[Nepomuk] Review Request 108997: Fix bug 292950 by deleting QCoreApplication only after Storage->ServerCore.

[Simeon Bird]

> On March 7, 2013, 6:25 p.m., Vishesh Handa wrote:
> > I'm quite against the concept of adding something which *has* to be implemented in all other services as well. Mainly cause there are other services outside our control - telepathy and some others who might/might not patch this up. And even if we add this in the documentation, there is no guarantee people will read it when implementing new services.
> > 
> > But I don't know how else to fix it right now. Hence the delay. Do you have any other ideas?

Hm, I didn't even know there was a telepathy nepomuk service. You're quite right though, but I do have one other idea...


- Simeon


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/108997/#review28768
-----------------------------------------------------------


On Feb. 17, 2013, 6:01 p.m., Simeon Bird wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://git.reviewboard.kde.org/r/108997/
> -----------------------------------------------------------
> 
> (Updated Feb. 17, 2013, 6:01 p.m.)
> 
> 
> Review request for Nepomuk and Vishesh Handa.
> 
> 
> Description
> -------
> 
> Fix bug 292950 by deleting QCoreApplication only after Storage->ServerCore.
> 
> The bug occurs because Storage->ServerCore may persist after
> QCoreApplication is gone. If the socket receives a remote disconnected
> message from virtuoso (which it might because this is shutdown and
> virtuoso is being killed) it tries to translate it, but can't because
> the translators went with QCoreApplication. Thus the crash.
> 
> This was made much less likely by Vishesh already with cea2870f04eec
> but we can make it even less likely by connecting
> QCoreApplication::quit() to ServerCore::destroyed() (so that the
> application is only quit after the ServerCore destructor starts to
> execute). The downside is that we have to explicitly do this
> for every other service as well.
> 
> BUG: 292950
> FIXED-IN: 4.10.1
> 
> 
> This addresses bug 292950.
>     http://bugs.kde.org/show_bug.cgi?id=292950
> 
> 
> Diffs
> -----
> 
>   services/fileindexer/fileindexer.cpp 0f81b44babe3eff121a354114c13c884569cb3c5 
>   services/filewatch/nepomukfilewatch.cpp 6081132469ee5d4d1af9c15d654cccb0e3c0c7ed 
>   services/storage/storage.cpp 16f7fd483eeb462227287c797c4619c78c23092d 
>   servicestub/servicecontrol.cpp 97f96d4ed33c452fdc91ea666c91b6fbced1e461 
> 
> Diff: http://git.reviewboard.kde.org/r/108997/diff/
> 
> 
> Testing
> -------
> 
> Compiled, checked shutdown and restart
> 
> 
> Thanks,
> 
> Simeon Bird
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/nepomuk/attachments/20130308/de404f93/attachment.html>