Try also setting mipmap to true: https://doc.qt.io/qt-5/qml-qtquick-image.html#mipmap-prop.
Maybe it will give better results (disabled by default).

Simple Ruby script to remove black outlines in OpenMoji flags:

require 'rexml/xpath'
require 'rexml/document'

USA flag:

Black outline in OpenMoji can be simply hidden by this CSS rule:

<style>
  #line { display: none; }
</style>

Or removed with use of any XML library - by removing tag <g id="line">.

Nice 👍

Patch won't apply directly, but you've done the work finding the bug, I can sort that.

Just updated the commit.

Unneeded prefix akonadi/ removed.

Another approach is to remove prefix akonadi/ from all calls to the StandardDirs::locateResourceFile (second parameter):

src/widgets/selftestdialog.cpp
320:    const QString globalConfig = StandardDirs::locateResourceFile("config", QStringLiteral("akonadi/mysql-global.conf"));
335:    const QString localConfig  = StandardDirs::locateResourceFile("config", QStringLiteral("akonadi/mysql-local.conf"));

Thanks!

https://stackoverflow.com/questions/16562121/git-diff-head-vs-staged

No problem. I'm closing revision then.
Thanks for the feedback.

Just found another bug on the KDE bugzilla related to this: https://bugs.kde.org/show_bug.cgi?id=410357.

Wrong. Still present in KF frameworks 5.61:

Bug still present in plasma 5.16.4.

In D22528#497803, @sitter wrote:

Looks good to me. The slave indeed must issue an exit state there.

What email address would you like to have associated with the git commit?

Patch was made against version 5.60.0 (rev 8513ca9).
Sorry that I didn't mention it earlier.

One more small thing: it's better to replace sCopyFile in call to error with something like dest.path(),
to display server path rather than local source.

New signal slot syntax