pdfce
Open-source PDF editor for Windows. Pre-1.0, actively developed.
- Commits
- 296
- Stars
- 0
- Created
- August 9, 2026
A native PDF editor for Windows written in Rust. No web server, no browser runtime, no network listener—it runs as a single-folder desktop application. The goal is feature parity with Adobe Acrobat Pro; it is not there yet (pre-1.0), and the docs plainly say what works and what does not.
Working today: opening and rendering PDFs, page operations (merge, split, extract, reorder, rotate), text extraction and editing with reflow, AcroForm field creation and flattening, FDF/XFDF import/export, markup annotations, redaction, Bates numbering, PDF/A validation and conversion, digital-signature inspection, vector object editing, measurement authoring, image placement from PNG/JPEG/BMP/TIFF.
Not built yet: OCR, JavaScript, XFA, encrypted documents, and a long tail of Acrobat’s surface area. Some capabilities exist in the core library and CLI but have no GUI yet.
It ships with both a GUI and pdfce-cli, a first-class command-line interface with 60 subcommands—deliberately not a debug tool, but a scriptable peer to the graphical editor. Everything is done locally; no upload, no cloud processing.
Architecture
Four crates, load-bearing separation: pdfce-core (PDF engine, zero GUI dependencies), pdfce-render (headless rasterizer), pdfce-cli (scriptable shell), pdfce-gui (desktop application via egui/eframe). The GUI and CLI are two independent front ends over one engine. This split keeps a WebAssembly build a shell-crate swap instead of a rewrite, and keeps the core testable in isolation.
Local processing only
No HTTP client, no TLS stack, no telemetry, no analytics, no crash reporting, no update check. The docs list every dependency—all permissive open-source, no copyleft code (GPL/AGPL PDF engines are deliberately excluded). Documents never leave your machine.
What’s built, what’s not
docs/FEATURES.md is the source of truth—a capability matrix with per-surface (core/CLI/GUI) checkboxes, updated on every feature land, not at release time. It distinguishes implemented features (page operations, text editing, markup, redaction, measurements, vector editing), planned features (OCR, JavaScript, XFA, encryption), features that cannot be built (cloud sync, Adobe Document Security, Sensei-backed AI services—architectural constraints), and features that will not be built (dynamic XFA, barcode form fields, GPL-based shortcuts).
296 commits. Brand new. Fast.
Where this stands
Better at
- Local-only processing: no cloud upload, no network calls, no telemetry or update checks—everything runs on your machine.
- Ships with both GUI and a first-class CLI (`pdfce-cli`) with 60 subcommands, not a debug tool but a scriptable peer to the editor.
- Honest feature documentation: `docs/FEATURES.md` distinguishes what's implemented, what's planned, what cannot be built, and what will not be built.
- Heavily tested for its age: 212 test files and 2,623 test functions, second only to ScripTree on this site.
- Single-folder portable download with no installer, dependencies included, runs as a native Windows application.
- Digital-signature inspection, Bates numbering, PDF/A validation and conversion, and vector object editing are built; the feature set is broader than typical lightweight PDF editors.
Worse at
- No OCR, no JavaScript support, no XFA (forms), encrypted document support not yet built.
- No cloud storage, Document Cloud sync, Acrobat Sign service, or Adobe's Sensei-powered auto-tagging and AI Assistant features.
- Pre-1.0 software; some capabilities exist in the core library and CLI but lack GUI implementation (e.g., form field editing CLI exists, GUI surface is partial).
- Brand new — the repository went public on 2026-08-09, and pre-1.0 means interfaces can still move under you.
Pick something else when…
- Adobe Acrobat ProYou need OCR, JavaScript execution, XFA forms, or cloud-based collaboration and e-signature workflows.
- Adobe Acrobat CloudYou need Document Cloud storage, automatic backup, or multi-device sync.
- Lightweight PDF readers like SumatraPDF or PDF-XChange ViewerYou only need viewing, annotation and basic page operations. They have been around far longer than this has, and for that job the extra editing depth here buys you nothing.