Aha! Now I'm seeing what I misunderstood. Whoops. We should probably back
out those two changes from the inbox.
Thanks Bert.

On Mon, Sep 7, 2009 at 5:47 PM, Bert Freudenberg <bert at freudenbergs.de>wrote:

>
> On 08.09.2009, at 02:16, Ronald Spengler wrote:
>
>  Looking at this method, I don't really see why it needs to be on Object
>> (feels like if it goes anywhere it should be on Morph.)
>>
>> It has three senders in my trunk image, and one of them is mine:
>>
>> IconicButton, PartsBin, and my own Pointy.
>>
>> I decided to follow the rabbit down the hole and move the method from
>> Object to Morph; after that the object palette and my own class (which uses
>> IconicButtons to do more less what the object palette does) started saying
>> the method wasn't found on Object's class side.
>>
>> I sense metaprogramming, but I haven't figured out where it's happening.
>> As an experiment, I moved the method over to the class side of Morph, and
>> voila, everything works.
>>
>
> Not quite everything I would guess. This method is used to launch any
> object from a parts bin, e.g. the supplies/widgets/tools flaps and the
> object catalog, not just morphs. It typically is sent to a class but could
> very well be used with other objects too. So moving is seems inappropriate.
>
> - Bert -
>
>
>  I am not sure if I'm doing something terribly wrong with this, so please
>> apply with caution.
>>
>> In the inbox:
>>
>> Morphic-rss.173
>> Kernel-rss.245
>>
>> Which would be applied in that order. If this actually works out for
>> folks, I may also try moving #launchPartVia: out too.
>>
>> --
>> Ron
>>
>>
>
>
>
>


-- 
Ron
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20090907/f96cd1e0/attachment.htm