FROM : Dent John
DATE : Sun Jan 13 11:47:48 2008
Filed as bug # 5685715 as the API docs don't say either way.
The equivalent (but process-oriented) method, setSyncAlertToolPath:,
does claim to retain it's arguments, so I think clarification in the
documentation would be good.
The workaround of referencing the object from a dummy outlet in the
delegate worked.
Thanks,
denty.
On Jan 9, 2008, at 13:54, Clark S. Cox III wrote:
>
>
> Clark Cox III
> Clark.<email_removed>
> Sent from my iPhone
>
> On Jan 9, 2008, at 5:24, Dent John <<email_removed>> wrote:
>
>> Hi Jonathon,
>>
>> So, thanks for the quick response there.
>>
>> The object in question is instantiated in my NIB and, though is not
>> visible, it has outlets to other objects (specifically an array
>> controller and a spinning circle thing). I figured that would be
>> strong enough to prevent collection.
>
> That is not enough. The object needs pointers pointing *to* it in
> order to be kept around.
>
>>
>>
>> Also, the self is obviously stored in SyncServices as it knows that
>> it's a callback-able class for when of tries to
>> client:mightWantToSyncEntityNames:.
>
> If the syncservices docs claim to retain the object, then I'd file a
> bug. Otherwise, you cannot assume that it will keep your object
> around.
>
>>
>>
>> I implemented the finalize method just now and, sure enough, the
>> object is getting collected.
>>
>> I'll try making the object pop itself into a global to see if it
>> helps.
>>
>> d.
>> --
>> Sent from my iPhone
>>
>> On 9 Jan 2008, at 11:57, Jonathon Mah <<email_removed>> wrote:
>>
>>> Hi,
>>>
>>> On 2008-01-09, at 21:15, Dent John wrote:
>>>
>>>> For example, in a classes init method, I get it to NSLog the self
>>>> pointer. Then, after some time, I see errors relating to the class.
>>>
>>>
>>> Are you storing a reference to that instance in a "strong"
>>> location, such as a global variable? Otherwise the garbage
>>> collector has no idea that it might be used later on.
>>>
>>>
>>>
>>> Jonathon Mah
>>> <email_removed>
>>>
>>>
>>> _______________________________________________
>>>
>>> Cocoa-dev mailing list (<email_removed>)
>>>
>>> Please do not post admin requests or moderator comments to the list.
>>> Contact the moderators at cocoa-dev-admins(at)lists.apple.com
>>>
>>> Help/Unsubscribe/Update your Subscription:
>>> http://lists.apple.com/mailman/options/cocoa-dev/<email_removed>
>>>
>>> This email sent to <email_removed>
>> _______________________________________________
>>
>> Cocoa-dev mailing list (<email_removed>)
>>
>> Please do not post admin requests or moderator comments to the list.
>> Contact the moderators at cocoa-dev-admins(at)lists.apple.com
>>
>> Help/Unsubscribe/Update your Subscription:
>> http://lists.apple.com/mailman/options/cocoa-dev/
>> <email_removed>
>>
>> This email sent to <email_removed>
--
http://www.QQdd.eu/
DATE : Sun Jan 13 11:47:48 2008
Filed as bug # 5685715 as the API docs don't say either way.
The equivalent (but process-oriented) method, setSyncAlertToolPath:,
does claim to retain it's arguments, so I think clarification in the
documentation would be good.
The workaround of referencing the object from a dummy outlet in the
delegate worked.
Thanks,
denty.
On Jan 9, 2008, at 13:54, Clark S. Cox III wrote:
>
>
> Clark Cox III
> Clark.<email_removed>
> Sent from my iPhone
>
> On Jan 9, 2008, at 5:24, Dent John <<email_removed>> wrote:
>
>> Hi Jonathon,
>>
>> So, thanks for the quick response there.
>>
>> The object in question is instantiated in my NIB and, though is not
>> visible, it has outlets to other objects (specifically an array
>> controller and a spinning circle thing). I figured that would be
>> strong enough to prevent collection.
>
> That is not enough. The object needs pointers pointing *to* it in
> order to be kept around.
>
>>
>>
>> Also, the self is obviously stored in SyncServices as it knows that
>> it's a callback-able class for when of tries to
>> client:mightWantToSyncEntityNames:.
>
> If the syncservices docs claim to retain the object, then I'd file a
> bug. Otherwise, you cannot assume that it will keep your object
> around.
>
>>
>>
>> I implemented the finalize method just now and, sure enough, the
>> object is getting collected.
>>
>> I'll try making the object pop itself into a global to see if it
>> helps.
>>
>> d.
>> --
>> Sent from my iPhone
>>
>> On 9 Jan 2008, at 11:57, Jonathon Mah <<email_removed>> wrote:
>>
>>> Hi,
>>>
>>> On 2008-01-09, at 21:15, Dent John wrote:
>>>
>>>> For example, in a classes init method, I get it to NSLog the self
>>>> pointer. Then, after some time, I see errors relating to the class.
>>>
>>>
>>> Are you storing a reference to that instance in a "strong"
>>> location, such as a global variable? Otherwise the garbage
>>> collector has no idea that it might be used later on.
>>>
>>>
>>>
>>> Jonathon Mah
>>> <email_removed>
>>>
>>>
>>> _______________________________________________
>>>
>>> Cocoa-dev mailing list (<email_removed>)
>>>
>>> Please do not post admin requests or moderator comments to the list.
>>> Contact the moderators at cocoa-dev-admins(at)lists.apple.com
>>>
>>> Help/Unsubscribe/Update your Subscription:
>>> http://lists.apple.com/mailman/options/cocoa-dev/<email_removed>
>>>
>>> This email sent to <email_removed>
>> _______________________________________________
>>
>> Cocoa-dev mailing list (<email_removed>)
>>
>> Please do not post admin requests or moderator comments to the list.
>> Contact the moderators at cocoa-dev-admins(at)lists.apple.com
>>
>> Help/Unsubscribe/Update your Subscription:
>> http://lists.apple.com/mailman/options/cocoa-dev/
>> <email_removed>
>>
>> This email sent to <email_removed>
--
http://www.QQdd.eu/
| Related mails | Author | Date |
|---|---|---|
| Dent John | Jan 9, 11:45 | |
| Jonathon Mah | Jan 9, 12:57 | |
| Dent John | Jan 9, 14:24 | |
| Clark S. Cox III | Jan 9, 14:54 | |
| Jonathan Dann | Jan 9, 15:40 | |
| mmalc crawford | Jan 9, 16:55 | |
| Dent John | Jan 10, 10:27 | |
| Bill Bumgarner | Jan 10, 17:58 | |
| Clark Cox | Jan 10, 22:20 | |
| Dent John | Jan 13, 11:47 |






Cocoa mail archive

