Process-specific state broken and uncomplete

> Suppose you have a domain object which provides following ways to
> access the session:
>
> session: forkedSession
>      Processor activeProcess environmentAt: #session put: forkedSession
>
> session
>      ^ Processor activeProcess environmentAt: #session
>
>
> now, here what i'd like to write:
>
> | session |
>
>   session := MySession new .
>   session initialize blabla.
>
> "good fork"
> good := [
>     self session: session.
>    [ self process stuff ] ensure: [ session close ]
> ] newProcess.
>
> good resume.
> 1 minute asDelay wait.
> good terminate.
> self assert: (session isClosed).

> On 11/8/2010 7:39 PM, Igor Stasenko wrote:
>>
>> Suppose you have a domain object which provides following ways to
>> access the session:
>>
>> session: forkedSession
>>     Processor activeProcess environmentAt: #session put: forkedSession
>>
>> session
>>     ^ Processor activeProcess environmentAt: #session
>>
>>
>> now, here what i'd like to write:
>>
>> | session |
>>
>>  session := MySession new .
>>  session initialize blabla.
>>
>> "good fork"
>> good := [
>>    self session: session.
>>   [ self process stuff ] ensure: [ session close ]
>> ] newProcess.
>>
>> good resume.
>> 1 minute asDelay wait.
>> good terminate.
>> self assert: (session isClosed).
>
> You'll be happy to hear that Eliot's fixes address this very problem. I just
> wrote a simple test to illustrate the issue more clearly:
>
> ProcessTest>>testFaithfulTerminate
>        "Ensure that we have process-faithful termination and unwind"
>
>        | p |
>        p := [
>                [Semaphore new wait] ensure:[self assert: Processor
> activeProcess == p].
>        ] forkAt: Processor activePriority + 1.
>        p terminate.
>
> This test passes fine in our images that have the process-faithful debugging
> fixes.
>

> On 11/9/2010 11:41 AM, Casimiro de Almeida Barreto wrote:
>> Em 08-11-2010 17:23, Andreas Raab escreveu:
>>> Try it now. The method directoryContentsFor:do: was added to 4.2 and
>>> created a conflict when loading TweakCore. I've removed it from Tweak
>>> so hopefully it will work now.
>>>
>>> Cheers,
>>>    - Andreas
>> Ok. Installed by hand (all latest). It complains about CProjectBuilder&
>> other stuff.
>
> Well, it is Tweak->Core<- meaning only the essential parts of the API.
> Try evaluating
>
> CProjectMorph open: CSampleApp.
>
> to see a simple example.
>
> Cheers,
>    - Andreas
>
>