This is the API that must be implemented by any interpreter that wants to have its run loop integrated with the SMOP Run Loop.

• Running operations
• has_next($interpreter: ) -- called before next to see if there is something to be evaluated.
• next($interpreter: ) -- called before eval to prepare the next node.
• eval($interpreter: ) -- evaluate the current instruction.
• setr($interpreter: $value) -- sets the result of the current instruction to the given value, usefull to pass arguments and return values.
• Introspection operators
• backtrace($interpreter: ) -- returns the backtrace of the current continuation (the continuation that precedes this one, being it a subroutine call or not). This must not be confused with stacktrace. The semantics might be different. There's no such thing as a native stacktrace in smop, if a language is stack-oriented, the backtrace must be interpreted to provide a sane stacktrace.
• debug($interpreter: ) -- returns debug information about the current execution.
• lexical($interpreter: ) -- returns lexical information about the current execution.
• jail($interpreter: ) -- returns the jail information that any exception-like operator would check to loop for a safe point.

The Debug API

This is completely vague now. The only thing that are presumably interesting to be available are:

• file name
• line number
• character position range (if available)
• source code

The Lexical API

This is also completely vague now.

• lookup($lexical: $name) -- returns a variable from the lexical information