A few more fixes that have straggled in before 3.7gamma.  I didn't have
time to put together a gamma image last week, but I have a bit of time
this week so I should be able to do it in the next few days.

- Doug


------------------------------

5964FileList2preferrenceFix -- Karl Ramberg -- 20 June 2004
Seems like there was a bug in this update with a call to startsWith:
which was not in my image. I changed that to beginsWith:. I also added a
postscript to update the Tools flap"

5965WaveEditorInitFix-nk -- Ned Konz -- 22 June 2004
The default WaveEditorMorph in the Objects catalog wasn't getting its
data initialized correctly. Because of this, it wouldn't store on disk
properly.
This change set changes the initialization of the default
WaveEditorMorph to use a real sound instead.

5966SuppressCtrlKeyBrowser-nk -- Ned Konz -- 21 June 2004
v2: fixed what to show in MVC.
Now that we have the 'what to show' menu, the use of the control key to
display decompilation without temps and the use of the shift key to
display decompilation can be replaced with a single modifier key in the
'show decompiled' code holder views.
This avoids the surprising and mystifying behavior upon accidentally
accepting a method with Ctrl-Alt-S, which can on some platforms display
a decompiled method with no temp names.
This change set:
* removes sensitivity to Ctrl and Shift keys in all but the 'show
decompiled' mode of the various browsers
* factors all of the decompilation display logic into a single method
* retains backwards compatibility with other browsers
* makes the 'what to show' menu also work in MVC (so as not to lose the
ability to display decompiled methods in the MVC environment)
* makes the display of decompiled code *without* temp names be the only
special key sequence: if you're already displaying decompiled methods
and hold the shift down when you select a method, you will see the
decompilation without the temp names.

5967nilUUIDFix-pk -- Pavel K?ivánek -- 28 June 2004
Reposted by:	Ned Konz
CS 5919KCPInitializeNewFix broke UUID>>nilUUID because UUID defined an
initialize method and did not expect to have it called from its new:
method.
This change set fixes that problem.
Pavel's comments:
    UUID nilUUID -> de6309fd-188f-0643-9c90-0edb825f2f46
    UUID nilUUID isNilUUID -> false
Fixed:
    UUID nilUUID -> 00000000-0000-0000-0000-000000000000
    UUID nilUUID isNilUUID -> true
Ned's comments:
I have scanned the image and verified that UUID is the only variable
class that defines an initialize method.
variableClasses := Behavior allSubInstances select: [ :ea | ea
isVariable ].
variableClasses select: [ :ea | ea includesSelector: #initialize ].




<This automated message brought to you by the UpdateIncorporationTool.>
<See the UpdateIncorporationTool package on SqueakMap for details.>