2016-04-15 15:46:06 +02:00
|
|
|
/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
|
|
|
|
/*
|
|
|
|
* This file is part of the LibreOffice project.
|
|
|
|
*
|
|
|
|
* This Source Code Form is subject to the terms of the Mozilla Public
|
|
|
|
* License, v. 2.0. If a copy of the MPL was not distributed with this
|
|
|
|
* file, You can obtain one at http://mozilla.org/MPL/2.0/.
|
|
|
|
*/
|
|
|
|
|
2017-04-14 08:42:15 +10:00
|
|
|
#include <memory>
|
2016-04-15 15:46:06 +02:00
|
|
|
#include <cppunit/TestAssert.h>
|
|
|
|
#include <cppunit/extensions/HelperMacros.h>
|
|
|
|
#include <cppunit/plugin/TestPlugIn.h>
|
|
|
|
|
|
|
|
#include <com/sun/star/table/BorderLineStyle.hpp>
|
|
|
|
|
|
|
|
#include <drawinglayer/geometry/viewinformation2d.hxx>
|
|
|
|
#include <drawinglayer/primitive2d/borderlineprimitive2d.hxx>
|
|
|
|
#include <drawinglayer/primitive2d/polypolygonprimitive2d.hxx>
|
borderline: correct problems with border display
Borderline display with direct paint and with primitive direct
paint has quite some errors in the current state. Started to
unify usages, check deeper with creation/usage.
borderline: deep changes to BorderLine
Found basic error in determining the offset values
for BorderLinePrimitive creation, these were not
centered on the lines. Corrected that. This makes
it possible to remove the formally used clipping
which seems to have been used to correct that. Also
allows to go back to a 'normal' decomposition that
creates line primitives as expected. That again can
then be painted quite normally.
Also added view-dependent case to the decomposition
to guarantee a gap of one discrete unit (pixel).
Removed the direct painter, too. Checked and corrected
stroking.
borderline: Adapted previews to primitives
Added code to use the primitive representation in
all dialogs and apps using tables. The edit views
use these mostly, so the preview should do that,
too. Currently missing is a good visualization of
diagonals, but this is also true for edit views.
Checked all apps and table usages to not get worse
borderline: correct line dash visualization
When a dashed line is used, a factor of 10.0 was applied in the
original coded, added that. Also the orientation of vertical
borders was inverted since it was simpler to exchange Start/End,
but this also mirrors the line dash visualisation, corrected that
Change-Id: I4c1b380a76cb37389fab1259a53fb7cc9da982d1
e95e246d5563360617a2a2213e4d5ec7d0e736b9
62369b4de58fb0264aeb710ec6983ceddca5701d
77418cc6c84ebb0632f8c3448976e82ce612d6b6
b4eb28dc86ce05eb89b26517167305b994158ef8
borderline: adapt cppunittest and clang
2017-07-04 16:28:58 +02:00
|
|
|
#include <drawinglayer/primitive2d/polygonprimitive2d.hxx>
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
#include <drawinglayer/processor2d/baseprocessor2d.hxx>
|
|
|
|
#include <drawinglayer/processor2d/processorfromoutputdevice.hxx>
|
2016-04-15 18:28:16 +02:00
|
|
|
#include <rtl/ref.hxx>
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
#include <test/bootstrapfixture.hxx>
|
|
|
|
#include <vcl/vclptr.hxx>
|
|
|
|
#include <vcl/virdev.hxx>
|
2017-03-16 12:44:19 +02:00
|
|
|
#include <editeng/borderline.hxx>
|
2017-09-08 13:39:29 +02:00
|
|
|
#include <svtools/borderhelper.hxx>
|
2016-04-15 15:46:06 +02:00
|
|
|
|
|
|
|
using namespace com::sun::star;
|
|
|
|
|
|
|
|
namespace
|
|
|
|
{
|
|
|
|
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
class DrawinglayerBorderTest : public test::BootstrapFixture
|
2016-04-15 15:46:06 +02:00
|
|
|
{
|
|
|
|
public:
|
|
|
|
void testDoubleDecompositionSolid();
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
void testDoublePixelProcessing();
|
2016-04-15 15:46:06 +02:00
|
|
|
|
|
|
|
CPPUNIT_TEST_SUITE(DrawinglayerBorderTest);
|
|
|
|
CPPUNIT_TEST(testDoubleDecompositionSolid);
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
CPPUNIT_TEST(testDoublePixelProcessing);
|
2016-04-15 15:46:06 +02:00
|
|
|
CPPUNIT_TEST_SUITE_END();
|
|
|
|
};
|
|
|
|
|
|
|
|
void DrawinglayerBorderTest::testDoubleDecompositionSolid()
|
|
|
|
{
|
|
|
|
// Create a border line primitive that's similar to the one from the bugdoc:
|
|
|
|
// 1.47 pixels is 0.03cm at 130% zoom and 96 DPI.
|
|
|
|
basegfx::B2DPoint aStart(0, 20);
|
|
|
|
basegfx::B2DPoint aEnd(100, 20);
|
2017-09-08 13:39:29 +02:00
|
|
|
double const fLeftWidth = 1.47;
|
2017-06-22 16:05:11 +02:00
|
|
|
double const fDistance = 1.47;
|
|
|
|
double const fRightWidth = 1.47;
|
|
|
|
double const fExtendLeftStart = 0;
|
|
|
|
double const fExtendLeftEnd = 0;
|
|
|
|
double const fExtendRightStart = 0;
|
|
|
|
double const fExtendRightEnd = 0;
|
2016-04-15 15:46:06 +02:00
|
|
|
basegfx::BColor aColorRight;
|
|
|
|
basegfx::BColor aColorLeft;
|
2017-09-08 13:39:29 +02:00
|
|
|
const std::vector<double> aDashing(svtools::GetLineDashing(SvxBorderLineStyle::DOUBLE, 10.0));
|
|
|
|
const drawinglayer::attribute::StrokeAttribute aStrokeAttribute(aDashing);
|
|
|
|
std::vector< drawinglayer::primitive2d::BorderLine > aBorderlines;
|
|
|
|
|
|
|
|
aBorderlines.push_back(
|
|
|
|
drawinglayer::primitive2d::BorderLine(
|
|
|
|
drawinglayer::attribute::LineAttribute(
|
|
|
|
aColorLeft,
|
|
|
|
fLeftWidth),
|
|
|
|
fExtendLeftStart,
|
|
|
|
fExtendLeftStart,
|
|
|
|
fExtendLeftEnd,
|
|
|
|
fExtendLeftEnd));
|
|
|
|
|
|
|
|
aBorderlines.push_back(
|
|
|
|
drawinglayer::primitive2d::BorderLine(fDistance));
|
|
|
|
|
|
|
|
aBorderlines.push_back(
|
|
|
|
drawinglayer::primitive2d::BorderLine(
|
|
|
|
drawinglayer::attribute::LineAttribute(
|
|
|
|
aColorRight,
|
|
|
|
fRightWidth),
|
|
|
|
fExtendRightStart,
|
|
|
|
fExtendRightStart,
|
|
|
|
fExtendRightEnd,
|
|
|
|
fExtendRightEnd));
|
|
|
|
|
2017-07-27 16:03:48 +02:00
|
|
|
rtl::Reference<drawinglayer::primitive2d::BorderLinePrimitive2D> aBorder(
|
|
|
|
new drawinglayer::primitive2d::BorderLinePrimitive2D(
|
|
|
|
aStart,
|
|
|
|
aEnd,
|
2017-09-08 13:39:29 +02:00
|
|
|
aBorderlines,
|
|
|
|
aStrokeAttribute));
|
2016-04-15 15:46:06 +02:00
|
|
|
|
|
|
|
// Decompose it into polygons.
|
|
|
|
drawinglayer::geometry::ViewInformation2D aView;
|
2016-10-25 09:54:25 +02:00
|
|
|
drawinglayer::primitive2d::Primitive2DContainer aContainer;
|
|
|
|
aBorder->get2DDecomposition(aContainer, aView);
|
2016-04-15 15:46:06 +02:00
|
|
|
|
|
|
|
// Make sure it results in two borders as it's a double one.
|
|
|
|
CPPUNIT_ASSERT_EQUAL(static_cast<std::size_t>(2), aContainer.size());
|
|
|
|
|
borderline: correct problems with border display
Borderline display with direct paint and with primitive direct
paint has quite some errors in the current state. Started to
unify usages, check deeper with creation/usage.
borderline: deep changes to BorderLine
Found basic error in determining the offset values
for BorderLinePrimitive creation, these were not
centered on the lines. Corrected that. This makes
it possible to remove the formally used clipping
which seems to have been used to correct that. Also
allows to go back to a 'normal' decomposition that
creates line primitives as expected. That again can
then be painted quite normally.
Also added view-dependent case to the decomposition
to guarantee a gap of one discrete unit (pixel).
Removed the direct painter, too. Checked and corrected
stroking.
borderline: Adapted previews to primitives
Added code to use the primitive representation in
all dialogs and apps using tables. The edit views
use these mostly, so the preview should do that,
too. Currently missing is a good visualization of
diagonals, but this is also true for edit views.
Checked all apps and table usages to not get worse
borderline: correct line dash visualization
When a dashed line is used, a factor of 10.0 was applied in the
original coded, added that. Also the orientation of vertical
borders was inverted since it was simpler to exchange Start/End,
but this also mirrors the line dash visualisation, corrected that
Change-Id: I4c1b380a76cb37389fab1259a53fb7cc9da982d1
e95e246d5563360617a2a2213e4d5ec7d0e736b9
62369b4de58fb0264aeb710ec6983ceddca5701d
77418cc6c84ebb0632f8c3448976e82ce612d6b6
b4eb28dc86ce05eb89b26517167305b994158ef8
borderline: adapt cppunittest and clang
2017-07-04 16:28:58 +02:00
|
|
|
// Get the inside line, now a PolygonStrokePrimitive2D
|
|
|
|
auto pInside = dynamic_cast<const drawinglayer::primitive2d::PolygonStrokePrimitive2D*>(aContainer[0].get());
|
2016-04-15 15:46:06 +02:00
|
|
|
CPPUNIT_ASSERT(pInside);
|
|
|
|
|
|
|
|
// Make sure the inside line's height is fLeftWidth.
|
borderline: correct problems with border display
Borderline display with direct paint and with primitive direct
paint has quite some errors in the current state. Started to
unify usages, check deeper with creation/usage.
borderline: deep changes to BorderLine
Found basic error in determining the offset values
for BorderLinePrimitive creation, these were not
centered on the lines. Corrected that. This makes
it possible to remove the formally used clipping
which seems to have been used to correct that. Also
allows to go back to a 'normal' decomposition that
creates line primitives as expected. That again can
then be painted quite normally.
Also added view-dependent case to the decomposition
to guarantee a gap of one discrete unit (pixel).
Removed the direct painter, too. Checked and corrected
stroking.
borderline: Adapted previews to primitives
Added code to use the primitive representation in
all dialogs and apps using tables. The edit views
use these mostly, so the preview should do that,
too. Currently missing is a good visualization of
diagonals, but this is also true for edit views.
Checked all apps and table usages to not get worse
borderline: correct line dash visualization
When a dashed line is used, a factor of 10.0 was applied in the
original coded, added that. Also the orientation of vertical
borders was inverted since it was simpler to exchange Start/End,
but this also mirrors the line dash visualisation, corrected that
Change-Id: I4c1b380a76cb37389fab1259a53fb7cc9da982d1
e95e246d5563360617a2a2213e4d5ec7d0e736b9
62369b4de58fb0264aeb710ec6983ceddca5701d
77418cc6c84ebb0632f8c3448976e82ce612d6b6
b4eb28dc86ce05eb89b26517167305b994158ef8
borderline: adapt cppunittest and clang
2017-07-04 16:28:58 +02:00
|
|
|
const double fLineWidthFromDecompose = pInside->getLineAttribute().getWidth();
|
|
|
|
|
2016-04-15 15:46:06 +02:00
|
|
|
// This was 2.47, i.e. the width of the inner line was 1 unit (in the bugdoc's case: 1 pixel) wider than expected.
|
borderline: correct problems with border display
Borderline display with direct paint and with primitive direct
paint has quite some errors in the current state. Started to
unify usages, check deeper with creation/usage.
borderline: deep changes to BorderLine
Found basic error in determining the offset values
for BorderLinePrimitive creation, these were not
centered on the lines. Corrected that. This makes
it possible to remove the formally used clipping
which seems to have been used to correct that. Also
allows to go back to a 'normal' decomposition that
creates line primitives as expected. That again can
then be painted quite normally.
Also added view-dependent case to the decomposition
to guarantee a gap of one discrete unit (pixel).
Removed the direct painter, too. Checked and corrected
stroking.
borderline: Adapted previews to primitives
Added code to use the primitive representation in
all dialogs and apps using tables. The edit views
use these mostly, so the preview should do that,
too. Currently missing is a good visualization of
diagonals, but this is also true for edit views.
Checked all apps and table usages to not get worse
borderline: correct line dash visualization
When a dashed line is used, a factor of 10.0 was applied in the
original coded, added that. Also the orientation of vertical
borders was inverted since it was simpler to exchange Start/End,
but this also mirrors the line dash visualisation, corrected that
Change-Id: I4c1b380a76cb37389fab1259a53fb7cc9da982d1
e95e246d5563360617a2a2213e4d5ec7d0e736b9
62369b4de58fb0264aeb710ec6983ceddca5701d
77418cc6c84ebb0632f8c3448976e82ce612d6b6
b4eb28dc86ce05eb89b26517167305b994158ef8
borderline: adapt cppunittest and clang
2017-07-04 16:28:58 +02:00
|
|
|
CPPUNIT_ASSERT_DOUBLES_EQUAL(fLeftWidth, fLineWidthFromDecompose, basegfx::fTools::getSmallValue());
|
2016-04-15 15:46:06 +02:00
|
|
|
}
|
|
|
|
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
void DrawinglayerBorderTest::testDoublePixelProcessing()
|
|
|
|
{
|
|
|
|
// Create a pixel processor.
|
|
|
|
ScopedVclPtrInstance<VirtualDevice> pDev;
|
|
|
|
drawinglayer::geometry::ViewInformation2D aView;
|
|
|
|
std::unique_ptr<drawinglayer::processor2d::BaseProcessor2D> pProcessor(drawinglayer::processor2d::createBaseProcessor2DFromOutputDevice(*pDev, aView));
|
|
|
|
CPPUNIT_ASSERT(pProcessor);
|
|
|
|
GDIMetaFile aMetaFile;
|
|
|
|
// Start recording after the processor is created, so we can test the pixel processor.
|
|
|
|
aMetaFile.Record(pDev);
|
|
|
|
|
|
|
|
// Create a border line primitive that's similar to the one from the bugdoc:
|
|
|
|
// 1.47 pixels is 0.03cm at 130% zoom and 96 DPI.
|
|
|
|
basegfx::B2DPoint aStart(0, 20);
|
|
|
|
basegfx::B2DPoint aEnd(100, 20);
|
2017-06-22 16:05:11 +02:00
|
|
|
double const fLeftWidth = 1.47;
|
|
|
|
double const fDistance = 1.47;
|
borderline: correct problems with border display
Borderline display with direct paint and with primitive direct
paint has quite some errors in the current state. Started to
unify usages, check deeper with creation/usage.
borderline: deep changes to BorderLine
Found basic error in determining the offset values
for BorderLinePrimitive creation, these were not
centered on the lines. Corrected that. This makes
it possible to remove the formally used clipping
which seems to have been used to correct that. Also
allows to go back to a 'normal' decomposition that
creates line primitives as expected. That again can
then be painted quite normally.
Also added view-dependent case to the decomposition
to guarantee a gap of one discrete unit (pixel).
Removed the direct painter, too. Checked and corrected
stroking.
borderline: Adapted previews to primitives
Added code to use the primitive representation in
all dialogs and apps using tables. The edit views
use these mostly, so the preview should do that,
too. Currently missing is a good visualization of
diagonals, but this is also true for edit views.
Checked all apps and table usages to not get worse
borderline: correct line dash visualization
When a dashed line is used, a factor of 10.0 was applied in the
original coded, added that. Also the orientation of vertical
borders was inverted since it was simpler to exchange Start/End,
but this also mirrors the line dash visualisation, corrected that
Change-Id: I4c1b380a76cb37389fab1259a53fb7cc9da982d1
e95e246d5563360617a2a2213e4d5ec7d0e736b9
62369b4de58fb0264aeb710ec6983ceddca5701d
77418cc6c84ebb0632f8c3448976e82ce612d6b6
b4eb28dc86ce05eb89b26517167305b994158ef8
borderline: adapt cppunittest and clang
2017-07-04 16:28:58 +02:00
|
|
|
double const fRightWidth = 1.47;
|
2017-06-22 16:05:11 +02:00
|
|
|
double const fExtendLeftStart = 0;
|
|
|
|
double const fExtendLeftEnd = 0;
|
|
|
|
double const fExtendRightStart = 0;
|
|
|
|
double const fExtendRightEnd = 0;
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
basegfx::BColor aColorRight;
|
|
|
|
basegfx::BColor aColorLeft;
|
2017-09-08 13:39:29 +02:00
|
|
|
const std::vector<double> aDashing(svtools::GetLineDashing(SvxBorderLineStyle::DOUBLE, 10.0));
|
|
|
|
const drawinglayer::attribute::StrokeAttribute aStrokeAttribute(aDashing);
|
|
|
|
std::vector< drawinglayer::primitive2d::BorderLine > aBorderlines;
|
|
|
|
|
|
|
|
aBorderlines.push_back(
|
|
|
|
drawinglayer::primitive2d::BorderLine(
|
|
|
|
drawinglayer::attribute::LineAttribute(
|
|
|
|
aColorLeft,
|
|
|
|
fLeftWidth),
|
|
|
|
fExtendLeftStart,
|
|
|
|
fExtendLeftStart,
|
|
|
|
fExtendLeftEnd,
|
|
|
|
fExtendLeftEnd));
|
|
|
|
|
|
|
|
aBorderlines.push_back(
|
|
|
|
drawinglayer::primitive2d::BorderLine(fDistance));
|
|
|
|
|
|
|
|
aBorderlines.push_back(
|
|
|
|
drawinglayer::primitive2d::BorderLine(
|
|
|
|
drawinglayer::attribute::LineAttribute(
|
|
|
|
aColorRight,
|
|
|
|
fRightWidth),
|
|
|
|
fExtendRightStart,
|
|
|
|
fExtendRightStart,
|
|
|
|
fExtendRightEnd,
|
|
|
|
fExtendRightEnd));
|
|
|
|
|
|
|
|
rtl::Reference<drawinglayer::primitive2d::BorderLinePrimitive2D> aBorder(
|
2017-07-27 16:03:48 +02:00
|
|
|
new drawinglayer::primitive2d::BorderLinePrimitive2D(
|
|
|
|
aStart,
|
|
|
|
aEnd,
|
2017-09-08 13:39:29 +02:00
|
|
|
aBorderlines,
|
|
|
|
aStrokeAttribute));
|
2017-07-27 16:03:48 +02:00
|
|
|
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
drawinglayer::primitive2d::Primitive2DContainer aPrimitives;
|
2017-09-08 13:39:29 +02:00
|
|
|
aPrimitives.push_back(drawinglayer::primitive2d::Primitive2DReference(aBorder.get()));
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
|
|
|
|
// Process the primitives.
|
|
|
|
pProcessor->process(aPrimitives);
|
|
|
|
|
borderline: correct problems with border display
Borderline display with direct paint and with primitive direct
paint has quite some errors in the current state. Started to
unify usages, check deeper with creation/usage.
borderline: deep changes to BorderLine
Found basic error in determining the offset values
for BorderLinePrimitive creation, these were not
centered on the lines. Corrected that. This makes
it possible to remove the formally used clipping
which seems to have been used to correct that. Also
allows to go back to a 'normal' decomposition that
creates line primitives as expected. That again can
then be painted quite normally.
Also added view-dependent case to the decomposition
to guarantee a gap of one discrete unit (pixel).
Removed the direct painter, too. Checked and corrected
stroking.
borderline: Adapted previews to primitives
Added code to use the primitive representation in
all dialogs and apps using tables. The edit views
use these mostly, so the preview should do that,
too. Currently missing is a good visualization of
diagonals, but this is also true for edit views.
Checked all apps and table usages to not get worse
borderline: correct line dash visualization
When a dashed line is used, a factor of 10.0 was applied in the
original coded, added that. Also the orientation of vertical
borders was inverted since it was simpler to exchange Start/End,
but this also mirrors the line dash visualisation, corrected that
Change-Id: I4c1b380a76cb37389fab1259a53fb7cc9da982d1
e95e246d5563360617a2a2213e4d5ec7d0e736b9
62369b4de58fb0264aeb710ec6983ceddca5701d
77418cc6c84ebb0632f8c3448976e82ce612d6b6
b4eb28dc86ce05eb89b26517167305b994158ef8
borderline: adapt cppunittest and clang
2017-07-04 16:28:58 +02:00
|
|
|
// Double line now gets decomposed in Metafile to painting four lines
|
|
|
|
// with width == 0 in a cross pattern due to real line width being between
|
|
|
|
// 1.0 and 2.0. Count created lines
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
aMetaFile.Stop();
|
|
|
|
aMetaFile.WindStart();
|
borderline: correct problems with border display
Borderline display with direct paint and with primitive direct
paint has quite some errors in the current state. Started to
unify usages, check deeper with creation/usage.
borderline: deep changes to BorderLine
Found basic error in determining the offset values
for BorderLinePrimitive creation, these were not
centered on the lines. Corrected that. This makes
it possible to remove the formally used clipping
which seems to have been used to correct that. Also
allows to go back to a 'normal' decomposition that
creates line primitives as expected. That again can
then be painted quite normally.
Also added view-dependent case to the decomposition
to guarantee a gap of one discrete unit (pixel).
Removed the direct painter, too. Checked and corrected
stroking.
borderline: Adapted previews to primitives
Added code to use the primitive representation in
all dialogs and apps using tables. The edit views
use these mostly, so the preview should do that,
too. Currently missing is a good visualization of
diagonals, but this is also true for edit views.
Checked all apps and table usages to not get worse
borderline: correct line dash visualization
When a dashed line is used, a factor of 10.0 was applied in the
original coded, added that. Also the orientation of vertical
borders was inverted since it was simpler to exchange Start/End,
but this also mirrors the line dash visualisation, corrected that
Change-Id: I4c1b380a76cb37389fab1259a53fb7cc9da982d1
e95e246d5563360617a2a2213e4d5ec7d0e736b9
62369b4de58fb0264aeb710ec6983ceddca5701d
77418cc6c84ebb0632f8c3448976e82ce612d6b6
b4eb28dc86ce05eb89b26517167305b994158ef8
borderline: adapt cppunittest and clang
2017-07-04 16:28:58 +02:00
|
|
|
sal_uInt32 nPolyLineActionCount = 0;
|
|
|
|
|
2016-04-19 13:55:33 +02:00
|
|
|
for (std::size_t nAction = 0; nAction < aMetaFile.GetActionSize(); ++nAction)
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
{
|
|
|
|
MetaAction* pAction = aMetaFile.GetAction(nAction);
|
borderline: correct problems with border display
Borderline display with direct paint and with primitive direct
paint has quite some errors in the current state. Started to
unify usages, check deeper with creation/usage.
borderline: deep changes to BorderLine
Found basic error in determining the offset values
for BorderLinePrimitive creation, these were not
centered on the lines. Corrected that. This makes
it possible to remove the formally used clipping
which seems to have been used to correct that. Also
allows to go back to a 'normal' decomposition that
creates line primitives as expected. That again can
then be painted quite normally.
Also added view-dependent case to the decomposition
to guarantee a gap of one discrete unit (pixel).
Removed the direct painter, too. Checked and corrected
stroking.
borderline: Adapted previews to primitives
Added code to use the primitive representation in
all dialogs and apps using tables. The edit views
use these mostly, so the preview should do that,
too. Currently missing is a good visualization of
diagonals, but this is also true for edit views.
Checked all apps and table usages to not get worse
borderline: correct line dash visualization
When a dashed line is used, a factor of 10.0 was applied in the
original coded, added that. Also the orientation of vertical
borders was inverted since it was simpler to exchange Start/End,
but this also mirrors the line dash visualisation, corrected that
Change-Id: I4c1b380a76cb37389fab1259a53fb7cc9da982d1
e95e246d5563360617a2a2213e4d5ec7d0e736b9
62369b4de58fb0264aeb710ec6983ceddca5701d
77418cc6c84ebb0632f8c3448976e82ce612d6b6
b4eb28dc86ce05eb89b26517167305b994158ef8
borderline: adapt cppunittest and clang
2017-07-04 16:28:58 +02:00
|
|
|
|
|
|
|
if (MetaActionType::POLYLINE == pAction->GetType())
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
{
|
borderline: correct problems with border display
Borderline display with direct paint and with primitive direct
paint has quite some errors in the current state. Started to
unify usages, check deeper with creation/usage.
borderline: deep changes to BorderLine
Found basic error in determining the offset values
for BorderLinePrimitive creation, these were not
centered on the lines. Corrected that. This makes
it possible to remove the formally used clipping
which seems to have been used to correct that. Also
allows to go back to a 'normal' decomposition that
creates line primitives as expected. That again can
then be painted quite normally.
Also added view-dependent case to the decomposition
to guarantee a gap of one discrete unit (pixel).
Removed the direct painter, too. Checked and corrected
stroking.
borderline: Adapted previews to primitives
Added code to use the primitive representation in
all dialogs and apps using tables. The edit views
use these mostly, so the preview should do that,
too. Currently missing is a good visualization of
diagonals, but this is also true for edit views.
Checked all apps and table usages to not get worse
borderline: correct line dash visualization
When a dashed line is used, a factor of 10.0 was applied in the
original coded, added that. Also the orientation of vertical
borders was inverted since it was simpler to exchange Start/End,
but this also mirrors the line dash visualisation, corrected that
Change-Id: I4c1b380a76cb37389fab1259a53fb7cc9da982d1
e95e246d5563360617a2a2213e4d5ec7d0e736b9
62369b4de58fb0264aeb710ec6983ceddca5701d
77418cc6c84ebb0632f8c3448976e82ce612d6b6
b4eb28dc86ce05eb89b26517167305b994158ef8
borderline: adapt cppunittest and clang
2017-07-04 16:28:58 +02:00
|
|
|
auto pMPLAction = static_cast<MetaPolyLineAction*>(pAction);
|
|
|
|
|
|
|
|
if (0 == pMPLAction->GetLineInfo().GetWidth() && LineStyle::Solid == pMPLAction->GetLineInfo().GetStyle())
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
{
|
borderline: correct problems with border display
Borderline display with direct paint and with primitive direct
paint has quite some errors in the current state. Started to
unify usages, check deeper with creation/usage.
borderline: deep changes to BorderLine
Found basic error in determining the offset values
for BorderLinePrimitive creation, these were not
centered on the lines. Corrected that. This makes
it possible to remove the formally used clipping
which seems to have been used to correct that. Also
allows to go back to a 'normal' decomposition that
creates line primitives as expected. That again can
then be painted quite normally.
Also added view-dependent case to the decomposition
to guarantee a gap of one discrete unit (pixel).
Removed the direct painter, too. Checked and corrected
stroking.
borderline: Adapted previews to primitives
Added code to use the primitive representation in
all dialogs and apps using tables. The edit views
use these mostly, so the preview should do that,
too. Currently missing is a good visualization of
diagonals, but this is also true for edit views.
Checked all apps and table usages to not get worse
borderline: correct line dash visualization
When a dashed line is used, a factor of 10.0 was applied in the
original coded, added that. Also the orientation of vertical
borders was inverted since it was simpler to exchange Start/End,
but this also mirrors the line dash visualisation, corrected that
Change-Id: I4c1b380a76cb37389fab1259a53fb7cc9da982d1
e95e246d5563360617a2a2213e4d5ec7d0e736b9
62369b4de58fb0264aeb710ec6983ceddca5701d
77418cc6c84ebb0632f8c3448976e82ce612d6b6
b4eb28dc86ce05eb89b26517167305b994158ef8
borderline: adapt cppunittest and clang
2017-07-04 16:28:58 +02:00
|
|
|
nPolyLineActionCount++;
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
borderline: correct problems with border display
Borderline display with direct paint and with primitive direct
paint has quite some errors in the current state. Started to
unify usages, check deeper with creation/usage.
borderline: deep changes to BorderLine
Found basic error in determining the offset values
for BorderLinePrimitive creation, these were not
centered on the lines. Corrected that. This makes
it possible to remove the formally used clipping
which seems to have been used to correct that. Also
allows to go back to a 'normal' decomposition that
creates line primitives as expected. That again can
then be painted quite normally.
Also added view-dependent case to the decomposition
to guarantee a gap of one discrete unit (pixel).
Removed the direct painter, too. Checked and corrected
stroking.
borderline: Adapted previews to primitives
Added code to use the primitive representation in
all dialogs and apps using tables. The edit views
use these mostly, so the preview should do that,
too. Currently missing is a good visualization of
diagonals, but this is also true for edit views.
Checked all apps and table usages to not get worse
borderline: correct line dash visualization
When a dashed line is used, a factor of 10.0 was applied in the
original coded, added that. Also the orientation of vertical
borders was inverted since it was simpler to exchange Start/End,
but this also mirrors the line dash visualisation, corrected that
Change-Id: I4c1b380a76cb37389fab1259a53fb7cc9da982d1
e95e246d5563360617a2a2213e4d5ec7d0e736b9
62369b4de58fb0264aeb710ec6983ceddca5701d
77418cc6c84ebb0632f8c3448976e82ce612d6b6
b4eb28dc86ce05eb89b26517167305b994158ef8
borderline: adapt cppunittest and clang
2017-07-04 16:28:58 +02:00
|
|
|
|
|
|
|
// Check if all eight (2x four) simple lines with width == 0 and
|
|
|
|
// solid were created
|
|
|
|
const sal_uInt32 nExpectedNumPolyLineActions = 8;
|
|
|
|
|
|
|
|
CPPUNIT_ASSERT_EQUAL(nExpectedNumPolyLineActions, nPolyLineActionCount);
|
tdf#99315 VclPixelProcessor2D: fix double border line width
Regression from commit 2c91cb08d65cd35fa8ef6eaca3677aa82fb58cbe (better
drawing support for borders of different width, fdo#33634, 2012-04-04),
the problem is that previously the width of inner/outer double border
lines got rounded to integer values quite early, but after the commit
they are kept at a double precision for much longer, which needs pixel
correction in VclPixelProcessor2D.
Example: if the border with is 1.47, and the line gets moved by 0.2
pixels, then the inner and outer edge of the line will be 0.2 and 1.67,
which gets rounded to 0 -> 2 in the pixel processor. Previously the
input was rounded to 1, so moving by 0.2 resulted in 0.2 -> 1.2, which
got rounded to 0 -> 1. The result is that sometimes the line width is 1
pixel wider than expected.
Fix the problem by allowing VclPixelProcessor2D to request pixel
correction from BorderLinePrimitive2D. It wouldn't be possible to do
pixel correction only in VclPixelProcessor2D, as it has no idea what to
correct: it only gets polygons, so it has no idea if e.g. the top of a
polygon is the outer edge of a top border line or an inner edge of a
bottom border line.
Change-Id: I1971f3a952fbcdc598ab46c659e12d976c13cbe6
Reviewed-on: https://gerrit.libreoffice.org/24221
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2016-04-18 16:53:48 +02:00
|
|
|
}
|
|
|
|
|
2016-04-15 15:46:06 +02:00
|
|
|
CPPUNIT_TEST_SUITE_REGISTRATION(DrawinglayerBorderTest);
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
CPPUNIT_PLUGIN_IMPLEMENT();
|
|
|
|
|
|
|
|
/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
|