Age | Commit message (Collapse) | Author |
|
...and implement export of all missing commands,
use existing viewBox if suitable,
use one EnhancedCustomShape2d
move WriteCustomGeometryPoint to protected,
make GetCustomGeometryPointValue local
The fix solves tdf#100390 too.
Without the fix the entire enhanced-path was exported as one element
<a:path>. The command F was applied to the whole drawing but should
affect only the subpath. The implementation is changed so that each
subpath gets its own element <a:path> and command F acts only on its
subpath.
Support for export of handles and equations is still a long way off.
Thus there is no reason to tread shapes with and without handles
different. Shapes with handles had used method WritePolyPolygon, but
that is not able to handle subpaths. So have desided to use method
WriteCustomGeometry for all cases.
To get shapes exported regardless of path commands I have added the
export for the missing commands.
I have removed the no longer used method WritePolyPolygon.
The special treatment of shapes "moon" and "mso-spt89" (right up arrow)
in export is no longer needed. Related code parts are removed. The
unit test testFlipAndRotateCustomShape is adapted.
In case the method WriteCustomGeometry fails, the enhanced-path is
invalid. In that case an minimal custGeom is written in case of docx.
Shapes whose drawing does not touch all edges of the snap rectangle
were exported with wrong position and size of the drawing inside the
snap rectangle. That is fixed by using an existing ViewBox for the
OOXML path size. The old way of creating a path size from point
coordinates is only used if the shape has no suitable ViewBox.
The point values in unit test SdOOXMLExportTest2::testTdf111798 are
adapted to path size 21600 x 21600 and traverse direction of the points
is corrected. The resulting shape outline is still the same as before.
The expected xml is updated for file tdf92001.odp in
SdImportTest::testDocumentLayout. The resulting shape outline is the
same, because the shape touches the edges of the snap rectangle.
The case, that the shape outline does not touch a edge of the snap
rectangle is tested in SdOOXMLExportTest3::testEnhancedPathViewBox.
Still missing is the case, that ViewBox has other left,top than 0,0.
In that case all coordinates would have to be shifted because the path
size in OOXML has only width and height but not left,top. That will
not be included in this patch.
Change-Id: Ib1736d6a08371f4d98411d2769275f0580cd0030
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131837
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Do this only in case there are no effects on the color, as that is not
yet handled.
Note that the theme color was already stored in the grab-bag, so this is
primarily interesting in case the theme color was changed or the source
format was ODP.
Change-Id: Ia4995be68d5f243875632eec4aaf8afbb8f4d5cc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131984
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Corrects copy/paste residue with the correct content
Change-Id: Ie10a4db728de154ebbdaf6c7902b28c71c815456
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131957
Tested-by: Jenkins
Reviewed-by: Sarper Akdemir <sarper.akdemir@collabora.com>
|
|
Change-Id: Ifcdae0106380de978c608f46e82155c6c9eed20b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/130292
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Fixes paragraphs made from a single new line, not getting the
correct sz(text size) value in EndParagraphRunProperties on pptx
export
Change-Id: I31ebb5735ad392e081aa2f43b0b60a845e4de9c2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/128265
Tested-by: Jenkins
Reviewed-by: Andras Timar <andras.timar@collabora.com>
|
|
Change-Id: Ib14e5357830b587921b633fb57fa74b1314aa1e4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129975
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
It appears that placeholder shapes with service names:
- com.sun.star.presentation.TitleTextShape
- com.sun.star.presentation.DateTimeShape
- com.sun.star.presentation.FooterShape
- com.sun.star.presentation.SlideNumberShape
...
Are unable to have custom shapes in Impress. (i.e. can't be
ellipse, triangle etc.). These presets get specified in OOXML
with <a:prstGeom prst="ellipse"/> inside spPr (shape properties).
Therefore with similar results to the PPT import, a workaround
is applied where slide footers which have non default shapes
are imported with com.sun.star.drawing.CustomShapes service.
The layout/master footers are left as is since if they were to
be imported as CustomShapes they would appear on each slide
even if the slide had those footers enabled or not.
Change-Id: Ic8a8ab3f6dfb7367ecd2c619ce888bf77abef460
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129889
Tested-by: Jenkins
Reviewed-by: Jan Holesovsky <kendy@collabora.com>
|
|
Change-Id: I6a571f12f2e194a6d72a457622505cca72bd8eb1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/128021
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Expands previous idea from a9c5c0d814a266096483572b84c72875ef8efd77
(tdf#133037 OOXML shape import: camera rotation along Z)
and uses it also for shapes that have a true bUseRotationTransform flag
Also fixes same Z rotation exporting twice from InteropGrabBag to
both spPr and textBody causing text overrotating on roundtrip.
Change-Id: If0f192af029ca86b932a63613f961be1f5003c5b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/127880
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Map it to (the UNO API of) GraphicDrawMode::Watermark, similar to what
the binary import does in SvxMSDffManager::ImportGraphic() and how the
drawingML import does it in
oox::drawingml::GraphicProperties::pushToPropMap().
On export, the drawingML export is used, and that already maps
GraphicDrawMode::Watermark to <a:lum bright="70000" contrast="-70000">.
Change-Id: I33986a03bf3d3863da5c5b1f0a2e0da0fa595c9e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126908
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
Handle luminance modulation and offset (lighter and darker colors in
PowerPoint); not handling tinting/shading for now, as that seems to be
not used in DrawingML (only in WordprocessingML), and this code is for
shape text only at the moment.
Change-Id: I5e97f890d3072c7ef282ed4fb971362b3ddaaa4d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126400
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
As a start, do this only in case there are no effects used. If there is
no theme color or there are effects, fall back to the old code.
Also move parseExportStream() from SdModelTestBaseXML up to MacrosTest,
so oox/ test code can use it as well.
Change-Id: Ia76581dcef110341f6c3e60f22c34818ed0dcabc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126215
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
This was already handled in oox/ at import-time: this adds it to the doc
model, including UNO API and PPTX import.
This is a dependency PPTX export and UI, and also the last transform
which is easy to generate from the PowerPoint UI and we didn't import to
the doc model.
Change-Id: Ica4e738d8dc8e0409160ceab941a82f2475ddc68
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125749
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
This was already handled in oox/ at import-time: this adds it to the doc
model, including UNO API and PPTX import.
This is a dependency PPTX export and UI.
Change-Id: I5d875b53d715beb10c13ef616d06cf958d43f70f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125684
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
This was already handled by converting them to raw colors at import
time. This commit imports the color scheme contents (the 12 colors) into
the doc model.
This is a dependency to export them back to PPTX and to be able to
update these colors on the UI by picking a different theme.
Change-Id: I177de4f15d5f0e628669998d1cda7db24220b2eb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125651
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
We had doc model for this, but the UNO API and the PPTX import was
missing.
Change-Id: I199e9cc235a783d91700ce74f17d442f41d3c3f8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125532
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
A problem since commit 08818d8a45e034ad825c7fafbb76766f106f1d1d
(bnc#882383: Do not ignore themeOverride for charts in .pptx,
2014-07-04), an override for one chart should not affect later drawingML
objects.
Change-Id: I22b70c8c82e8e8520179c628f566d7f6663c887f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125218
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
PowerPoint aligns the bottom of the text to the path for prstTxWarp
'textCircle'. That preset type is mapped to 'fontwork-circle-curve'
on import with additional attribute ScaleX=true. Currently the property
TextVerticalAlign is only evaluated in case ScaleX=true. Therefore I
have written the condition similar as the already existing with
'fontwork-arch-up-curve'. If it will be necessary later, all those
conditions can be changed to use rPresetType instead of rClass.
The rendering is slightly different compared to PowerPoint, because
descenders and paragraphs line-spacing are handled differently.
The rendering has still the problem, that in PowerPoint the letters
are placed without gap, but in LO they have additional distances from
each other. The needed ODF attribute draw:text-path-mode is not yet
implemented in LO. Its value 'normal' would be needed here.
Change-Id: I1f03d4845312885eff9ee8dbe1d51ddd437ed8e4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/123726
Tested-by: Jenkins
Reviewed-by: Regina Henschel <rb.henschel@t-online.de>
|
|
Change-Id: I15d56d133cf464a3cb6483be785b1259c7f35b43
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/123120
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
When exporting custom shape arrows LO used to fall back to writing
out their plain vertex coordinates, losing the customizability of
the shape after loading from file. This commit changes the export
of some of the most commonly used arrow custom shapes to proper
adjustment value export.
Follow-up to commit 63cd67e5e18f01aca303131e148c80398a181a41
"tdf#92525 tdf#142398: fix export of simple custom shapes".
Change-Id: If248556764bdb7e00cfde4ebe5b32bb380b1fa19
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/121901
Tested-by: László Németh <nemeth@numbertext.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
The points in file source might have units. Decode was missing.
maWidth and maHeight are used in ShapeBase::convertAndInsert(), and
moCoordSize is used in PolyLineShape::implConvertAndInsert(). So
ShapeContext needs to provide both in case of importing a polyline.
That was missing.
Word writes the size into the coordsize attribute of the v:polyline
element. But from VML specification it need not exist, but bounding
rectangle of points has to be used. That is added too.
Unclosed polyline cannot be filled in LO and ODF, a fill is only
possible for a closed polygon. LO defines a sequence of points as
being closed, if first point == last point. The import is adapted
to that behavior.
Word allows an unclosed polyline to have filling. That cannot be
represented with a simple PolyPolygonShape but would need a
CustomShape. Because the user cannot yet edit points in a CustomShape
but can do that in a PolyPolygonShape, the v:polyline element is
not converted to a CustomShape on import and not to DML on export.
LO would first need an UI for editing points of a CustomShape.
Change-Id: I23d08fda2005f8b36488e1d9ba32e565d8a0f803
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/121042
Tested-by: Jenkins
Reviewed-by: Regina Henschel <rb.henschel@t-online.de>
|
|
The export had used the bound rectangle of PolyPolygonBezier. But that
contains control points. Use API position and size instead.
I have not incorporated the changes into existing WritePolyPolygon,
but have made an own version for SdrPathObj, because I find it easier
to read and maintain, than having a lot of case distinctions depending
on the shape type.
Change-Id: I2e646c4f5fa37174c4979855212ca72f2dfa447e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/120407
Tested-by: Jenkins
Reviewed-by: Regina Henschel <rb.henschel@t-online.de>
|
|
Regression from commit e9986153e44d7ec6ca9c5f1373971de74dcbacda (PPTX
import: import SmartArt drawing into single GroupShape, 2019-03-14), the
problem was that oox::ppt::PPTShape::addShape() and
oox::drawingml::Shape::addShape() were not in sync.
PPTShape unconditionally maps SmartArt to shapes, while the shared Shape
class defaults to converting it to a non-editable metafile. The above
commit changed the handling of in-groupshape SmartArts to go via
Shape::addShape() instead of PPTShape::addShape(), which exposed the
underlying problem that the convert-to-metafile mechanism is currently
only working in the DOCX case.
Fix the problem by again ignoring the convert-to-metafile flag for the
PPTX import case.
This also exposed a previously hidden problem:
make -C oox -sr CppunitTest_oox_drawingml CPPUNIT_TEST_NAME="testGroupShapeSmartArt testTdf131082"
started to make testTdf131082 fail. The tweak in
Shape::createAndInsert() fixes the testcase failure, but likely there is
a deeper problem there, unrelated to the regression.
Change-Id: I4e1e9645eaa266d0d7560767c3c59ba9549ccdb4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/120122
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
...in export to docx.
Rotated child elements need a correction to the position values,
because LO uses left/top of snap rectangle as position, but Word uses
left/top of unrotated shape. For the group itself it is done in
DocxSdrExport::startDMLAnchorInline. But child elements' position
is exported in DrawingML::WriteShapeTransformation. And there this
correction was missing.
Position is relative to anchor in Writer and relative to group in
Word. The adaption is contained in WriteShapeTransformation. But
PolyPolygon and Connector have no explicit rotation and therefore they
do not use DrawingML::WriteShapeTransformation but call directly
DrawingML::WriteTransformation. So they missed this adaption. I have
added the adapation directly to these shapes.
Unittest testDmlTextshapeB in sw-ooxmlexport6: I have adapted the
values. The position of the connectors relative to the group is better
now. You see this, if you compare it with a screenshot of the original
file in Word. The handles of the connectors are still wrong and the
whole group is still shifted.
The patch does not fix the wrong position of rotated legacy ovals,
because that error exists independent of groups.
Change-Id: Iaf843dcf04bac596427dd35ccfa6cd20e3a4cdc8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/118748
Tested-by: Jenkins
Reviewed-by: Regina Henschel <rb.henschel@t-online.de>
|
|
See tdf#42949 for motivation.
Change-Id: I559eb3b41a5a0fdb37db6c45d73211ca223384fe
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/117193
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
sc/ could potentially benefit from this as well, that's not yet done
here.
Change-Id: I03d0b4afa21a02c74d34aab6e03ab53991df29dd
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116679
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
The test had multiple failures on Windows with:
oox/qa/unit/vml.cxx(79) : error : Assertion
Test name: tdf137314_vml_rotation_unit_fd::TestBody
equality assertion failed
- Expected: 1490
- Actual : 1491
So add a 1px margin to all value checks.
Change-Id: I70298db253299a57cc37eed482c0816d902fbeab
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116192
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
|
|
Fixes wrong rotation in tdf#109129 and tdf#142432 too, but they have
further, unrelated errors.
Change-Id: I7bd56876bb42b261fe425f80cf9beb639c3ac276
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116078
Tested-by: Jenkins
Reviewed-by: Regina Henschel <rb.henschel@t-online.de>
|
|
Change-Id: I41d4fe581842cb7822ae899dc6ee6a43e485d211
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/115856
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
and implement special resize handling for rotation angles larger 45deg.
This solves tdf#93952 and tdf#141953 too.
Change-Id: I798f6d2cea29c4a5285f530e9cf7bb10e7f6c41d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/115296
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Regression from commit cfb5b20cdc230320ff9f864d1cfd81aaea221da0
(DocxAttributeOutput::OutputFlyFrame_Impl: enable DML export by default,
2013-12-18), there were two problems here.
First, <a:chOff> and <a:chExt> was not written for docx group shapes.
This can be done for toplevel shapes just by writing what would be the
shape position and size (but for docx, we don't write the size).
Second, (poly)polygon shapes used the bounding rectangle of their points
as size, which doesn't necessarily match the shape size. Given that the
group shape is meant to simply contain its children in LibreOffice (and
not have an own size), switch to using the UNO API for polygon shapes as
well, that way the two sizes will always match.
Change-Id: I4406ddefe5f6105aa2fc74d805359add452936bb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/114305
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Change-Id: Ic79d81387867f028eb8dc9553fb87f5961d6c771
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/112364
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
MacrosTest::loadFromDesktop itself asserts on its return value.
Thus, the additional checks in unit tests are redundant, and only create
noise unrelated to the tested functionality.
Change-Id: If616001b296afdde38f5a23ececee3d44b4a395d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/111290
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
Change-Id: I4884bfb67a061b865e8cf38b2fea6de0cb1bc3d6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/109057
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I044dd21b63d7eb03224675584fa143009c6b6008
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/108418
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
See the comment at the top of compilerplugins/clang/stringliteralvar.cxx for
details.
(Turned some affected variables in included files into inline variables, to
avoid GCC warnings about unused variables.)
Change-Id: Ie77219e6adfdaaceaa8b4e590b08971f2f04c83a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/108239
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Custom shapes export shadow as part of WriteShapeEffects(), so use the
same for table shapes as well.
This needs fixing the effect export up a bit, because table shapes have
no interop grab-bag, glow or soft edge properties, but the rest of the
code can be shared.
Change-Id: Icf0b90c5b44e3d9c4115c9f3b0d56ba0852ab640
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/107836
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Change-Id: I7535065d37a5d56ace0fec90c4f146ff4ce2d753
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/107586
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
PPTX export and ODP filter is still missing.
Change-Id: I451b334ada80d9d228b7d7f36b5f26473b575ef6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/107529
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
VML v:shape/v:textbox element was imported only as
a text frame, losing (otherwise recognized) preset
shape geometry, i.e. replacing a callout bubble
(wedgeRectCallout) and other special shapes with a
plain rectangle.
Thanks to Attila Bakos for the initial help.
Change-Id: I03a608822ed54a20ed07406a08c3539e72958f5b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105299
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
Change-Id: Ie061189450e0f9004ca503bb28164885812f2acc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105694
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
Tested-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
|
|
Found with bin/find-unneeded-includes
Only removal proposals are dealt with here.
Change-Id: I886b6f446293d3b1cfbf4ae05e8dbd7fabab9f20
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105510
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Change-Id: Ia28e96cf1f6ec476f202e99877fa80e93d691278
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105314
Tested-by: Jenkins
Reviewed-by: Andras Timar <andras.timar@collabora.com>
|
|
Regression from commit f4ba484183a1e7b9824f10580d633466c266828f (ooxml
import: supprt cropping to shape, 2019-05-13), which changed the type of
a cropped-to-shape image from drawing.GraphicObjectShape to
drawing.CustomShape.
drawing.GraphicObjectShape worked because GraphicImport::lcl_attribute()
in writerfilter/ had a check for drawing.GraphicObjectShape and did an
explicit setPosition().
Doing the same for bitmap-filled custom shapes would be an option, but
it would be ugly: scaling/translation/rotation/mirroring can only work
together if they are only applied once, and that should happen in oox/,
that's why we already have a mechanism to send the position from
writerfilter/ to oox/ for WPS shapes. (<a:xfrm> contains the size, but
not the position of the shape, so oox/ in itself could not know the
position.)
Fix the problem by improving ShapeContextHandler instead the pass the
position from writerfilter/ to oox/ for <pic:pic> as well, the same is
done for <wps:wsp> already since commit
6c4f737ec88a4f4dc5da8b2295ca5e7de2d4c24f (DOCX drawingML shape import:
fix position when CustomShapeGeometry is set, 2013-11-21).
Change-Id: I74a60136d0ca8383e58948711b47858823f42437
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105263
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
Instead of implementing proper OOXML 3D rotation (which would be
an entirely new feature if I understand correctly), I merely
interpret attribute "rev" of the rotation element a:camera/a:rot
as a directive to rotate the entire shape the usual 2D way. That
is already implemented and works well. This isn't the same thing
Word does, but it might be good enough for now. This is kind of a
mock solution, but it will be very easy to revert if it turns out
to cause problems.
Note: the export worked well previously, too (moreover, reloading
the first LO export fixed the import).
Change-Id: I2a99c119880bbed1c5b6430c4638cfbd10b7ac06
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103627
Tested-by: László Németh <nemeth@numbertext.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
Regression from commit 10bb02efd8afd42e633e370480104e2575546d8e
(tdf#129685 PPTX import: fix unexpected centering of shape text,
2020-09-18), now the problem was that some text should be left aligned,
but was centered.
Fix the problem by reverting most of the above commit: XML changes,
changes to SdImportTest::testTdf113198() (manual testing show that this
change is not needed after all) and changes to the
TextBodyPropertiesContext ctor in oox/ (but not the testcase itself).
Fix tdf#113198 again, this time in Shape::createAndInsert(), which is
meant to be closer to what the binary PPT import does.
With this, all cases from tdf#104722, tdf#113198, tdf#129685 and
tdf#137023 are meant to be handled correctly at the same time.
Change-Id: Id785252c26fc407cd74c9cfb55624091798d7773
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103996
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Tested before split here:
https://gerrit.libreoffice.org/c/core/+/103464
Change-Id: Iadc9dd49762ec63bd8b3edba322bcbf5d0f862a3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103487
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Regression from commit 89f0af144c18efafe2573801641689a1432c0cae (tdf#113198 set
default shape paragraph alignment.., 2019-11-19), the old bugdoc had this
markup:
<a:bodyPr ... anchor="ctr"/> (centered)
The new bugdoc has 2 shapes with text:
<a:bodyPr .../> (aligned to left)
<a:bodyPr ... anchorCtr="1"/> (should be centered)
"anchor" is about vertical, "anchorCtr" is about horizontal centering of text.
Checking what the binary filter does, it maps horizontal centering to
TextHorizontalAdjust, so fix the original bug differently, by leaving
ParaAdjust alone, and tweaking TextHorizontalAdjust intead: this keeps the old
bugdoc working but fixes the new one.
This caused a number of "change detector" XML-based tests to fail: all of them
are unchanged visually, so the XML files are adapted to the new state.
The tdf#113198 fix itself was fixing a regression from tdf#104722, and that
commit had no testcase, I tested that we don't regress there, manually.
Change-Id: I81a7b3e8c76bfbce5c5569d16d5238958ac20f75
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103012
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
ComplexShape::implConvertAndInsert() returns early in the graphic object
shape case, so stroke model is not applied at all.
Also fix a problem in ShapeBase::finalizeFragmentImport(), where the
shape type had no stroke, but the shape itself had, and the later should
win.
The warning in OleObjectGraphicDataContext::onCreateContext() now points
out that <mc:AlternateContent> is ignored as a child of <a:graphicData>,
which probably should be addressed at some stage, but it's not required
to fix the missing stroke.
Change-Id: I4ab43b4c6d40d9f43caad22b85f5b885fa8b4ef1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/100952
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
Impress core only support a single step, improve the conversion from
multi-step to single step to take transparency into account explicitly.
Once we select the widest segment, look backwards and forward if there
are other next segments which have the same RGB color, just different
transparency and include them. This helps in general, because in case a
0% or 100% transparency is mishandled, that's very visible.
Change-Id: I11d593c01a6a4b16149ce74c1408c2a39895e941
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/100231
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|