In the Gist, I have explained my problem, and I have provided my tokenizer as well. If you can help me with the parser, that would be great. Also if my tokenizer is not very well-structured, please tell me.

btw the tokenizer is written in extremely portable Scheme. I think Chez is the best interpreter for Scheme.