Re: [evolution-patches] fix for the bug #274406 [calendar]

[Rodrigo Moya <rodrigo novell com>]

On Wed, 2005-04-27 at 08:15 +0530, chenthill wrote:
> This occurs if the status of the received in cal_opened_cb is not OK. So
> i think we can then remove the extra unref present there. This causes
> the problem. As soon as the signal is emitted the cal_opened_cb is
> called which does an additional unref and thus causes a crash in
> async_signal_idle_cb. Have attached the modified patch.
> 
right, this one looks ok. The extra _unref call was not needed, since we
are creating the hash table with g_hash_table_new_full, which means
g_object_unref is called when you remove an item from the hash table.

I guess this should go to both HEAD and 2.2 branch.
-- 
Rodrigo Moya <rodrigo novell com>