[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-panel-devel
Subject:    Re: Review Request: krunner not running apps / crashing after hitting
From:       "Aaron Seigo" <aseigo () kde ! org>
Date:       2009-03-29 18:52:34
Message-ID: 20090329185234.13734.77644 () localhost
[Download RAW message or body]


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://reviewboard.kde.org/r/452/#review698
-----------------------------------------------------------

Ship it!


looks good...

- Aaron


On 2009-03-27 22:20:05, wilder wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://reviewboard.kde.org/r/452/
> -----------------------------------------------------------
> 
> (Updated 2009-03-27 22:20:05)
> 
> 
> Review request for Plasma.
> 
> 
> Summary
> -------
> 
> Calling KHistoryComboBox::addToHistory( item ),  in some cases (depending on the \
> position of the item in history) causes KLineEdit to emit userTextChanged; this in \
> turns triggers queryTextChanged("") in interface, which resets the context. When \
> the runner is asked to run, the context query is "", so either the runner either \
> doesn't do anything (#181453) or it crashes (#188186 #186036) [at least before \
> today's fix in the libs by dfaure]. This behaviour is compatible with br 181453 \
> i.e. if one repeats the same query two times in a row, the second time it's \
> guaranteed to work.  
> This patch moves addToHistory after the run, therefore the context can't possibly \
> be reset before the match is run. The patch goes together with a patch in kdelibs \
> which prevents the signal to be emitted in some cases (the ones observed for this \
> situation), but, after discussing with dfaure, we can't block the signals \
> altogheter, so it might still happen that userTextChanged is emitted in some weird \
> edge cases I can't think of. 
> In any case, the patch is quite harmless and it moves the call in the right place \
> in case, later on, we might want to make sure that the run actually works before \
> adding the query term to history (right now my history is polluted with typos) 
> Too bad I found this after tagging. We can't do anything about it, can we?
> 
> 
> This addresses bugs 181453, 186036 and 188186.
> https://bugs.kde.org/show_bug.cgi?id=181453
> https://bugs.kde.org/show_bug.cgi?id=186036
> https://bugs.kde.org/show_bug.cgi?id=188186
> 
> 
> Diffs
> -----
> 
> branches/KDE/4.2/kdebase/workspace/krunner/interfaces/default/interface.cpp 945379 
> 
> Diff: http://reviewboard.kde.org/r/452/diff
> 
> 
> Testing
> -------
> 
> The patch solves the problem. So far krunner always launched what it was asked to \
> run without any crash 
> 
> Thanks,
> 
> wilder
> 
> 

_______________________________________________
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel


[prev in list] [next in list] [prev in thread] [next in thread]