Review Request: Fix pretty printer for QString.

[Maciek Zarzycki]

> On Oct. 5, 2011, 7:09 a.m., Niko Sams wrote:
> > Seems ok. Though this would be probably needed for every printer? (better: gdb should handle exceptions gracefully)
> > 
> > Would be great if you could add a unit test for an uninitialized string.
> 
> Niko Sams wrote:
>     see also this bug: http://sourceware.org/bugzilla/show_bug.cgi?id=11407
>     it addresses this issue, and seems to be fixed for 7.3. I haven't tried it yet though.
>     
>     which gdb version did you test with?

That is strange, cause I'm running gdb-7.3.1 on gentoo. I didn't have problems with other variable types though, only QString.

As for the unit test, i never got a 100% reproducable crash so I don't think I can UT it.


- Maciek


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/102778/#review7098
-----------------------------------------------------------


On Oct. 4, 2011, 11:07 p.m., Maciek Zarzycki wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://git.reviewboard.kde.org/r/102778/
> -----------------------------------------------------------
> 
> (Updated Oct. 4, 2011, 11:07 p.m.)
> 
> 
> Review request for KDevelop.
> 
> 
> Description
> -------
> 
> The printer may be ran when the QString object is not yet initialized.
> In this case we might get memory access violation resulting from undefined value of QString::size.
> Wrap the printer inside try/except to prevent this.
> 
> 
> This addresses bug 280201.
>     http://bugs.kde.org/show_bug.cgi?id=280201
> 
> 
> Diffs
> -----
> 
>   debuggers/gdb/printers/qt4.py 51f3a11 
> 
> Diff: http://git.reviewboard.kde.org/r/102778/diff/diff
> 
> 
> Testing
> -------
> 
> Ran unit tests.
> 
> 
> Thanks,
> 
> Maciek Zarzycki
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kdevelop-devel/attachments/20111005/c7ee669e/attachment.html>