In D28932#657011, @ngraham wrote:

I'll get around to reviewing this soon. I'm trying to figure out of I think the loss is acceptable.

Ping!

Ping!

Ping!

Ping!

In T8569#227200, @ngraham wrote:

In T8569#227195, @bruns wrote:

I like the inline password, though I have a small concern - grey on white is low contrast, what about persons with bad eyesight?

We should probably fix that in our mockup toolkit. The actual looks of the widgets and icons itself isn't a part of this task and isn't scheduled to change AFAIK.

In D28590#656133, @meven wrote:

In D28590#654139, @bruns wrote:

Do not create m_storageAccess in the constructor

Hmm, you told me the f (m_displayName.isEmpty()) block belongs here, I don't see why instantiating m_storageAccess here is bad but mehh.

Bruns's comment about right-handed mice (gradient) is due to the quite asymmetrical shape many modern mice use for ergonomic reasons (RSI).

The new one looks very oldschool. Some ideas how to improve that:

Do not create m_storageAccess in the constructor

In D20739#653647, @rjvbb wrote:

any bug reports for the Windows platform will be rejected immediately then.

Because they aren't already given that baloo only works on Linux and *BSD?

This is the sort of attitude that always makes my eyes roll whenever I see a reference to the idea of support in most KDE software.

In D20739#653485, @rjvbb wrote:

Qt 5.10. is required since KF 5.55: https://kde.org/announcements/kde-frameworks-5.55.0.php.

True, but not relevant here because the issue is not with KF5 APIs but with QMap. The frameworks relevant for baloo all build against Qt 5.9 either without or with minimal patching (which in essence reintroduce removed code).

In D20739#653376, @rjvbb wrote:

Hi,

I'm updating some of my packaging for a set-up that should meet the minimum requirements that were current when this commit was made (Qt 5.9 and KF5 5.60.0; baloo-widgets 19.08.3 requires Qt 5.8 and KF5 5.58.0). Evidently this change wasn't checked against those requirements; the constKeyValue* iterators were introduced in Qt 5.10 (this lack of testing is also apparent in filemetadatadatedisplaytest.cpp:57!)

I've tried to roll a Qt 5.9 implementation of the new algorithm, but apparently failed:

Can I ask some feedback (from @astippich maybe)?

Ping!

Ping!

In D28995#652764, @ngraham wrote:

Aren't Imperial and US the same thing?The USA is the only country that still officially uses the imperiaI system.

In D28980#652102, @mgallien wrote:

To be clear, I am fine with this change going in.

I would still like to work on a proper solution to allow an application to monitor moved files or similar stuff without having to add file system watchers itself (duplicate from Baloo owns watchers).

Do you have an idea ?

In D28980#652087, @mgallien wrote:

The unusual method to get notified is about avoiding signaling moved files if nobody is watching. Baloo will only callback if applications have registered themselves.

I like the inline password, though I have a small concern - grey on white is low contrast, what about persons with bad eyesight?

Please just remove the writeEntry.

In D28956#651795, @apol wrote:

You're using QDeadlineTimer weird. Why don't you use QDeadlineTimer::remainingTime() and ::setDeadline?

Am I missing something, or is this essentially dead code?

See D28940

Thanks for the notice.

add missing tests

whitespace

extend summary

Extend commit message, reduce idle timeout

add line break before closing brace in lambda, use 2 minutes

In D28862#649411, @broulik wrote:

It used to be 2 minutes, now it's 1?