[prev in list] [next in list] [prev in thread] [next in thread]
List: kwin
Subject: Re: Review Request 128824: Add KWindowSystem::closeWindow(WId) and its unit test
From: Harald Sitter <sitter () kde ! org>
Date: 2016-09-12 8:15:00
Message-ID: 20160912081500.10006.95114 () mimi ! kde ! org
[Download RAW message or body]
--===============1318536772626678786==
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
> On Sept. 8, 2016, 2:30 p.m., Martin Gräßlin wrote:
> > Overall it's good and I changed my mind compared to the previous review request \
> > and think we can add it.
> > Only change is the new virtual call in the private API. Yeah it's private, so we \
> > can change it, but nevertheless it's going to cause problems for distros.
> > The better way might be to have a small new class e.g. CloseWindowRequester which \
> > has the new virtual function. Adding Harald as internal ABI maintenance hero to \
> > comment on it.
>
> Thomas Lübking wrote:
> If you preserve ABI (eg. through a generic function or object getter) you'll end up \
> with broken behavior on the target systems. Ie. in ths case, the client code must \
> be able to query whether the feature is provided by the present backend and \
> otherwise show a warning instead of the button.
> => tag plugin versions and refuse loading of outdated ones to unstash the conflict? \
> (yes, debian will manage to introduce symbol namespaces... ;-)
> Martin Gräßlin wrote:
> > tag plugin versions and refuse loading of outdated ones to unstash the conflict?
>
> That wouldn't work. It would mean that e.g. plasma-workspace's kwayland-integration \
> plugin won't be loaded once the plugin interfaces changes. That's not what we want. \
> But I don't think that there is a problem with it being "broken". If I compare to \
> Qt there are quite a few api calls which just result in a qWarning "not implemented \
> in the qpa plugin".
It's true though that a client is left high and dry if the used plugin doesn't \
actually support an extension. OTOH that's fairly cheap to change if desirede since \
you internally need to know whether a plugin supports a given feature or not, so it's \
just a matter of having API so a client can query its support. Which could be a new \
`class ExtensionQueryInterface` or simply an `bool isCloseWindowSupported()` on the \
KWMS class.
- Harald
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/128824/#review99000
-----------------------------------------------------------
On Sept. 9, 2016, 2:33 a.m., Paulo Lieuthier wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/128824/
> -----------------------------------------------------------
>
> (Updated Sept. 9, 2016, 2:33 a.m.)
>
>
> Review request for kwin, Martin Gräßlin and Harald Sitter.
>
>
> Repository: kwindowsystem
>
>
> Description
> -------
>
> This missing function will enable LXQt's taskbar to not depend directly on \
> NETRootInfo or KWayland. It's weird you haven't added it before, as Plasma's old \
> and new taskbars call the base libraries directly as far as I can see.
>
> Diffs
> -----
>
> autotests/kwindowsystemx11test.cpp 363c805
> src/kwindowsystem.h a282ecd
> src/kwindowsystem.cpp fda1682
> src/kwindowsystem_dummy_p.h 736a6f2
> src/kwindowsystem_p.h 5f30042
> src/platforms/osx/kwindowsystem.cpp 1758829
> src/platforms/wayland/kwindowsystem.cpp a5dfa93
> src/platforms/wayland/kwindowsystem_p_wayland.h 14aa73e
> src/platforms/xcb/kwindowsystem.cpp 091c71a
> src/platforms/xcb/kwindowsystem_p_x11.h 72df642
>
> Diff: https://git.reviewboard.kde.org/r/128824/diff/
>
>
> Testing
> -------
>
> I installed on my system and tested it through LXQt's taskbar.
>
>
> Thanks,
>
> Paulo Lieuthier
>
>
--===============1318536772626678786==
MIME-Version: 1.0
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: 8bit
<html>
<body>
<div style="font-family: Verdana, Arial, Helvetica, Sans-Serif;">
<table bgcolor="#f9f3c9" width="100%" cellpadding="12" style="border: 1px #c9c399 \
solid; border-radius: 6px; -moz-border-radius: 6px; -webkit-border-radius: 6px;"> \
<tr> <td>
This is an automatically generated e-mail. To reply, visit:
<a href="https://git.reviewboard.kde.org/r/128824/">https://git.reviewboard.kde.org/r/128824/</a>
</td>
</tr>
</table>
<br />
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: \
10px;"> <p style="margin-top: 0;">On September 8th, 2016, 2:30 p.m. UTC, <b>Martin \
Gräßlin</b> wrote:</p> <blockquote style="margin-left: 1em; border-left: 2px solid \
#d0d0d0; padding-left: 10px;"> <pre style="white-space: pre-wrap; white-space: \
-moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: \
break-word;"><p style="padding: 0;text-rendering: inherit;margin: 0;line-height: \
inherit;white-space: inherit;">Overall it's good and I changed my mind compared to \
the previous review request and think we can add it.</p> <p style="padding: \
0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">Only \
change is the new virtual call in the private API. Yeah it's private, so we can \
change it, but nevertheless it's going to cause problems for distros.</p> <p \
style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: \
inherit;">The better way might be to have a small new class e.g. CloseWindowRequester \
which has the new virtual function. Adding Harald as internal ABI maintenance hero to \
comment on it.</p></pre> </blockquote>
<p>On September 9th, 2016, 11:09 a.m. UTC, <b>Thomas Lübking</b> wrote:</p>
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: \
10px;"> <pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: \
-pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;"><p style="padding: \
0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">If \
you preserve ABI (eg. through a generic function or object getter) you'll end up with \
broken behavior on the target systems. Ie. in ths case, the client code must be able \
to query whether the feature is provided by the present backend and otherwise show a \
warning instead of the button.</p> <p style="padding: 0;text-rendering: \
inherit;margin: 0;line-height: inherit;white-space: inherit;">=> tag plugin \
versions and refuse loading of outdated ones to unstash the conflict? (yes, debian \
will manage to introduce symbol namespaces... ;-)</p></pre> </blockquote>
<p>On September 9th, 2016, 11:16 a.m. UTC, <b>Martin Gräßlin</b> wrote:</p>
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: \
10px;"> <pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: \
-pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;"><blockquote \
style="text-rendering: inherit;padding: 0 0 0 1em;border-left: 1px solid \
#bbb;white-space: normal;margin: 0 0 0 0.5em;line-height: inherit;"> <p \
style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: \
inherit;">tag plugin versions and refuse loading of outdated ones to unstash the \
conflict?</p> </blockquote>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: \
inherit;white-space: inherit;">That wouldn't work. It would mean that e.g. \
plasma-workspace's kwayland-integration plugin won't be loaded once the plugin \
interfaces changes. That's not what we want.</p> <p style="padding: 0;text-rendering: \
inherit;margin: 0;line-height: inherit;white-space: inherit;">But I don't think that \
there is a problem with it being "broken". If I compare to Qt there are quite a few \
api calls which just result in a qWarning "not implemented in the qpa \
plugin".</p></pre> </blockquote>
</blockquote>
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: \
-pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;"><p style="padding: \
0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">It's \
true though that a client is left high and dry if the used plugin doesn't actually \
support an extension. OTOH that's fairly cheap to change if desirede since you \
internally need to know whether a plugin supports a given feature or not, so it's \
just a matter of having API so a client can query its support. Which could be a new \
<code style="text-rendering: inherit;color: #4444cc;padding: 0;white-space: \
normal;margin: 0;line-height: inherit;">class ExtensionQueryInterface</code> or \
simply an <code style="text-rendering: inherit;color: #4444cc;padding: 0;white-space: \
normal;margin: 0;line-height: inherit;">bool isCloseWindowSupported()</code> on the \
KWMS class.</p></pre> <br />
<p>- Harald</p>
<br />
<p>On September 9th, 2016, 2:33 a.m. UTC, Paulo Lieuthier wrote:</p>
<table bgcolor="#fefadf" width="100%" cellspacing="0" cellpadding="12" style="border: \
1px #888a85 solid; border-radius: 6px; -moz-border-radius: 6px; \
-webkit-border-radius: 6px;"> <tr>
<td>
<div>Review request for kwin, Martin Gräßlin and Harald Sitter.</div>
<div>By Paulo Lieuthier.</div>
<p style="color: grey;"><i>Updated Sept. 9, 2016, 2:33 a.m.</i></p>
<div style="margin-top: 1.5em;">
<b style="color: #575012; font-size: 10pt;">Repository: </b>
kwindowsystem
</div>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Description </h1>
<table width="100%" bgcolor="#ffffff" cellspacing="0" cellpadding="10" \
style="border: 1px solid #b8b5a0"> <tr>
<td>
<pre style="margin: 0; padding: 0; white-space: pre-wrap; white-space: \
-moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: \
break-word;"><p style="padding: 0;text-rendering: inherit;margin: 0;line-height: \
inherit;white-space: inherit;">This missing function will enable LXQt's taskbar to \
not depend directly on NETRootInfo or KWayland. It's weird you haven't added it \
before, as Plasma's old and new taskbars call the base libraries directly as far as I \
can see.</p></pre> </td>
</tr>
</table>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Testing </h1>
<table width="100%" bgcolor="#ffffff" cellspacing="0" cellpadding="10" style="border: \
1px solid #b8b5a0"> <tr>
<td>
<pre style="margin: 0; padding: 0; white-space: pre-wrap; white-space: \
-moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: \
break-word;"><p style="padding: 0;text-rendering: inherit;margin: 0;line-height: \
inherit;white-space: inherit;">I installed on my system and tested it through LXQt's \
taskbar.</p></pre> </td>
</tr>
</table>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Diffs</b> </h1>
<ul style="margin-left: 3em; padding-left: 0;">
<li>autotests/kwindowsystemx11test.cpp <span style="color: \
grey">(363c805)</span></li>
<li>src/kwindowsystem.h <span style="color: grey">(a282ecd)</span></li>
<li>src/kwindowsystem.cpp <span style="color: grey">(fda1682)</span></li>
<li>src/kwindowsystem_dummy_p.h <span style="color: grey">(736a6f2)</span></li>
<li>src/kwindowsystem_p.h <span style="color: grey">(5f30042)</span></li>
<li>src/platforms/osx/kwindowsystem.cpp <span style="color: \
grey">(1758829)</span></li>
<li>src/platforms/wayland/kwindowsystem.cpp <span style="color: \
grey">(a5dfa93)</span></li>
<li>src/platforms/wayland/kwindowsystem_p_wayland.h <span style="color: \
grey">(14aa73e)</span></li>
<li>src/platforms/xcb/kwindowsystem.cpp <span style="color: \
grey">(091c71a)</span></li>
<li>src/platforms/xcb/kwindowsystem_p_x11.h <span style="color: \
grey">(72df642)</span></li>
</ul>
<p><a href="https://git.reviewboard.kde.org/r/128824/diff/" style="margin-left: \
3em;">View Diff</a></p>
</td>
</tr>
</table>
</div>
</body>
</html>
--===============1318536772626678786==--
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic