[Cuis] Newbie question

Hari Balaraman hbalaraman at gmail.com
Wed May 13 15:17:29 CDT 2015 

Hi Hannes,
Thank you for your reply. My responses are inline.
Best regards,
Hari

> On May 13, 2015, at 4:01 PM, H. Hirzel <hannes.hirzel at gmail.com> wrote:
> 
> Hari,
> 
> A suggestion:
> 
> Maybe you can write some learner's notes written from your perspective
> (questions and answers) we may add then to as a file to
> 
> https://github.com/Cuis-Smalltalk/Cuis-Smalltalk-Dev/tree/master/Documentation

I really like that idea. I was thinking the same thing this afternoon. They will also serve to remind me of the things I (assume) I am learning. Will get back to you on that.

> Tentative title
> 
> Some suggestions for Cuis Morphic experiments for learners.
> 
> BTW here a have a series of Layout experiments which were fine one year ago
> 
> https://github.com/hhzl/Cuis-Add-Ons/blob/master/Add-Ons-LayoutDemo.pck.st
> 
> I have not checked if they are still fine with the current Cuis. The
> layout experiments start very simple and one builds on the other.

Thanks! I will check them out. Right now I am trying to avoiding layouts (and innerpluggablemorphs etc.) like the plague :) and rebuilding my UI with the most basic pieces and looking to switch behaviour on and off as required.  I like direct access when it comes to morphs and submorphs (and a sort of WYSIWIG in the ownership chain) and am still very suspicious of the value of having multiple levels of morphs contained in each other.

> --Hannes
> 
> 
> 
> On 5/13/15, Hari Balaraman <hbalaraman at gmail.com> wrote:
>> Hi Juan,
>> Thank you for your reply.
>> My replies to specific points are inline.
>> 
>>> On May 13, 2015, at 9:53 AM, Juan Vuletich <juan at jvuletich.org> wrote:
>>> 
>>> Hi Hari,
>>> 
>>> I think it is great that you experiment. For sure you'll learn and have
>>> fun. And maybe you can contribute useful code.
>> 
>> Hopefully soon! At this point, the only benefit to you is seeing how a
>> newbie tackles the system.
>> 
>>>> On 5/12/2015 7:19 PM, Hari Balaraman wrote:
>>>> Hi Ken,
>>>> 
>>>> Thanks for your reply. It came in just as I finished a slightly improved
>>>> version of the code I presented earlier. This code moves the embedded
>>>> morph in relation to the enclosing morph (by moving the centre of the
>>>> embedded morph in proportion to the centre of the container morph.) I
>>>> moved the variables and methods I used in the attached test class
>>>> (including the stepAt method) upstream to RectangleLikeMorph to test and
>>>> attach some screenshots (of that testing) when I embed an EllipseMorph
>>>> into a BorderedRectMorph and switch ‘proportionalresize' off and on
>>>> within the embedded morph while resizing the container morph. Thus, I
>>>> maintain control over layout on a per embedded morph basis.
>>>> ...
>>>> I don’t yet have a sense of what load the additional variables and
>>>> methods put on the system when added at the top of the hierarchy (other
>>>> than my laptop fan not going insane), but since the tax is on the
>>>> individual embedded morph for maintaining layout, I hope it is small.
>>>> 
>>>> Best regards,
>>>> 
>>>> Hari
>>>> 
>>> 
>>> Well, the cost is not so much loading the system, but the extra stuff that
>>> everybody needs to keep in our heads.
>>> 
>>> Besides, there are yet more possibilities, besides Layouts in Ken's
>>> example. For instance, in a workspace evaluate:
>>> 
>>>    g _ RectangleLikeMorph new openInWorld.
>>>    f _ EllipseMorph new.
>>>    g morphExtent: 400 at 400; color: Color orange.
>>>    g addMorph: f.
>>>     f morphPosition: 50 at 50; color: Color blue.
>>>    g morphPosition:70 at 70.
>>> 
>>> Then, in the same workspace, try:
>>>    g rotation: 0.0 scale: 0.5.
>>> 
>>> or:
>>>    g rotation: 0.0 scale: 2.
>>> 
>>> When Morphic3 is ready, this will also work:
>>>    g rotation: 1.0 scale: 1.0.
>>> The idea is that each morph can define its own coordinate system.
>> 
>> That does sound interesting.
>> I'd like to explore how morphs can be aware of and organise themselves
>> around other morphs as peers (either when told about them or not.) As
>> organic an approach as possible.
>> 
>>> Cheers,
>>> Juan Vuletich
>>> 
>>> 
>>>>>> On May 12, 2015, at 5:52 PM, Ken.Dickey <Ken.Dickey at whidbey.com>
>>>>>> wrote:
>>>>>> 
>>>>>> On Tue, 12 May 2015 00:24:06 -0400
>>>>>> Hari Balaraman <hbalaraman at gmail.com> wrote:
>>>>>> 
>>>>>> Hi Juan et al,
>>>>>> 
>>>>>> I’ve been pondering on how to position and size morphs (based on Juan's
>>>>>> comment below about automatic layout in an earlier exchange) without
>>>>>> having an intermediate layer of layoutMorphs.
>>>>>> 
>>>>>> Attached is a fileout of a test class with a class method “example 1”
>>>>>> showing a rectangle that varies its size based on its owner (I’m also
>>>>>> writing the method to reset the location of a morph if it wishes to be
>>>>>> remain in the same location relative to the midpoints of two sides of
>>>>>> the enclosing morph.)
>>>>>> 
>>>>>> This way a morph can be positioned and at any time sent a message to
>>>>>> relocate or resize itself based on its then owner. That way there are
>>>>>> only two actors, the morph and its owner, and a morph knows (based on
>>>>>> two messages, whether it should maintain its relative place and size
>>>>>> within its owner. Although I have got a bit of the hang of
>>>>>> layoutMorphs, I’m exploring whether positioning and repositioning a lot
>>>>>> of elements in my UI may be easier with this approach.
>>>>>> 
>>>>>> I was wondering what you think of this approach.
>>>>> Hari,
>>>>> 
>>>>> I think we all need to get more experience building things to see what
>>>>> works and is as simple as we can make it.
>>>>> 
>>>>> Please continue to build things and share them!
>>>>> 
>>>>> Here is the example using aLayoutMorpk and a LayoutSpec.
>>>>> 
>>>>> example2
>>>>> "
>>>>>   self example2.
>>>>> "	| f g |
>>>>> 	g := LayoutMorph initializedInstance openInWorld.
>>>>> 	g padding: 0.5. "center"
>>>>> 	f := RectangleLikeMorph new.
>>>>> 	g morphExtent:400 at 400; color: Color orange.
>>>>> 	g addMorph: f layoutSpec: (LayoutSpec proportionalWidth: 0.25
>>>>> proportionalHeight: 0.25).
>>>>> 	f morphPosition:150 at 150; color: Color blue.
>>>>> 	g morphPosition:70 at 70.
>>>>> 
>>>>> 
>>>>> Cheers,
>>>>> -KenD
>>>> 
>>>> _______________________________________________
>>>> Cuis mailing list
>>>> Cuis at jvuletich.org
>>>> http://jvuletich.org/mailman/listinfo/cuis_jvuletich.org
>>> 
>> 
> 
> _______________________________________________
> Cuis mailing list
> Cuis at jvuletich.org
> http://jvuletich.org/mailman/listinfo/cuis_jvuletich.org

More information about the Cuis mailing list