[Konsole-devel] Review Request: split view: automatically close the split area if it becomes empty.

[Francesco Cecconi]

> On May 4, 2012, 3:08 p.m., Kurt Hindenburg wrote:
> > With this patch and following the steps in bko, I get this assert when I finish the d-n-d.
> > 
> > ASSERT: "containers().contains(container)" in file /mnt/kde/kde4/trunk/src/konsole/src/ViewSplitter.cpp, line 126
> >
> 
> Francesco Cecconi wrote:
>     Thanks for the test Kurt.
> 
> Kurt Hindenburg wrote:
>     Francesco, have you had anytime to look at this further?
> 
> Francesco Cecconi wrote:
>     Hi Kurt,
>     sorry but at moment i haven't much spare time and the portion of code 
>     relative to the issue is changed in the master branch.
>     
>     however, on the next week I would like to start the work on the new patch.
>     
>     Francesco
> 
> Francesco Cecconi wrote:
>     The new patch is ready, this should remove the issue with the assert and disable the "split menu" correctly.
>     
>     I've tested the patch with d-n-d and with a single window (no detach tab).
> 
> Kurt Hindenburg wrote:
>     AFter #4 in the bko, I get crashes.  I'll test on another system to see if I can get a backtrace.
>     
>     4. drag the tab on view B1 to window A
> 
> Francesco Cecconi wrote:
>     Thanks Kurt, the backtrace would be useful because I can't reproduce the crash on my two test pcs with debian.
> 
> Kurt Hindenburg wrote:
>     well it works on my kubuntu laptop; it crashes all the time on my Mac; I'll try some VMs.
> 
> Kurt Hindenburg wrote:
>     I should note that Konsole/KDE often act a little weird on my Mac - there are still some issues/bugs in the core kdelibs I think
>     
>     
>     Program received signal EXC_BAD_ACCESS, Could not access memory.
>     Reason: 13 at address: 0x0000000000000000
>     0x000000010167b1ac in QDrag::setMimeData ()
>     (gdb) bt
>     #0  0x000000010167b1ac in QDrag::setMimeData ()
>     #1  0x0000000101614589 in QDragManager::drag ()
>     #2  0x000000010167b3d9 in QDrag::exec ()
>     #3  0x000000010167b408 in QDrag::exec ()
>     #4  0x00000001000c68f6 in Konsole::TabbedViewContainer::startTabDrag (
>         this=0x104922ad0, 
>         tab=<value temporarily unavailable, due to optimizations>)
>         at /Volumes/OWC500GB/Projects/KDE/kde4/trunk/src/kde/kde-baseapps/konsole/src/ViewContainer.cpp:477
>     #5  0x00000001000c851b in Konsole::TabbedViewContainer::qt_static_metacall (
>         _o=0x104922ad0, _c=<value temporarily unavailable, due to optimizations>, 
>         _id=<value temporarily unavailable, due to optimizations>, 
>         _a=<value temporarily unavailable, due to optimizations>)
>         at ViewContainer.moc:251
>     #6  0x00000001013311f1 in QMetaObject::activate ()
>     
>

I agree with you Kurt, the only reference is the call drag->exec();


- Francesco


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/104717/#review13430
-----------------------------------------------------------


On June 25, 2012, 9:29 p.m., Francesco Cecconi wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://git.reviewboard.kde.org/r/104717/
> -----------------------------------------------------------
> 
> (Updated June 25, 2012, 9:29 p.m.)
> 
> 
> Review request for Konsole.
> 
> 
> Description
> -------
> 
> automatically close blank split, after that last tab is moved.
> 
> 
> This addresses bug 275771.
>     http://bugs.kde.org/show_bug.cgi?id=275771
> 
> 
> Diffs
> -----
> 
>   src/ViewContainer.h d5bdf73 
>   src/ViewContainer.cpp 980c130 
>   src/ViewContainerTabBar.h 5b97abc 
>   src/ViewContainerTabBar.cpp 29805f3 
>   src/ViewManager.h 7d1079e 
>   src/ViewManager.cpp 40ed915 
> 
> Diff: http://git.reviewboard.kde.org/r/104717/diff/
> 
> 
> Testing
> -------
> 
> - one tab + split + one tab -> move split right/left, no blank area
> - from bug: after step 4 no blank area
> 
> seems to work well but more test are necessary. ;)
> 
> 
> Thanks,
> 
> Francesco Cecconi
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/konsole-devel/attachments/20120627/f6de45d9/attachment.html>