Cooper.NativeInterpGrammar (Cooper v0.1.0)

casc_interp.aether, compiled the same way Cooper.NativeGrammar is (see its own moduledoc for the full ahead-of-time-generation rationale). This sub-grammar is invoked dynamically -- once per double-quoted string literal encountered during parsing, not once overall -- but that's an argument for native codegen here, not against it: under Grammar.VM (Cooper.Test.VMParity.run_interp_vm/1, test/support only), every one of those calls re-parses casc_interp.aether's own source text from scratch; here it's compiled exactly once, ahead of Cooper's own build, regardless of how many string literals a given file has.

Unlike Cooper.NativeGrammar, run/2's discarded final context is never needed here -- Cooper.InterpActions never mutates context (interpolated string content only ever builds Cooper.Ref.* nodes) -- so Cooper.InterpGrammar.run/1 calls straight through to this module's own run/2, no Ichor.Actions.evaluate/5-plus-capture-shapes workaround needed the way Cooper.NativeGrammar's does.

Regenerate whenever priv/grammar/casc_interp.aether changes:

mix ichor.gen priv/grammar/casc_interp.aether \
  --module Cooper.NativeInterpGrammar.Native \
  --actions Cooper.InterpActions \
  --out lib/cooper/native_interp_grammar/native.ex

Summary

Functions

parse(input, context \\ nil)

Matches input against the grammar's root rule with no Ichor.Actions involved -- a bare recognizer.

run(input, initial_context \\ nil)

Matches and evaluates input through Cooper.InterpActions -- what Cooper.InterpGrammar.run/1 calls.

run_sequence(input, initial_context)

Like run/2, but evaluates input as a sequence of top-level matches, threading context from each into the next.

tokenize(input, context \\ nil)

Tokenizes input. See Native.tokenize/2's own doc for context.