Whisker Browser

> Hi Doug,
>
> I'm a great fan of your Whisker Browser. I really can't fancy  
> developing
> without it any more. It has it glitches, though. Small ones.  
> Nothing to
> worry about. But now in a 3.9 image with the latests release (1.1)  
> things
> have gone a little bit worse. And I would like to do something  
> about it.
> So question is what should I do? Go on and try to fix it and send you
> changesets or just send a bug report?
>
> Anyway, here are the two things that bug me at the moment:
>
> Whenever I open a collapsed Whisker Window I get a workback because  
> the
> background window does not have a layout frame. Well, I wondered  
> why we
> need it anyway and disabled it. Seems to solve the problem (at the  
> cost of
> a slightly less aesthetic look when no method pane is open - which is
> almost never the case).
>
> Second thing is that the panes for classes and methods cover the  
> system
> window border at the buttom and to the right. This makes it  
> impossible to
> use the resize handle at the lower right corner. I haven't looked into
> this, yet.
>
> Martin
>

> Hi Martin, sorry I didn't reply sooner.  I've just put a new "1.2alpha"
> version of Whisker on SqueakMap, which fixes the most serious bug you
> pointed out -- when opening a collapsed Whisker window in 3.9-final.
>
> There are a few other minor cleanups/fixes, and I also converted the
> method category names to be Monticello-compatible: "*whisker" for method
> extensions, and "*whisker-override" for the few overrides.  (Is
> xxx-override still used for overrides in MC?)  I haven't taken the final
> step of saving it on SqueakMap (or SqueakSource) as a Monticello .mcz
> just yet, because when I converted it from a .cs.gz to an .mcz, the file
> was about twice as large... maybe that's normal, or I may not have
> converted it properly.
>
> Anyway, this version should work reasonably in 3.9-final now.  I didn't
> tackle a complete fix for the new pane resizers in 3.9 yet, though, I
> just added a LayoutFrame for that background rectangle so that no
> walkback error occurs.  So there's still the problem with the
> lower-right resizing widget, and mainly the fact that you can't resize
> the class or method panes as you could in 3.8.  If you or someone wanted
> to look into that, that would be great. :)  Feel free to enter a bug in
> Mantis to help tracking.  If you want to look into it, I'd suggest
> starting at WhiskerWindow>>insertClassPane:atIndex:, where it sends
> addMorph:inFrontOf:frame: as an example.  Hmm, maybe just the frame
> coordinates need to be adjusted to allow room for the resizing widgets.
> Also, maybe that background rectangle is in front of (thus hiding) the
> resizing widgets when it should be behind.  And see how a regular system
> Browser does it correctly (Browser>>addAListPane:to:...).
>
> - Doug
>
> ----- Original message -----
> From: "Martin Kuball" <[hidden email]>
> To: [hidden email]
> Date: Sun, 4 Feb 2007 17:22:09 +0100
> Subject: Whisker Browser
>
> Hi Doug,
>
> I'm a great fan of your Whisker Browser. I really can't fancy developing
> without it any more. It has it glitches, though. Small ones. Nothing to
> worry about. But now in a 3.9 image with the latests release (1.1)
> things
> have gone a little bit worse. And I would like to do something about it.
> So question is what should I do? Go on and try to fix it and send you
> changesets or just send a bug report?
>
> Anyway, here are the two things that bug me at the moment:
>
> Whenever I open a collapsed Whisker Window I get a workback because the
> background window does not have a layout frame. Well, I wondered why we
> need it anyway and disabled it. Seems to solve the problem (at the cost
> of
> a slightly less aesthetic look when no method pane is open - which is
> almost never the case).
>
> Second thing is that the panes for classes and methods cover the system
> window border at the buttom and to the right. This makes it impossible
> to
> use the resize handle at the lower right corner. I haven't looked into
> this, yet.
>
> Martin