Good afternoon Jafar,
The major difference between your test and mine is that I am using an
expression (containing all the required code) whereas you have used a
procedure call. I will try to see if this makes any difference. What
I;ll do is convert the individual expression sequences to a procedure
call instead and see if this makes any difference with what I get back
when running my code.
thanks for the testing you have undertaken at this time.
regards
Bruce Rennie
On 22/02/13 07:16, Jafar Al-Gharaibeh wrote:
Bruce,
I ran a quick test to see if there is anything wrong with
&source/¤t depending in where the co-expression gets created or
activated. Nothing is suspicious at my end. Both &source and ¤t
give the correct values.
This is just a sanity test with a simple scenario. It could be the
case that a more complex use case scenario like the one you have
breaks something.
My test and its output are attached.
Cheers,
Jafar
On Thu, Feb 21, 2013 at 7:03 AM, Bruce & Breeanna Rennie
<[email protected] <mailto:[email protected]>> wrote:
Good evening Clint,
Yes it is somewhat complicated. Its purpose is to test the idea of
generating lambda functions in code using some of the ideas found in
"The Generator". It is also implementing a static environment for the
lambda functions using a table of tables to manage this. I am
currently
working on a side project that requires these facilities and I want to
implement it in Unicon. I am seeing how easy it is to implement the
Kernel Programming Language by John Shutt in Unicon.
What I found I had to do in code was to specify the calling
co-expression as being &main when the lambda was activated. &source is
returning the initial creation/activating co-expression which has
passed
the new co-expression back to &main for further activations. I have
tried looking through the source code for the runtime but at this
point
I have not be able to make much sense of where &source is being set on
activation.
I thought it best to pass the problem back to you and in the meantime
use my workaround for my testing.
Just a further question for you and your team. I have been thinking
about what it would require to get an IDE up that would
incorporate the
monitoring/debugging tools as they exist at this point as well as
being
able to access the contents of all packages that exist in the main
Unicon system as well as any others that the programmer may
develop. My
question: Has there been any thought about this at all from your
perspective? It looks like a major task but might give Unicon a
bit more
traction in the main programming community. What are your thoughts
about
this matter?
regards
Bruce Rennie
On 21/02/13 20:22, Clinton Jeffery wrote:
> Hi Bruce,
>
> I am sorry that I was not more timely in answering. Your example
> program is kind of complicated to follow. I may not be able to
trace
> it down until I get a break from my current teaching load. It
might in
> fact inspire me to improve the ability to monitor and/or debug
> co-expressions, which are currently under-supported by the
monitoring
> facilities.
>
> Each co-expression has its own &source and it is set whenever that
> co-expression is activated. &source is not local in the sense of
> changing when a procedure call or return occurs, but of course, your
> example is doing some weird stuff. The language has both explicit
> activation via @ and implicit activation via producing a result.
>
>
------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_feb
_______________________________________________
Unicon-group mailing list
[email protected]
<mailto:[email protected]>
https://lists.sourceforge.net/lists/listinfo/unicon-group
------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_feb
_______________________________________________
Unicon-group mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/unicon-group
------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_feb
_______________________________________________
Unicon-group mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/unicon-group
