Board index » delphi » Re: RAD Studio Roadmap Updated

2008-04-27 05:00:51 AM
delphi12
Mikko Laiho writes:
msdn2.microsoft.com/en-us/library/0yw3tz5k(VS.80).aspx
I guess it is a safe bet that Delphi's anonymous methods will have much
in common with C#'s anonymous methods.
--
Rudy Velthuis [TeamB] www.teamb.com
"Modern capitalism is not about free markets, it is about
building sufficient mass that the market gravitationally
collapses around you." -- Alisdair Meredith
 

Robert Giesecke writes:
Tiberion, which sounds like some ancient Greek name. <g>
--
Rudy Velthuis [TeamB] www.teamb.com
"The world is a tragedy to those who feel, but a comedy to those
who think." -- Horace Walpole (1717-1797)
 

Delphinian writes:
in handy for MVC, but this principle is old enough to have existed
before multicast delegates, or delegates at all, existed.
--
Rudy Velthuis [TeamB] www.teamb.com
"Real punks help little old ladies across the street because
it shocks more people than if they spit on the sidewalk."
-- Unknown
 

Rudy Velthuis [TeamB] writes:
- have existed for a good long time. At least since the 50s and Lisp,
and implicitly in the Church lambda calculus before that (i.e. 1930s).
When you have a way of creating closures, writing a "combine" method
which wraps two closure calls into a single forwarding closure becomes
trivial.
-- Barry
--
barrkel.blogspot.com/
 

On Sun, 27 Apr 2008 02:37:23 +0100, Barry Kelly (CodeGear) writes:
generic method to do this. I sure hope there would be a way.
(FWIW, I use multicast delegates all the time to reduce cohesion in my
code. I'd use them more often if they weren't so much work)
--
Marc Rohloff [TeamB]
marc -at- marc rohloff -dot- com
 

Eric Grange writes:
There's no reason why you can not have a 32 bit IDE launch a 64 bit
de{*word*81} as a separate process.
--
Mike Swaim
MD Anderson Division of Quantitative Sciences
XXXX@XXXXX.COM or XXXX@XXXXX.COM
 

Does Win32 implementation Allow Syntax Like that
Button1.OnClick := begin
ShowMessage(Inttostr(ExternalToMethodVar);
end;
What about the Lifetime of that closure object?
Who Controls IT?
Oxffff from Russia.
"Rudy Velthuis [TeamB]" <XXXX@XXXXX.COM>???????/???????? ?
???????? ?????????: news:XXXX@XXXXX.COM...

65536 writes:
solution for the lifetime problem. They found one for strings and
dynamic arrays too.
--
Rudy Velthuis [TeamB] www.teamb.com
"640K ought to be enough for anybody."
-- Bill Gates (1955-), in 1981
 

It is unlikely.
Suppose the next operation
(OnClick):=nil;
There is no way to finalize closure object because TMethod is not finalized
i.e. have none of finalized members.
Only if you use some tricks manually
var Method:TMethod;
begin
Method:=Tmethod(OnClick);
Tobject(Method.Data).free;
end;
But it is very unsafe.
There is not guarantee that Data member of TMethod is an object. And no
RTTI for DATA member.
Does Implementation consist of some magic finalized wrappers(Custom
Variants, Interfaces)?
BUT!!! In that case programmer have to maintain Life for Wrapper object
Until there is no Ref to Closure object, i.e. determinate it.
"Rudy Velthuis [TeamB]" <XXXX@XXXXX.COM>???????/???????? ?
???????? ?????????: news:XXXX@XXXXX.COM...

John Kaster (CodeGear) writes:
speed things along?
 

"Rudy Velthuis [TeamB]" <XXXX@XXXXX.COM>???????/???????? ?
???????? ?????????: news:XXXX@XXXXX.COM...
TClosureObject000=class
private
ExternalToMethodVar:DWORD
public
procedure Implementation(Sender: TObject);
begin
showmessage(inttostr(ExternalToMethodVar));
end;
end;
Or may be it looks a little efficient. Like
TClosureObject001=record
ExternalToMethodVar:DWORD
procedure Implementation(Sender: TObject);
begin
showmessage(inttostr(ExternalToMethodVar));
end;
end;
PClosureObject000=^TClosureObject001;
So the next sample is still not valid for Win32
Button1.OnClick := begin
ShowMessage(Inttostr(ExternalToMethodVar);
end;
May be Anonymous methods it is just the short form declaration of nested
procedures and functions?
 

Sergey Antonov aka oxffff writes:
under .NET.
Anyway, instead of a class they could use something reference counted
in Delphi for Win32, but otherwise use the same mechanism: local
variables are not local, they become part of the class instance (or
whatever it is) and are accessed from there. Right?
--
Rudy Velthuis [TeamB] www.teamb.com
"I want to die in my sleep like my grandfather...
not screaming and yelling like the passengers in his car..."
-- bumper sticker
 

Andy Syms writes:
--
Rudy Velthuis [TeamB] www.teamb.com
"Cholesterol is your natural defence against excessive
circulation of {*word*76}, which can carry venoms, poisons and other
toxins around your body." -- Michael Warner, in bpot
 

model. The VES is
responsible for loading and running programs written for the CLI. It
provides the services needed to execute
managed code and data using the metadata to connect separately generated
modules together at runtime. The
VES is also known as the Execution Engine.
Look for ECMA-335
Let's process the next code
Button1.onlick:=begin
showmessage(ExternalVar)
end;
that is translated to something like this:
1. Alloc instance (Closure object)
_Reference_Counted_Essence.Allocate; ->Getmem(InstanceSize+HiddenSystemFields);
_Reference_Counted_Essence.Initialize(ExternalVar); ->
self.ExternalVar:=ExternalVar;
2. And ImplicitConversion
_Reference_Counted_Essence._Addref ->
Something Like InterlockedIncrement(P.refcnt);
TMethod(A).Data:=_Reference_Counted_Essence;
TMethod(A).Code:=_Reference_Counted_Essence.Compiler_generated_method_for_
showmessage
3. Assign A to Button1.onlick property.
Right?
And now question for one million dollars.
How have Closure object been notived after next operation
Button1.onlick:=nil.