Struct TableResult 

pub struct TableResult {
    pub bbox: BoundingBox,
    pub table_type: TableType,
    pub classification_confidence: Option<f32>,
    pub structure_confidence: Option<f32>,
    pub cells: Vec<TableCell>,
    pub html_structure: Option<String>,
    pub cell_texts: Option<Vec<Option<String>>>,
    pub structure_tokens: Option<Vec<String>>,
}

Result of table recognition.

Fields

bbox: BoundingBox

Bounding box of the table in the original image

table_type: TableType

Table type (wired or wireless)

classification_confidence: Option<f32>

Confidence score for table type classification (None if classifier wasn’t configured/run)

structure_confidence: Option<f32>

Confidence score for table structure recognition (None if structure recognition failed)

cells: Vec<TableCell>

Detected table cells

html_structure: Option<String>

HTML structure of the table (if available)

cell_texts: Option<Vec<Option<String>>>

OCR text content for each cell (if OCR was integrated)

structure_tokens: Option<Vec<String>>

Structure tokens from table structure recognition (used for HTML generation after stitching)

Implementations

impl TableResult

pub fn new(bbox: BoundingBox, table_type: TableType) -> Self

Creates a new table result.

pub fn with_classification_confidence(self, confidence: f32) -> Self

Sets the classification confidence.

pub fn with_structure_confidence(self, confidence: f32) -> Self

Sets the structure recognition confidence.

pub fn with_cells(self, cells: Vec<TableCell>) -> Self

Sets the table cells.

pub fn with_html_structure(self, html: impl Into<String>) -> Self

Sets the HTML structure.

pub fn with_cell_texts(self, texts: Vec<Option<String>>) -> Self

Sets the cell texts from OCR.

pub fn with_structure_tokens(self, tokens: Vec<String>) -> Self

Sets the structure tokens for later HTML generation.

pub fn confidence(&self) -> Option<f32>

Returns the best available confidence score for this table.

This method provides a unified confidence API for callers who want to filter tables by confidence without caring whether classification or structure recognition was used. Priority:

1. If both classification and structure confidence are available, returns the minimum (most conservative estimate)
2. If only structure confidence is available (common when classifier isn’t configured), returns that
3. If only classification confidence is available, returns that
4. Returns None only if neither confidence is available (stub result)

pub fn has_structure(&self) -> bool

Returns true if this table has valid structure data.

A table is considered valid if it has either cells or an HTML structure. Stub results (created when structure recognition fails) will return false.

Trait Implementations

impl Clone for TableResult

fn clone(&self) -> TableResult

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for TableResult

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for TableResult

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for TableResult

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

where __S: Serializer,

Serialize this value into the given Serde serializer.