Struct ChunkMetadata 

pub struct ChunkMetadata {
    pub chapter: Option<String>,
    pub section: Option<String>,
    pub subsection: Option<String>,
    pub topic: Option<String>,
    pub keywords: Vec<String>,
    pub summary: Option<String>,
    pub structural_level: Option<u8>,
    pub position_in_document: Option<f32>,
    pub heading_path: Vec<String>,
    pub confidence: Option<f32>,
    pub custom: HashMap<String, String>,
}

Metadata associated with a text chunk, providing semantic context

This structure enriches chunks with information about their position in the document hierarchy, extracted keywords, automatically generated summaries, and other contextual information useful for retrieval and understanding.

Fields

chapter: Option<String>

Chapter name/title this chunk belongs to

Examples: “Chapter 1: Introduction”, “Part I”

section: Option<String>

Section name/title within the chapter

Examples: “1.1 Background”, “Introduction”

subsection: Option<String>

Subsection name/title within the section

Examples: “1.1.1 Historical Context”, “Early Development”

topic: Option<String>

Automatically detected or manually assigned topic

Examples: “Machine Learning”, “Neural Networks”, “Data Processing”

keywords: Vec<String>

Extracted keywords from the chunk content (TF-IDF or similar)

Ordered by relevance/importance, typically 3-10 keywords

summary: Option<String>

Automatically generated summary of the chunk content

Typically 1-3 sentences capturing the main points

structural_level: Option<u8>

Hierarchical level in document structure (0 = root/chapter, 1 = section, 2 = subsection, etc.)

Used to understand the depth of this chunk in the document hierarchy

position_in_document: Option<f32>

Relative position in the document (0.0 to 1.0)

0.0 = beginning, 0.5 = middle, 1.0 = end Useful for positional weighting in retrieval

heading_path: Vec<String>

Full heading path from document root to this chunk

Example: [“Chapter 1”, “Section 1.1”, “Subsection 1.1.1”] Provides complete context of the chunk’s location in document hierarchy

confidence: Option<f32>

Confidence score for metadata extraction (0.0 to 1.0)

Indicates how confident the system is about the assigned metadata

custom: HashMap<String, String>

Custom metadata key-value pairs

Allows for extensibility with domain-specific metadata

Implementations

impl ChunkMetadata

pub fn new() -> Self

Create a new empty metadata instance

pub fn with_chapter(self, chapter: String) -> Self

Create metadata with chapter information

pub fn with_section(self, section: String) -> Self

Create metadata with section information

pub fn with_subsection(self, subsection: String) -> Self

Create metadata with subsection information

pub fn with_keywords(self, keywords: Vec<String>) -> Self

Create metadata with keywords

pub fn with_summary(self, summary: String) -> Self

Create metadata with summary

pub fn with_structural_level(self, level: u8) -> Self

Create metadata with structural level

pub fn with_position(self, position: f32) -> Self

Create metadata with position in document

pub fn with_heading_path(self, path: Vec<String>) -> Self

Create metadata with heading path

pub fn add_custom(self, key: String, value: String) -> Self

Add a custom metadata field

pub fn has_structure_info(&self) -> bool

Check if metadata has any structural information (chapter, section, or subsection)

pub fn has_semantic_info(&self) -> bool

Check if metadata has semantic enrichment (keywords or summary)

pub fn get_deepest_heading(&self) -> Option<&String>

Get the deepest level heading (subsection > section > chapter)

pub fn get_hierarchy_string(&self) -> Option<String>

Get full hierarchical context as a formatted string

Example: “Chapter 1 > Section 1.1 > Subsection 1.1.1”

pub fn completeness_score(&self) -> f32

Calculate completeness score (0.0 to 1.0) based on populated fields

Higher scores indicate more complete metadata

Trait Implementations

impl Clone for ChunkMetadata

fn clone(&self) -> ChunkMetadata

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ChunkMetadata

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

Formats the value using the given formatter.

impl Default for ChunkMetadata

fn default() -> ChunkMetadata

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for ChunkMetadata

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for ChunkMetadata

fn eq(&self, other: &ChunkMetadata) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for ChunkMetadata

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for ChunkMetadata