// the take
A single-file C library (plus header) for generating PDFs from scratch, with no external dependencies. It's aimed at embedding basic PDF output into C/C++ programs — think firmware, CLI tools, or print systems — rather than full-blown document layout work.
Genuinely dependency-free and embeddable — one .c/.h pair you can drop into any build. Feature set is wide for something this small: TrueType text, bezier/polygon primitives, several barcode formats (Code-128, EAN, QR), and JPEG/PNG/BMP image embedding. It's also unusually serious about hardening for its size — AFL-style fuzz harnesses for font, image, and header parsing, plus CI running Coverity scan and Coveralls coverage, which matters a lot for a C library parsing untrusted binary formats (TTF, JPEG, PNG).
There's no text layout engine — no word wrap, no flow, you position every text run by hand, so anything beyond a simple label/report requires you to build layout logic yourself. Encryption support is RC4 40-bit, which is cryptographically broken by modern standards and only useful for basic obscurity, not real document protection. PNG support explicitly drops alpha channels, so transparent images get flattened. It's a single-maintainer project under the Unlicense (public domain, no warranty), and as a C library doing manual memory management over several binary parsers, its safety is only as good as the fuzz coverage — solid for a hobby-scale project, but worth pressure-testing before trusting it with user-supplied images or fonts in production.
We read the README, the directory tree and a key source file in October 2026. Nobody pays to be listed. How we judge →