[gtg-refactor] tags drag and drop broken

[Luca Invernizzi]

You'll probably already know, but anyways...

Traceback (most recent call last):
  File "/home/luca/Projects/gtg/gtg-refactor/GTG/taskbrowser/tagtree.py", line 416, in on_drag_data_received
    tagtree_model.move_tag(par_iter_tagtree, drag_iter_tagtree)
  File "/home/luca/Projects/gtg/gtg-refactor/GTG/taskbrowser/tagtree.py", line 221, in move_tag
    c = c.get_parent()

[Luca Invernizzi (invernizzi)]

Lionel, please see commit 695: there is a hack in tasktree.py to update a task when its parents change, but it doesn't work properly. How could this be resolved? This is really bug #537932

[Luca Invernizzi (invernizzi)]

Well, perhaps is not bug #537932, but it's still annoying :-)

[Lionel Dricot (ploum-deactivatedaccount)]

Luca > I reviewed your commit and everything looks really fine for me (I didn't tested it, just read the code). Nice work Luca.

I will investigate a bit more later the refresh thingy.

[Pablo Olmos de Aguilera Corradini (pablox)]

I'm using the latest revision r745 and I get the same error when I try to drag-n-drop a tag to create a sub-tag (the only difference is the line):

Traceback (most recent call last):
  File "/usr/lib/python2.6/site-packages/GTG/taskbrowser/tagtree.py", line 463, in on_drag_data_received
    tagtree_model.move_tag(par_iter_tagtree, drag_iter_tagtree)
  File "/usr/lib/python2.6/site-packages/GTG/taskbrowser/tagtree.py", line 263, in move_tag
    c = c.get_parent()
AttributeError: 'NoneType' object has no attribute 'get_parent'

I used to use this feature on older versions...

[Pablo Olmos de Aguilera Corradini (pablox)]

I'm using r748 and the bug it's still there. Is kind of annoying... any plans to fix it? Should I create a new bug report?

[Luca Invernizzi (invernizzi)]

This bug is fine, there's no need to open a new one. Since we had a major rewrite, we are proceeding to fix bugs from the core code out. Drag and drop will be fixed soon, but it's not a top priority.

 May I ask why you use our development branch instead of the released code (http://edge.launchpad.net/gtg/0.2/0.2.4/+download/gtg-0.2.4.tar.gz)? That code is stable and has almost no bugs, and those which are reported usually get fixed in a timely manner. You can't complain for code which hasn't been released yet :)

[Pablo Olmos de Aguilera Corradini (pablox)]

On 10 May 2010 20:27, Luca Invernizzi <email address hidden> wrote:
> This bug is fine, there's no need to open a new one. Since we had a
> major rewrite, we are proceeding to fix bugs from the core code out.
> Drag and drop will be fixed soon, but it's not a top priority.

Nice, thanks :)

>  May I ask why you use our development branch instead of the released
> code
> (http://edge.launchpad.net/gtg/0.2/0.2.4/+download/gtg-0.2.4.tar.gz)?
> That code is stable and has almost no bugs, and those which are reported
> usually get fixed in a timely manner. You can't complain for code which
> hasn't been released yet :)

Oh, okay, my idea was to help finding new bugs, but if it's too soon,
you could prolly tell me when it's ""done"" so I can help.

Regards!
--
Pablo Olmos de Aguilera Corradini - @PaBLoX
http://www.glatelier.org/
http://www.chilemoz.org/
http://friendfeed.com/pablox/
http://www.linkedin.com/in/pablooda/
Linux User: #456971 - http://counter.li.org/

[Bryce Harrington (bryce)]

On Tue, May 11, 2010 at 12:42:48AM -0000, Pablo Olmos de Aguilera Corradini wrote:
> Oh, okay, my idea was to help finding new bugs, but if it's too soon,
> you could prolly tell me when it's ""done"" so I can help.

Reporting bugs against the development version is fine, but make sure to
be wearing your 'tester' hat rather than your 'user' hat. So for
example, try seeing how to reproduce the bug in a few different ways
(such as when using the ./scripts/debug.sh tool to launch gtg), and
itemize exact steps to reproduce the error. That makes things a lot
easier for the developers.

In this case, though, I think it's pretty well known that drag-and-drop
is busted, so is just a matter of someone having some time to code up a
solution. You're welcome to submit a patch if you're anxious to see a
fix and know coding (or have an interest in learning python). I've done
this for a few bugs I've found, and my patches were warmly received by
the devs.

[Luca Invernizzi (invernizzi)]

This bug is "fixed", in the sense that the incorrect behavior that still exist is due to a series of regression bugs (tagged regression).