About the project¶
Why this exists¶
Rustaveli.Pdf exists to give .NET a PDF tool that is:
- Free — MIT-licensed, for any project, with no revenue limits, keys or paid tiers.
- Open — every line of source public, to read, change and ship.
- Complete — from a one-page invoice to a PDF/A archive or an accessible, tagged document, so you never need a second library.
- Dependable — tested against independent validators, and released with signed provenance.
It is a design of its own:
- A vocabulary drawn from print and typesetting — trims, margins, running heads, folios, inks (glossary) — so the API reads like the documents it makes.
- Plain .NET at the core. The PDF writer, fonts, images and text engine are managed code with no native dependencies; native libraries come in only through the optional packages that need them.
- Layout kept apart from output. The layout engine draws through a single surface, so producing a PDF, a page image or anything else is a decision at the edge rather than one built into the core (how layout works).
History¶
The first release was built in ten phases, each merged only once it met every quality gate. The parity checklist tracks each capability against the phase that delivered it, and the architecture decision records explain the choices behind the plan. Every phase below is done.
| Phase | Delivered |
|---|---|
| 0 | Groundwork: quality gates, pipelines, conformance and property tests, benchmarks |
| 1 | The print and typesetting vocabulary (ADR 0002); Ink colour |
| 2 | A managed PDF writer with font subsetting and parallel rendering |
| 3 | The text engine: line breaking, justification, fallback, OpenType features, complex scripts |
| 4 | Layout and styling parity, and named style sheets |
| 5 | Images and SVG |
| 6 | Output formats and conformance: page images, CMYK and spot colour, PDF/A, PDF/UA |
| 7 | Document operations: merge, overlay, attachments, encryption |
| 8 | A live preview tool with hot reload |
| 9 | Documentation, and the first release |
Inspiration¶
The idea of describing a document as a fluent tree of content that breaks into pages by itself comes from QuestPDF. How the two libraries relate — that inspiration, what QuestPDF is used for in this repository, and the clean-room rewrite of this library's earliest code — is set out in Rustaveli.Pdf and QuestPDF.