Todd,

> I've completely given up on trying to fix Morph itself as it is  
> *FRAGILE* and making little changes in it frequently results in a  
> frozen system (as does trying to write new layout managers).

A word of advise: As always, before you start hacking classes that are used
throughout the system, make a COPY and hack this copy. In this case you can
decide which morphs should use it and you have a cleanroom environment which
you can use for experimentation.

Also, Morphs should actually be rather robust in this area - #fullBounds
contains an error handler which will fix up its bounds if a problem is
encountered in the layout process before passing the error. Therefore,
unless you hack any of the existing layout managers this should be pretty
robust. If it isn't, I'd like to see an example where it locks up.

Cheers,
  - Andreas

> -----Original Message-----
> From: squeak-dev-bounces at lists.squeakfoundation.org 
> [mailto:squeak-dev-bounces at lists.squeakfoundation.org] On 
> Behalf Of tblanchard at mac.com
> Sent: Saturday, March 08, 2003 7:06 PM
> To: The general-purpose Squeak developers list
> Subject: Re: [Q] [Morphic] SystemWindow
> 
> 
> FWIW,  I've been working on ProtoMorph, trying to get a new 
> base Morph  
> class that is more straightforward to use.  The number one "issue" I  
> run into when trying to build composite morphs is layout.  It is my  
> opinion that the current layout schemes are a little too 
> "tense" - IOW  
> they execute too much code that sometimes conflicts and this 
> makes them  
> unpredictable.
> 
> I am making slow progress getting a ProtoMorph that can be composed  
> using a struts and springs scheme like you find in NextStep.
> 
> I've completely given up on trying to fix Morph itself as it is  
> *FRAGILE* and making little changes in it frequently results in a  
> frozen system (as does trying to write new layout managers).
> 
> My plan is to get ProtoMorph working properly including doing struts  
> and springs layout, then start reimplementing many of our popular  
> morphs on top of ProtoMorph rather than Morph.  When SystemWindow is  
> done, it should be really easy to make a new tool with conventional  
> resizing behavior.
> 
> I just thought I'd mention what I'm doing in case anyone else is  
> thinking about tackling the same issue.
> 
> On Saturday, March 8, 2003, at 07:03  AM, Chris Burkert wrote:
> 
> > Hello Squeakers
> >
> > I want to build a nice SystemWindow for my application. The 
> problem is  
> > resizing. For example take a Package Browser and resize it 
> as small as  
> > possible. There will be Morphs displayed outside the SystemWindow.
> >
> >
> >
> >
> > My code with the same problem would look like this (a DoIt should  
> > work):
> >
> > | window playfield bar |
> > bar _ (RectangleMorph new) width: 500.
> > window _ (SystemWindow labelled: 'my App').
> > playfield _ (RectangleMorph new)
> > 	layoutPolicy:		TableLayout new;
> > 	hResizing:		#spaceFill;
> > 	vResizing:		#spaceFill.
> > playfield addMorph: bar.
> > window
> > 	addMorph: playfield fullFrame: (LayoutFrame fractions: 
> (0 at 0 corner:  
> > 1 at 1) offsets: (0 at 0 corner: 0 at 0));
> > 	openInWorld.
> >
> >
> >
> >
> > So I tried the following:
> >
> > | window playfield bar |
> > bar _ (RectangleMorph new) width: 500.
> > window _ (SystemWindow labelled: 'my App')
> > 	layoutPolicy:		TableLayout new;
> > 	hResizing:		#shrinkWrap;
> > 	vResizing:		#shrinkWrap.
> > playfield _ (RectangleMorph new)
> > 	layoutPolicy:		TableLayout new;
> > 	hResizing:		#spaceFill;
> > 	vResizing:		#spaceFill.
> > playfield addMorph: bar.
> > window
> > 	addMorphBack: playfield;
> > 	openInWorld.
> >
> >
> >
> >
> > Now the Titelbar and some colors are displayed completely 
> wrong _but_  
> > it resizes correct. I believe this is a problem with Alignment. Is  
> > there a way around so that the window displays itself 
> correct _and_ it  
> > resizes to #shrinkWrap?
> >
> > Should SystemWindow be refactored so that it supports the 
> newer layout  
> > Policies?
> >
> > thanks
> >            Chris Burkert
> > --  
> > 
> --------------------------------------------------------------
> --------- 
> > -
> > Student of applied Computer Science at Chemnitz University of  
> > Technology
> >      http://www.chrisburkert.de/            chris at chrisburkert.de
> > 
> --------------------------------------------------------------
> --------- 
> > -
> > "I invented the term Object-Oriented, and I can tell you I 
> did not have
> >  C++ in mind." - Alan Kay --> http://www.whysmalltalk.com/
> >
> >
> 
>