Review Request 129057: RFC: Make sense out of Plasma plugin metadata loading

Aleix Pol Gonzalez aleixpol at kde.org
Wed Sep 28 13:29:42 UTC 2016



> On Sept. 28, 2016, 11:10 a.m., Marco Martin wrote:
> > src/plasma/applet.cpp, line 92
> > <https://git.reviewboard.kde.org/r/129057/diff/1/?file=481388#file481388line92>
> >
> >     don't we have already a kpluginmetadata, built in the private ctor?

I guess? I'm trying not to change the logic too much for the moment. If you think it can go, I'll drop it.


- Aleix


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/129057/#review99608
-----------------------------------------------------------


On Sept. 28, 2016, 4:44 a.m., Aleix Pol Gonzalez wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/129057/
> -----------------------------------------------------------
> 
> (Updated Sept. 28, 2016, 4:44 a.m.)
> 
> 
> Review request for KDE Frameworks and Plasma.
> 
> 
> Repository: plasma-framework
> 
> 
> Description
> -------
> 
> I've been looking into what we're actually doing nowadays when we learn about a plasmoid at boot. I realized that we were doing a bit of a roundabout and there's definitely shorter paths.
> 
> As far as I understood, what we're doing is:
> 
> 1. We list all plugins and construct their KPluginMetaData
> 2. We get the metadata file path and send it to corona, that will create a KService::Ptr and then construct a KPluginInfo from it.
> 3. KPluginInfo in turn parses the file again. Since recently KPluginInfo is parsing the files twice
> 3.1. using KPluginMetaData
> 3.2. using KDesktopFileParser
> 
> This patch, tries to simplify 2 and 3 into a KPluginMetaData construction call.
> 
> Furthermore, this will allow us eventually (in a separate iteration, and everything works according to the plan) to load plugins from metadata.json which would make it possible to get rid of the internal usage of the desktopfileparser within kcoreaddons. (With something like this https://paste.kde.org/pth9wmasc)
> 
> 
> Diffs
> -----
> 
>   src/plasma/applet.h 449f761 
>   src/plasma/applet.cpp 5e278dc 
>   src/plasma/private/applet_p.h c3fe1d2 
>   src/plasma/private/applet_p.cpp 0f37fe5 
>   src/plasma/scripting/appletscript.cpp 42c0395 
> 
> Diff: https://git.reviewboard.kde.org/r/129057/diff/
> 
> 
> Testing
> -------
> 
> My `plasmashell` starts beautifully, `CoronaTest::startupCompletion` doesn't for some reason I need to investigate further, but yes it's that late.
> 
> 
> Thanks,
> 
> Aleix Pol Gonzalez
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-frameworks-devel/attachments/20160928/33a82ced/attachment.html>


More information about the Kde-frameworks-devel mailing list