Re: Review Request 129380: remove kcm docbooks which have their code in other repos from plasma-desktop

2016-11-12 Thread Burkhard Lück 

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/129380/
---

(Updated Nov. 13, 2016, 12:06 a.m.)


Status
--

This change has been marked as submitted.


Review request for Documentation, KDE Frameworks and Plasma.


Changes
---

Submitted with commit e9bf567e4531ab01c7d0d73274771797dd397125 by Burkhard Lück 
to branch master.


Bugs: 350619
https://bugs.kde.org/show_bug.cgi?id=350619


Repository: plasma-desktop


Description
---

remove kcm docbooks for cache, proxy, useragent, cookies, netpref, 
webshortcuts, smb
these docbooks will be imported into frameworks/kio from kde-runtime master, 
see https://git.reviewboard.kde.org/r/128912/
remove kcm docbook for filetypes, this docbook will be imported into 
kde-cli-tools from kde-runtime master, see 
https://git.reviewboard.kde.org/r/128914/


Diffs
-

  doc/kcontrol/CMakeLists.txt c057288 
  doc/kcontrol/cache/CMakeLists.txt 7728027 
  doc/kcontrol/cache/index.docbook a03658e 
  doc/kcontrol/cookies/CMakeLists.txt 8f86430 
  doc/kcontrol/cookies/index.docbook a2379b9 
  doc/kcontrol/filetypes/CMakeLists.txt c6a6b24 
  doc/kcontrol/filetypes/index.docbook bf40a12 
  doc/kcontrol/netpref/CMakeLists.txt 42010fd 
  doc/kcontrol/netpref/index.docbook 7db99a6 
  doc/kcontrol/proxy/CMakeLists.txt 37b19c9 
  doc/kcontrol/proxy/index.docbook a7bc293 
  doc/kcontrol/smb/CMakeLists.txt 3285ccf 
  doc/kcontrol/smb/index.docbook f42f9a4 
  doc/kcontrol/useragent/CMakeLists.txt 35165d2 
  doc/kcontrol/useragent/index.docbook 6a816ab 
  doc/kcontrol/webshortcuts/CMakeLists.txt 7324220 
  doc/kcontrol/webshortcuts/index.docbook bde3028 

Diff: https://git.reviewboard.kde.org/r/129380/diff/


Testing
---

builds


Thanks,

Burkhard Lück

Re: Review Request 129380: remove kcm docbooks which have their code in other repos from plasma-desktop

2016-11-12 Thread Luigi Toscano 

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/129380/#review100807
---


Ship it!




Inviala!

- Luigi Toscano


On Nov. 11, 2016, 7:43 p.m., Burkhard Lück wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/129380/
> ---
> 
> (Updated Nov. 11, 2016, 7:43 p.m.)
> 
> 
> Review request for Documentation, KDE Frameworks and Plasma.
> 
> 
> Bugs: 350619
> https://bugs.kde.org/show_bug.cgi?id=350619
> 
> 
> Repository: plasma-desktop
> 
> 
> Description
> ---
> 
> remove kcm docbooks for cache, proxy, useragent, cookies, netpref, 
> webshortcuts, smb
> these docbooks will be imported into frameworks/kio from kde-runtime master, 
> see https://git.reviewboard.kde.org/r/128912/
> remove kcm docbook for filetypes, this docbook will be imported into 
> kde-cli-tools from kde-runtime master, see 
> https://git.reviewboard.kde.org/r/128914/
> 
> 
> Diffs
> -
> 
>   doc/kcontrol/CMakeLists.txt c057288 
>   doc/kcontrol/cache/CMakeLists.txt 7728027 
>   doc/kcontrol/cache/index.docbook a03658e 
>   doc/kcontrol/cookies/CMakeLists.txt 8f86430 
>   doc/kcontrol/cookies/index.docbook a2379b9 
>   doc/kcontrol/filetypes/CMakeLists.txt c6a6b24 
>   doc/kcontrol/filetypes/index.docbook bf40a12 
>   doc/kcontrol/netpref/CMakeLists.txt 42010fd 
>   doc/kcontrol/netpref/index.docbook 7db99a6 
>   doc/kcontrol/proxy/CMakeLists.txt 37b19c9 
>   doc/kcontrol/proxy/index.docbook a7bc293 
>   doc/kcontrol/smb/CMakeLists.txt 3285ccf 
>   doc/kcontrol/smb/index.docbook f42f9a4 
>   doc/kcontrol/useragent/CMakeLists.txt 35165d2 
>   doc/kcontrol/useragent/index.docbook 6a816ab 
>   doc/kcontrol/webshortcuts/CMakeLists.txt 7324220 
>   doc/kcontrol/webshortcuts/index.docbook bde3028 
> 
> Diff: https://git.reviewboard.kde.org/r/129380/diff/
> 
> 
> Testing
> ---
> 
> builds
> 
> 
> Thanks,
> 
> Burkhard Lück
> 
>

Re: Review Request 129373: Plasma Framework changes for #129372

2016-11-12 Thread René Fürst 


> On Nov. 10, 2016, 8:58 p.m., Kai Uwe Broulik wrote:
> > Thanks for your patch!
> > 
> > I'm not fully sure about this, but let's wait for other Plasma developer 
> > feedback :) Especially, since I couldn't yet look at your patch that 
> > actually makes use of this API in icontasks.

I've uploaded it to the other review request. 

If those APIs are not added then this basically means that taskmanager needs to 
be reworked to not use tooltips in order to fix the referenced bug. Basically 
this calls for a definition of use cases where tooltips should and where 
tooltips should not be used ;)


- René


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/129373/#review100763
---


On Nov. 10, 2016, 8:47 p.m., René Fürst wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/129373/
> ---
> 
> (Updated Nov. 10, 2016, 8:47 p.m.)
> 
> 
> Review request for Plasma and Eike Hein.
> 
> 
> Repository: plasma-framework
> 
> 
> Description
> ---
> 
> Plasma Framework changes for #129372
> 
> 
> Diffs
> -
> 
>   src/declarativeimports/core/tooltip.h d38c49b 
>   src/declarativeimports/core/tooltip.cpp ffe1064 
>   src/declarativeimports/core/tooltipdialog.h d4e0ff0 
>   src/declarativeimports/core/tooltipdialog.cpp 28ba9be 
> 
> Diff: https://git.reviewboard.kde.org/r/129373/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> René Fürst
> 
>

Re: Review Request 129373: Plasma Framework changes for #129372

2016-11-12 Thread René Fürst 


> On Nov. 10, 2016, 9:59 p.m., David Edmundson wrote:
> > src/declarativeimports/core/tooltipdialog.cpp, line 109
> > 
> >
> > there's a bug in this logic
> > 
> >  - Display Tooltip Dialog
> >  - Move mouse over the dialog
> >  - Move Tooltip Dialog
> > 
> > 
> > contains mouse will still be true even though you're not containing the 
> > mouse; and you won't get a leave event until you mouse in and out again.

Are tooltips supposed to be moveable by mouse? If yes, do we have one around?


- René


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/129373/#review100765
---


On Nov. 10, 2016, 8:47 p.m., René Fürst wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/129373/
> ---
> 
> (Updated Nov. 10, 2016, 8:47 p.m.)
> 
> 
> Review request for Plasma and Eike Hein.
> 
> 
> Repository: plasma-framework
> 
> 
> Description
> ---
> 
> Plasma Framework changes for #129372
> 
> 
> Diffs
> -
> 
>   src/declarativeimports/core/tooltip.h d38c49b 
>   src/declarativeimports/core/tooltip.cpp ffe1064 
>   src/declarativeimports/core/tooltipdialog.h d4e0ff0 
>   src/declarativeimports/core/tooltipdialog.cpp 28ba9be 
> 
> Diff: https://git.reviewboard.kde.org/r/129373/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> René Fürst
> 
>

Re: Review Request 129380: remove kcm docbooks which have their code in other repos from plasma-desktop

2016-11-12 Thread Burkhard Lück 

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/129380/#review100801
---



filetypes kcm docbook was imported into kde-cli-tools with commit 
775cc72ed9a529de02607e277b379ed6260399ef
kcm docbooks for cache, proxy, useragent, cookies, netpref, webshortcuts, smb 
were imported into kio with commit e0c3945a88fd041f36ff3873f6fe0ca5cfa00158
so nothing blocks this RR anymore and the docbooks should be immediately removed

- Burkhard Lück


On Nov. 11, 2016, 6:43 nachm., Burkhard Lück wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/129380/
> ---
> 
> (Updated Nov. 11, 2016, 6:43 nachm.)
> 
> 
> Review request for Documentation, KDE Frameworks and Plasma.
> 
> 
> Bugs: 350619
> https://bugs.kde.org/show_bug.cgi?id=350619
> 
> 
> Repository: plasma-desktop
> 
> 
> Description
> ---
> 
> remove kcm docbooks for cache, proxy, useragent, cookies, netpref, 
> webshortcuts, smb
> these docbooks will be imported into frameworks/kio from kde-runtime master, 
> see https://git.reviewboard.kde.org/r/128912/
> remove kcm docbook for filetypes, this docbook will be imported into 
> kde-cli-tools from kde-runtime master, see 
> https://git.reviewboard.kde.org/r/128914/
> 
> 
> Diffs
> -
> 
>   doc/kcontrol/CMakeLists.txt c057288 
>   doc/kcontrol/cache/CMakeLists.txt 7728027 
>   doc/kcontrol/cache/index.docbook a03658e 
>   doc/kcontrol/cookies/CMakeLists.txt 8f86430 
>   doc/kcontrol/cookies/index.docbook a2379b9 
>   doc/kcontrol/filetypes/CMakeLists.txt c6a6b24 
>   doc/kcontrol/filetypes/index.docbook bf40a12 
>   doc/kcontrol/netpref/CMakeLists.txt 42010fd 
>   doc/kcontrol/netpref/index.docbook 7db99a6 
>   doc/kcontrol/proxy/CMakeLists.txt 37b19c9 
>   doc/kcontrol/proxy/index.docbook a7bc293 
>   doc/kcontrol/smb/CMakeLists.txt 3285ccf 
>   doc/kcontrol/smb/index.docbook f42f9a4 
>   doc/kcontrol/useragent/CMakeLists.txt 35165d2 
>   doc/kcontrol/useragent/index.docbook 6a816ab 
>   doc/kcontrol/webshortcuts/CMakeLists.txt 7324220 
>   doc/kcontrol/webshortcuts/index.docbook bde3028 
> 
> Diff: https://git.reviewboard.kde.org/r/129380/diff/
> 
> 
> Testing
> ---
> 
> builds
> 
> 
> Thanks,
> 
> Burkhard Lück
> 
>

[Differential] [Commented On] D2680: [Task Manager] Activate tasks by Meta + number shortcuts

2016-11-12 Thread cfeck (Christoph Feck) 
cfeck added a comment.


  monthly ping: What is the status of this? Ony unresolved issues?

REPOSITORY
  rPLASMADESKTOP Plasma Desktop

REVISION DETAIL
  https://phabricator.kde.org/D2680

EMAIL PREFERENCES
  https://phabricator.kde.org/settings/panel/emailpreferences/

To: broulik, #plasma, #plasma:_design, hein
Cc: cfeck, colomar, ivan, graesslin, lbeltrame, mart, plasma-devel, lesliezhai, 
ali-mohamed, jensreuterberg, abetts, sebas

Re: Review Request 129377: Remove trash kcm docbook from kio-extras

2016-11-12 Thread Burkhard Lück 

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/129377/
---

(Updated Nov. 12, 2016, 2:25 p.m.)


Status
--

This change has been marked as submitted.


Review request for Documentation, Plasma and Aleix Pol Gonzalez.


Changes
---

Submitted with commit f7a6bc8721c90e992e7535436929cd9580d46efc by Burkhard Lück 
to branch master.


Repository: kio-extras


Description
---

the trash kcm docbook  will be imported from kde-runtime master with history 
into frameworks/kio where the code is, see 
https://git.reviewboard.kde.org/r/128912/
this docbook was imported into kio-extras with 
https://cgit.kde.org/kio-extras.git/commit/doc/kcontrol/trash/index.docbook?id=09cd27cca7daa2339960efa0ab852ad492b5ac97
 but why?


Diffs
-

  doc/CMakeLists.txt 0afa5b5 
  doc/kcontrol/CMakeLists.txt fe7c777 
  doc/kcontrol/trash/CMakeLists.txt 6790c61 
  doc/kcontrol/trash/index.docbook d550b1b 

Diff: https://git.reviewboard.kde.org/r/129377/diff/


Testing
---

builds


Thanks,

Burkhard Lück

Re: Review Request 128912: Import kcontrol docbooks for code in kio from kde-runtime master

2016-11-12 Thread Burkhard Lück 

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/128912/
---

(Updated Nov. 12, 2016, 2:13 p.m.)


Status
--

This change has been marked as submitted.


Review request for Documentation, KDE Frameworks and Plasma.


Changes
---

Submitted with commit e0c3945a88fd041f36ff3873f6fe0ca5cfa00158 by Burkhard Lück 
to branch master.


Repository: kio


Description
---

doc/kcontrol/cache doc/kcontrol/proxy doc/kcontrol/useragent 
doc/kcontrol/cookies/ doc/kcontrol/useragent/ doc/kcontrol/netpref/ 
doc/kcontrol/ebrowsing/ doc/kcontrol/smb/
splitted from kde-runtime master using https://github.com/ajdruff/git-splits

Not included in build so far, the docbooks are imported as they were splitted

To be done next:
Move to existing subdir docs, adapt CMakeLists.txt, change dtd to kf5, 
rename ebrowsing to webshortcuts
reapply a few updates done in plasma-desktop
remove these imported kcontrol docbooks from plasma-desktop
adapt scripty to these changes


Diffs
-

  docs/CMakeLists.txt 350bce3 
  docs/kcontrol5/CMakeLists.txt PRE-CREATION 
  docs/kcontrol5/cache/CMakeLists.txt PRE-CREATION 
  docs/kcontrol5/cache/index.docbook PRE-CREATION 
  docs/kcontrol5/cookies/CMakeLists.txt PRE-CREATION 
  docs/kcontrol5/cookies/index.docbook PRE-CREATION 
  docs/kcontrol5/netpref/CMakeLists.txt PRE-CREATION 
  docs/kcontrol5/netpref/index.docbook PRE-CREATION 
  docs/kcontrol5/proxy/CMakeLists.txt PRE-CREATION 
  docs/kcontrol5/proxy/index.docbook PRE-CREATION 
  docs/kcontrol5/smb/CMakeLists.txt PRE-CREATION 
  docs/kcontrol5/smb/index.docbook PRE-CREATION 
  docs/kcontrol5/trash/CMakeLists.txt PRE-CREATION 
  docs/kcontrol5/trash/index.docbook PRE-CREATION 
  docs/kcontrol5/useragent/CMakeLists.txt PRE-CREATION 
  docs/kcontrol5/useragent/index.docbook PRE-CREATION 
  docs/kcontrol5/webshortcuts/CMakeLists.txt PRE-CREATION 
  docs/kcontrol5/webshortcuts/index.docbook PRE-CREATION 
  src/ioslaves/trash/kcmtrash.desktop c8ff4ac 
  src/kcms/kio/cache.desktop f0bd261 
  src/kcms/kio/cookies.desktop f91b5cd 
  src/kcms/kio/netpref.desktop 963ef3f 
  src/kcms/kio/proxy.desktop c11451d 
  src/kcms/kio/smb.desktop c323614 
  src/kcms/kio/useragent.desktop 856e9cc 
  src/kcms/webshortcuts/webshortcuts.desktop 24fbb2a 

Diff: https://git.reviewboard.kde.org/r/128912/diff/


Testing
---

history looks good, see git-log-doc-kcontrol.output, same log in kde-runtime 
master


File Attachments


git-log-doc-kcontrol.output
  
https://git.reviewboard.kde.org/media/uploaded/files/2016/09/15/94d16f46-6d99-4597-b7a4-3ba4add1dee0__git-log-doc-kcontrol.output
kio-log-docs.txt
  
https://git.reviewboard.kde.org/media/uploaded/files/2016/11/11/911e5e67-3094-4d47-aa94-95129f5545fe__kio-log-docs.txt


Thanks,

Burkhard Lück

Re: [Discussion] - Plasma Panels Future Enhancements?

2016-11-12 Thread Michail Vourlakos 

David, thanks a lot for the quick reply...


How will that help your use case of the dock?
You control the containment in a panel, but you're not changing 
Panel.qml itself.


Everything should be exposed to the client already as the "user 
setting its value" is just the panel config, another QML file with 
access to the same panelview properties as panel does.


In terms of Panel dimensions  we refer to thickness and length.
When the panel is at the top or bottom of the screen, thickness is 
height and length is width. When the panel is on the left or right 
side, it's the opposite.


Thickness is self-explanatory

Length can be confusing:
 - length should be the implicit size of the contents of the panel 
containment want to be.

 - min/max length are the

panelview will set the panel to length, but constrained to min/max 
length and then also bound to the size of the screen.





David, the Panel.qml file as I have understood it, is a layer between 
the panelview (c++ code) and the client (qml) implementation.
From my dock I can not find any way to alter the the panel's thickness 
(the Layout.preferredHeight which corrresponds in some
cases to thickness (horizontal) is not exposed). I need it because the 
user can set the icon size and the zoomfactor for the hovering
animation, these two are used to calculate the thickness which is needed 
in order to not have animations that are not shown. Currently

the user must try to set the optimal thickness manually

What I tried yesterday and it worked is the following, I added in 
Panel.qml the following but I dont know if this is going to be accepted in

the plasma source code:

...
property Item panelClient
...
onContainmentChanged: {
if (!containment) {
return;
}
containment.parent = containmentParent;
containment.visible = true;
containment.anchors.fill = containmentParent;
containment.locationChanged.connect(adjustPrefix);
adjustPrefix();

if (containment.children[0]) {  //With this I 
access the client's panel implementation

panelClient = containment.children[0]
}
}

 If the client's panel implementation has set a thickness 
variable then this binding is activated

     e.g. property int thickness (for the client code)
Binding {
target: panel
property: "thickness"
when: panelClient && panelClient.thickness!==undefined
value: {
if (!panelClient || !panelClient.thickness) {
return;
}

return panelClient.thickness;
}
}
.


3. "Real" Panel Transparency

Docks are panels that not only have a visual transparent
background but also they pass the mouse events to go underneath
their transparent areas. This is also the way Plank is doing it.
It creates a big window that for the transparent background leaves
all the mouse events go underneath. I believe plasma panels are Qt
windows managed from KWin, I think Qt is offering this feature by
setting some flags for the window. Do you have any ideas for this?


Plasma panels are just normal windows, so a mask could be set. Would 
have to be on the C++ side in panelview


I have already the c++ code for this and I am using it!!! :) I have set 
a QRect maskArea for panelview which can be set
and it works !!! :). I am sending from my dock the correct maskArea and 
everything works ok... but, in order to achieve
the communication between my implementation and the panelview I added in 
the Panel.qml the following:



 If the client's panel implementation has set a maskArea 
variable then this binding is activated

     e.g. property rect  mouseArea (for the client code)

Binding {
target: panel
property: "maskArea"
when: panelClient && panelClient.maskArea!==undefined
value: {
if (!panelClient || !panelClient.maskArea) {
return;
}

return panelClient.maskArea;
}
}
.


what do you think?

Re: [Discussion] - Plasma Panels Future Enhancements?

2016-11-12 Thread David Edmundson 
On Fri, Nov 11, 2016 at 4:15 PM, Michail Vourlakos 
wrote:

> Hello everyone,
>
>
> I am working in the plasma panels codepage these days, trying to enhance
> their features and I would like your opinion in some aspects of it. This
> effort is based a lot in the Now Dock Panel and its needs. I have sent some
> feature requests and lately I have sent some patches for review. I have the
> opinion that all the future enhancements should go inside plasma and kwin
> instead of keeping all the workarounds outside. I have done that mistake in
> the past :). So I am here and I am willing to work on the following but I
> will need your knowledge and guidance in some of them. Of course you are
> them who are going to decide if this is worth it...
>
>
> 1. Transparent Panels
>
> https://bugs.kde.org/show_bug.cgi?id=368384
>
> https://phabricator.kde.org/D3282
>
> https://phabricator.kde.org/D3283
>
>
> 2. Change Panel's Thickness from client side
>
> As I have understood it, the plasma panels change their thickness only by
> the user setting its value. I would like this to be able to be changed from
> the panel's client qml codepage. I tried to expose thickness with
> Layout.preferredWidth and preferredHeight but this didnt go quite well
> because it breaks current implementatation. I tried
> containment.implicitWidth... but this isnt exposed at all. Do you have any
> ideas how this should be implemented in order to not break current
> implementation?
>
>
> How will that help your use case of the dock?
You control the containment in a panel, but you're not changing Panel.qml
itself.

Everything should be exposed to the client already as the "user setting its
value" is just the panel config, another QML file with access to the same
panelview properties as panel does.

In terms of Panel dimensions  we refer to thickness and length.
When the panel is at the top or bottom of the screen, thickness is height
and length is width. When the panel is on the left or right side, it's the
opposite.

Thickness is self-explanatory

Length can be confusing:
 - length should be the implicit size of the contents of the panel
containment want to be.
 - min/max length are the

panelview will set the panel to length, but constrained to min/max length
and then also bound to the size of the screen.


3. "Real" Panel Transparency
>
> Docks are panels that not only have a visual transparent background but
> also they pass the mouse events to go underneath their transparent areas.
> This is also the way Plank is doing it. It creates a big window that for
> the transparent background leaves all the mouse events go underneath. I
> believe plasma panels are Qt windows managed from KWin, I think Qt is
> offering this feature by setting some flags for the window. Do you have any
> ideas for this?
>
> Plasma panels are just normal windows, so a mask could be set. Would have
to be on the C++ side in panelview



> regards,
>
> michail
>
>
> P.S. I am willing to work on the above and I am not in any hurry. I would
> like just the above to be offered from our main codepage.
>
>
>

[Differential] [Commented On] D3210: make scrollbar size configurable

2016-11-12 Thread hpereiradacosta (Hugo Pereira Da Costa) 
hpereiradacosta added a comment.


  I fear there is still some misunderstanding here. Maybe it is due to the 
original animation send for the other modification. 
  Maybe to make it clear to colomar:
  for a vertical scrollbar, the current animation does _not_ change the width 
of the scrollbar. The only thing it does is to make the scrollbar "groove" 
below it (the grey area that indicates the range along which the handle can 
move) appear gradually.
  The other dimension is unchanged. (and the hit area is unchanged too)
  Consequently, changes on the groove visibility and on the scrollbar width, to 
me at least, are completely uncorrelated. 
  I think this is why Marco wanted two options from the beginning, and I was 
arguing with either two or no option (but certainly not "one" option)
  
  Now the current situation, for which both the animation (the appearance of 
the groove) and the width of the scrollbar are controlled by the "enable 
animation" option, indeed sounds quite like magic to me (why would animations 
control the width of an object ?). I don't think this is satisfactory. Options 
must do what they are supposed to.
  
  So (and sorry for all the back and forth). I would make the "enable 
animation" option control *only* whether the groove fades-in and out on 
mouse-over, or is permanently drawn. 
  For the scrollbar width: either we have a separate option for that ("use thin 
scrollbars"), or we make the thin scrollbar the one and only scrollbar design. 
(And I would advocate for the second solution, keeping the full scrollbar width 
- including the emty sides- unchanged, as well as the corresponding hit area).

REPOSITORY
  rBREEZE Breeze

REVISION DETAIL
  https://phabricator.kde.org/D3210

EMAIL PREFERENCES
  https://phabricator.kde.org/settings/panel/emailpreferences/

To: mart, #plasma, #vdg, hpereiradacosta
Cc: colomar, alex-l, plasma-devel, lesliezhai, ali-mohamed, jensreuterberg, 
abetts, sebas

Re: [Discussion] - Plasma Panels Future Enhancements?

2016-11-12 Thread Michail Vourlakos 



2. Change Panel's Thickness from client side

As I have understood it, the plasma panels change their thickness only 
by the user setting its value. I would like this to be able to be 
changed from the panel's client qml codepage. I tried to expose 
thickness with Layout.preferredWidth and preferredHeight but this 
didnt go quite well because it breaks current implementatation. I 
tried containment.implicitWidth... but this isnt exposed at all. Do 
you have any ideas how this should be implemented in order to not 
break current implementation?




for this I found a solution by adding to the client plasma panel  a

property int thickness /// variable

and binding it correctly with the panelview




3. "Real" Panel Transparency

Docks are panels that not only have a visual transparent background 
but also they pass the mouse events to go underneath their transparent 
areas. This is also the way Plank is doing it. It creates a big window 
that for the transparent background leaves all the mouse events go 
underneath. I believe plasma panels are Qt windows managed from KWin, 
I think Qt is offering this feature by setting some flags for the 
window. Do you have any ideas for this?





for this I found a solution by adding in the panel view a new variable

QRect m_maskArea

which is exposed and lets the client code to set the mask for the panel,

it uses setMask from QWindow...


what do you think?