Loading Morphic Wrappers change set

Stewart MacLean stewart.maclean at nzhis.govt.nz
Thu Sep 9 06:07:48 UTC 1999


Dan,

Thanks for your reply.

OK I tried that, checking to see if runTo: was alread defined (and putting
in a full stop after the bracket).

It filed in OK. Then I loaded the Cannon example (with WorldMorph changed to
PasteUpMorph) and ran the example method.

The system froze. I control breaked, and it got fouled trying to remove a
process from a linked list. I tried this a couple of times with the same
results.
(Is there any way of dumping the stack trace? - would like to have included
it here)

I guess I'll have to wait until the Maths guys convert their stuff to 2.5
:(.

Thanks,

Stewart

>  -----Original Message-----
>  From: Dan Ingalls [mailto:Dan.Ingalls at disney.com]
>  Sent: Thursday, September 09, 1999 4:53 PM
>  To: Stewart MacLean
>  Cc: recipient.list.not.shown; @cs.uiuc.edu
>  Subject: Re: Loading Morphic Wrappers change set
>
>
>  > !WorldMorph methodsFor: 'morphic wrappers' stamp: 'LC
>  12/31/1998 13:20'!
>  >runTo: aMorph
>  > "A world cannot run."
>  > self shake: 6! !
>  >
>  >where WorldMorph is undeclared in my 2.5 image.
>  >
>  >Being a Mophic newbie I'm not sure whether it's supposed to
>  pre-exist of
>  >come with your package?
>  >
>  >Any help appreciated (I wanted to fire that cannon!)
>
>  Stewart -
>  In 2.5, WorldMorphs were replaced by PasteUpMorphs, so that
>  they would make a reasonable unit of morphic storage whether
>  or not they were a whole world.  All pasteUpMorphs respond to
>  #isWorldMorph with true or false if you need to know what's what.
>
>  For this particular case, I believe the following would be an
>  effective replacement:
>
>  !PasteUpMorph methodsFor: 'morphic wrappers' stamp: 'DI
>  9/9/1999 11:11'!
>  runTo: aMorph
>  	self isWorldMorph ifTrue:
>  		["A world cannot run."
>  		^ self shake: 6]
>  	^ super runTo: aMorph! !
>
>  Note that if runTo: is elsewhere defined in PasteUpMorph,
>  then that def should replace the line with super in it.
>
>  The WorldMorph change may have fouled up MathMorphs a lot,
>  but we managed to keep it compatible enough that we read in
>  all the changes while the system was running in Morphic.  My
>  guess is that it shouldn't take too many fixes like this to
>  bring it up in 2.5.
>
>  Hope this helps.
>
>  	- Dan
>
>


Content-Type: application/octet-stream;
	name="CannonMorph.st"
Content-Disposition: attachment;
	filename="CannonMorph.st"

Attachment converted: Anon:CannonMorph.st (????/----) (00013040)





More information about the Squeak-dev mailing list