Browse all 9 articles in this open-source series. A hands-on series on PDF classification, layout extraction, and selective OCR.
Featured starting point
A source-backed overview of Rust parsing, layout extraction, selective OCR, bindings, and benchmarks.
Suggested learning path
01 → 09
A source-backed overview of Rust parsing, layout extraction, selective OCR, bindings, and benchmarks.
Run detection, Markdown extraction, table handling, and selective OCR through Python, Node, Rust, or CLI.
Choose local CLI, service, browser WebAssembly, or native bindings while keeping PDFium and ONNX optional.
Follow one PDF through single-load parsing, content streams, fonts, layout, tables, Markdown, and selective OCR.
Use fixtures to read detector, extractor, font decoding, layout, tables, bindings, and OCR boundaries.
Reproduce corpus speed and quality while separating native parsing, OCR, memory, storage, and hosted fallbacks.
Protect local documents, OCR runtimes, hosted fallbacks, output provenance, and service boundaries.
Choose a PDF ingestion path by layout quality, OCR need, privacy, latency, cost, and operational burden.
Design a local document-ingestion project with classification manifests, selective OCR, provenance, and accessible outputs.
