Struct TextSpan 

pub struct TextSpan {

    pub text: String,
    pub x: f64,
    pub y: f64,
    pub width: f64,
    pub height: f64,
    pub font_size: f64,
    pub complete_run: bool,
    pub font_name: Option<String>,
    pub is_bold: bool,
    pub is_italic: bool,
    pub color: Option<[u8; 4]>,
    pub width_source: WidthSource,
    pub char_bounds: Vec<[f64; 4]>,
    pub transform: Option<[f64; 6]>,
    pub font_weight: Option<u16>,
    pub is_serif: Option<bool>,
    pub is_monospace: Option<bool>,
    pub render_mode: Option<u8>,
    pub is_ocr_layer: bool,
    pub font_metrics: Option<FontMetrics>,
    /* private fields */
}

A single text span at a specific position.

Fields

text: String

The extracted text.

x: f64

X position in user space.

y: f64

Y position in user space.

width: f64

Approximate bounding-box width in user space.

height: f64

Approximate bounding-box height in user space.

font_size: f64

Font size (approximate, from transform).

complete_run: bool

Whether a content run ends after this span: a consumer joining same-style spans must not merge past it (#643). Only set on the raw path; the default path leaves every span open, as before.

font_name: Option<String>

PostScript name of the font, with any 6-character subset prefix stripped (e.g. Helvetica-Bold, TimesNewRomanPS-BoldMT). None for Type1 standard-14 fonts and Type3 fonts where no embedded font data is available through the public pdf-interpret API.

is_bold: bool

Inferred bold style: weight >= 700 or PostScript name suggests bold (“bold”, “demi”, “semibold”, “heavy”, “black”). Defaults false when no descriptor data is reachable.

is_italic: bool

Inferred italic style: FontDescriptor /Italic flag set or PostScript name suggests italic/oblique/slant. Defaults false when no descriptor data is reachable.

color: Option<[u8; 4]>

Fill color as sRGB RGBA, derived from Paint::Color(c).to_rgba().to_rgba8() at the moment of glyph paint. None for Paint::Pattern (tiling / shading) — the editor falls back to “auto” in that case.

width_source: WidthSource

Whether glyph widths were measured from real font advance data or estimated.

char_bounds: Vec<[f64; 4]>

Per-glyph bounding boxes in user-space, one entry per source glyph. [x0, y0, x1, y1] with y0 < y1 (PDF coordinate frame).

transform: Option<[f64; 6]>

Full affine transform of the span’s first glyph in user space, [a, b, c, d, e, f] (kurbo coeffs of CTM × text-matrix). Captures rotation and shear that (x, y, font_size) discards. Text placed by a rectangle rather than drawn by a glyph (a form field’s value with no appearance stream, #549) carries the placement in the same form. None for spans built from neither (e.g. synthetic markers).

font_weight: Option<u16>

Numeric font weight (~100–900) from the embedded font’s OS/2 table when available. None for non-embedded / standard-14 fonts.

is_serif: Option<bool>

Whether the font is serif, when determinable from embedded font data. None when no embedded descriptor is reachable.

is_monospace: Option<bool>

Whether the font is monospace, when determinable from embedded font data. None when no embedded descriptor is reachable.

render_mode: Option<u8>

Coarse PDF text render mode: 0 fill, 1 stroke, 3 invisible — the only three values the renderer’s GlyphDrawMode expresses (the fill+stroke / clip modes 2 and 4–7 are not distinguished). Reflects the span’s first glyph, and 0 for text placed by a rectangle rather than drawn (#549). None for spans built from neither.

is_ocr_layer: bool

Whether the span was set in the engine’s own OCR layer resource (pdf_ocr::pipeline::OCR_FONT_NAME). The hidden-text drop keeps these spans while author-hidden text goes, because the layer is the engine’s own voice, not ink the author hid.

font_metrics: Option<FontMetrics>

Vertical font metrics (ascent/descent, optional cap/x-height) in /1000 em from the embedded font binary. None for non-embedded fonts.

Implementations

impl TextSpan

pub fn is_hidden(&self) -> bool

Whether this span is text the page hides: drawn with render mode 3 (invisible) or painted with zero alpha (lane L1).

The render mode comes from the span’s first glyph and the alpha from its fill paint; a span with no paint information counts as visible.

Trait Implementations

impl Clone for TextSpan

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for TextSpan

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

Formats the value using the given formatter.

impl Default for TextSpan

fn default() -> Self

Returns the “default value” for a type.

impl From<&TextSpan> for TextSpanInfo

fn from(s: &TextSpan) -> Self

Build the canonical wire form from a borrowed TextSpan (clones).

impl From<TextSpan> for TextSpanInfo

fn from(s: TextSpan) -> Self

Build the canonical wire form from an owned engine TextSpan.

Centralises the conversions that previously lived in each binding: the ½-em width fallback, the RGBA-u8 → RGB-f32 colour normalisation, and the height-reports-font-size contract.