build joex image

.github/workflows/container-images.yml

@@ -0,0 +1,31 @@
+name: Container Images
+
+on:
+  workflow_dispatch:
+  push:
+    branches:
+      - main
+  create:
+    tags:
+      - v*
+
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+    - name: Login to GitHub Container Registry
+      uses: docker/login-action@v3
+      with:
+        registry: ghcr.io
+        username: ${{ github.actor }}
+        password: ${{ secrets.GITHUB_TOKEN }}
+    - name: Build the Docker image
+      run: |
+        tag="${GITHUB_REF##*/}"
+        if [ "$tag" = 'main' ]; then tag='latest'; fi
+        image_name="ghcr.io/docspell/joex:${tag}"
+        docker build --cache-from type=gha --cache-to type=gha,mode=max -t "$image_name" .
+        docker push "$image_name"
+

Dockerfile.joex

@@ -0,0 +1,48 @@
+FROM ubuntu:oracular AS base
+
+
+RUN apt update \
+ && apt full-upgrade -y
+
+RUN apt install -y \
+    $(apt list 2>/dev/null | grep 'tesseract-ocr-' | cut -d'/' -f1) \
+    tzdata \
+    openjdk-21-jre \
+    ghostscript \
+    unpaper \
+    weasyprint \
+    unoconv \
+    python3-distutils-extra \
+    fonts-droid-fallback \
+    fonts-dejavu \
+    fonts-freefont-ttf \
+    fonts-noto \
+    pngquant \
+    ocrmypdf \
+    qpdf
+
+FROM base AS download
+
+RUN apt install -y \
+    curl \
+    unzip
+
+WORKDIR /opt
+
+ARG DOCSPELL_VERSION
+
+RUN curl -sLfo joex.zip "https://github.com/eikek/docspell/releases/download/v${DOCSPELL_VERSION}/docspell-joex-${DOCSPELL_VERSION}.zip" && \
+  unzip joex.zip && \
+  rm joex.zip && \
+  mv docspell-joex-* docspell-joex && \
+  rm docspell-joex/conf/docspell-joex.conf
+
+FROM base
+
+COPY --from=download /opt/docspell-joex /opt/docspell-joex
+
+COPY joex-entrypoint.sh /opt/joex-entrypoint.sh
+
+ENTRYPOINT ["/opt/joex-entrypoint.sh"]
+EXPOSE 7878
+

joex-entrypoint.sh

@@ -0,0 +1,7 @@
+#!/bin/sh
+
+echo "Starting unoconv listener"
+unoconv -l &
+
+/opt/docspell-joex/bin/docspell-joex "$@"
+