Struct UnicodeEncoder 

pub struct UnicodeEncoder { /* private fields */ }

Available on crate feature font-subset only.

Accumulates glyph usage for one page, then writes the font.

Encoding and embedding are deliberately one object: the character codes written into the content stream are post-subset glyph indices, so they are only meaningful next to the font program produced by the same remapper.

Implementations

impl UnicodeEncoder

pub fn new(font: UnicodeFont) -> UnicodeEncoder

Start accumulating for font.

pub fn encode(&mut self, text: &str) -> Result<Vec<u8>, ManipError>

Encode text as Identity-H character codes (2 bytes per glyph, big-endian), registering every glyph it uses.

Fails if the font has no glyph for a character, naming the characters rather than silently substituting .notdef — a row of blank boxes in a translated document is worse than a refusal the caller can act on.

pub fn width_em(&self, text: &str) -> f64

Advance width of text in em units (1.0 = the font size), using this font’s own metrics.

Used to decide whether a replacement fits where the original sat. The substitute font is never exactly as wide as the one it replaces, so a translated block needs measuring rather than assuming.

pub fn is_empty(&self) -> bool

Whether anything was encoded, i.e. whether a font needs writing.

pub fn embed( self, doc: &mut Document, page_num: u32, ) -> Result<String, ManipError>

Subset the program, write the Type0 font, and register it in the page’s resources. Returns the resource name to use in Tf.