Struct ClassificationEngineConfig 

pub struct ClassificationEngineConfig {
    pub use_llm: bool,
    pub llm_model: String,
    pub llm_provider: String,
    pub openrouter_api_key: Option<String>,
    pub confidence_threshold: f64,
    pub weighted_sum: WeightedSumConfig,
}

Runtime configuration for the ClassificationEngine.

Why: classification tiers (especially LLM) need provider / model / threshold tuning; bundling those knobs into a config struct keeps the engine constructor signature stable as new tiers are added. What: holds LLM toggles (use_llm, llm_model, llm_provider, openrouter_api_key) plus a confidence_threshold shared by all tiers, and the weighted_sum config for Tier 2.5 (added in 1.3.0). Test: every classifier test in classify::tests builds one via Default::default() or with explicit overrides.

Fields

use_llm: bool

Whether to engage the LLM tier when tiers 1–3 fail.

llm_model: String

LLM model identifier (provider-specific).

llm_provider: String

LLM provider: "openrouter", "openai", or "auto".

openrouter_api_key: Option<String>

Optional OpenRouter API key. If None, the env var OPENROUTER_API_KEY is consulted at engine-build time.

confidence_threshold: f64

Minimum confidence required to accept a verdict.

Verdicts below this threshold are returned as-is (so the caller can still inspect them), but their confidence informs filtering in downstream reports.

weighted_sum: WeightedSumConfig

Configuration for the weighted-sum tier (Tier 2.5). Added in 1.3.0.

Why: the pipeline builds the engine from ClassificationEngineConfig; threading the weighted-sum config here keeps all per-run tier knobs in one place without changing the engine constructor’s arity. What: forwards to WeightedSumClassifier::new at engine-build time. Test: covered indirectly by classify_sync tests that exercise Tier 2.5.

Trait Implementations

impl Clone for ClassificationEngineConfig

fn clone(&self) -> ClassificationEngineConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ClassificationEngineConfig

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

Formats the value using the given formatter.

impl Default for ClassificationEngineConfig

fn default() -> Self

Returns the “default value” for a type.