thanks

LGTM, is the new function available since old CUPS versions? Just to make sure it won't break usage with old versions.

Well I still think that this applet rewrite is very useless, very little care about their printers, it's the print jobs that matter, still the patch looks fine, although I see the buttons showing up before the selection is fully expanded.

Looks Great To Me!

LGTM

Well I think your approach is probably the best to this, I'd only change the sanitize loop to use iterators and erase as afair they are a little faster.

But the remove old jobs code would also need fixing to account with the dups

Oh, now I see the problem with this, I tried many times reproduce but never managed to see the issue with the dups IDs.

Lgtm thanks

Ok, while not an ideal change, these options aren't the kind of options that should be changing that often so I'm ok. Thanks

Yes, the idea is to be up to date, the issue it seems to be that some CUPS installations keep asking the server password on each query instead of having it cached.

thanks

Thanks +1

thanks

I ended up compiling it using vcpkg toolchain directly on Windows, which showed this isn't required.

I'm not entirely sure this is really needed on Windows, but when cross-compiling this allowed to search for the katehighlightingindexer.exe which wouldn't run because I was on Linux...

LGTM Thanks!

LGTM

LGTM

Well the issue will probably depend on the server,
for example Apache2 with Nextcloud it returns forbidden for both operations,
in Cloudlyst which does:
QUrl::fromEncoded(QStringLiteral("file:///speçiál").toLatin1()).toLocalFile();
you get a file with a different file name as URL.
Since 'ç' is extended ASCII this "fails" only because URL are sopposed
to use only printable ASCII chars

Test plan:

Ping? This is a pretty obvious fix, desturl is QString which later becomes latin1 loosing characters...

In D14006#289680, @ngraham wrote:

Does this fix any of the bugs here? https://bugs.kde.org/buglist.cgi?component=webdav&list_id=1529637&product=kio&resolution=---

LGTM I only confused about why do you need QRegExp :P

AFAIK this is the wrong way of handling signals, don't recall exactly what the problem is, but Qt documentation says to create a pipe, attach a QSocketNotifier to it, and write data to it when the signal is caught, then the QSN signal will call qApp->quit();

LGTM

OK, thanks I'll probably make it mandatory for 1.1.0 as it's a simple dep and the less ifdefs the better

master please

oops my bad

Thanks, I've been thinking into making this dependency mandatory, could you please tell me why would you like to build without AppStream or are you just fixing the bug? thanks

LGTM