it pointed to the page
https://www.davidrevoy.com/article193/compile-krita-from-source-code-on-linux-for-cats
which itself points to the more recent
https://docs.krita.org/en/untranslatable_pages/building_krita.html

When the value of a QDoubleSpinBox is changed it will be round down to the amount of decimals that are assigned to that spin box. The KisDoubleParseSpinBox would check the new unrounded value with the current rounded value, which resulted in an incorrect comparison.

Only in this mode will Qt respect the contents of the ComboBox popup. This is what Fusion uses.

BUG: 406595

Differential Revision: https://phabricator.kde.org/D25739

The crash happens only on systems that have asserts enabled, that
is, don't have NDEBUG defined.

Binding uninitialized '*m_savedTransformArgs' to a cont-reference
generated valid c++ code (a reference initialized with nullptr),
and given that finishStrokeImpl() didn't try to use/dereference
this reference, the code worked fine and didn't crash on systems
without asserts.

But on systems with asserts enabled (non NDEBUG), boost::optional
triggered a sanity check assert for dereferencing nullptr and crashed
the application.

The patch removes entire code for resetting m_savedTransformArgs in
cancelStrokeCallback(). It was actually an artifact of some initial
refactoring.

Thanks Fredrik Hansson for pointing out the problem!
!197

BUG:414672

Because if we do, it won't be deleted before Qt removes all fonts
on destroying QApplication, and then QFontCombBox will try to get
an active window.

BUG:414718

This returns QSize(0,0) in all cases, so it's useless to call it.
And it's going to be deprecated anyway.

This adjust the diferent position of KisPaintInformation
from a line to be inside a line between starting and
ending points. This is particularly important as brushes
that use 1px perfect pixels require the line to be perfect
in order to avoid looking jagged. In 1px size with
sharpen active we only use the start and end
KisPaintInformation points.

This patch also adjusts the outine on 1px brushes to
adjust to the grid to better preview where the paint is
going to be applyied

BUG:407405

Before this commit, there was no maximum set on the double
spin box which holded the value. Because by default it would
be filled with half of the width and height of the layer
extent, there will be some maximum value set bigger
than 99 which is the default for double spin boxes.
This commit fixes this behaviour by setting 100 000
as a maximum explicitely.

BUG:414625

Using onion skins on layers without any content keyframes
is possible if one is using opacity keyframes.
Before this commit, turning onion skins on would cause a crash.
After this commit, turning onion skins on will do nothing.

BUG:414668

Before this commit there was a mistake in calculating
the border of the selection. It was caused by misreading
the original calculation equation while fixing the
CID issue (dividing by 0).
This commit fixes the behaviour by ensuring that
the value is being multiplied, not divided by sqrt.

BUG:414647

This is the backtrace:

0  0x00007ffff38a11cc in KoColorSet::getGroup(QString const&) () from
/usr/lib64/libkritapigment.so.18
1  0x00007ffff3d2aa69 in KisPaletteModel::index(int, int, QModelIndex
const&) const () from /usr/lib64/libkritawidgets.so.18
2  0x00007ffff5812796 in
QAbstractItemView::setModel(QAbstractItemModel*) () from
/usr/lib64/libQt5Widgets.so.5
3  0x00007ffff582e749 in QHeaderView::setModel(QAbstractItemModel*) ()
from /usr/lib64/libQt5Widgets.so.5
4  0x00007ffff5871493 in QTableView::setModel(QAbstractItemModel*) ()
from /usr/lib64/libQt5Widgets.so.5
5  0x00007ffff3d31e17 in
KisPaletteView::setPaletteModel(KisPaletteModel*) () from
/usr/lib64/libkritawidgets.so.18
6  0x00007ffff3d2691e in
KisDlgInternalColorSelector::KisDlgInternalColorSelector(QWidget*,
KoColor, KisDlgInternalColorSelector::Config, QString const&,
KoColorDisplayRendererInterface const*) () from
/usr/lib64/libkritawidgets.so.18
7  0x00007ffff7000d67 in
KoDualColorButton::Private::init(KoDualColorButton*) () from
/usr/lib64/libkritaui.so.18
8  0x00007ffff7002c5b in KoDualColorButton::KoDualColorButton(KoColor
const&, KoColor const&, KoColorDisplayRendererInterface const*,
QWidget*, QWidget*) ()
  from /usr/lib64/libkritaui.so.18
9  0x00007ffff6e2b5b5 in KisControlFrame::setup(QWidget*) () from
/usr/lib64/libkritaui.so.18
10 0x00007ffff71014ea in KisViewManager::KisViewManager(QWidget*,
KActionCollection*) () from /usr/lib64/libkritaui.so.18
11 0x00007ffff70cb72b in KisMainWindow::KisMainWindow(QUuid) () from
/usr/lib64/libkritaui.so.18
12 0x00007ffff70d8352 in KisPart::createMainWindow(QUuid) () from
/usr/lib64/libkritaui.so.18
13 0x00007ffff70d85de in KisPart::startBlankSession() () from
/usr/lib64/libkritaui.so.18
14 0x00007ffff70995b9 in KisApplication::start(KisApplicationArguments
const&) () from /usr/lib64/libkritaui.so.18
15 0x0000555555e4e229 in main ()

The KisPaletteModel::index() call is inside an Q_ASSERT_X in
QAbstractItemView::setModel(), so you have to compile Qt in debug mode
to reproduce this crash.

m_rowGroupNameMap is empty during the startup.

BUG: 41425

Weirdly enough, this code has been here for 13 years, so I guess
it never got executed...

qCompare doesn't allow different types, related to commit #7a709c54

It's broken, and I don't know why.

Krita doesn't support it, so we shouldn't show it. Especially not
on Windows, where the Windows Store quality people object to it
not working.

Hopefully, fake paint device will never be used, but we cannot
guarantee it.

This reverts commit a1c74034.

When doing KisPaintDevice::prepareClone(), we should first set
new defaultBounds object, and only after that clone the data.
The reason for this: switching defaultBounds can change current
level-of-detail/frameId, and therefore, switch currentData().

BUG:412361

CCBUG:412361

It is needed for debugging/sanity-check purposes only. It returns an
abstract pointer to the source object, where default bounds takes its
data from. One can ckeck if the pointer is null or compare two pointers
of different default bounds objects.

CCBUG:412361

Before KisFileLayer loaded its image (or when the image is not
available), file layer uses a fake paint device. This paint device
should also have proper default bounds set.

CCBUG:412361

In case of conflict in i18n, keep the version of the branch "ours"
To resolve a particular conflict, "git checkout --ours path/to/file.desktop"

We need to take layer's masks' change rect into account. We cannot
use normal changeRect() call, because it never shrinks the update rect
and adds projection device bounds (obviously, because it is supposed
to render stuff and not just shwo a decoration).

That is why the patch introduces KisLayer::tightUserVisibleBounds(), which
calculates tight bounding rect of the layer, taking all the masks into
account.

BUG:410163