Every DrawMask variant calls on a final DrawMask - we duplicate code
and there is an unnecessary private ImplDrawMask function in
OutputDevice. We should just forward calls to DrawMask.
Change-Id: Ice24598e3a437ca51a047f6006acc5a66198ff22
The primary reason is not performance: on export currently we loose some
EMF+ properties. While it would be good to fix all those problems one by
one, this is a small fix to avoid all kind of roundtrip problems in this
situation.
Change-Id: If5e1c1eabd8290f36b538e374c9707ae17097786
...and rather live with macro calls not terminated by semicolons until those
macros are gone for good anyway. Completely remove unused
DECLARE_UNO3_XCOMPONENT_DEFAULTS.
Change-Id: I0d4eb5a91736004b0da2b09fce860e57f6d1ddbd
When the environment variable DRAW_INCREMENTALLY_FROM is set to a
number, we want TiledLibreOffice to loop, initially performing only
that number of drawing operations in AquaSalGraphics, then wait for
some seconds, and redraw. Next time perform one operation
more. Repeat.
Implemented in vcl by surrounding the entry and exit(s) of the drawing
functions in AquaSalGraphics with macros.
All this is active only for iOS and in a dbgutil build.
Moved DoesActionHandleTransparency and IsTransparentAction back to
being ocal functions in an anonymous namespace so they aren't
exported.
Change-Id: Ic8335155c1a32035b02caacb29821bbd26197e49
Microsoft's new .doc file format specification says to ignore
sprmTFCantSplit90 in favour of (the older!) sprmTFCantSplit.
This seems to have confused the OO developer who must have been trying
to reverse engineer this without documentation: the table parsing code
gets sprmTFCantSplit90 and sprmTFCantSplit the wrong way around.
This commit removes handling of sprmTFCantSplit90 according to the
instruction in the Microsoft file format specification.
http://msdn.microsoft.com/en-us/library/dd951612.aspx
Change-Id: Ic9d3dddf102d6af0a061d2e8b68de3cf5bd9bae1
Reviewed-on: https://gerrit.libreoffice.org/8885
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
I understand that Libreoffice's names for SPRM IDs come from old
documentation for WW8 (Word '97), plus some later additions named by
OpenOffice developers as they encountered newer SPRMs in the wild.
Meanwhile Microsoft has released newer documentation which supplies
names for these previously undocumented SPRMs, plus it renames some
WW8 properties to have the suffix "80" to make room for newer
versions of those properties.
This commit aims to bring LibreOffice's SPRM ID names in line with
the current file format specification from Microsoft.
http://msdn.microsoft.com/en-us/library/dd923581.aspx
Change-Id: I904c1db17a776e2d9e6076f87369160e3b80f2e4
Reviewed-on: https://gerrit.libreoffice.org/8884
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
Also rename them to something saner. ImplIsActionSpecial in particular
is not a particularly useful function name, so updating this for
clarity.
Functions renamed:
+ ImplIsActionHandlingTransparency() -> DoesActionHandleTransparency()
+ ImplIsActionSpecial() -> IsTransparentAction()
I've also updated some of the comments to clarify.
Change-Id: I6c9d0c33d96ce40af2c877b52da66de17ed8ce78
Firstly, IsFormat is unused so there's no "escape" of
bFmt back into the outside world.
Then bFmt has two purposes.
Purpose 1 is 'not be a format dialog, be a format dialog, be a format dialog but hide standard button'
so, lets just add an explicit "Hide standard button" method and call
it in the (apparently) one place where it's necessary.
Purpose 2 is to flag that "BaseFmtHdl" was called from clicking
the "Standard" button at which point its set to 2.
SfxTabDialog::Init_Impl had...
"
// bFmt = temporary Flag passed on in the Constructor(),
// if bFmt == 2, then also sal_True,
// additional suppression of the standard button,
// after the Initializing set to sal_True again
if ( bFmtFlag != 2 )
m_pBaseFmtBtn->Show();
else
bFmtFlag = sal_True;
"
but the variable acted on is bFmtFlag a copy of bFmt, and is never read again
after that line, so setting it to sal_True is meaningless. The comment suggests
that the intent is to reset bFmt to true if it was 2 during initialization,
which fits with the later use of bFmt == 2 to indicate that the standard button was
clicked, i.e. reset bFmt back to its standard value.
So make bFmt a simple toggle of dialog as a format dialog or not, add a way
to remove the standard button and add a second variable to indicate the standard
button got pressed.
Change-Id: I98a441f5f314845abe243e05b6d92fd71d7b0b04
