Release Notes
Changed
- A document with no extractable text now points at the tool that fixes it. The message used to
say a scan "needs --ocr" — a flag DeepDoc does not have, and rejects with exit 2. It now names
DeepOCR, which writes an invisible text layer over the
page image; the result is born-digital and extracts normally. Exit code 4 is unchanged, so an
ingestion pipeline keeps its routing signal.
- The README documents that loop under Scanned documents. OCR stays deliberately outside this
parser: recognition is probabilistic, and mixing it in would cost the determinism DeepDoc is
built on. A --ocr feature that links DeepOCR's engine in directly remains planned and off by
default, so the released binary stays small and model-free.
Install deepdoc 0.1.1
Install prebuilt binaries via shell script
curl --proto '=https' --tlsv1.2 -LsSf https://github.com/deeplabua/deepdoc/releases/download/v0.1.1/deepdoc-installer.sh | sh
Install prebuilt binaries via powershell script
powershell -ExecutionPolicy Bypass -c "irm https://github.com/deeplabua/deepdoc/releases/download/v0.1.1/deepdoc-installer.ps1 | iex"
Install prebuilt binaries via Homebrew
brew install deeplabua/tap/deepdoc
Download deepdoc 0.1.1
