use truncate() instead of pop() for unused and shadowed variables

[ehwan]

No description provided.

[Copilot]

Pull Request Overview

This PR optimizes the parser code generation by replacing individual pop() calls with more efficient truncate() operations for unused and shadowed variables on parser stacks. This reduces unnecessary work when variables are not referenced in the reduce action code.

Key changes:

• Implements analysis to detect unused location variables in reduce actions
• Replaces multiple consecutive pop() operations with single truncate() calls
• Handles variable shadowing by nullifying earlier declarations of the same variable name

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

File	Description
rusty_lr_parser/src/parser/parser_expanded.rs	Generated parser code updated to use truncate() instead of pop() for unused variables
rusty_lr_parser/src/emit.rs	Code generation logic enhanced to analyze variable usage and emit optimized stack operations
SYNTAX.md	Documentation updated to clarify variable access patterns in reduce actions

[Copilot]

The StackName enum definition is placed inside the loop, which means it gets redefined for every iteration. This should be moved outside the loop to avoid redundant type definitions.

Suggested change

	for (token_index_from_end, token) in rule.tokens.iter().rev().enumerate() {
	let stack_name = token_to_stack_name(token.token);
	let tag_name = stack_name.unwrap_or(&empty_tag_name);
	#[derive(PartialEq, Eq, PartialOrd, Ord)]
	enum StackName {
	DataStack(Ident),
	LocationStack,
	}
	impl StackName {
	pub fn to_token_stream(&self) -> TokenStream {
	match self {
	StackName::DataStack(name) => quote! {__data_stack.#name},
	StackName::LocationStack => quote! {__location_stack},
	}
	}
	}
	if let Some(stack_name) = stack_name {
	#[derive(PartialEq, Eq, PartialOrd, Ord)]
	enum StackName {
	DataStack(Ident),
	LocationStack,
	}
	impl StackName {
	pub fn to_token_stream(&self) -> TokenStream {
	match self {
	StackName::DataStack(name) => quote! {__data_stack.#name},
	StackName::LocationStack => quote! {__location_stack},
	}
	}
	}
	for (token_index_from_end, token) in rule.tokens.iter().rev().enumerate() {
	let stack_name = token_to_stack_name(token.token);
	let tag_name = stack_name.unwrap_or(&empty_tag_name);

[Copilot]

The tokenstream_contains_ident function is defined inside the loop, causing it to be redefined on every iteration. This helper function should be moved outside the loop or to module level for better performance and code organization.

Suggested change

	fn tokenstream_contains_ident(
	stream: TokenStream,
	ident: &Ident,
	) -> bool {
	for t in stream {
	match t {
	proc_macro2::TokenTree::Ident(i) if &i == ident => {
	return true
	}
	proc_macro2::TokenTree::Group(g) => {
	if tokenstream_contains_ident(g.stream(), ident) {
	return true;
	}
	}
	_ => {}
	}
	}
	false
	}