Docist
18.1k lines of code · mid-size island
Self-hosted document toolkit — merge PDFs with automatic conversion of ~19 input formats, convert files between ~150 format pairs, page tools, OCR, watermarking and AES-256 password protection, in six browser tools plus a REST API.
Live demo
The demo at docist-production.up.railway.app sits behind a shared password — docist-demo-4c9fd6 — to keep the converter from becoming a free public utility. Log in and try it with any files you like.
Overview
Docist (formerly “PDF Merger”) is a self-hosted document toolkit that grew from one merge page into six browser tools and a versioned REST API: Merge & Convert, Convert Files, Page Tools, Print Prep, Export, and Watermark & Security. Non-PDF inputs — Markdown, Word documents, spreadsheets, HTML, SVG, and a dozen image formats — are converted automatically on the way in.
Plugin architecture
Converters (anything → PDF) and transforms (anything → anything) are auto-discovered modules — adding a format means dropping in one file, no registry to edit. When no direct transform exists, the registry pivots through PDF automatically, so DOCX → PNG works without anyone writing it.
Highlights
- Page thumbnails throughout — every PDF in the merge list shows its first page, and Page Tools renders a clickable grid of the whole document that fills the page-range box for you.
- Duplex-friendly merging — automatic page numbering, bookmarks per source file, and blank-page insertion so every document starts on the front of a sheet.
- OCR — make scanned PDFs searchable via OCRmyPDF/Tesseract, or extract text straight from images.
- Hardened for deployment — optional login gate, per-IP rate limiting, magic-byte content sniffing on uploads, traversal-proof downloads, and per-request temp directories.
- 767 tests — 559 backend (pytest) and 208 frontend (
node:test), covering every converter, transform, route and hardening rule.