Struct EvalConfig 

pub struct EvalConfig {

    pub enable_parallel: bool,
    pub max_threads: Option<usize>,
    pub max_vertices: Option<usize>,
    pub max_eval_time: Option<Duration>,
    pub max_memory_mb: Option<usize>,
    pub workbook_seed: u64,
    pub volatile_level: VolatileLevel,
    pub range_expansion_limit: usize,
    pub stripe_height: u32,
    pub stripe_width: u32,
    pub enable_block_stripes: bool,
    pub spill: SpillConfig,
    pub use_dynamic_topo: bool,
    pub pk_visit_budget: usize,
    pub pk_compaction_interval_ops: u64,
    pub max_layer_width: Option<usize>,
    pub pk_reject_cycle_edges: bool,
    pub sheet_index_mode: SheetIndexMode,
    pub warmup: WarmupConfig,
    pub arrow_storage_enabled: bool,
    pub arrow_fastpath_enabled: bool,
    pub delta_overlay_enabled: bool,
    pub write_formula_overlay_enabled: bool,
    pub date_system: DateSystem,
    pub defer_graph_building: bool,
}

Configuration for the evaluation engine

Fields

enable_parallel: bool

max_threads: Option<usize>

max_vertices: Option<usize>

max_eval_time: Option<Duration>

max_memory_mb: Option<usize>

workbook_seed: u64

Stable workbook seed used for deterministic RNG composition

volatile_level: VolatileLevel

Volatile granularity for RNG seeding and re-evaluation policy

range_expansion_limit: usize

Ranges with size <= this limit are expanded into individual Cell dependencies

stripe_height: u32

Height of stripe blocks for dense range indexing

stripe_width: u32

Width of stripe blocks for dense range indexing

enable_block_stripes: bool

Enable block stripes for dense ranges (vs row/column stripes only)

spill: SpillConfig

Spill behavior configuration (conflicts, bounds, buffering)

use_dynamic_topo: bool

Use dynamic topological ordering (Pearce-Kelly algorithm)

pk_visit_budget: usize

Maximum nodes to visit before falling back to full rebuild

pk_compaction_interval_ops: u64

Operations between periodic rank compaction

max_layer_width: Option<usize>

Maximum width for parallel evaluation layers

pk_reject_cycle_edges: bool

If true, reject edge insertions that would create a cycle (skip adding that dependency). If false, allow insertion and let scheduler handle cycles at evaluation time.

sheet_index_mode: SheetIndexMode

Sheet index build strategy for bulk loads

warmup: WarmupConfig

Warmup configuration for global pass planning (Phase 1)

arrow_storage_enabled: bool

Enable Arrow-backed storage reads (Phase A)

arrow_fastpath_enabled: bool

Enable Arrow fast paths in builtins (Phase B)

delta_overlay_enabled: bool

Enable delta overlay for Arrow sheets (Phase C)

write_formula_overlay_enabled: bool

Mirror formula scalar results into Arrow overlay for Arrow-backed reads This enables Arrow-only RangeView correctness without Hybrid fallback.

date_system: DateSystem

Workbook date system: Excel 1900 (default) or 1904.

defer_graph_building: bool

Defer dependency graph building: ingest values immediately but stage formulas for on-demand graph construction during evaluation.

Trait Implementations

impl Clone for EvalConfig

fn clone(&self) -> EvalConfig

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for EvalConfig

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for EvalConfig

fn default() -> Self

Returns the “default value” for a type.