[squeak-dev] Re: Polymorph in a trunk image

Ronald Spengler ron.spengler at gmail.com
Sat Sep 19 06:27:10 UTC 2009


Bingo -- I think.

Polymorph-ToolBuilder-Morphic introduces PSToolBuilder>>buildPluggableWindow:

This does has one sender, PluggableWindowSpec>>buildWith:

Removing PSToolBuilder>>buildPluggableWindow: makes the browser open
to a proper size again.

It's looking like you were on the money, Andreas:)

Only catch is, I'm not sure what the method did or why because I
removed it without quarter. I'm going to play around with trying to
figure out what it was doing and why.

On Fri, Sep 18, 2009 at 11:02 PM, Ronald Spengler
<ron.spengler at gmail.com> wrote:
> I'm not seeing changes to either such method, but I am certain that
> the problem happens somewhere inside of
> Polymorph-ToolBuilder-gvc.10.mcz, because (after tiny bit of hacking)
> I can load Polymorph-Widgets-gvc.97.mcz, and the problem doesn't
> reproduce.
>
> I'm going to have to keep looking.
>
> On Fri, Sep 18, 2009 at 8:03 PM, Andreas Raab <andreas.raab at gmx.de> wrote:
>> I've seen this before. Check if there are overrides in any of the buildWith:
>> methods (very likely) and revert them to see if that fixes it. In addition
>> check Polymorph's ToolBuilder>>open: method and compare it to
>> MorphicToolBuilder's version. It's got to be one of those.
>>
>> Cheers,
>>  - Andreas
>>
>> Ronald Spengler wrote:
>>>
>>> What's different about the way Monticello decides how big it's system
>>> windows should be when new ones are open from the way the system
>>> browser does it?
>>>
>>> With Polymorph in a trunk image, something is quite wrong with the
>>> system browser that isn't wrong with Monticello. I'm hoping this is a
>>> clue to the bug.
>>>
>>> If you're interested, see the affixed image.
>>>
>>> To be clear:
>>>
>>> I have loaded:
>>>
>>> Polymorph-Widgets-gvc.97.mcz
>>> Polymorph-ToolBuilder-gvc.10.mcz
>>>
>>> Gut says it's related to something in the ToolBuilder. I'm going to
>>> start looking at what changed there in the package.
>>>
>>
>>
>>
>
>
>
> --
> Ron
>



-- 
Ron



More information about the Squeak-dev mailing list