[Help-Smalltalk] GNU Smalltalk Sprint

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
23 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Re: GNU Smalltalk Sprint

Holger Freyther
On 10/17/2010 02:11 PM, Paolo Bonzini wrote:
> - a coverage tool (for method-coverage, it can be based on gst-profile)

This can be ported from Pharo's SUnit, or at least used as inspiration.

_______________________________________________
help-smalltalk mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/help-smalltalk
Reply | Threaded
Open this post in threaded view
|

Re: GNU Smalltalk Sprint

Stefan Schmiedl
In reply to this post by Paolo Bonzini-2
On Sat, 16 Oct 2010 13:08:34 +0200
Paolo Bonzini <[hidden email]> wrote:

> On 10/16/2010 12:18 PM, Nicolas Petton wrote:
> >> >  Allow VisualGST to hook up to a headless gst like gst-remote
> >> >  does. Just imagine how much fun working on a running Iliad app
> >> >  you could have with this.
> > +1 (not sure it's should be on the top of the list but it's very cool
> > anyway)
>
> True, it would really be cool.  You need some kind of RPC, basically, so
> that would be the "real" task.

Isn't this already implemented with gst-remote?
I may be naive, but, theoretically, there should be
a point where GUI actions are resolved into commands
that right now are being sent to the local VM.
Could they not just as well be sent to another VM as well?

Curious,
s.

_______________________________________________
help-smalltalk mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/help-smalltalk
Reply | Threaded
Open this post in threaded view
|

Re: GNU Smalltalk Sprint

Paolo Bonzini-2
On 10/17/2010 03:04 PM, Stefan Schmiedl wrote:

> On Sat, 16 Oct 2010 13:08:34 +0200
> Paolo Bonzini<[hidden email]>  wrote:
>
>> On 10/16/2010 12:18 PM, Nicolas Petton wrote:
>>>>>   Allow VisualGST to hook up to a headless gst like gst-remote
>>>>>   does. Just imagine how much fun working on a running Iliad app
>>>>>   you could have with this.
>>> +1 (not sure it's should be on the top of the list but it's very cool
>>> anyway)
>>
>> True, it would really be cool.  You need some kind of RPC, basically, so
>> that would be the "real" task.
>
> Isn't this already implemented with gst-remote?
> I may be naive, but, theoretically, there should be
> a point where GUI actions are resolved into commands
> that right now are being sent to the local VM.
> Could they not just as well be sent to another VM as well?

It is slow.  You want to pass object references and messages back and
forth, not messages.

If you can avoid subtleties like exception handling, it is basically on
the client side

doesNotUnderstand: aMessage
     oneObjectDumper dump: self; dump: aMessage.
     ^anotherObjectDumper loadProxy "this doesn't exist :)"

and on the server side

[anotherObjectDumper dumpProxy: "also doesn't exist :)"
     (oneObjectDumper load perform: oneObjectDumper load]

on the other side.

Paolo

_______________________________________________
help-smalltalk mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/help-smalltalk
12