Merge pull request #119 from OCR-D/dockerfile

Fix dockerfile

.github/workflows/docker.yml

@@ -0,0 +1,44 @@
+name: CD
+
+on:
+  push:
+    branches: [ "master" ]
+  workflow_dispatch: # run manually
+
+jobs:
+
+  build:
+    runs-on: ubuntu-latest
+    permissions:
+      packages: write
+      contents: read
+    steps:
+    - name: Checkout
+      uses: actions/checkout@v4
+      with:
+        # we need tags for docker version tagging
+        fetch-tags: true
+        fetch-depth: 0
+    - # Activate cache export feature to reduce build time of images
+      name: Set up Docker Buildx
+      uses: docker/setup-buildx-action@v3
+    - name: Login to GitHub Container Registry
+      uses: docker/login-action@v3
+      with:
+        registry: ghcr.io
+        username: ${{ github.actor }}
+        password: ${{ secrets.GITHUB_TOKEN }}
+    - name: Log in to Docker Hub
+      uses: docker/login-action@v3
+      with:
+        username: ${{ secrets.DOCKERIO_USERNAME }}
+        password: ${{ secrets.DOCKERIO_PASSWORD }}
+    - name: Build the Docker image
+      # build both tags at the same time
+      run: make docker DOCKER_TAG="docker.io/ocrd/calamari -t ghcr.io/ocr-d/calamari"
+    - name: Test the Docker image
+      run: docker run --rm ocrd/calamari ocrd-calamari-recognize -h
+    - name: Push to Dockerhub
+      run: docker push docker.io/ocrd/calamari
+    - name: Push to Github Container Registry
+      run: docker push ghcr.io/ocr-d/calamari

Dockerfile

@@ -1,20 +1,35 @@
-FROM ocrd/core
-MAINTAINER OCR-D
+ARG DOCKER_BASE_IMAGE
+FROM $DOCKER_BASE_IMAGE
+ARG VCS_REF
+ARG BUILD_DATE
+LABEL \
+    maintainer="https://ocr-d.de/kontakt" \
+    org.label-schema.vcs-ref=$VCS_REF \
+    org.label-schema.vcs-url="https://github.com/OCR-D/ocrd_calamari" \
+    org.label-schema.build-date=$BUILD_DATE \
+    org.opencontainers.image.vendor="DFG-Funded Initiative for Optical Character Recognition Development" \
+    org.opencontainers.image.title="ocrd_calamari" \
+    org.opencontainers.image.description="OCR-D compliant workspace processor for the functionality of Calamari OCR" \
+    org.opencontainers.image.source="https://github.com/OCR-D/ocrd_calamari" \
+    org.opencontainers.image.documentation="https://github.com/OCR-D/ocrd_calamari/blob/${VCS_REF}/README.md" \
+    org.opencontainers.image.revision=$VCS_REF \
+    org.opencontainers.image.created=$BUILD_DATE \
+    org.opencontainers.image.base.name=$DOCKER_BASE_IMAGE
 ENV DEBIAN_FRONTEND noninteractive
 ENV PYTHONIOENCODING utf8
 ENV LC_ALL C.UTF-8
 ENV LANG C.UTF-8
 
-WORKDIR /build
+WORKDIR /build/calamari
 COPY Makefile .
-COPY setup.py .
+COPY pyproject.toml .
 COPY ocrd-tool.json .
 COPY requirements.txt .
 COPY README.md .
-COPY ocrd_calamari ocrd_calamari
+COPY ocrd_calamari ./ocrd_calamari
+RUN make install
+RUN rm -rf /build/calamari
 
-RUN pip3 install --upgrade pip && \
-    pip3 install . && \
-    pip3 check
 
-ENTRYPOINT ["/usr/local/bin/ocrd-calamari-recognize"]
+WORKDIR /data
+VOLUME ["/data"]

Makefile

@@ -8,6 +8,9 @@ EXAMPLE = actevedef_718448162.first-page+binarization+segmentation
 
 # BEGIN-EVAL makefile-parser --make-help Makefile
 
+DOCKER_BASE_IMAGE = docker.io/ocrd/core-cuda-tf2:v2.70.0
+DOCKER_TAG = 'ocrd/calamari'
+
 help:
 	@echo ""
 	@echo "  Targets"
@@ -21,13 +24,16 @@ help:
 	@echo "    assets-clean     Remove symlinks in test/assets"
 	@echo "    test             Run unit tests"
 	@echo "    coverage         Run unit tests and determine test coverage"
+	@echo "    docker           Build Docker image"
 	@echo ""
 	@echo "  Variables"
 	@echo ""
-	@echo "    PYTHON       '$(PYTHON)'"
-	@echo "    PIP_INSTALL  '$(PIP_INSTALL)'"
-	@echo "    GIT_CLONE    '$(GIT_CLONE)'"
-	@echo "    MODEL        '$(MODEL)'"
+	@echo "    PYTHON            '$(PYTHON)'"
+	@echo "    PIP_INSTALL       '$(PIP_INSTALL)'"
+	@echo "    GIT_CLONE         '$(GIT_CLONE)'"
+	@echo "    MODEL             '$(MODEL)'"
+	@echo "    DOCKER_TAG        '$(DOCKER_TAG)'"
+	@echo "    DOCKER_BASE_IMAGE '$(DOCKER_BASE_IMAGE)'"
 
 # END-EVAL
 
@@ -91,4 +97,11 @@ coverage: test/assets $(MODEL)
 	coverage report
 	coverage html
 
-.PHONY: install assets-clean deps-test test coverage $(MODEL) example
+docker:
+	docker build \
+	--build-arg DOCKER_BASE_IMAGE=$(DOCKER_BASE_IMAGE) \
+	--build-arg VCS_REF=$$(git rev-parse --short HEAD) \
+	--build-arg BUILD_DATE=$$(date -u +"%Y-%m-%dT%H:%M:%SZ") \
+	-t $(DOCKER_TAG) .
+
+.PHONY: install assets-clean deps-test test coverage $(MODEL) example docker

pyproject.toml

@@ -40,6 +40,9 @@ Repository = "https://github.com/OCR-D/ocrd_calamari.git"
 dependencies = {file = ["requirements.txt"]}
 optional-dependencies.dev = {file = ["requirements-dev.txt"]}
 
+[tool.setuptools.package-data]
+"*" = ["*.json"]
+
 [tool.setuptools.packages.find]
 where = ["."]
 include = ["ocrd_calamari"]