Looks like you fixed the rounded corner issues, nice. This does look much better IMO. +1 from me.

In D26739#596363, @ndavis wrote:

You can't shrink the size of it to fit within the blue border?

Yeah. I don't disagree. But if I just ensure that the highlight effect is the highest item, the borders still get obscured by the player controls and album art. And the rounded corners of the highlight leave a pixel or two of the art to escape the bounds too.

I'm not 100% sold on drawing the highlight on top and tinting everything, and can revert that part if everyone else hates it.

Yeah, it might be nice to extract this into a framework that can bee used by all three apps so it can't diverge this way again.

Abandoning in favor of those patches. Thanks @meven!

Does this fix https://bugs.kde.org/show_bug.cgi?id=403499?

Works for me FWIW.

The "Dans une application fondée sue le contenu d l'URL" setting is actually something I proposed removing in D17372, and it got approval in theory, but it seems like this is quite a far-reaching change that requires many considerations.

Address review comments

@broulik discovered that this needs qRegisterMetaType<Baloo::IndexerState>(); added

Re-added the namespace but the state checking still doesn't seem to work properly.

Re-add namespace

The goal is to have a sane default size.

Adjust to changes in D25743

Baloo::Suspended doesn't seem to be exported properly or set. Also it seems like toggleSuspendState() only ever suspends and never resumes. Looking at the code in monitor.cpp, it seems like these may be related.

Rebase

Friendly Dolphin ping. :)

Ping

Ping

@vinzenzv are you interested in finishing this, or should someone else take it over?

@vinzenzv are you interested in finishing this, or should someone else take it over?

In T12221#218075, @vinzenzv wrote:

The "advanced users" bit is targeting at openSUSE, not at Gnome, Xfce or Plasma. We chose Ubuntu as basis for various reasons and went with Budgie as some simple, classic DE especially for newbies. This decision dates from early 2017 and is something we currently re-evaluate. Some details still need to be clarified but everything points towards Plasma as our default DE on all devices (ultrabooks, gaming laptops, mini & regular PCs).

LGTM but please wait for confirmation from @sandsmark

Please always add screenshots for patches that result in UI changes.

Looks good, and this should go on the plasma/5.18 branch to make sure it gets into Plasma 5.18 so this KCM isn't the odd one out.

Good point, I totally forgot to consider that. I have no idea how it works and will need to look into it.

I suppose it doesn't hurt anything to add these even though they're still "up for discussion."

When set to false, it makes the label look less lighter without manually setting the opacity to some value that we'll often forget to be consistent on which would (and does) result in labels of opacity values varying between 0.3 and 0.6.

All right, no problem!

Selecting the "Other..." option for this one takes me to the filetype editor for inode/directory instead of the app chooser as in the other patches. Is that intentional?

Yep.

The whole point was to save some space on the user's machine. However trading space savings for load time maybe isn't the best trade-off? 18 MB isn't really all that much of a savings.

Thanks Noah.

Note that I found the code here very complicated so it's possible I may have done this in a sub-optimal way.

IIRC we used to do this, but changed it to run-on-demand to save some memory on startup.

No I see it as a consistency/polish/bugfix thing. The rest of the KCMs got ported to use the new dialog for Plasma 5.18; it would be a bit weird if this one didn't get the same treatment.

Even better would be if we can use the new QML dialogs too. See D26544

Very nice first patch. May it be the first of many! :)

no in a comment is fine

+ ten million, this is a fantastic idea. By showing this, we close the loop for regular users who want to know how to get our software, and we also highlight partners and incentivize future hardware vendors to ship our stuff because of the free publicity they'll get.

@chroniceel Can you provid an email address so we can land this patch with correct authorship information?

I'm so sorry, this was my mistake. Somehow KMail got re-installed without my knowledge. Probably https://bugzilla.suse.com/show_bug.cgi?id=1035802 biting me again.

Hmm, I don't have KMail installed yet it shows up in the combobox anyway. I don't think that makes sense.

Here's my suggestions for what to display in the applet for various conditions:

Would you mind rebasing this? It seems like we're moving forward with the idea of iterating on Breeze rather than making a whole new theme, and since Plasma 5.18 just branched, we have some space to hack on master now. I'd like to see if we can move forward with this.

Hmm, I'm not totally sure I agree. If the text becomes focusable by tabbing, then I think that just clutters up the focus chain with non-interactive UI elements that nobody is actually going to care about doing something with 99% of the time.

Oh never mind, I see it now. Quite subtle! Nice fix.

Much nicer.

Hmm, I don't notice a difference before and after this patch. Maybe I'm not noticing the original problem

I don't know of any netbooks shorter with screens shorter than 600px, which is why I chose 600. Also, the window is resizable to be smaller.

Very nice. It occurs to me that by showing the full list of what gets sent, we no longer need to show the details text. It's rendered obsolete by showing the actual list of what gets sent. Can we remove the description from the KCM here without touching KUSerFeedback again?

Yep, go ahead.

LGTM. @ndavis?

Will we still need the LayoutMirroring.enabled ? logic to account for when using an RTL language where the scrollbar is on the left side?

Strings can't be changed during the beta.

Ah I didn't even notice, perfect.

Push it! Push it real good! :)

I think we can call this done. There are a few enhancements such as syncing the window decoration button orders (https://bugs.kde.org/show_bug.cgi?id=414113) and presence or absence of the circle behind the close button but I think we can do those later since they weren't already implemented. The original scope is definitely done now.