James R. Barlow
7565d20c0a
Ubuntu version changed on suspicion of Ghostscript 10, but that issue is resolved.
69 lines
1.6 KiB
Docker
69 lines
1.6 KiB
Docker
# SPDX-FileCopyrightText: 2023 James R. Barlow
|
|
# SPDX-License-Identifier: MPL-2.0
|
|
|
|
# Note: Alpine 3.20 builds tesseract with --enable-opencl, which is not
|
|
# supported by anyone. OCRmyPDF is not compatible with Alpine 3.20.0
|
|
# through 3.20.3. The Alpine issue should be fixed in 3.21.0. It is
|
|
# not clear if 3.20.4+ will have the fix.
|
|
# Details
|
|
# https://gitlab.alpinelinux.org/alpine/aports/-/issues/16143
|
|
# https://github.com/ocrmypdf/OCRmyPDF/issues/1395
|
|
FROM alpine:3.19 AS base
|
|
|
|
ENV LANG=C.UTF-8
|
|
ENV TZ=UTC
|
|
|
|
RUN apk add --no-cache \
|
|
python3 \
|
|
zlib
|
|
|
|
FROM base AS builder
|
|
|
|
RUN apk add --no-cache \
|
|
ca-certificates \
|
|
git \
|
|
python3-dev \
|
|
curl
|
|
|
|
COPY . /app
|
|
|
|
WORKDIR /app
|
|
|
|
COPY --from=ghcr.io/astral-sh/uv:0.5.5 /uv /uvx /bin/
|
|
|
|
ENV UV_COMPILE_BYTECODE=1 UV_LINK_MODE=copy
|
|
|
|
# Instead of restarting the shell, use uv directly from its installed location.
|
|
RUN uv sync --extra test --extra webservice --extra watcher
|
|
|
|
FROM base
|
|
|
|
RUN apk add --no-cache \
|
|
ghostscript \
|
|
jbig2dec \
|
|
jbig2enc \
|
|
pngquant \
|
|
tesseract-ocr \
|
|
tesseract-ocr-data-chi_sim \
|
|
tesseract-ocr-data-deu \
|
|
tesseract-ocr-data-eng \
|
|
tesseract-ocr-data-fra \
|
|
tesseract-ocr-data-osd \
|
|
tesseract-ocr-data-por \
|
|
tesseract-ocr-data-spa \
|
|
ttf-droid \
|
|
unpaper \
|
|
&& rm -rf /var/cache/apk/*
|
|
|
|
WORKDIR /app
|
|
|
|
COPY --from=builder --chown=app:app /app /app
|
|
|
|
RUN rm -rf /app/.git && \
|
|
ln -s /app/misc/webservice.py /app/webservice.py && \
|
|
ln -s /app/misc/watcher.py /app/watcher.py
|
|
|
|
ENV PATH="/app/.venv/bin:${PATH}"
|
|
|
|
ENTRYPOINT ["/app/.venv/bin/ocrmypdf"]
|