[comp.sys.mac.system] Bizarre INIT conflict of the week

kenh@hscfsas1.harvard.edu (Ken Hancock) (10/16/90)

Ok...let's see if someone can figure out the rhyme and reason to
this one...I'm stumped.
 
Mac SE/30
System 6.0.5
Multifinder 6.1b9
Desktop Manager
IconWrap 1.2
HyperCard 2.0
 
Click in the arrows of any scrolling field in HyperCard 2.0 and the
application exits with the typical "out of memory."  IconWrap is the
culprit here.  Remove it and everything runs fine, whether I'm running
one INIT or dozens.

And I always thought it was perfectly harmless...

Ken

-- 
Ken Hancock                   | INTERNET: kenh@hscfsas1.harvard.edu 
Isle Systems                  | Disclaimer: My opinions are mine,  
Macintosh Consulting          | your opinions are yours.  Simple, isn't it?

thecloud@dhw68k.cts.com (Ken McLeod) (10/23/90)

In article <4435@husc6.harvard.edu> kenh@hscfsas1.harvard.edu (Ken Hancock) writes:
>Ok...let's see if someone can figure out the rhyme and reason to
>this one...I'm stumped.
>
>Mac SE/30
>System 6.0.5
>Multifinder 6.1b9
>Desktop Manager
>IconWrap 1.2
>HyperCard 2.0
>
>Click in the arrows of any scrolling field in HyperCard 2.0 and the
>application exits with the typical "out of memory."  IconWrap is the
>culprit here.  Remove it and everything runs fine, whether I'm running
>one INIT or dozens.
 
 
 I've heard the same thing now from another source. It's certainly bizarre,
since I'm running an identical software setup on my IIcx and haven't been
able to duplicate the problem. I've tried various combinations of INITs,
running with and without 32-Bit QuickDraw, etc.
 
 What makes the problem really interesting is that the original version of
IconWrap had problems with HC 1.x fields, just as you describe. This was
because that version of IconWrap tail-patched _CopyBits, interfering with
the wonderful and strange return-address checking dependencies in the
system. Version 1.2 was a "clean" rewrite, at least in terms of structure.
 
 The only thing I can think of at the moment is that either your version
of "IconWrap 1.2" is not *really* 1.2 (2878 bytes, 4/4/89), or your copy
of HyperCard 2.0 is different (I'm running the "gold" 2.0 from the 5-disk
set, created 8/31/90.)  I know, they're the same. Just checking.
 
 If anybody else is having this problem, please e-mail me with info about
your system configuration, list of INITs, etc.  I'd like to nail this
down and stomp on it.
 
>And I always thought it was perfectly harmless...
>
>Ken
 
 No, HyperCard is actually quite dangerous, which is why naive users have
to be protected by restricting their user level choices in the Home stack.
(*groan*)
 
-"ken"

-- 
==========     .......     =============================================
Ken McLeod    :.     .:    UUCP: ...{spsd,zardoz,felix}!dhw68k!thecloud
==========   :::.. ..:::   INTERNET: thecloud@dhw68k.cts.com
                ////       =============================================