Cyrille I tried to understand what you need because minExtent seems to returns the minimum size that is required to display a morph. I tried your example and open the two morphs and everything looks fine. Did you check the implementation of minExtent because may be you can use it as a basis for optimalExtent? Stef On May 30, 2011, at 1:50 PM, Cyrille Delaunay wrote:

Thanks gary. Indeed #minExtent give something interesting according the 'resizing strategy' of the submorphs (#rigid #spaceFill #shrinkWrap). If resizing of submorphs are #rigid, it works well. For example :

|tmpPanelMorph| tmpPanelMorph := PanelMorph new changeTableLayout;hResizing: #shrinkWrap; vResizing: #shrinkWrap; listDirection: #leftToRight; yourself.

tmpPanelMorph addMorphBack: ( (PluggableListMorph on: (ListModel new list: #(adededed ) ;yourself) list: #list selected: #selectionIndex changeSelected: #selectionIndex: ) hResizing: #rigid; vResizing: #rigid ; yourself) . tmpPanelMorph addMorphBack: ( (PluggableListMorph on: (ListModel new list: #(a b c d e f g h i j k l m n ) ;yourself) list: #list selected: #selectionIndex changeSelected: #selectionIndex: ) hResizing: #rigid; vResizing: #rigid ; yourself).

tmpPanelMorph minExtent.

The minExtent is 'correct'. Now if you try with:

|tmpPanelMorph| tmpPanelMorph := PanelMorph new changeTableLayout;hResizing: #shrinkWrap; vResizing: #shrinkWrap; listDirection: #leftToRight; yourself.

tmpPanelMorph addMorphBack: ( (PluggableListMorph on: (ListModel new list: #(adededed ) ;yourself) list: #list selected: #selectionIndex changeSelected: #selectionIndex: ) hResizing: #spaceFill; vResizing: #spaceFill ; yourself) . tmpPanelMorph addMorphBack: ( (PluggableListMorph on: (ListModel new list: #(a b c d e f g h i j k l m n ) ;yourself) list: #list selected: #selectionIndex changeSelected: #selectionIndex: ) hResizing: #spaceFill; vResizing: #spaceFill ; yourself).

tmpPanelMorph minExtent.

It answers something smaller. So I don't know if it would make sense to be able to retreive the same extent whatever the 'extending stratgies' of the submorphs ? Maybe another method that would always return this value (optimalExtent like you said) ?

2011/5/23 Gary Chambers <gazzaguru2@btinternet.com> Try #minExtent, though that won't cater for things with scrollbars. Perhaps worth re-implementing #optimalExtent to deal with those.

Regards, Gary ----- Original Message ----- From: Cyrille Delaunay To: pharo-project Sent: Friday, May 20, 2011 2:34 PM Subject: [Pharo-project] How to know 'real' extent of a PanelMorph

Hello,

I have a panelMorph composed by several row/columns with different kind of morphs. What I would like to know is, what is the extent / size that I should use to FULLY display this panel (without scrollbars). Each time I ask for the extent of a panelMorph, it answers 50@40, whatever what is inside. If you openInWorld you will have something very very small. Is there a way to retrieve an extent with which the panel morph will be 'well' displayed (without having scrollbars everywhere)?

In other words, what I really want to know is: Given two panelMorph, is there a way to say that one need more place than the other to be displayed?

I would like a method that always return something like the first result (from the first example) , whatever the 'extending stratgies' of the submorphs (in my two examples, both would return the 'first' result). 2011/5/30 Stéphane Ducasse <stephane.ducasse@inria.fr>

cyrille

I was rereading your mail and I could not really understand what you are really looking for.

Stef

On May 20, 2011, at 3:34 PM, Cyrille Delaunay wrote:

...

Hello,

I have a panelMorph composed by several row/columns with different kind of morphs. What I would like to know is, what is the extent / size that I should use to FULLY display this panel (without scrollbars). Each time I ask for the extent of a panelMorph, it answers 50@40, whatever what is inside. If you openInWorld you will have something very very small. Is there a way to retrieve an extent with which the panel morph will be 'well' displayed (without having scrollbars everywhere)?

In other words, what I really want to know is: Given two panelMorph, is there a way to say that one need more place than the other to be displayed?

2011/5/31 Stéphane Ducasse <stephane.ducasse@inria.fr>

so I suggest that you just try to add optimalExtent based on minimal it does not seem that an existing method compute what you want.

For me I still did not clearly get it. It seems that you want the potential space that a morph should occupy in the space of its container?

Yes :)

Stef On May 31, 2011, at 9:51 AM, Cyrille Delaunay wrote:

...

I would like a method that always return something like the first result (from the first example) , whatever the 'extending stratgies' of the submorphs (in my two examples, both would return the 'first' result).

2011/5/30 Stéphane Ducasse <stephane.ducasse@inria.fr> cyrille

I was rereading your mail and I could not really understand what you are really looking for.

Stef

On May 20, 2011, at 3:34 PM, Cyrille Delaunay wrote:

...

Hello,

I have a panelMorph composed by several row/columns with different kind of morphs. What I would like to know is, what is the extent / size that I should use to FULLY display this panel (without scrollbars). Each time I ask for the extent of a panelMorph, it answers 50@40, whatever what is inside. If you openInWorld you will have something very very small. Is there a way to retrieve an extent with which the panel morph will be 'well' displayed (without having scrollbars everywhere)?

In other words, what I really want to know is: Given two panelMorph, is there a way to say that one need more place than the other to be displayed?