Re: Bad bug in gsequence

[Alexander Larsson <alexl redhat com>]

On Sat, 2006-12-02 at 22:56 +0100, Soeren Sandmann wrote:
> There is a bad bug in the version of EggSequence that nautilus is
> using. Basically it will sometimes cast the data pointer you insert
> into it to an EggSequence *, then attempt to read the first member.
> If that read succeeds, it will then compare the data to the
> end-iterator for the sequence. If they match, various bad things will
> happen.
> 
> I believe for nautilus this is likely harmless as the dereferenced
> memory is valid, and the items inserted are FileEntries whose first
> member is a pointer to NautilusFile, which can never match the
> end-iterator for the sequence. 
> 
> The attached patch, by James Livingston, fixes it. It should certainly
> be applied to HEAD, and probably also stable.

I applied this to HEAD and gnome-2-16. Thanks for the heads up.

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
 Alexander Larsson                                            Red Hat, Inc 
                   alexl redhat com    alla lysator liu se 
He's a Nobel prize-winning flyboy cowboy with acid for blood. She's a foxy 
insomniac mechanic from beyond the grave. They fight crime!