RE: [boost] SUBMISSION: Signal/slot library

From: Borgerding, Mark A. (MarkAB_at_[hidden])
Date: 2000-04-04 08:52:46

Sorry, I don't have many comments on the code itself, mainly on the concepts
and presentation.

Let me start off by showing my ignorance.  I am not really familiar with the
concept of signal and slot.

I found the Qt documentation about it and it sheds a little more light.
http://doc.trolltech.com/signalsandslots.html

It would help clarify your explanation if you noted the strengths and
inadequacies of the Qt approach (like the fact that it seems to need a
special compiler is a *major* drawback).  And then comment on how your
approach builds on the strengths and addresses the inadequacies.

I have a hard time thinking up some good examples.  Some code and narrative
explanation might help to clarify.  

It may just be my unfamiliarity with signals and slots, but I question
whether something that is based on callbacks is the right idiom to promote
for C++.  

By the way, I think that (Qt's) choice of the term "signal" was a poor one.

- Mark

> -----Original Message-----
> From: tomasz_at_[hidden] [mailto:tomasz_at_[hidden]]On
> Behalf Of Tomasz Kowalczyk
> Sent: Tuesday, April 04, 2000 8:09 AM
> To: boost_at_[hidden]
> Subject: [boost] SUBMISSION: Signal/slot library 
> 
> 
> Dear Boost Members,
> 
> I have written a library which deals with so-called signal/slot
> connections. I would like to ask you for review and comments. Since
> boost libraries contain lots of beautiful work, I thought it would be
> the best place to have my own code criticized.
> 
> The source can be found in:
> http://www2.gol.com/users/tomek/signals.tar.gz
> 
> and a short introduction:
> http://www2.gol.com/users/tomek/signals.html
> 
> Although the library is far from complete, I decided to post 
> it because
> I believe the techniques employed are very powerful and extremely
> useful, especially in event-driven environments (such as most GUI
> toolkits). I hope to find here new ideas, advice, and maybe gain some
> interest in the further development.
> 
> The library you can find in the above URL was a side product of an
> OpenGL-based viewer for 3D geometry. That's why the 
> abstractions may not
> be as general as they could be (this is especially true for slots and
> and connection predicates).
> 
> 
> Please let me explain in a few words what the whole signal idea is
> about, as I can see it now.
> 
> Say you have an class template called signal. For example:
> 
> signal<A,B> sig;  // equivalent to a function  void sig(A,B)
> 
> where A and B are some types. You may store in this signal some
> lambda-expressions for further use:
> 
> sig.connect( <lambda-expression-1> );
> sig.connect( <lambda-expression-2> );
> ...
> 
> Some time later in a different place you can execute all previously
> stored code with your choice of arguments:
> 
> sig( a, b );
> 
> Note the difference when compared to usage of lambda expressions in
> algorithms, such as std::for_each, where the expressions are composed
> and executed immediately. Signals are capable to delay the execution
> until requested.
> 
> GUI toolkits use the mentioned mechanisms to manage notification
> callbacks. However, it is very likely that there exist other potential
> applications than just callbacks (I am hoping to find some 
> ideas in this
> list :)
> 
> 
> The library provides also another kind of signal, which I call a
> delegate. It can store only one expression, but in turn can return
> values (returned from that expression). I find it convenient 
> to think of
> delegates as virtual functions, which implementation can be 
> bound at run
> time. For example:
> 
> delegate<void,char> put;   // equivalent to  void put(char)
> 
> put.connect( closure( &cout, &cout.put ) );
> put('a');   // a printed to cout
> 
> put.connect( function(fputc) );
> put('a');   // a printed to stdout
> 
> Delegates can be used as a way to achieve non-intrusive polymorphism,
> because implementation can be provided by completely unrelated code.
> 
> 
> As the next step, I am currently thinking about adaptation of the
> signals to be able to use boost's lambda-calculus library, instead of
> the limited choice of slot classes.
> 
> I hope you will like it :)
> 
> Thank you
> Tomasz
> 
> --------------------------------------------------------------
> ----------
> Save up to 54% on Quest & Kelty tents, backpacks, sleeping bags and
> outdoor gear.  FREE Shipping and a 30 Day Money-Back Guarantee at
> screaminghotdeals.com
> http://click.egroups.com/1/2716/2/_/9351/_/954849615/
> --------------------------------------------------------------
> ----------
> 
>