RE: coroutines & timeout

["wilfried.verachtert@..." <wilfried.verachtert@...>]

Hi,

Thanks for your input.

> I would think in general this is not such a hot idea for Lua. For
> instance, this will not work in the middle of a function called under
> pcall or a metatable. Thus there is no way to enforce it to begin with. 

I'm not quite sure what you mean with that. I did see some similar remarks
in some other threads (about threads :-) that seem to suggest the same
thing. I've done a few test using metatable functions and everything seems
to be working as expected. (I added my script and the output of it at the
end of this mail) So, I might be lucky or I'm missing something. Since I'm
pretty new at LUA, I would guess that the latter is more likely but I
wouldn't know.

Primitive functions written in C, those are not captured by the way I'm
forcing a 'yield' in the VM. But that's something I could live with. 

> On 
> the other hand it is hard to imagine the usefullness of this accross the
> board as it seems from my experience the scripts want to yield in a
> particular spot, rather than with a given time regularity in the middle of
> nowhere (imagine driving animation, simulation, state machine, a slide
> show, etc. for instance). All of them will need yilding occur in some
> orderly fashion in places dictated by logic rather than time. Using the
> debug hooks for yeilding have caused more problems for us than it solved
> as it had been highly unreliable from my experience (this may change in
> the future)... opcode hooks are especially prone to nasty side
> effects/bugs when yielded from (at least they were a year or so ago). Line
> hooks have also suffered from similar dificiencies but to a lesser degree.

That's a valid point, but there are some cases where you would rather have
some time-out control over child processes. I'm thinking of writing a
server that can serve multiple clients. (like a web-server, telnet-server,
...) I would like to have multiple processes running in parallel, but I
would like to control the 'fairness' of the processing time given to each
process. I don't want to enforce everybody to put 'yields' at specific
places in their code. I've worked with such a model in the past, and for
this kind of code it's a terrible thing to maintain. Having a little
scheduler that can rely on time-outs is much easier in this case.

In some other cases (like the ones you mention), I would rather use the
explicit yielding of the child processes.

In short, I would like to have both of them. 
 
> I use a number sleep functions in my scheduler and people use them
> where/how they see fit in their scripts. And we have non-programming staff
> using Lua on a daily basis with no complaines about this particular aspect
> of it.
> 
> Just my .02
> Alex

The script:

--------------------------------------------------------------
function fib(n)
  if n<2 then
    return n
  else
    return fib(n-1)+fib(n-2)
  end
end

object = {}

object.__index = function (a,b)
  co = coroutine.create(fib)
  s, v = coroutine.timedresume(co, 1000, 35)
  print(v)
  return object[b]
end

function object:classname ()
  return "<object>"
end

function object:displaystring ()
  return self:classname()
end

function object:display ()
  fib(32) -- just to make this function take some time to calculate
  print(self:displaystring())
end

person = {}
setmetatable(person, object)

function person:classname ()
  return "<person>"
end

print(">>")
person:display()
print("<<")

co_1 = coroutine.create(function () person:display() end)
co_2 = coroutine.create(function () object:display() end)

repeat
  s1 = coroutine.timedresume(co_1, 1000)
  print("-")
  s2 = coroutine.timedresume(co_2, 1000)
  print("+")
until (not s1) and (not s2)

print(coroutine.timedresume(co, 8000))
print(coroutine.timedresume(co, 8000))

--------------------------------------------------------------
... and the output ...
--------------------------------------------------------------
>>
nil
nil
<person>
<<
nil
-
+
-
+
-
+
-
<object>
+
nil
<person>
-
+
-
+
true
true    9227465
--------------------------------------------------------------



--------------------------------------------------------------------
mail2web - Check your email from the web at
http://mail2web.com/ .