D22540: Port ResultDelegate to use PlasmaComponents 3.0

[Kai Uwe Broulik]

broulik requested changes to this revision.
broulik added a comment.
This revision now requires changes to proceed.


  The highlight isn't using Plasma theme anymore
  F7118110: Screenshot_20190730_084841.png 

  The layout explodes for long category names, they should be elided.
  F7118127: Screenshot_20190730_085343.png 

  
  The tooltips look quite bad and crammed because as I said they're not 
separate windows anymore. Also, they show up immediately instead of after a 
delay.
  Clicking results doesn't do anything now, and there can now be both a 
highlight following the mouse and a selection following the keyboard.

INLINE COMMENTS

> ResultDelegate.qml:233
> +if (index === 0) { // Shift+Return will invoke first 
> action
> +text = i18ndc("milou", "placeholder is action e.g. 
> run in terminal, in parenthesis is shortcut", "%1 (Shift+Return)", text)
> +}

QQC2 tooltip doesn't seem to handle mnemonics, leading to text like "Run in 
&Terminal"

> apol wrote in ResultDelegate.qml:142
> There's no separator svg. you suggest creating a new one?

Use whatever the `PlasmaComponents.ListItem` uses

REPOSITORY
  R112 Milou

REVISION DETAIL
  https://phabricator.kde.org/D22540

To: apol, #plasma, broulik
Cc: davidedmundson, broulik, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D22817: Port KRunner to Plasma Components 3

[Kai Uwe Broulik]

broulik added a comment.


  The padding on the left of the `TextField` looks off and what's up with that 
clear button?
  F7118082: Screenshot_20190730_084115.png 

  No tooltip shows up for the configure and close buttons.

INLINE COMMENTS

> RunCommand.qml:56
>  Layout.alignment: Qt.AlignTop
> +height: queryField.implicitHeight
> +

Why?

> RunCommand.qml:59
>  PlasmaComponents.ToolButton {
> -iconSource: "configure"
> +implicitHeight: queryField.implicitHeight
> +implicitWidth: queryField.implicitHeight

Why?

> RunCommand.qml:226
>  keyNavigationWraps: true
> -highlight: PlasmaComponents.Highlight {}
>  highlightMoveDuration: 0

Why remove this? It breaks highlighting of history items.

REPOSITORY
  R120 Plasma Workspace

REVISION DETAIL
  https://phabricator.kde.org/D22817

To: apol, #plasma
Cc: broulik, plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, ragreen, 
Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, abetts, 
sebas, apol, mart

D22818: Avoid crash if kicker plugin is used outside plasma context

[David Edmundson]

davidedmundson created this revision.
davidedmundson added a reviewer: Plasma.
Herald added a project: Plasma.
Herald added a subscriber: plasma-devel.
davidedmundson requested review of this revision.

REVISION SUMMARY
  Code currently gets the applet interface object passed in explicitly,
  from this it gets the configuration via the metaobject system
  
  We always guard the appletConfig not being available, but never the
  appletInterface, resulting in crashes if used from qmlscene or any
  potential tests.

TEST PLAN
  Used from qmlscene

REPOSITORY
  R119 Plasma Desktop

REVISION DETAIL
  https://phabricator.kde.org/D22818

AFFECTED FILES
  applets/kicker/plugin/appsmodel.cpp

To: davidedmundson, #plasma
Cc: plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, ragreen, Pitel, 
ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, 
apol, mart

D22762: Take in consideration flat() for the All Applications menu

[David Edmundson]

davidedmundson added a comment.


  m_flat currently behaves a bit specially. It doesn't affect the toplevel, it 
only affects sub models created within it.
  
  If we used AppsModel directly to fix the original bug, we would need to fix 
AppsModel to follow m_flat when showing an empty m_entryPath
  
  So in addition to a property for entry path we would add something like:
  
AppsModel.cpp:481
-  if (m_entryPath.isEmpty())
+if (m_entryPath.isEmpty() && !m_flat) {

(bringing us into the else path that uses processServiceGroup which follows 
m_flat)
  
  Which fixes the bug giving us a flat AppsModel. But it now that introduces a 
regression into the dash that's currently in kdeplasma-addons.
  
  We're now implicitly affecting the RootModel top level to be flat, meaning 
the sidebar on the right hand side now lists every application not just groups 
- which obviously isn't OK.
  
  We would need to introduce a property "flatButForRealThisTime" property on 
AppsModel which IMHO is going in a messy direction compared to only affecting 
allModel.

REPOSITORY
  R119 Plasma Desktop

REVISION DETAIL
  https://phabricator.kde.org/D22762

To: tcanabrava, davidedmundson, hein
Cc: hein, plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, ragreen, 
Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, abetts, 
sebas, apol, mart

D22817: Port KRunner to Plasma Components 3

[Aleix Pol Gonzalez]

apol created this revision.
apol added a reviewer: Plasma.
Herald added a project: Plasma.
Herald added a subscriber: plasma-devel.
apol requested review of this revision.

REVISION SUMMARY
  Using the new, stable components. Biggest change is that now we don't
  have a clear button and we need to add it by hand.

TEST PLAN
  Been using it for a few days, forgot to send the patch

REPOSITORY
  R120 Plasma Workspace

BRANCH
  krunner-pc3

REVISION DETAIL
  https://phabricator.kde.org/D22817

AFFECTED FILES
  lookandfeel/contents/runcommand/RunCommand.qml

To: apol, #plasma
Cc: plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, ragreen, Pitel, 
ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, 
apol, mart

D20972: [sidebar] Make search field not glued to the separator

[Filip Fila]

filipf added a comment.


  So Discover seems to have greater issues, but the added margin looks better 
than no added margin in system settings.
  
  F7115662: image.png 
  
  @davidedmundson I can't find a better solution..

REPOSITORY
  R124 System Settings

REVISION DETAIL
  https://phabricator.kde.org/D20972

To: filipf, #vdg, ngraham, davidedmundson
Cc: abetts, davidedmundson, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, sebas, apol, mart

D22176: Rework system tray settings

[Filip Fila]

filipf added a comment.


  In D22176#489327 , @ngraham wrote:
  
  > Honestly I would support staying with the QQC1 TableView for now. Trying to 
move to QQC2 without a replacement for it just means we're rolling our own 
table view, which seems even hackier.
  
  
  +1, we've already stuck with QQC1 TableView elsewhere.

REPOSITORY
  R120 Plasma Workspace

REVISION DETAIL
  https://phabricator.kde.org/D22176

To: nicolasfella, #plasma, #vdg
Cc: filipf, broulik, ngraham, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D22316: Avoid garbling the sound if the volume slider is moved fast

[Filip Fila]

filipf added a comment.


  Removed VDG as the reviewer, seems like a technical issue all together. Also 
wanted to contribute, but I don't really perceive any issues when moving the 
slider fast :/

REPOSITORY
  R115 Plasma Audio Volume Applet

REVISION DETAIL
  https://phabricator.kde.org/D22316

To: kezik, #plasma, drosca
Cc: filipf, ngraham, davidedmundson, plasma-devel, kezik, LeGast00n, jraleigh, 
fbampaloukas, GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D22359: Revert "[showdesktop][minimizeall] Reduce the maximum panel icon size"

[Filip Fila]

filipf added a comment.


  I don't feel too wise regarding this, pinging #plasma 
 in case someone has advice.

REPOSITORY
  R119 Plasma Desktop

REVISION DETAIL
  https://phabricator.kde.org/D22359

To: ndavis, #vdg, #plasma
Cc: filipf, plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, ragreen, 
Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, abetts, 
sebas, apol, mart

D22493: [Notifications] Move history items' icons over to the left a bit

[Filip Fila]

filipf requested changes to this revision.
filipf added a comment.
This revision now requires changes to proceed.


  Yeah this affects popups as well
  
  Before:
  F7115595: image.png 
  
  After: 
  F7115590: image.png 

REPOSITORY
  R120 Plasma Workspace

REVISION DETAIL
  https://phabricator.kde.org/D22493

To: ngraham, broulik, #vdg, #plasma, filipf
Cc: filipf, alexde, ndavis, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D22533: [KCM] Slightly imptove tab bar's integration with frame

[Filip Fila]

filipf accepted this revision.
filipf added a comment.
This revision is now accepted and ready to land.


  Looks better. I would also add the bug no. to the comment so all instances of 
workarounds can be easily grepped when the bug is closed.

REPOSITORY
  R115 Plasma Audio Volume Applet

BRANCH
  slightly-better-tab-bar (branched from master)

REVISION DETAIL
  https://phabricator.kde.org/D22533

To: ngraham, drosca, #vdg, filipf
Cc: filipf, plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, ragreen, 
Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, abetts, 
sebas, apol, mart

D22589: Embrace X-KDE-Wayland-Interfaces

[Aleix Pol Gonzalez]

apol updated this revision to Diff 62771.
apol added a comment.


  include keystates

REPOSITORY
  R120 Plasma Workspace

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D22589?vs=62130&id=62771

BRANCH
  master

REVISION DETAIL
  https://phabricator.kde.org/D22589

AFFECTED FILES
  krunner/CMakeLists.txt
  krunner/krunner.desktop
  krunner/krunner.desktop.cmake
  shell/CMakeLists.txt
  shell/org.kde.plasmashell.desktop
  shell/org.kde.plasmashell.desktop.cmake

To: apol, #plasma, #kwin, davidedmundson
Cc: davidedmundson, zzag, mvourlakos, plasma-devel, LeGast00n, jraleigh, 
fbampaloukas, GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D22813: Don't give up if no results arrive after 500 ms, again.

[Kai Uwe Broulik]

This revision was automatically updated to reflect the committed changes.
Closed by commit R112:82b1ed82848c: Don't give up if no results arrive 
after 500 ms, again. (authored by broulik).

REPOSITORY
  R112 Milou

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D22813?vs=62765&id=62766

REVISION DETAIL
  https://phabricator.kde.org/D22813

AFFECTED FILES
  lib/runnerresultsmodel.cpp

To: broulik, #plasma, fvogt
Cc: plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, ragreen, Pitel, 
ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, 
apol, mart

D22814: Remove colon (:) prefix when looking up dictionary word

[Kai Uwe Broulik]

broulik created this revision.
broulik added reviewers: Plasma, mart, kossebau.
Herald added a project: Plasma.
Herald added a subscriber: plasma-devel.
broulik requested review of this revision.

REVISION SUMMARY
  The dictionary engine lets us specify a server and dictionary but we don't 
need any of that, and the engine can deal with just a word (these days?).
  More importantly, the key inserted into the `m_lockers` map is without a 
colon and since `sourceRemoved` is also called without it, the wait condition 
is never met and we time out eventually.
  The colon has been in there forever but from current dictionary engine code I 
don't see why it would be neccessary.

TEST PLAN
  5.16 branch?
  
  - Together with D21605  (or on 5.16 with 
old milou model) dictionary runner finds stuff again "define test"

REPOSITORY
  R114 Plasma Addons

REVISION DETAIL
  https://phabricator.kde.org/D22814

AFFECTED FILES
  runners/dictionary/dictionarymatchengine.cpp

To: broulik, #plasma, mart, kossebau
Cc: plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, ragreen, Pitel, 
ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, 
apol, mart

D22813: Don't give up if no results arrive after 500 ms, again.

[Kai Uwe Broulik]

broulik updated this revision to Diff 62765.
broulik added a comment.


  - `doppelthältbesser--`

REPOSITORY
  R112 Milou

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D22813?vs=62762&id=62765

REVISION DETAIL
  https://phabricator.kde.org/D22813

AFFECTED FILES
  lib/runnerresultsmodel.cpp

To: broulik, #plasma, fvogt
Cc: plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, ragreen, Pitel, 
ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, 
apol, mart

D22813: Don't give up if no results arrive after 500 ms, again.

[Fabian Vogt]

fvogt requested changes to this revision.
fvogt added inline comments.
This revision now requires changes to proceed.

INLINE COMMENTS

> runnerresultsmodel.cpp:56
> +
> +m_hasMatches = false;
>  }

Redundant

REPOSITORY
  R112 Milou

REVISION DETAIL
  https://phabricator.kde.org/D22813

To: broulik, #plasma, fvogt
Cc: plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, ragreen, Pitel, 
ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, 
apol, mart

D22813: Don't give up if no results arrive after 500 ms, again.

[Kai Uwe Broulik]

broulik created this revision.
broulik added reviewers: Plasma, fvogt.
Herald added a project: Plasma.
Herald added a subscriber: plasma-devel.
broulik requested review of this revision.

REVISION SUMMARY
  Port the fix added by D21605  to the new 
model and add a comment so I don't end up cleaning up that thing again a year 
from now.

TEST PLAN
  Context is no longer cleared for the dict runner

REPOSITORY
  R112 Milou

REVISION DETAIL
  https://phabricator.kde.org/D22813

AFFECTED FILES
  lib/runnerresultsmodel.cpp

To: broulik, #plasma, fvogt
Cc: plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, ragreen, Pitel, 
ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, 
apol, mart

D22540: Port ResultDelegate to use PlasmaComponents 3.0

[Aleix Pol Gonzalez]

apol updated this revision to Diff 62753.
apol added a comment.


  forgot opacity

REPOSITORY
  R112 Milou

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D22540?vs=62752&id=62753

BRANCH
  pc3 (branched from master)

REVISION DETAIL
  https://phabricator.kde.org/D22540

AFFECTED FILES
  lib/qml/ResultDelegate.qml
  lib/qml/ResultsView.qml

To: apol, #plasma
Cc: davidedmundson, broulik, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D22540: Port ResultDelegate to use PlasmaComponents 3.0

[Aleix Pol Gonzalez]

apol updated this revision to Diff 62752.
apol added a comment.


  Rebase on top of kai's patch

REPOSITORY
  R112 Milou

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D22540?vs=62517&id=62752

BRANCH
  pc3 (branched from master)

REVISION DETAIL
  https://phabricator.kde.org/D22540

AFFECTED FILES
  lib/qml/ResultDelegate.qml
  lib/qml/ResultsView.qml

To: apol, #plasma
Cc: davidedmundson, broulik, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D17464: [Timer applet] Minor fixes for the applet

[Alexander Kernozhitsky]

gepardo added a comment.


  Here is how it looks on tall panels:
  F7114427: clock2.png 

REPOSITORY
  R114 Plasma Addons

REVISION DETAIL
  https://phabricator.kde.org/D17464

To: gepardo, muhlenpfordt, mmazur, friedreich, #plasma
Cc: ngraham, davidedmundson, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D17464: [Timer applet] Minor fixes for the applet

[Alexander Kernozhitsky]

gepardo updated this revision to Diff 62751.
gepardo added a comment.


  Now the widget doesn't become too wide if the panel is tall. The size formula 
now looks similar to the size formula of the clock widget. Expanatory comment 
is updated.

REPOSITORY
  R114 Plasma Addons

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D17464?vs=62750&id=62751

BRANCH
  arcpatch-D17464

REVISION DETAIL
  https://phabricator.kde.org/D17464

AFFECTED FILES
  applets/timer/package/contents/ui/TimerDigit.qml
  applets/timer/package/contents/ui/TimerView.qml
  applets/timer/package/contents/ui/main.qml

To: gepardo, muhlenpfordt, mmazur, friedreich, #plasma
Cc: ngraham, davidedmundson, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D17464: [Timer applet] Minor fixes for the applet

[Alexander Kernozhitsky]

gepardo added a comment.


  > I don't follow how it makes them the same when the other number is 
different.
  
  Timer applet is rendered using SVGs, clock applet is rendered using text. So, 
as the SVGs and font glyphs have different sizes, we need different 
coefficients. The coefficient 0.86 itself is obtained by trying different other 
"magic" numbers and finding the nearest one visually.
  
  Here is how it looks:
  F7114370: clock.png 
  
  Or, I can just give up and remove 0.86 from the code :)
  F7114374: clock1.png 

REPOSITORY
  R114 Plasma Addons

REVISION DETAIL
  https://phabricator.kde.org/D17464

To: gepardo, muhlenpfordt, mmazur, friedreich, #plasma
Cc: ngraham, davidedmundson, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D17464: [Timer applet] Minor fixes for the applet

[Alexander Kernozhitsky]

gepardo updated this revision to Diff 62750.
gepardo added a comment.


  Change magic value to 0.86, as it looks nearer to the clock font size

REPOSITORY
  R114 Plasma Addons

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D17464?vs=62730&id=62750

BRANCH
  arcpatch-D17464

REVISION DETAIL
  https://phabricator.kde.org/D17464

AFFECTED FILES
  applets/timer/package/contents/ui/TimerDigit.qml
  applets/timer/package/contents/ui/TimerView.qml
  applets/timer/package/contents/ui/main.qml

To: gepardo, muhlenpfordt, mmazur, friedreich, #plasma
Cc: ngraham, davidedmundson, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D6313: Support Icon Scale from Icon naming specification 0.13

[Kai Uwe Broulik]

broulik added inline comments.

INLINE COMMENTS

> volkov wrote in kiconloader.cpp:1368
> Looks like a rebase: 
> https://phabricator.kde.org/D6313?vs=31197&id=34779&whitespace=ignore-most#toc

Feel free to submit a patch to restore this

REPOSITORY
  R302 KIconThemes

REVISION DETAIL
  https://phabricator.kde.org/D6313

To: broulik, kde-frameworks-devel, #plasma, #vdg
Cc: volkov, hein, rkflx, acrouthamel, ngraham, elvisangelaccio, mart, 
kvermette, cfeck, davidedmundson, plasma-devel, LeGast00n, sbergeron, jraleigh, 
fbampaloukas, GB_2, ragreen, Pitel, michaelh, ZrenBot, bruns, himcesjf, 
lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, apol

D6313: Support Icon Scale from Icon naming specification 0.13

[Alexander Volkov]

volkov added inline comments.

INLINE COMMENTS

> broulik wrote in kiconloader.cpp:1368
> Dunno, probably oversight or forgotten to rebase..

Looks like a rebase: 
https://phabricator.kde.org/D6313?vs=31197&id=34779&whitespace=ignore-most#toc

REPOSITORY
  R302 KIconThemes

REVISION DETAIL
  https://phabricator.kde.org/D6313

To: broulik, kde-frameworks-devel, #plasma, #vdg
Cc: volkov, hein, rkflx, acrouthamel, ngraham, elvisangelaccio, mart, 
kvermette, cfeck, davidedmundson, plasma-devel, LeGast00n, sbergeron, jraleigh, 
fbampaloukas, GB_2, ragreen, Pitel, michaelh, ZrenBot, bruns, himcesjf, 
lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, apol

D6313: Support Icon Scale from Icon naming specification 0.13

[Kai Uwe Broulik]

broulik added inline comments.

INLINE COMMENTS

> volkov wrote in kiconloader.cpp:1368
> Why this line reverts D12002 ?

Dunno, probably oversight or forgotten to rebase..

REPOSITORY
  R302 KIconThemes

REVISION DETAIL
  https://phabricator.kde.org/D6313

To: broulik, kde-frameworks-devel, #plasma, #vdg
Cc: volkov, hein, rkflx, acrouthamel, ngraham, elvisangelaccio, mart, 
kvermette, cfeck, davidedmundson, plasma-devel, LeGast00n, sbergeron, jraleigh, 
fbampaloukas, GB_2, ragreen, Pitel, michaelh, ZrenBot, bruns, himcesjf, 
lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, apol

D6313: Support Icon Scale from Icon naming specification 0.13

[Alexander Volkov]

volkov added inline comments.

INLINE COMMENTS

> kiconloader.cpp:1368
>  
> -if (group >= 0 && group < KIconLoader::LastGroup) {
> +if (group >= 0) {
>  img = d->mpEffect.apply(img, group, state);

Why this line reverts D12002 ?

REPOSITORY
  R302 KIconThemes

REVISION DETAIL
  https://phabricator.kde.org/D6313

To: broulik, kde-frameworks-devel, #plasma, #vdg
Cc: volkov, hein, rkflx, acrouthamel, ngraham, elvisangelaccio, mart, 
kvermette, cfeck, davidedmundson, plasma-devel, LeGast00n, sbergeron, jraleigh, 
fbampaloukas, GB_2, ragreen, Pitel, michaelh, ZrenBot, bruns, himcesjf, 
lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, apol

D19947: systray: show a small animation on the icons when activated

[David Edmundson]

davidedmundson added a comment.


  Code wise, it's fine.
  
  Personally I like it. If VDG approve, go for it.

REPOSITORY
  R120 Plasma Workspace

REVISION DETAIL
  https://phabricator.kde.org/D19947

To: apol, #plasma
Cc: anthonyfieroni, davidedmundson, ngraham, plasma-devel, LeGast00n, jraleigh, 
fbampaloukas, GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D17464: [Timer applet] Minor fixes for the applet

[David Edmundson]

davidedmundson added a comment.


  > The rationale is to make the clock font size and the timer font size 
identical. Setting 0.84 as timer height multiplier indeed reaches this goal. 
But two other problems remain:
  
  I don't follow how it makes them the same when the other number is different.

REPOSITORY
  R114 Plasma Addons

REVISION DETAIL
  https://phabricator.kde.org/D17464

To: gepardo, muhlenpfordt, mmazur, friedreich, #plasma
Cc: ngraham, davidedmundson, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D22804: [System Tray] Do not send ContextMenu signal twice

[Konrad Materka]

kmaterka updated this revision to Diff 62738.
kmaterka added a comment.


  I simplified the patch. It moves handling of onPressed to PlasmoidItem. This 
way we can have context menu when mouse is pressed for plasmoids and on click 
for SNI.

REPOSITORY
  R120 Plasma Workspace

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D22804?vs=62728&id=62738

REVISION DETAIL
  https://phabricator.kde.org/D22804

AFFECTED FILES
  applets/systemtray/package/contents/ui/items/AbstractItem.qml
  applets/systemtray/package/contents/ui/items/PlasmoidItem.qml

To: kmaterka, #plasma_workspaces
Cc: broulik, apol, plasma-devel, #plasma_workspaces, LeGast00n, jraleigh, 
fbampaloukas, GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, mart

D19947: systray: show a small animation on the icons when activated

[Aleix Pol Gonzalez]

apol added a comment.


  Ping? I still miss this...

REPOSITORY
  R120 Plasma Workspace

REVISION DETAIL
  https://phabricator.kde.org/D19947

To: apol, #plasma
Cc: anthonyfieroni, davidedmundson, ngraham, plasma-devel, LeGast00n, jraleigh, 
fbampaloukas, GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D22804: [System Tray] Do not send ContextMenu signal twice

[Konrad Materka]

kmaterka added a comment.


  In D22804#503733 , @broulik wrote:
  
  > > Yes, probably onPressed should be handled in PasmoidItem, not in the 
AbstractItem to avoid this kind of problems.
  >
  > Indeed, that would be the best solution.
  
  
  Should this be done together with this change? I shouldn't be hard, I can 
move onPressed to PlasmoidItem, which will simplify the fix also?
  
  >> I would personally opt for consistency, though. As far as I know GTK is 
showing context menu on mouse click, not pressed. The same is for MS Windows 
(this does not mean we must follow this or reimplement Windows, no). What is 
your opinion?
  > 
  > Gimp, Firefox, Chrome, Gedit, etc, they open it on press here, and so does 
Qt and the rest of KDE software, so this must be consistent to here, regardless 
of what someone else does.
  
  Oh, right! I focused on tray icons only... That makes perfect sense.

REPOSITORY
  R120 Plasma Workspace

REVISION DETAIL
  https://phabricator.kde.org/D22804

To: kmaterka, #plasma_workspaces
Cc: broulik, apol, plasma-devel, #plasma_workspaces, LeGast00n, jraleigh, 
fbampaloukas, GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, mart

D22804: [System Tray] Do not send ContextMenu signal twice

[Kai Uwe Broulik]

broulik added a comment.


  > Yes, probably onPressed should be handled in PasmoidItem, not in the 
AbstractItem to avoid this kind of problems.
  
  Indeed, that would be the best solution.
  
  > I would personally opt for consistency, though. As far as I know GTK is 
showing context menu on mouse click, not pressed. The same is for MS Windows 
(this does not mean we must follow this or reimplement Windows, no). What is 
your opinion?
  
  Gimp, Firefox, Chrome, Gedit, etc, they open it on press here, and so does Qt 
and the rest of KDE software, so this must be consistent to here, regardless of 
what someone else does.

REPOSITORY
  R120 Plasma Workspace

REVISION DETAIL
  https://phabricator.kde.org/D22804

To: kmaterka, #plasma_workspaces
Cc: broulik, apol, plasma-devel, #plasma_workspaces, LeGast00n, jraleigh, 
fbampaloukas, GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, mart

Addendum to 29/07/2019 meeting

[David Redondo]

Hi,

I landed https://phabricator.kde.org/D22121 which allows sorting the wallpaper 
slideshow in a specific order. This is managed by a QSortFilterProxyModel. 
This model is now also used for the actual slideshow and the config dialog (No 
more QStringlist of the images that we want to show/are excluded). Also if you 
have a different order than Random the slideshow will continue where it 
stopped if you restart Plasmashell.
At the onboarding sprint David, David and I investigated why clipboard with 
larger data is broken under Wayland (for example images). David Edmundson has 
a qtwayland patch for that: https://codereview.qt-project.org/c/qt/qtwayland/
+/268352

David

D22804: [System Tray] Do not send ContextMenu signal twice

[Konrad Materka]

kmaterka added a comment.


  @apol SNI applications (including XembedSniProxy) are using 
StatusNotifierItem.qml, which already has onClick handled for right button. 
That why we had ContextMenu event's duplicated.
  @broulik Yes, probably onPressed should be handled in PasmoidItem, not in the 
AbstractItem to avoid this kind of problems. I would personally opt for 
consistency, though. As far as I know GTK is showing context menu on mouse 
click, not pressed. The same is for MS Windows (this does not mean we must 
follow this or reimplement Windows, no). What is your opinion?

REPOSITORY
  R120 Plasma Workspace

REVISION DETAIL
  https://phabricator.kde.org/D22804

To: kmaterka, #plasma_workspaces
Cc: broulik, apol, plasma-devel, #plasma_workspaces, LeGast00n, jraleigh, 
fbampaloukas, GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, mart

D22804: [System Tray] Do not send ContextMenu signal twice

[Kai Uwe Broulik]

broulik added a comment.


  Maybe the behavior must be split between `PlasmoidItem` and 
`StatusNotifierItem`. Ideally, all menus opened on press but that has shown to 
be problematic for XembedSniProxy

REPOSITORY
  R120 Plasma Workspace

REVISION DETAIL
  https://phabricator.kde.org/D22804

To: kmaterka, #plasma_workspaces
Cc: broulik, apol, plasma-devel, #plasma_workspaces, LeGast00n, jraleigh, 
fbampaloukas, GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, mart

D22804: [System Tray] Do not send ContextMenu signal twice

[Aleix Pol Gonzalez]

apol added a comment.


  How will non-plasmoids (e.g. SNI) get contextMenu calls then?

REPOSITORY
  R120 Plasma Workspace

REVISION DETAIL
  https://phabricator.kde.org/D22804

To: kmaterka, #plasma_workspaces
Cc: apol, plasma-devel, #plasma_workspaces, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, mart

D17464: [Timer applet] Minor fixes for the applet

[Alexander Kernozhitsky]

gepardo updated this revision to Diff 62730.
gepardo added a comment.


  Minor change in the comment for the "magic" value 0.84

REPOSITORY
  R114 Plasma Addons

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D17464?vs=62660&id=62730

BRANCH
  arcpatch-D17464

REVISION DETAIL
  https://phabricator.kde.org/D17464

AFFECTED FILES
  applets/timer/package/contents/ui/TimerDigit.qml
  applets/timer/package/contents/ui/TimerView.qml
  applets/timer/package/contents/ui/main.qml

To: gepardo, muhlenpfordt, mmazur, friedreich, #plasma
Cc: ngraham, davidedmundson, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D17464: [Timer applet] Minor fixes for the applet

[Alexander Kernozhitsky]

gepardo added a comment.


  The rationale is to make the clock font size and the timer font size 
identical. Setting 0.84 as timer height multiplier indeed reaches this goal. 
But two other problems remain:
  
  - the fonts are different itself, as clock uses the real font, and timer uses 
SVGs to render digits
  - if the panel is large enough, the vertical position of the text differs 
(timer's top and bottom margins are equal, but clock's top and bottom margins 
have different size)
  
  Should I fix some of these issues also?

REPOSITORY
  R114 Plasma Addons

REVISION DETAIL
  https://phabricator.kde.org/D17464

To: gepardo, muhlenpfordt, mmazur, friedreich, #plasma
Cc: ngraham, davidedmundson, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D22804: [System Tray] Do not send ContextMenu signal twice

[Konrad Materka]

kmaterka added a comment.


  This fixes 27b1030756002e91b60ba51483efe9c2c477d16e 
,
 which reverts b9af99140365c8c8cf4e3744af7e13b35c35fa18 
,
 but not fully.

REPOSITORY
  R120 Plasma Workspace

REVISION DETAIL
  https://phabricator.kde.org/D22804

To: kmaterka, #plasma_workspaces
Cc: plasma-devel, #plasma_workspaces, LeGast00n, jraleigh, fbampaloukas, GB_2, 
ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, 
abetts, sebas, apol, mart

D22804: [System Tray] Do not send ContextMenu signal twice

[Konrad Materka]

kmaterka created this revision.
kmaterka added a reviewer: Plasma: Workspaces.
kmaterka added a project: Plasma: Workspaces.
Herald edited projects, added Plasma; removed Plasma: Workspaces.
Herald added a subscriber: plasma-devel.
kmaterka requested review of this revision.

REVISION SUMMARY
  System Tray sends two "ContextMenu" events which breaks focus.
  One is send on mouse pressed, second on mouse clicked. As a result
  right click on the system tray icon causes focus to lock on the icon.
  If later user clicks somewhere else event is send to the first icon.
  
  BUG: 376277
  BUG: 409649

TEST PLAN
  The best to test with Skype for Linux with SNI bridge.
  
  To reproduce:
  
  - Right click on the Skype icon, click on menu item (can be Show Skype)
  - Right click on any other icon in system tray (or even any other place on 
the panel)
  - Skype menu will appear
  
  After fix:
  
  - Right click on the Skype icon, click on menu item (can be Show Skype)
  - Right click on any other icon in system tray (or even any other place on 
the panel)
  - Correct menu should show

REPOSITORY
  R120 Plasma Workspace

REVISION DETAIL
  https://phabricator.kde.org/D22804

AFFECTED FILES
  applets/systemtray/package/contents/ui/items/AbstractItem.qml
  applets/systemtray/package/contents/ui/items/PlasmoidItem.qml

To: kmaterka, #plasma_workspaces
Cc: plasma-devel, #plasma_workspaces, LeGast00n, jraleigh, fbampaloukas, GB_2, 
ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, 
abetts, sebas, apol, mart

D17464: [Timer applet] Minor fixes for the applet

[David Edmundson]

davidedmundson added a comment.


  I still can't accept a patch with an unexplained magic number.
  
  If the rationale is that it's to match the clock, then it has to match the 
clock. If there's something else, please can the comment say what.

REPOSITORY
  R114 Plasma Addons

REVISION DETAIL
  https://phabricator.kde.org/D17464

To: gepardo, muhlenpfordt, mmazur, friedreich, #plasma
Cc: ngraham, davidedmundson, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D22803: Improved OSD behavior

[Kai Uwe Broulik]

broulik created this revision.
broulik added reviewers: Plasma, VDG.
Herald added a project: Plasma.
Herald added a subscriber: plasma-devel.
broulik requested review of this revision.

REVISION SUMMARY
  Don't ask the user for a screen configuration on login, instead, apply a sane 
default.
  When connecting a new screen, apply a default configuration so the user gets 
immediate feedback "something's happening" and then prompt for a configuration 
via the OSD in case the chosen default doesn't fit.

TEST PLAN
  The login case is especially annoying as it would usually choose a sane 
default and then the user would choose Cancel which won't save a config and 
then kscreen asks again next time until the user choses anyhting but Cancel.
  
  Nuked my kscreen config, restarted kded, two screens attached, had kscreen 
apply a default config for my two screens without asking.
  Nuked my kscreen config, unplugged one monitor, restarted kded, plugged in 
second screen, KScreen applied a sane default (though it cannot know that the 
screens are swapped), and showed the OSD which I then used to swap the monitors 
"extend left" to how they should be

REPOSITORY
  R104 KScreen

REVISION DETAIL
  https://phabricator.kde.org/D22803

AFFECTED FILES
  kded/daemon.cpp
  kded/daemon.h

To: broulik, #plasma, #vdg
Cc: plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, ragreen, Pitel, 
ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, 
apol, mart

D22767: Calculate first clickable point, from the top-left

[David Edmundson]

davidedmundson added inline comments.

INLINE COMMENTS

> sniproxy.cpp:576
>  event->root = QX11Info::appRootWindow();
>  event->root_x = x;
>  event->root_y = y;

This point was marked as closed, but not changed.  Probably my fault for not 
expanding.

We move the window so that it's at x-clickPoint
The relative point is clickPoint.x

The absolute position of the click should be offset by the same amount. I would 
have expected

> event->root_x = x + clickPoint.x
> event->root_y = y + clickPoint.y

in both places

REVISION DETAIL
  https://phabricator.kde.org/D22767

To: kmaterka, #plasma_workspaces
Cc: davidedmundson, plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, 
ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, 
abetts, sebas, apol, mart

notes for 29/7/2019 monday meeting

[Marco Martin]

Kai Uwe:
* Wrote a new model backing Milou/KRunner which avoids resetting the
entire list every time results change
* Improves performance, especially when dealing with results that
hardly change (e.g. calculator)
* Keeps the selection steady as results come in delayed and shift
the results around
* Fixed a few issues in qqc2-desktop-style, `ComboBox` closes when
clicking on itself, just like in Qt Widgets
* Merged my huge backlog of plasma-browser-integration stuff
* Attended KDE Connect sprint in Nürnberg at SUSE
* Wrote a little browser extension for volker|office that checks for
machine-readable metadata of websites for Itinerary, e.g. schema.org
annotations, json-ld
* https://cgit.kde.org/scratch/broulik/itinerary-research.git/
* Poked dfaure about changing Qt dependency policy for Frameworks to
accomodate distros better
* Removed KAuth session persistency from Kate/KTextEditor's root
helper for security
* Please review D22213 D22506 D22531 D22569 D22596
Diff 22213 "[Tab] Fix sizing" [Needs Review] https://phabricator.kde.org/D22213
Diff 22506 "Read BAMF_DESKTOP_FILE_HINT from environment" [Needs
Review] https://phabricator.kde.org/D22506
Diff 22531 "[KColorSchemeManager] Optimize preview generation" [Needs
Review] https://phabricator.kde.org/D22531
Diff 22569 "[Dialog] Apply window type after changing flags" [Needs
Review] https://phabricator.kde.org/D22569
Diff 22596 "[MPRIS Data Engine] Ignore non-standards compliant
players" [Needs Review] https://phabricator.kde.org/D22596

Jonathan:
* 2020 release schedule

Nico:
* Attended the Megasprint
* kill the Phonon KCM. The remaining settings are most likely barely
used. D22616
   Diff 22616 "Remove phonon KCM" [Needs Review]
https://phabricator.kde.org/D22616
* That would allow us to patch PulseAudio to stop loading
module-device-manager for us and instead use switch-on-connect by
default
* Which is what most people prefer
* Poked some Franmworks to see what we can deprecate to remove it in KF6
* Mostly worked on my pet project, KTrip, a public transport assistant
for Plasma Mobile and Android
* "[RFC] Deprecate KPassivePopup" [Needs Review]
https://phabricator.kde.org/D22544

David:
* been to KWin mini sprint part of the bigger sprint
* There's lots of notes from the kwin sprint. Mostly in note form:
https://notes.kde.org/p/kwin-sprint-2019
* we assigned quite a few action tasks
* in wayland code, I started support for klipper
* and kde connect
* and kde connect's clipboard
* most important change is the code patch from aleix which at least
unblocks all the secuirty related wayland topics
* that plan being that I should just implement it
* Kai and I have a patch to add a global animation speed slider

Marco:
Kirigami:
* Fixed sevaral bugs in ActionToolBar component

Plasma:
* Landed desktop layout rewrite \m/
* added support for configurationrequired button in the new layout
* desktop scripting now supports applets args

Frameworks
* pushed the new layout for config dialogs

D22540: Port ResultDelegate to use PlasmaComponents 3.0

[Aleix Pol Gonzalez]

apol added a comment.


  Ping? This is blocking D22514 

REPOSITORY
  R112 Milou

REVISION DETAIL
  https://phabricator.kde.org/D22540

To: apol, #plasma
Cc: davidedmundson, broulik, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D20093: Show a context menu on pressAndHold

[Aleix Pol Gonzalez]

apol abandoned this revision.
apol added a comment.


  Will need a better solution

REPOSITORY
  R119 Plasma Desktop

REVISION DETAIL
  https://phabricator.kde.org/D20093

To: apol, #plasma, mart
Cc: GB_2, davidedmundson, hein, plasma-devel, LeGast00n, jraleigh, 
fbampaloukas, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D22738: Remove some uses on QDesktopWidget

[Aleix Pol Gonzalez]

apol updated this revision to Diff 62725.
apol added a comment.


  Add eike's braces

REPOSITORY
  R119 Plasma Desktop

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D22738?vs=62582&id=62725

BRANCH
  master

REVISION DETAIL
  https://phabricator.kde.org/D22738

AFFECTED FILES
  applets/pager/plugin/pagermodel.cpp
  containments/desktop/plugins/folder/foldermodel.cpp
  kcms/desktoppaths/globalpaths.cpp
  kcms/fonts/previewrenderengine.cpp

To: apol, #plasma, davidedmundson
Cc: hein, davidedmundson, plasma-devel, LeGast00n, jraleigh, fbampaloukas, 
GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, 
jensreuterberg, abetts, sebas, apol, mart

D22767: Calculate first clickable point, from the top-left

[Konrad Materka]

kmaterka updated this revision to Diff 62723.
kmaterka marked 3 inline comments as done.
kmaterka added a comment.


  Review changes. Call to the calculateClickPoint is local and on only on mouse 
click, not on every icon update.

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D22767?vs=62621&id=62723

REVISION DETAIL
  https://phabricator.kde.org/D22767

AFFECTED FILES
  xembed-sni-proxy/sniproxy.cpp
  xembed-sni-proxy/sniproxy.h

To: kmaterka, #plasma_workspaces
Cc: davidedmundson, plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, 
ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, 
abetts, sebas, apol, mart

Re: Renaming appdata and desktop file for GCompris

[Johnny Jazeix]

Hi,

thanks for the explanation. We'll go with the
QGuiApplication::setDesktopFileName solution as we don't use KF5 and
it seems the easiest way to do it.

Johnny

Le lun. 29 juil. 2019 à 08:03, Harald Sitter  a écrit :
>
> Hi,
>
> Do note that the appstream file name  and the desktop file name do not
> need to be the same. An appstream file tags a specific desktop file as
>  [1]. So technically you do not need to rename it, you only
> need to change the desktop file and the provides entry. If you want to
> rename it do note that the  tag may contain an  tag,
> that's where you specify your old name and the appstream tooling will
> know that your component has two names.
>
> Lastly, I *think* you do not technically need to rename anything. If
> you want you should also be able to use KAboutData::setDesktopFileName
> [2] respectively QGuiApplication::setDesktopFileName [3] to help the
> task manager resolve the desktop file from your process without having
> to rename anything.
>
> [1] 
> https://www.freedesktop.org/software/appstream/docs/chap-Metadata.html#sect-Metadata-GenericComponent
> [2] 
> https://api.kde.org/frameworks/kcoreaddons/html/classKAboutData.html#a112d2fc20c31e7847995930e030cc67b
> [3] https://doc.qt.io/qt-5/qguiapplication.html#desktopFileName-prop
>
> HS
>
> On Mon, Jul 29, 2019 at 12:02 AM Johnny Jazeix  wrote:
> >
> > Hi Plasma team,
> >
> > I'm a developer of GCompris and we recently had an issue with our
> > desktop file (https://bugs.kde.org/show_bug.cgi?id=409996).
> > Luigi proposed to ask you to know if changing the name of it would be
> > an issue or have some unseen consequences for application managers?
> >
> > I'm not subscribed to the list, please keep me in cc if you reply.
> >
> > Johnny
> >
> > -- Forwarded message -
> > De : Luigi Toscano 
> > Date: dim. 28 juil. 2019 à 14:29
> > Subject: Re: Renaming appdata and desktop file
> > To: KDE i18n-doc , Johnny Jazeix 
> >
> >
> > Johnny Jazeix ha scritto:
> > > Hi,
> > > for GCompris we named the files org.kde.gcompris.appdata.xml and
> > > org.kde.gcompris.desktop.
> > > However, our binary file is gcompris-qt. According to
> > > https://bugs.kde.org/show_bug.cgi?id=409996 (gcompris 0.96 has Wayland
> > > generic icon when you press alt+tab) and after some research
> > > (http://plfiorini.blogspot.com/2015/07/how-to-make-your-qt-application-icon.html),
> > > it seems we should rename our desktop file to
> > > org.kde.gcompris-qt.desktop
> > >
> > > Does changing the name is the good solution?
> >
> > I don't really have an answer for this. I would suggest asking the Plasma 
> > team
> > for confirmation. They probably know also whether renaming has can make
> > application managers think that this is a new program.
> >
> > > And if yes, do I just
> > > have to change the filenames (for both) and the translation can be
> > > automatically retrieved (or will I have to do some svn move)?
> >
> > A manual rename would be needed for the templates and the files, but I can
> > help with that.
> >
> > Ciao
> > --
> > Luigi

D22767: Calculate first clickable point, from the top-left

[David Edmundson]

davidedmundson added a comment.


  Thanks for doing some work on xembedsniproxy!

INLINE COMMENTS

> sniproxy.cpp:239
>  
> +calculateClickPoint(image);
> +

This is quite expensive to do every frame (some animate), I think we can do 
something else, see below.

> sniproxy.cpp:530
>   else
> - configVals[0] = static_cast(x);
> + configVals[0] = static_cast(x - m_clickPoint.x());
>   if (pointer->root_y > y + clientGeom->height)

I don't think clickpoint has to be a member variable, we only use it within the 
context of this method.

We can make it a local variable evaluated within sendClick. Relative to frame 
updates, clicks are quite rare.

> sniproxy.cpp:551
>  event->root = QX11Info::appRootWindow();
>  event->root_x = x;
>  event->root_y = y;

+ m_clickPoint ?

REPOSITORY
  R120 Plasma Workspace

REVISION DETAIL
  https://phabricator.kde.org/D22767

To: kmaterka, #plasma_workspaces
Cc: davidedmundson, plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, 
ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, 
abetts, sebas, apol, mart