Steven Friedrich posted on Sun, 10 Jul 2011 19:18:27 -0400 as excerpted:
Steven Friedrich posted on Sun, 10 Jul 2011 19:18:27 -0400 as excerpted:
> with KDE we had something called Superkaramba that allowed little apps
> on our desktops,
Talking about superkaramba... I've been meaning to post this, and this
gives me the opportunity as it includes a fine example of reasonably
easy starter jobs. Anyway, what is superkaramba's status?
When I first upgraded to kde4, I needed a replacement for the missing
ksysguard kicker applet (ksysguard was rather broken and AFAIK there's
STILL no full gamut ksysguard applet replacement plasmoid) and I found
both superkaramba, and the yasp-scripted plasmoid on kde-look. Back then
I chose yasp-scripted since its scripts were strictly order-of inclusion
positioned so it was easier to learn at a time when I was already having
to cope with and work around other brokenness of a still VERY alpha-
quality late kde 4.2 (4.3 being beta, 4.4 rc, and 4.5 finally reaching
what to experienced users finally looked like a normal x.0 release,
despite the claims about earlier versions).
But here recently I finally decided to switch my real-time system monitors
over to superkaramba, so I've been working with it.
Frustratingly, some of its sensors were missing functionality like cpu
wait-load and distinct cached and buffered memory values.
I don't claim to be a coder, but that code was clear and commented quite
well, and in some cases it already fetched the numbers I wanted, only
failing to make them available as output, so even without being a coder,
the patches were easy enough to make and test.
I submitted bugs with the patches as #274906 and 275070 (links below)
back in early June, but from the looks of things, superkaramba appears to
be semi-abandoned. A lot of the documentation is still kde3 era. The
documentation says that simply running a theme in superkaramba itself once
should make it available in the plasmoid explorer, but it doesn't,
etc. (I think it needs skz packaged first but techbase lacks instructions
for that (that I could find) and the instructions saying to run it in
superkaramba once imply it's not needed.)
I obviously believe superkaramba fills a niche if I'm using it and
submitting bugs for it, so I'm concerned.
https://bugs.kde.org/show_bug.cgi?id=274906
https://bugs.kde.org/show_bug.cgi?id=275070
Superkaramba techbase entry:
http://techbase.kde.org/Development/Tutorials/SuperKaramba
Some snaps of what my monitor dashboard looked like -- this is back with
yasp-scripted, I've not posted any showing superkaramba yet.
(This is a full-size of my yasp-scripted setup. The screenshot
is of the scripts I submitted, that are now included as samples in the
yasp-scripted package. I modified my setup bit after that, before
ultimately switching to superkaramba. Long link, watch the wrap.)
http://kde-look.org/content/preview.php?preview=3&id=109367&file1=109367-1.png&file2=109367-2.jpeg&file3=109367-3.jpg&name=Yasp-Scripted+%28Systemmonitor%29+v1.0.8a
(This 1/3 size bugzilla screenshot of my entire desktop gives you a feel
for what it looks like. It was was documenting a plasma positioning bug.
The placement of the folderview on the left is off, but otherwise, it's
about what my desktop looked like including all the yasp-scripted stuff
across the top, modified a bit from the shot above.)
http://bugsfiles.kde.org/attachment.cgi?id=59229
It should be obvious from those why the current system monitor plasmoids
simply don't cut it, for what I'm doing, the reason I need yasp-scripted
and/or superkaramba.
FWIW, I wanted traditional Unix 1/5/15 load average numbers as well, but
there wasn't an existing sensor for that and my rudimentary patching/
coding skills, such as they are, weren't /quite/ upto creating a new
module from whole cloth, even with copying much of the /proc/ file
parsing, etc, from the existing modules.
I was /almost/ able to do it, tho, and I strongly suspect that it's
something our original poster, Eric, could do quite easily, since he
already knows C++ at least to some degree, and I make no such claims. My
experience with the patches above was that the code was remarkably clear
and easy to follow, such that even someone far more comfortable with bash
than anything else could successfully patch the sources for trivial
functionality and rebuild. If I can do that, surely someone with even a
bit of real C++ knowledge could go the next step, adding a sensor for
traditional Unix 1/5/15 minute load averages.
--
Duncan - List replies preferred. No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master." Richard Stallman
___________________________________________________
This message is from the kde mailing list.
Account management: https://mail.kde.org/mailman/listinfo/kde.
Archives: http://lists.kde.org/.
More info: http://www.kde.org/faq.html.
