have a question for morphic Designer in pharo.

Thank you for the excellent problem description. Even though I don't know any Korean,
the case was easy to follow. You came already quite far in making the designer work
in Pharo. 

I tried reproducing it, but got stuck pretty fast.
In a Pharo 2.0 20628 image on a mac, I opened monticello and 
added the repository
  MCHttpRepository
location: 'http://www.hpi.uni-potsdam.de/hirschfeld/squeaksource/MetacelloRepository'
user: ''
password: ''

In your description, you show that it contains a version 30. The latest version I see is 24.
Is there anything you wrote in Korean that I missed to do?


The Morphic Designer is maintained by Marcel Taeumel (@ hpi.uni-potsdam.de) in Squeak, and though it is no longer 

Peter Yoo wrote:
>1. im look monticello now. can look more version now. 28, 30

Interesting. I still cannot see anything newer than .24 in
http://www.hpi.uni-potsdam.de/hirschfeld/squeaksource/MetacelloRepository
Are you sure that it is exactly the same repository? There is likely to be a
different one for designer itself.

>2. email to Marcel Taeumel already. but not receive return mail. mabe into spam?

Possible, but unlikely. The last days were the worst time of the year to reach someone
in Germany. The weeks around christmas (24-26 december) and newyear are family
holidays, so it's likely that he doesn't read his mail. He reacted to your stackoverflow
post though.

>3. latest squeak is cannot use too. squeak 4.1 image(in squeak 4.4 all in one vm)
>with morphic designer standard version can use. but class not same(squeak version
>and pharo version). im little panic.

It should work in the latest squeak image. It is not used enough/doesn't have enough
test coverage that all mistakes are immediately noticed, but Marcel fixes bugs that are
found. With the quality of issue report you provide the major problem should be his
availability/priorities. The next week he's likely to have holidays.

>4. im look already. pharo canvas class not same squeak canvas class.
>this problem not easy. i try more now. but im newbie user. maybe so hard work.

They started as the same, but Pharo has made more radical and drastic clean-ups.
Pharo will break existing applications in order to improve things. The advantage
is that innovation can be much faster in Pharo, a disadvantage is that applications
need to co-evolve. The support for co-evolving is gradually improving, the
continuous integration server provided for Pharo contributions gives fast feedback
on consequences of changes.

Squeak tries to keep compatible with older applications while gradually making
everything unloadable. That is a much slower and more difficult approach.

Quite a lot of code that was not used was removed. Once in a while we find out that
a project uses that code and needs to be reintroduced. Or should be in a separate
extension package, or be moved into the using project. The issue you are now running
into seems to be the cleaning up of event handling that has taken place in Pharo
and not yet or not in the same way in Squeak. A lot of the code deep in the system
was very difficult to understand, so yes, this can be difficult work.

>5. im look already another UI project(http://uipainter-gsoc.over-blog.com/).
>but not handsome(little joke). maybe can then i want use this project. now im trace
>another way(for UiItemViewport class. but mouse event is not same to squeak env).

Yes, this years GSoC project did not get as far as Marcel's one. And the event handling
has changed since Marcel wrote the Designer. Signals was pretty intrusive.

UITreeViewPort doesn't seem to correctly set the source of its dragSpec.
I tried to change dragDropSpec in UiItemViewPort to

dragDropSpec
        ^ dragDropSpec
                source: self;
                yourself

That allowed the drag-and-dropping of a lot of the widgets, but created errors
where the dragDropSpec was set to nil (Tree widget). Then I looked at where
UiItemViewDragDropSpec was used, and found out that the source was not
set in UIDesigner initialize.

I can do drag and drop of all widgets with

        self ui widgetList
                selectionMode: UiViewSingleSelection;
                dragDropSpec: (UiItemViewDragDropSpec new
                        source: (self ui widgetList);
                        dropEnabled: false;
                        dragTransferType: #widgetClass).

So, in all I needed to make the following changes to make the designer
work in 20628 after loading development of ConfigurationOfDesigner-mt.24.mcz:

UIDesigner>gridForm

        | form canvas |
        form := Form extent: 20@20 depth: 32.
        form floodFill2: (Color gray: 0.98) at: 0@0.
       
        canvas := FormCanvas on: form.
       
        form pixelValueAt: 0@0 put: (form pixelValueFor: (Color gray: 0.2)).
        ^ form

UIDesigner>initialize

        super initialize.
        self ui setupUi: self.
        self ui prop addPaneSplitters.
       
        self selectDefaultModel.
       
        self ui centralScroller
                borderWidth: 0;
                fillStyle: (BitmapFillStyle fromForm: self gridForm).
       
        "BAD: Viewport should not be accessed from the outside."
        self ui widgetList viewport dragEnabled: true.
        self ui widgetList
                selectionMode: UiViewSingleSelection;
                dragDropSpec: (UiItemViewDragDropSpec new
                        source: (self ui widgetList);
                        dropEnabled: false;
                        dragTransferType: #widgetClass).
       
         self ui filterEdit deferEditSignalTime: 300.
       
        propertyTable := self ui propertyScroller widget.
               
        self connect: self ui filterEdit signal: #textEdited: toSelector: #nameFilter:.
               
        self connect: self ui newBtn signal: #clicked: toSelector: #reset.
        self connect: self ui saveBtn signal: #clicked: toSelector: #save.
        self connect: self ui openBtn signal: #clicked: toSelector: #load.

UIAbstractWidgetModel>initialize

        super initialize.
        self updateWidgets.
       
        SystemAnnouncer uniqueInstance
                on: ClassAdded
                send: #updateWidgets
                to: self.









 
 

When trying to load Designer in Pharo 3.0 I find uncompilable code in Animations:

'updateCurrentValue: aValue

        self target ifNil: [^ self].
       
        self target
                perform: (self property, #:) asSymbol
                with: aValue.'

I guess that

#: should be #’:’

On 27 Dec 2013, at 19:44, Stephan Eggermont <[hidden email]> wrote:

This is really cool :)

I would love to see how we could change the event system used because using thisContext all the time
does not work and breaks the idea of not reifying the stack all the time.

Once it will be working, it would be really nice to see if we can generate spec specification.

Stef

> I was successfull in making Designer start in Pharo 3.0 30659, with some help from
> Tobias Pape.

Tx tobias.

> The following changes were needed (compared to what was needed for Pharo 2.0)
> - floodfill2:at: is gone from Form. I replaced it with a simple fillColor:
>  That it had gone missing was not clear from issue 11570
        Do you like method with floodFill23 in their name and not invoked at all :)?
       
> - outsetBy: is deprecated and replaced by expandBy: in a lot of places
        we also cleaned a lot the logic and added MorphMargin to support polymorphic margins. So may be this is beneficial

> - (some symbol or string or something else, #:) asSymbol is replaced by ... asMutator
>  in a lot of places.
> - TranslucentColor is replaced by Color
>
> <PastedGraphic-1.png>
>
> TextMorph handleInteraction:forEvent: no longer exists, but gives a different problem in Pharo 2 than in 3
> just replacing it by handleInteraction: does not work. That makes a mess of the selection in
> theUiTextLineMorphs in the object inspector.

That I have no idea.
>
> Stephan
>
>

On 28.12.2013, at 10:07, Stéphane Ducasse <[hidden email]> wrote:

> I guess that
>
> #: should be #’:’

That is one way.
In this particular case,
  self property asMutator
was the better version, tho.

Best
        -Tobias
PS: Just curious, what was the reason to no longer allow #: ?
    Or, where can I find the respective discussion, just want to lern.

On 28 Dec 2013, at 14:24, Tobias Pape <[hidden email]> wrote:

I do not remember but may be this is ansi.
In visualworks #’zork foo’ is the way to have symbol with string with space.

On 29.12.2013, at 08:56, Stéphane Ducasse <[hidden email]> wrote:
Yes, I know the #'anything' form and I am completely ok with it.
I just was wondering, because things like #> #< scan directly :)
Thanks anyway.

Best
        -Tobias

On 28 Dec 2013, at 14:25, Tobias Pape <[hidden email]> wrote:
Pharo3 uses the new compiler infrastructure by default. This means that
instead of the old (hand-written) parser, we now use RBParser (hand written, too).

It seems that RBParser does not implement it. The question is who was wrong: the old
parser or the new? I have no idea.
But the good news is that we now just have two Parsers, not three… (syntax highlighting implements it’s
own parser, too).

What we need is a real grammar… that can solve these differences which come from the implementation.

Another example is #9 —> both RB and the old compile it as an integer, there is special code in the Parser
for that even. But do we want that? If we would have a definition of our grammar, we could look there and
say “yes, it’s part of the definition”.

Even better would be an executable grammar… we should explore of we can use PetitParser in the future,
but there are some open questions (error handling, speed…).

The most beautiful would be to have just *one* parser that has a high level model of it’s grammar that is
reusable everywhere.

        Marcus

On 30.12.2013, at 10:44, Marcus Denker <[hidden email]> wrote:
Ansi has one, but I think it is unpractical, dividing “symbols” and “quoted selectors” :(
Hehe. I think, both PP and Ometa would fit that criterion but I agree
that this is hard-think for the core-Parser of a Smalltalk :)


Best
        -Tobias