Sign Up to our social questions and Answers Engine to ask questions, answer people’s questions, and connect with other people.
Login to our social questions & Answers Engine to ask questions answer people’s questions & connect with other people.
Lost your password? Please enter your email address. You will receive a link and will create a new password via email.
Please briefly explain why you feel this question should be reported.
Please briefly explain why you feel this answer should be reported.
Please briefly explain why you feel this user should be reported.
I’m creating a DSL in Clojure and have a choice between either:
a) Representing the DSL in symbolic form that can be converted to an AST later:
'(foo (bar (baz 1) (boo 3)))
b) Representing the DSL as pure functions that generate AST nodes:
(foo (bar (baz 1) (boo 3)))
=> [AST with foo at top level]
The AST will subsequently be compiled.
Is there any strong reason to prefer one approach vs. the other?
It looks like the approach a) is more flexible in the sense that the parser will parse the s-expressions and basically can move back and forward in the expression tree to generated the required code for ex: while parsing child of foo the parser can backtrack to foo or may be to foo’s parent to fetch some other required info etc. Whereas in the b) approach the DSL is normal function calls so boo call doesn’t know about its parent and so on and hence you cannot apply backtracking in that case.
In terms of complexity – the a) approach although is flexible can be a bit complex specially if backtracking is implemented whereas approach b) should be easy to implement,