In D17955#387893, @sars wrote:

Besides the processEvents() I'm OK with this! :)

Typos are fixed. Thanks!

How about that:
Ok for QDir(), but still for local files only
As for SaveLocation - let's just pop up m_saveLocation dialog till user enter a valid directory that we can create.

Display dirUrl.path() instead of dir in message as dir will have "file://" schema. It just looks better without it

Done, let me know if CI fails again.

Hi, Kåre
I won't be able to test this. It's hard to setup build chain in virtual machine and it depends on some other kf5 libs.
How about just commiting it? As Win compilation is already broken it won't make worse

In D11070#229018, @eje211 wrote:

One problem I could not crack is that adding a mangnet link does not work as
expected. The code that adds magnet links is the same as the callback from
the code that adds magnet links from the callback menus, but when my plugin
tries to add a magnet link, they just appear in the "Magnet Downloads" panel
and they there with zero peers. I have no idea why.

I doubt I could be useful as a reviewer here bcs of lack of web/JS knowledge. Just wan't to say that in my opinion security evaluation should be prioritised. For a last few months I saw at least two news about hacker attacks on linux-based torrents via their web interfaces.

Wording of a warning message is changed.
With dot at sentence end.
A screenshot of warning appearance is attached{F5491726}

@ngraham Feel free to do it yourself.

Ok, I've added a warning for this. Please check wording - english isn't my native language. Feel free to change it.
Attached are screenshots of Okular showing current warning for eng and rus locales.

warning message was added

In D8415#162014, @aacid wrote:

In D8415#160837, @trufanov wrote:

In D8415#159049, @aacid wrote:

Please consider fixing this in Qt before asking us to create a workaround.

I'm fine with workarounded gwenview as well as working GIMP and ImageMagick.

Is this your way of saying "i'm not interested in pursuing a fix in Qt"?

In D8415#159049, @aacid wrote:

Please consider fixing this in Qt before asking us to create a workaround.

In D8415#158804, @aacid wrote:

Wouldn't it make more sense to fix this in QImageReader and not in every user of QImageReader ?

Ok, I've edited summary. Didn't know that KDE Bugs could be referred in a such way in phabricator summary instead of commit message.
Also I heard I can add
Differential Revision: https://phabricator.kde.org/D8415
in commit message to automatically close phabricator review. Is it so?