[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20120812094844.GA15895@debian>
Date: Sun, 12 Aug 2012 13:48:44 +0400
From: Aleksey Cherepanov <aleksey.4erepanov@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: Aleksey's daily status report #6
On Sat, Aug 11, 2012 at 11:52:00PM +0200, Frank Dittrich wrote:
> On 08/11/2012 08:28 PM, Aleksey Cherepanov wrote:
> > Done
> >
> > - minor fixes (added menus, updated copyright)
> > - tarball
>
> I did not yet test the tarball, but I pulled your latest changes.
>
> $ make > /dev/null
> mainwindow.ui: Warning: Z-order assignment: '' is not a valid widget.
> mainwindow.ui: Warning: Z-order assignment: '' is not a valid widget.
> mainwindow.ui: Warning: Z-order assignment: 'verticalSpacer_4' is not a
> valid widget.
>
> I think I've seen these warnings before. Currently, I'm not worried
> about these, so no need for you to try fixing them during GSoC.
Ok. I did not touch them since last time you mentioned this problem.
> When I start johnny on the command line, I get this output in the terminal:
>
> $ ./johnny
> QMetaObject::connectSlotsByName: No matching signal for
> on_pushButton_clicked()
> QMetaObject::connectSlotsByName: No matching signal for
> on_pushButton_JohnStatus_clicked()
>
> I think I've also seen those before, or was it just one of them?.
I think there was only one.
> Googling for
>
> QMetaObject "No matching signal for on_pushButton_clicked"
>
> shows some results, but I am not familiar enough with Qt to judge how
> helpful they are.
>
> Does this need to be fixed? Not sure. Is this easy to fix? If not,
> ignore it for now, and work on the other tasks.
It is trivial problem. I need to comment these methods out.
In Qt there are signal and slots. Slots are methods. Signals reflect
events but looks and are called like methods. To handle signal one
could write connect(object1, SIGNAL(signal1(signature)), object2,
SLOT(slot1(signature))), after that when signal1 of object1 is emitted
slot1 of object2 is called.
But it is too long to write connections between our form's methods and
all buttons it holds. So there is a convenience function for auto
connection: it looks for all objects and all methods named like
on_objectName_signalName and connects them, so to do described
connection automatically I need to rename slot1 of object2 into
on_object1name_signal1 (every object (should) have unique name, name
of methods could be accesses through meta object system). I do not
have such function in code explicitly but I call it implicitly using
.ui file. So .ui defines a lot of objects that are connected with
handlers defined in my MainWindow class.
The problem occurred when I commented some buttons out but did not
comment their handlers out.
Should I fix it now or could it be postponed till 1.2?
Thanks!
--
Regards,
Aleksey Cherepanov
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
