battery shrinking problem in battery plasmoid

[Sebastian Kügler]

On Friday 06 February 2009 22:32:47 matthias sweertvaegher wrote:
> hi sebas
>
> sorry for my late reply, caught some fever this week
>
> On Tue, Feb 3, 2009 at 1:51 AM, Sebastian Kügler <sebas at kde.org> wrote:
> >> First, thanks for having the guts to create a whole new
> >> rearchitectured desktop environment from scratch. :)
> >> also I guess for devs, it must be extra motivating to be able to
> >> contribute to still such a "pure" project.
> >
> > Yup, the pristine code base and not having to deal with a lot of old
> > cruft is very nice to have. But that's true for most of KDE actually,
> > since 4.0 allowed us to clean up big time. :)
>
> yes, I realize it is not plasma specific and I originally wrote this
> paragraph more accurately but then realized it was too much 'intro/blabla'
> for a simple bug report mail ;) also, to be right, I would have to send it
> then to all kde devs and not just plasma, so I just cut most of the
> paragraph and stuck to the plasma subject =) ;)

Haha, good call :-)

> >> anyway, in trying to get involved with my number one favorite FOSS
> >> project (KDE of course ;)), I am starting an effort in trying to fix
> >> my own pet bugs =) One down in the ksysguard, next is battery applet
> >>
> >> :)
> >
> > two down :)
>
> heh. just like that? too easy! ;)
> I was expecting at least some more debugging work ;)
>
> >> Here are my current findings:
> >> - the bug only occurs after having used the configuration dialog
> >> - seems disconnecting doesn't work? the plasmoid thinks there are more
> >> batteries ( I printed the battery_num var), but only 2 valid ones are
> >> detected (which is correct of course)
> >>
> >> why does the reconnection have to happen anyway? so, of course, when I
> >> disable the disconnect/reconnect code after the configuration is done,
> >> everything is fine (except maybe some memory leak I'm not aware off
> >> ;)). I tried debugging the events, but I don't manage to pull a lot of
> >> information out of it. So, I'm stuck.
> >
> > Apparently the sourceAdded() connection wasn't disconnected correctly for
> > all battery sources. The reconnect thing is probably some relic of the
> > past, and it shouldn't be needed anymore. Your solution of removing the
> > disconnect magic sounds right to me, and it indeed fixes #178116. It also
> > results in a patch with
> >
> >  workspace/plasma/applets/battery/battery.cpp |   19 -------------------
> >  workspace/plasma/applets/battery/battery.h   |    1 -
> >  2 files changed, 0 insertions(+), 20 deletions(-)
> >
> > Which is the best kind of bugfix (without losing functionality) there is.
> > :)
> >
> :D I am puzzled though why some battery source wouldn't disconnect

We could certainly find out, but I rather spend my time on something else. :)

> properly.. but I guess one can't have all answers in life ;) I'll just
> have to take on another bug to suppress this question.
>
> > I've committed (and effectively undid your previous change), thanks for
>
> what do you mean, my previous change? I can't commit..

Ow, I'm talking about the added "this" in the reconnect lines. (I had 
committed that for you already.)

> anyway, thanks for the comments!
> and as a reporter, I am supposed to close the bug, right?

I've done so with the commit. Including "Bug:<number>" in the commitmessage 
will do it automatically.
-- 
sebas

 http://www.kde.org | http://vizZzion.org |  GPG Key ID: 9119 0EF9