fc67523f0d99ca981465709b934
The unit test checks three cases:
* attributes in run properties
* attributes in default run properties
* attributes in style definitions
Change-Id: Ie3224f41bd29e8b258dbd6823b7c9549e1ef5b1e
The default values for run properties are stored in the styles.xml
file, in the <w:rPrDefault> tag. The process of this tag is slightly
different from normal <rPr> tags.
First, we fix DocxAttributeOutput::OutputDefaultItem to be able to
output the contents of the character grab bag.
Second, we fix StyleSheetTable::applyDefaults to read the property
values using the GetPropertyValue method instead of using an iterator
directly, because the former creates the grab bags and returns them
as properties while the latter returns all the properties individually,
including those that should be inside a grab bag.
Change-Id: I91254ba20fd9ca5a1b02afb587f52a449a215d6c
The attribute eastAsiaTheme was not being processed properly (the
derived theme font was being assigned to complex-script instead of
eastAsia), so this patch fixes that and it also preserves the value
of the theme itself.
Change-Id: I1a91ab2080eb2bc79f41fda9ae4bf7afb05f69a3
Font theme attributes were saved to an InteropGrabBag on import, and
they have to be saved back to the document when exporting to docx.
If the user modified the font in a text fragment during edition, the
theme attribute should be discarded in favour of the font attribute.
The attributes in the grab bag are processed in the last place so we
can compare the font attributes with the original fonts to know if
they have changed.
Change-Id: I2502b4aa96213e74b46fccd1a20f99421f7cf98b
The attributes w:*Theme in the rFonts tag are saved to the character
InteropGrabBag so they can be saved back to the document on export.
Font attributes are also pushed into the grab bag because they will
be needed to check if the user has modified the fonts while editing
the document.
Change-Id: Ib247845a2d5543823798e1bbd8ed1958a5d9e884
...that would fall into the same trap as
50bd5c11f551f5274be9a4411c5ddcbd32bd9a03 "wsprintf is broken by design and never
writes more than 1024 characters" if the length were not currently bounded by
2 * MAX_PATH + x < 1024 anyway.
Change-Id: I2503ba0a9df960f2fdb51925d5fbbcff5bdb998a
we encounter both fonts, but all their properties that we compare
are the same, so we think they are two versions of the same font
and throw away the "Regular" version.
Change-Id: I24369f10c7a0edba1faa02621bb78616e880966e
Experimental only.
This avoids ,us deadlocking as the main thread tries to join it's children,
and the child threads wait to 'Yield' in the progress bar update.
Also it's generally safer to move progress reporting out of the other
threads, and to have the mainloop spinning here.
Finally this allows people to continue to use the LibreOffice suite while
large XLSX spreadsheets are loading.
Change-Id: Id41c18f3941d6fc5eea593f7cfcf6a8b7215b3f8
It turns out it's possible to nest this, e.g. <mc:Choice
Requires="wps">, and inside that, a <mc:Choice Requires="wp14">.
Change-Id: Icb6250dcc5059894112d8eeba9f327659a42280e
Drag'n'Drop is handled in a second thread, which tries to "yield
the display" while the main Qt thread probably is already
yielding. Both need the YieldMutex, which freezes the application
until the D'n'D thread times out.
Trying to yield the display throught the application yield
results in a recursive loop.
So this catches and breaks the recursion, but just inside the Qt
thread, so other processes can "yield on the display", instead
of simply disabling the Display::Yield.
Change-Id: Ifba91aa89fe5b0a89cc94820935dc996a065112f
Reviewed-on: https://gerrit.libreoffice.org/6750
Tested-by: Jan-Marek Glogowski <glogow@fbihome.de>
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
The "2" part was a temporary measure so an older version of the file,
that is long gone, can coexist.
Change-Id: I3a22b287bfb1351bb73e9dabbfa67100fe17bb97
It is unclear why the modified flag should ever be blindly overwritten hence
we no longer touch the modified flag here. (This change appears to have no
undesired side-effects.)
Change-Id: I246eb7d84249794fa4d52fcd9b008d8340bbc82d
Reviewed-on: https://gerrit.libreoffice.org/6632
Reviewed-by: Lionel Elie Mamane <lionel@mamane.lu>
Tested-by: Lionel Elie Mamane <lionel@mamane.lu>
Commit 840a8573c8cebe67ddd3c9fe106c7dbd789bb334 (Fix fdo#70220
Superscript not imported from pptx., 2013-10-07) made it possible to set
CharEscapementHeight even if moBaseline is not set, but this causes
problems in the docx importer + not necessary, according to the bugdoc;
so just don't do that.
Change-Id: Ib95ac449bd8fdf6376261ddc86108f0d23f2200e
