Keith Hodges wrote:
> Eliot Miranda wrote:
>   
>> On Mon, Mar 9, 2009 at 3:58 PM, Keith Hodges <keith_hodges at yahoo.co.uk
>> <mailto:keith_hodges at yahoo.co.uk>> wrote:
>>
>>     Keith Hodges wrote:
>>     > Keith Hodges wrote:
>>     >
>>     >>> I will at some point push out a 3.8.22 VM that should put up a
>>     more
>>     >>> helpful dialog message, but people who have older VMs underfoot
>>     >>> should realize this behaviour should be an indication that perhaps
>>     >>> your VM is too old.
>>     >>>
>>     >>>
>>     >> Unfortunately the machine that I have designated for Bob to be
>>     an auto
>>     >> build/test server is a power-pc machine. What are the chances of a
>>     >> power-pc version of the closures vm?
>>     >>
>>     >> just interested to know
>>     >>
>>     >> Keith
>>     >>
>>     > Ok I see it is designated beta1U for universal, so the problem lies
>>     > elsewhere!
>>     >
>>     > Keith
>>     >
>>     OK so assigining to closure temporaries is no-longer allowed and
>>     flags a
>>     syntax error (not my code you understand)
>>
>>
>> There is a preference that allows one to compile old code that does
>> assign to block arguments.  Its called allowBlockArgumentAssignment.
>>  Of course one can assign to closure temporaries.  e.g.
>>
>>     self do: [:i| | j | j := i + 1. ...
>>
>> But assigning to a block argument is dubious.
>>  
>>     
> Working in 3.10.2-Closures image.
>
> I dutifully rewrote the offending code (Installer-Formats-kph.4.mcz) and
> bootstrapped it using a bog standard fileIn from the snapshot/source.st
> file. However the first thing that happens is that MczInstaller
> recompiles (invoked from #newSubclassOf: newSuper type: type
> instanceVariables: instVars from: oldClass) not sure why since the class
> instVar layout hasnt changed at all.
>
> In this case I am probably better off removing the old class before I
> load the new one, however, this indicates the nature of potential
> problems to come.
>
> Keith
So basically for a quiet life enabling those two preferences is
essential in 3.10.2 until some form of audit of the size  of the problem
is carried out.

Bob is happily building away, and so 3.10.2-closures-lpf is on its way

Keith