acorde-render-svg

Pure Rust/WASM SVG rendering driven by acorde-core and acorde-layout. The renderer is deterministic, has no browser or DOM dependency, and never loads fonts. It exposes full-score, layout-aware, per-row, metadata, preflight, and annotation rendering APIs.

Notation is emitted as font-independent geometry. Text, lyrics, chord symbols, and host annotation marks remain SVG text, so callers own font choice, font metrics, embedding, and publication typography. The renderer validates non-finite geometry, invalid XML characters, unsupported glyph requests, malformed TAB positions, and bounded annotation/text inputs before emitting SVG.

use acorde_core::Score;
use acorde_render_svg::{render_svg, SvgRenderOptions};

let svg = render_svg(&Score::default(), &SvgRenderOptions::default())?;
# let _ = svg;
# Ok::<(), acorde_render_svg::RenderError>(())

Host integration

Interactive output includes stable note-address hooks. render_svg_metadata supplies the matching typed note, text, TAB, and technique data; call svg_contract_version() before consuming fields added by later releases. render_preflight() reports source-located capability issues before a host mounts SVG. Rows can be rendered independently for virtualized views.

RenderAnnotation permits bounded, deterministic host text marks. Hosts supply their own viewport geometry and can opt into the renderer's priority-based collision pass; the renderer does not infer text metrics. Missing authored TAB positions render as an explicit marker rather than a guessed string/fret.

Ties never cross a rest, microtones retain exact cents as explicit markers, and unpitched notes retain display placement without inventing a percussion sound identity. These are bounded semantic projections, not claims of font-aware publication engraving.

See the browser contract for the WASM boundary and the security contract for SVG and annotation input rules.

API documentation · Repository