I've been looking to embed Lisp in some C++ code. Two options I'm interested in is Bigloo Scheme and ECL (Common Lisp). Reading through the docs they seem to support a very similar feature set. Obviously Bigloo is Scheme and ECL is CLisp, but what other differences do they have?
In particular I'm interested in the following criteria:
- Ease of embedding (for C++, not just C). I don't want to write a bunch of boilerplate.
- Performance. Bigloo is performance based and has many compiler optimization options, although I can't find anything comparable for ECL.
- Style of coding. This one is more for Bigloo - is it more functional than ECL?
I'm targeting this question towards someone who has used both.
Ease of embedding: For C++ should be about the same. Bigloo has a better story for Java and C# embedding IIRC.
Performance: Really depends on what you are going to use it for. They both are "okay" performers as far as Lisps go. The ability to embed in C reduces the opportunities for a lot of optimizations.
Style of Coding: This is really the big difference between the two, and why a comment said it amounts to "should I learn Scheme or Common Lisp" Both are mature embeddable implementations of two very different standards (R5RS and ANSI Common Lisp).