Arc Forumnew | comments | leaders | submitlogin
2 points by lojic 6139 days ago | link | parent

I wonder how ccc in Arc/Scheme compares to return in Ruby with respect to efficiency. Hopefully ccc is very lightweight. It really does make me question why return wasn't defined to be part of the language.


8 points by almkglor 6139 days ago | link

Plenty efficient. Try reading the original lambda papers: http://library.readscheme.org/page1.html

The one you need to read is "Lambda: the Ultimate GOTO" http://repository.readscheme.org/ftp/papers/ai-lab-pubs/AIM-...

Basically the most important realization is that the return address on the stack may be modeled as the address for a different procedure, which by a staggering coincidence just so happens to point to the code after the function call. So basically a return is itself equivalent to a function call.

-----

7 points by sacado 6139 days ago | link

ccc is one of the strangest things in Scheme (and Arc now). Hard to get everything behind it, but it can be used to implement return statements, try-catch à la C++/Java, coroutines à la Lua, generators à la Python and of course continuation-based web apps à la Arc...

A very strange beast, and as far as I know the thing Scheme has that CL hasn't (and cannot trivially implement).

-----

5 points by KirinDave 6139 days ago | link

It should be no surprise continuations could do all these things. Continuations are the fundamental unit of all control flow in programming.

-----

4 points by sjs 6139 days ago | link

Arc doesn't use ccc for web apps, just good old closures.

-----

7 points by almkglor 6139 days ago | link

Actually, the form of closure used by Arc is highly similar to a continuation passing style, so although it doesn't use 'ccc, it does use continuations.

-----

2 points by eds 6139 days ago | link

cl-cont is a library that implements closures and runs on many CL implementations. So it's not a matter of CL can't do it, CL just has to use a library to do it.

-----

4 points by jbert 6139 days ago | link

I think there are conceptual problems mixing unwind-protect-like constructs and call/cc.

The cl-cont lib appears to only support a subset of CL (in particular unwind-protect is not supported).

So cl-cont doesn't demonstrate that CL can support it, only that a restricted subset of CL can.

I don't know if it is possible to sensibly accommodate unwind-protect and call/cc in the same language, it seems smarter people than I have avoided doing so.

-----

4 points by randallsquared 6138 days ago | link

Scheme's dynamic-wind serves the same purpose as unwind-protect (except with entry conditions as well, since a continuation could suddenly return into your protected code), and plays nicely with continuations. I'm not sure about the details, though.

-----

1 point by jbert 6138 days ago | link

Thanks for that, pretty interesting. I guess that makes writing entry and exit conditions a little more interesting, since they could end up being invoked multiple times.

-----

3 points by lojic 6139 days ago | link

That's good to here. Thanks for the links - great reference material.

-----

1 point by treef 6138 days ago | link

i am learning how to write scheme compilers now it it looks like you just jump around your CCC's like GOTO labels - very fascinating and lower over head then functions calls (if you compile to machine level)

-----