|
Top-level
@dziban A lot of this is correct, and a lot of it is simpler than you've specified. Variable assignment and expression evaluation isn't a built-in feature of the language, but instead considered a part of the standard library. Rule priority is source-order, and rules consume the tuples they match. The left-hand side of a rule is the tuples it consumes, while the right-hand side is the tuples it produces. I've updated the snippet with more code. https://gitlab.com/-/snippets/3716812 ⦶ 4 comments
@wryl so addition is specified in terms of tuple replacement using something like peano numbers? @dziban It is, indeed, just some rules. "Compilation" is concatenation of rule-sets. The current implementation of Nova allows for an "escape hatch" in the form of patterns that are prefixed with a '%'. For the current implementation, rules are parsed and compiled to a form of byte-code that incorporates a stack. So.. %$variable = 1 2 +. ..results in a tuple that looks like.. $variable = 3 The new implementation will not feature this, as this was done for demonstration purposes. |
Gregory's Server
@wryl how does this standard library work? is it literally some rules provided that operate on the same tuple system? how does addition work under this?