Struct GraphEvalContext 

pub struct GraphEvalContext<'a> {
    pub graph: &'a CodeGraph,
    pub plugin_manager: &'a PluginManager,
    pub workspace_root: Option<&'a Path>,
    pub disable_parallel: bool,
    pub subquery_cache: HashMap<(usize, usize), Arc<HashSet<NodeId>>>,
}

Context for graph-native predicate evaluation.

Encapsulates all dependencies needed for evaluating predicates against a CodeGraph. The context includes precomputed caches for performance.

Fields

graph: &'a CodeGraph

Reference to the code graph

plugin_manager: &'a PluginManager

Plugin manager for detecting plugin fields

workspace_root: Option<&'a Path>

Workspace root for relative path resolution in path: predicates

disable_parallel: bool

If true, disable parallel execution

subquery_cache: HashMap<(usize, usize), Arc<HashSet<NodeId>>>

Precomputed subquery result sets, keyed by (span.start, span.end). Populated by precompute_subqueries() before the per-node evaluation loop.

Implementations

impl<'a> GraphEvalContext<'a>

pub fn new(graph: &'a CodeGraph, plugin_manager: &'a PluginManager) -> Self

Creates a new evaluation context.

pub fn with_workspace_root(self, root: &'a Path) -> Self

Sets the workspace root for relative path resolution.

pub fn with_parallel_disabled(self, disabled: bool) -> Self

Disables parallel execution.

pub fn precompute_subqueries(&mut self, expr: &Expr) -> Result<()>

Precomputes all subquery result sets from the expression tree.

This must be called before evaluate_all to avoid O(N^2) behavior where each subquery is re-evaluated for every candidate node.

Errors

Returns an error if subquery evaluation fails.