Struct ParsingConfig 

pub struct ParsingConfig {

    pub document_type: DocumentType,
    pub section_and_hierarchy: SectionAndHierarchyConfig,
    pub spatial_clustering: SpatialClusteringConfig,
    pub section_patterns: Vec<String>,
    pub include_raw_tika: bool,
    pub pipeline: PipelineConfig,
    pub list_detection: ListDetectionConfig,
    pub size_enforcer: SizeEnforcerConfig,
    pub minimal_parse: bool,
    pub section_detection_v2: SectionDetectionV2Config,
    pub node_type_clustering: NodeTypeClusteringConfig,
    pub graph_sanity: GraphSanityConfig,
    pub dump_analytics: bool,
}

Fields

document_type: DocumentType

section_and_hierarchy: SectionAndHierarchyConfig

spatial_clustering: SpatialClusteringConfig

section_patterns: Vec<String>

include_raw_tika: bool

Include raw Tika XML/HTML output in graph metadata for debugging

pipeline: PipelineConfig

Pipeline configuration - defines which rules to run and in what order

list_detection: ListDetectionConfig

List detection configuration

size_enforcer: SizeEnforcerConfig

Size enforcement configuration

minimal_parse: bool

Minimal parse mode - bypasses all rule processing and returns only base conversion

section_detection_v2: SectionDetectionV2Config

Configuration for the V2 section detection rule (Block 03). Uses #[serde(default)] so existing YAML configs without this key still deserialize.

node_type_clustering: NodeTypeClusteringConfig

Configuration for the NodeTypeClustering rule (CR-29; Block 05b — renamed from ParagraphClustering once Section started flowing through the same rule). Uses #[serde(default)] so existing YAML configs without this key still deserialize. Also accepts the legacy paragraph_clustering: block via the migration shim on NodeTypeClusteringConfig deserialization (see below).

graph_sanity: GraphSanityConfig

Configuration for the graph sanity-check-and-correction pipe (CR-28). Runs post-graph-build; defaults are safe (enabled with all invariants in check + correct mode).

dump_analytics: bool

When true, the analytics pre-pass writes one JSON file per stat kind to {cache_dir}/stat/<stat_name>/<pdf_hash>.json after finalization. This is a sidecar for offline tooling — not a pipeline cache (output is not read back). Default true for development; flip off in production where the extra writes are unwanted.

Implementations

impl ParsingConfig

pub fn load_from_file(path: &str) -> Result<ParsingConfig, Error>

Load config from file path (functional approach)

pub fn load_with_fallback(path: Option<&str>) -> ParsingConfig

Load config with fallback to default

Trait Implementations

impl Clone for ParsingConfig

fn clone(&self) -> ParsingConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ParsingConfig

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

Formats the value using the given formatter.

impl Default for ParsingConfig

fn default() -> ParsingConfig

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for ParsingConfig

fn deserialize<__D>( __deserializer: __D, ) -> Result<ParsingConfig, <__D as Deserializer<'de>>::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for ParsingConfig

fn serialize<__S>( &self, __serializer: __S, ) -> Result<<__S as Serializer>::Ok, <__S as Serializer>::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.