rpptx-render

rpptx-render lowers resolved presentation content into the shared positioned page model. It assembles source-scoped relationships and media, lays out one page per slide, and offers deterministic bundled-font output for stable rendering pipelines.

Capabilities

RenderInput assembly for slides, media, fonts, and metadata.
Source-scoped slide, layout, and master relationships.
Shapes, text, tables, images, groups, and backgrounds lowered to page frames.
Structured diagnostics and LayoutResult output for downstream backends.
Charts, SmartArt text, hyperlinks, speaker notes, handout composition, and animation snapshots share deterministic font and media resolution.

Measured footprint and speed

Measurement	Value	Version	Platform	Build mode	Input	Command	Statistic	Measured on
Crates.io archive: rpptx-render	64,970 compressed bytes, 351,711 member bytes, 8 members	0.13.1	macOS 26.6.2, Apple M5 Max, arm64	`cargo package --locked --no-verify`	Tracked `rpptx-render` package inventory	`python3 scripts/readme_doctests.py --record-measurements`	gzip archive bytes, tar member bytes, tar member count	2026-10-04

Use it when

Use this crate when integrating presentation layout below the rpptx facade. Applications normally call deterministic output methods on rpptx::Presentation.

Relationship

It consumes rpptx-layout output and produces oxml-layout page frames. It does not write PDF or raster files. Those formats belong to downstream output backends and the high-level facade.

Example

use rpptx_render::{RelScope, RelScopes};

let relationships = RelScopes::default();
let missing = relationships.get(RelScope::Slide, "rId1");
assert!(missing.is_err());