In this week's KDE Commit-Digest: The beginnings of a network management applet in Plasma, work on containments, and improvements in the "RSS" data engine and "Devices" applet. Initial support for saving changes to documents in Okular. A Kross-based scripting plungin in KLinkStatus. "Tip-of-the-day" and "Qt methods to avoid" checkers for the Krazy code quality reporting system. First steps towards a C++ parser in Umbrello. Work on projection independence in Marble. More maintenance and bux fixing in Kooka. Continued work in KHTML. Expanded support for subtitles and audio channel selection in Phonon (prompted by the requirements of Dragon Player). A rewritten "Todo" view in KOrganizer. Further work towards Amarok 2.0. "HTTP webseeding" in KTorrent. Kross-based scripting in KPlato, with support for "find in multiple documents" in KOffice. Caching support extended to all KDE card games. Improvements in KTurtle, KNetwalk, and Kubrick. Initial imports and work on the "Killbots" and "Astrododge" games. Import of "RSSNOW", an alternate RSS feed reader Plasmoid. The Step physics package moves from playground/edu to kdereview so that it can move to the kdeedu module for KDE 4.1. Read the rest of the Digest here.
Comments
WebKit is free software and can be forked at any time if WebKit's SVN becomes "hostile" towards KDE. But since some KDE and Trolltech developers got SVN write accress hostility is unlikely.
A fork is no good if you've lost all of your developers.
Well, do you think merging two source code trees that have been diverged for years is that simple? KHTML developers always do merging whatever can be merged from Webkit tree but I guess not all are as straightforward as you think. You can always check the websvn to see what have been merged ;)
I got 65/100 points with todays svn version.
JavaScript performance also has been improved, I, and will be improved further if kjs-frostbyte branch has been merged to trunk.
Nice, kudos to KHTML team!
I'm especially interested in the network manager thingy. Is this going to replace knetworkmanager as a kind of "reincarnation" as a Plasma-applet? Will it be based on Solid? Does it provide a Plasma engine which can be used by other applets? Or is it directly based on natworkmanager/dbus/HAL ?
possibly, yes, yes, no :-)
So it will be
Plasmoid --> Plasma engine -> Solid --> NetworkManager ?
Very nice indeed. You'll get like a thousand places to feed the bugs.
Happy hunting.
4 <<< "a thousand".
Also, most of these places are very thin wrappers and unlikely to introduce many bugs - NetworkManager will do the "heavy-lifting".
But you'll have less bugs since you have less duplicated code and more of that well-tested. It's also easy to use networky features on other platforms without rewriting your plasmoid. Using networkmanager is right now the backend for 'networky' stuff used on Linux, though the plasmoid doesn't need to care about that, thanks to Solid. With solidshell, it's easy to find out where it's going wrong.
Of course no one keeps you from just porting knetworkmanager, if that works better for you.
Sure, if your wifi chipset cannot connect to WPA network you'll have to hunt the bug in Plasma, sure.
Be realistic, troll.
It could be a replacement for knetworkmanager but it doesn't have to be. The dataengine is based on Solid. Use cases for other apps include verification of network status and network usage stats.
Hi
as time goes, we are getting more and more stanalone plasmoids.i dont think that having 3 (and counting!) rss reader plasmoids or a standalone 'todo list' plasmoid is good idea.
I think plasmoids should be provided by apps, like KGet that includes a plasmoid.
so rss readers should go into Akregator.with a datanengine that gets data from Akregator...
this way 3 rss readers should go into akregator and notes applet should go into knotes.TODO list viewer should be provided by KOrganizer and...
(btw, the RSSNOW applet is so cool)
I never use akregator. Why should I have to download (Or, keep it installed since its a KDE application) a whole new application just to get a desktop widget? As long as the widget doesn't depend on the application (such as those provided by Amarok, for example) they shouldn't be bundled with the application.
I agree, that all those rss reader's should share their data with akgregator, but making those applets part of akgregator isn't the way to go. Since KDE4 has got akonadi for exactly these kinds of purposes. When there's a akonadi dataengine for plasma, all RSS data could be shared by all RSS applets/akgregator. And this way, nobody is forced to install akgregator if they just want one of those applets.
Glad to see you like RSSNOW btw. :)
If the 'TODO list viewer' is a reply to my comment above asking for one then I should clarify I wanted a plasma applet that was integrated with Kontact/KOrganizer. I use that for managing all my time and would prefer to have it available on the desktop rather than buried in the app.
Definitely don't want more standalone tools!
Akonadi is supposed to provide the data to both applications or applets. Till Adam has shown during the talk at last Summer's aKademy that it's relatively easy to have Plasma and Akonadi interact with each other. I'm pretty sure once Akonadi is in use, Plasmoids using that will pop up quickly. That's probably 4.2 timeframe, though.
i dunno ...
the fact that we have three rss readers sort of goes to show how easy it is to write these sorts of things now. the rss lib in kdepimlibs and plasma make it really trivial.
and part of the idea of plasma is that no one person knows exactly how you want to do it. so ... we make it as easy as possible to create your own things. magically, people are doing just that.
that said, i think it is great to see kalzium, kget and akonadi (to name three) shipping plasmoids and engines of their own. that kind of integration can only get us even further.
and thankfully, those two things (proliferation of MyPlasmoids and nicely integrated default plasmoids) are not mutually exclusive. we can do both.
Aaron,
I have only had a little play with a live CD of 4.0.0 so maybe I am missing some of the vision here but are these standalone plasmoids simply going to sit on the desktop? If I have a lot in use, won't they just clutter up the screen real estate and make it unusable? I know we can resize applets but I don't want to be doing that every time I need a function.
That would mean:
1. Minimise running applications in order to see the desktop
2. Locate and highlight the running plasmoid so that the border appears
3. Drag to resize so that I can read it
4. Use the applet
5. Resize it back down
6. Go back to using whichever application I want
This is quite a few steps and seems a little clumsy. Could we call applets to the forefront of the screen with shortcuts so that they appear "layered" over the top of running applications? That way, we wouldn't need to return to the desktop all the time.
Alternatively, is that zoom feature going to be used to simply zoom in on the plasmoid we need on the desktop?
Please explain as I am probably missing some of the "vision" for the KDE4 series here!
Thanks, mate.
You already can do this (from day 1), I believe you have to press ctrl-f8 or 9 or something like that. I don't use it that often cos I don't use plasmoids (yet...no must have plasmoids for me yet).
You can allready do this. Just press ctrl+f12, and your desktop with all it's applets will magically appear in front of your applications (which looks really cool with compositing enabled btw). And stuff where you allways want to keep an eye one, you can just put in the panel.
1 won't be necessary, see other comments.
2, 3 - zoom might help there.
4 - of course you need that ;-)
5 - zoom
6 just click on a window which is still visible behind the applets and you go back.
Thanks for all of your prompt replies, Gents. I'll have another play with KDE 4.0 very soon (when Fedora 9 is released!)
Cheers.
Hi
having standalone plasmoids has a few problems:
1)every standalone plasmoid needs a maintainer.so we will have the problem of 'unmaintained' plasmoids in future.
2)The process of playgground->kdereview->kdebase-workspace will be there just for little plasmoids.really, plasmoids are little, they are not full applications...
3)hunderds of bugzilla entries just for little plasmoids? reporting those bugs would be confusing...
im not saying all plasmoids 'should' be included in an application.sure, some of them should be standalone.for example minimize/show desktop or analog clocks...but look at current status: the new plasmoids for konqueror/kate/konsole profiles should be included in konqueror, kate and konsole.not kdebase or anywhere else.
about the fact that a little rss reader shouldnt depend on akregator, i must say that third party plasmoids will come for that.but plasmoids that are included in KDE should be intergrated.
Most plasmoids will be available from kde-look, just like styles, superkaramba applets etcetera...
Generally I'm in favour of making competing implementations of a thing, but a CPP parser (a proper, full one) is a very large and difficult task, and KDevelop's one seems to be quite far along and sounds like it will be top-notch, from what I hear. May I ask what is the rationale for the Umbrello devs to write their own one from scratch?
I Agree. I don't know what level of completeless they need, so maybe a simple parser will do for them, but if they want "real" code-flow understanding, they should better think about interfacing umbrello with KDevelop through a KDevelop-plugin.
All they need is to be able to parse .h headers for class, method and property names. Doxygen comments if they are feeling fancy. Nothing like KDevelop's parser.
Aha, that makes perfect sense - thanks, Ian!
> Nothing like KDevelop's parser.
Presumably they would like to be able to read project files as well though - which is necessary to figure out how all the code fits together.
KDevelop 4 has a lot of useful code for this.
What project files keep track of interdependencies? Anyways... that's not C++.
Does anyone know the status of KNetworkManager in KDE4 ?
I see work is being done on knetworkmanager in the work branch, but afaik this has nothing to do with KDE4(trunk).
I depend on this program every day with lots of different wlans and vpn connections, so if it is going to TRUNK I'm more than willing to contribute with some fixes where needed.
Isn't KNetworkManager replaced by Solid?
Well, kinda. Solid will use NetworkManager as the default backend on Linux but as far as I know it's not ready yet. Still, someone will have to write a Plasmoid which offers KNetworkManager's functionality via GUI.
oh no!
My impression of the NetworkManager guys is not very good after they have refused to implement the "Refresh wireless networks-list".
Come'on! This is an invaluable functionality. I took my laptop in the bus. When I arrived I had 200 networks in the knetworkmanager and the one I wanted to use wasn't discovered yet. That is just plain dumb!!!!!!
I also hate it.
There is a solution to work around it. It's not perfect, but it works.
For Kubuntu it's the following
/etc/dbus-1/event.d/25NetworkManager stop
/etc/dbus-1/event.d/25NetworkManager start
Is this actually a functionality missing in NetworkManager itself or is it just not exposed via a GUI in nm-applet/KNetworkManager?
It's with NetworkManager itself. The knetworkmanager guys would like to fix it. There is a bugreport for it somewhere in bugs.kde.org where they explained it.
Danm those gnomers ;-)
I build KDE trunk today and tested.
KNetworkManager 0.2 runs without a glitch in the systemtray, so I guess its really not a problem..
Still wonder if it will be ported to KDE4 though
The networkmanager applet will be a functional replacement for knetworkmanager. However, rather than AP based it will be profile based. So you'll be able to click on the applet, and select a profile such as "Work", "Home" or "Connect to any wireless network" Profile creation will be wizard or systemsettings based. The network management portion of Solid is being re-worked by Will Stephenson to better accomodate changes in NM 0.7 and be more extensible in the future.
the velocity of development in kde 4.{0,1} make me wonder if there is something like the C constant 10^8 m/sec for speed, 10^8 LOCs per release or whatever ...
keep going the good works heroes
:)
Is 4.0.3 still on for tag/release around the beginning of next month?
http://techbase.kde.org/Schedules/KDE4/4.0_Release_Schedule#26_March_200...
I tried out the 4.0.2 and although it fell apart to the point of unusability (some self inflicted) I'm looking forward to having another go at it. As soon as it's useable I'll be over full time and helping bugsquash :)
Any know if there's any chance of multiple panels soon?
Thanks for all the hard work & Danny for the digest.
Trunk has multiple panels. I'm not sure if that's been backported last weekend, so have a look at the commits by Riccardo. If it has been backported, it'll be part of 4.0.2 on April 2nd.
The commit digest on Easter Sunday. Now if anyone complains once more about delayed publication and alleges Danny of lazyness, I'll hunt them down and break their legs.
That's just awesome Danny! Major coolness.
well, actually it is a week late.
But I appreciate reading it as much as the next guy!
Thanks Danny
pascal, you better watch out for jakob now! :)
10 days late, retarded. Time Danny got the boot, and someone else given the chance to get these out timely.
Aaah, I guess you volunteer?
I hope you understand the Commit Digest will stop coming if Danny stops writing them, right? Nobody wants to do it because it is a huge amount of work... So if you consider 'never' timely, just keep on saying these things. Maybe Danny will quit and you can be happy cuz by then we won't have any news left (he's also a very important editor for the DOT).
They already "have the chance" (all information needed to prepare a Commit Digest is freely available) and yet, there is a curious absence of people stepping up. This is probably because you are the only person in the world who considers the Commit Digest running a few days late to be even worth commenting on, let alone spamming every edition about.
"This is probably because you are the only person in the world who considers the Commit Digest running a few days late to be even worth commenting on, let alone spamming every edition about.".
And the worst thing is that he uses a name similar to yours, only with a small "a". Can't you kick his ass?