Add instVar in Process
Michael Klein
Mklein at nts.net
Wed Oct 6 19:09:33 UTC 1999
Rather than naming processes, I have a method
Process >># initiatorBlock
than answers the block that got forked.
I find this is enough to identify most processes, even if they are
unnamed
(very usefull if you are debugging somebody else's code).
I ported from VW to Squeak easily (while composing this response),
mostly because both systems have 'thisContext'.
I am presently trying to port to VA, but the exposure of
the stack is causing all sorts of iritating problems.
You may find this easier and more usefull than adding a name to Process.
In Squeak 2.5,
(Process allInstances collect: [:p| p initiatorBlock ]) reject: [:each
| each isNil ]
answers:
(
[] in Delay class>>DoIt
[] in WeakArray class>>DoIt
[] in InputSensor>>installInterruptWatcher
[] in ProcessorScheduler class>>startUp
[] in SystemDictionary>>installLowSpaceWatcher
[] in ControlManager>>activeController:
)
The first to process come from the bad habit of releasing images with
processes
forked from a workspace or from changed code. The Delay process should
be:
[] in Delay class>>startTimerInterruptWatcher
This can be remedied by doing a:
Delay startTimerInterruptWatcher.
I'll figure out the WeakArray one later.
If anyone cares, I'll clean the code and send it on up.
-- Mike Klein
Squeak versus VisualAge Hmmm, tough call.... BlockClosures would tip
the scale for me.
On Wednesday, October 06, 1999 10:53 AM, David N. Smith (IBM)
[SMTP:dnsmith at watson.ibm.com] wrote:
> At 9:38 -0700 10/5/99, Xu Wang wrote:
> >I noticed Squeak dose not allow me to add extra instVars in Process. Is
> >there a special reason for it (e.g. hardcode in VM etc.)?
> >I need to add a process context which holds user defined process
> >dependent info.
>
> I was looking at adding one yesterday, but backed off before I tried
> it. In another Smalltalk that I use, processes can be named, and I
> find it useful when a lot of processes are running around.
>
> Dave
> _______________________________
> David N. Smith
> IBM T J Watson Research Center
> Hawthorne, NY
> _______________________________
> Any opinions or recommendations
> herein are those of the author
> and not of his employer.
>
More information about the Squeak-dev
mailing list
|