Merge pull request #216 from bertsky/migrate-ocrd-v3

Migrate ocrd v3

Author: bertsky

.circleci/config.yml

@@ -22,12 +22,16 @@ jobs:
       - run: make install
       - run: make deps-test
       - run: mkdir test-results
-      - run: make test PYTEST_ARGS=--junitxml=test-results/test.xml
+      - run:
+          name: run tests
+          command: make test PYTEST_ARGS="-vv --junitxml=test-results/test.xml"
+          no_output_timeout: 30m
       - store_test_results:
           path: test-results
       - run: make test-cli
       - run: make coverage
       - codecov/upload
+    resource_class: large
 
   build-docker:
     docker:
@@ -75,7 +79,7 @@ workflows:
       - test-python:
           matrix:
             parameters:
-              python-version: ['3.7', '3.8', '3.9', '3.10']
+              python-version: ['3.8', '3.9', '3.10', '3.11']
       - build-docker
 
   deploy:

.dockerignore

@@ -1,17 +1,3 @@
-*
-!.git
-!.gitmodules
-!Makefile
-!setup.py
-!requirements.txt
-!requirements_test.txt
-!LICENSE
-!README.md
-!repo/tesserocr
-!repo/tesseract
-
-# avoid .git and __pycache__ etc:
-!ocrd_tesserocr/**/*.py
-!ocrd_tesserocr/**/*.json
-!test/**/*.py
-
+test/
+dist/
+build/

.pylintrc

@@ -21,8 +21,5 @@ disable =
     wrong-import-order,
     duplicate-code
 
-# allow indented whitespace (as required by interpreter):
-no-space-check=empty-line
-
 # allow non-snake-case identifiers:
 good-names=n,i

CHANGELOG.md

@@ -5,6 +5,10 @@ Versioned according to [Semantic Versioning](http://semver.org/).
 
 ## Unreleased
 
+Changed:
+
+ * adapt to ocrd 3.0, #216
+
 ## [0.19.1] - 2024-07-01
 
 Fixed:

Dockerfile

@@ -1,52 +1,54 @@
-FROM docker.io/ocrd/core:v2.67.2 AS base
-# set proper locales
-ENV LANG C.UTF-8
-ENV LC_ALL C.UTF-8
+ARG DOCKER_BASE_IMAGE
+FROM $DOCKER_BASE_IMAGE
 # install ocrd-tesserocr (until here commands for installing tesseract-ocr)
 ARG VCS_REF
 ARG BUILD_DATE
 LABEL \
     maintainer="https://ocr-d.de/kontakt" \
     org.label-schema.vcs-ref=$VCS_REF \
     org.label-schema.vcs-url="https://github.com/OCR-D/ocrd_tesserocr" \
-    org.label-schema.build-date=$BUILD_DATE
+    org.label-schema.build-date=$BUILD_DATE \
+    org.opencontainers.image.vendor="DFG-Funded Initiative for Optical Character Recognition Development" \
+    org.opencontainers.image.title="ocrd_tesserocr" \
+    org.opencontainers.image.description="Tesseract OCR bindings" \
+    org.opencontainers.image.source="https://github.com/OCR-D/ocrd_tesserocr" \
+    org.opencontainers.image.documentation="https://github.com/OCR-D/ocrd_tesserocr/blob/${VCS_REF}/README.md" \
+    org.opencontainers.image.revision=$VCS_REF \
+    org.opencontainers.image.created=$BUILD_DATE \
+    org.opencontainers.image.base.name=ocrd/core
 
-ENV PYTHONIOENCODING utf8
 
 # set frontend non-interactive to silence interactive tzdata config
-ARG DEBIAN_FRONTEND=noninteractive
+ENV DEBIAN_FRONTEND noninteractive
+# set proper locales
+ENV PYTHONIOENCODING utf8
+ENV LANG C.UTF-8
+ENV LC_ALL C.UTF-8
 
 # set proper date and timezone in container
 RUN echo "Europe/Berlin" > /etc/timezone
 RUN ln -sf /usr/share/zoneinfo/Europe/Berlin /etc/localtime
 RUN dpkg-reconfigure -f noninteractive tzdata
-
 # diagnostic output - check timezone settings
 # RUN cat /etc/timezone
 
 # avoid HOME/.local/share (hard to predict USER here)
 # so let XDG_DATA_HOME coincide with fixed system location
 # (can still be overridden by derived stages)
 ENV XDG_DATA_HOME /usr/local/share
+# avoid the need for an extra volume for persistent resource user db
+# (i.e. XDG_CONFIG_HOME/ocrd/resources.yml)
 ENV XDG_CONFIG_HOME /usr/local/share/ocrd-resources
 ENV TESSDATA_PREFIX $XDG_DATA_HOME/tessdata
 
 WORKDIR /build/ocrd_tesserocr
-COPY setup.py .
-COPY ocrd_tesserocr/ocrd-tool.json .
-COPY README.md .
-COPY requirements.txt .
-COPY requirements_test.txt .
-COPY .git .git
-COPY .gitmodules .
-COPY ocrd_tesserocr ocrd_tesserocr
-COPY repo/tesserocr repo/tesserocr
-COPY repo/tesseract repo/tesseract
-COPY Makefile .
+COPY . .
+# prepackage ocrd-tool.json as ocrd-all-tool.json
+RUN ocrd ocrd-tool ocrd_tesserocr/ocrd-tool.json dump-tools > $(dirname $(ocrd bashlib filename))/ocrd-all-tool.json
+# install everything and reduce image size
 RUN make deps-ubuntu \
-    && make -j4 install-tesseract \
-    && make -j4 install-tesseract-training \
-    && make deps install \
+    && make -j4 install GIT_SUBMODULE=: \
+    && make -j4 install-tesseract-training GIT_SUBMODULE=: \
     && rm -rf /build/ocrd_tesserocr \
     && apt-get -y remove --auto-remove g++ libtesseract-dev make
 

Makefile

@@ -3,6 +3,7 @@ export
 SHELL = /bin/bash
 PYTHON = python3
 PIP = pip3
+GIT_SUBMODULE = git submodule
 LOG_LEVEL = INFO
 PYTHONIOENCODING=utf8
 LC_ALL = C.UTF-8
@@ -27,6 +28,7 @@ PYTEST_ARGS =
 
 # Docker container tag
 DOCKER_TAG = 'ocrd/tesserocr'
+DOCKER_BASE_IMAGE = docker.io/ocrd/core:v3.1.0
 
 help:
 	@echo ""
@@ -38,6 +40,7 @@ help:
 	@echo "    install-tesserocr Compile and install Tesserocr"
 	@echo "    deps              Install Tesseract/Tesserocr and all Python dependencies"
 	@echo "    install           Install this package with all dependencies and download minimal models"
+	@echo "    build             Build source and binary distribution"
 	@echo "    deps-test         Install Python deps for test via pip"
 	@echo "    test              Run unit tests"
 	@echo "    coverage          Run unit tests and determine test coverage"
@@ -100,6 +103,7 @@ deps-test:
 # Build docker image
 docker: repo/tesseract repo/tesserocr
 	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) .
@@ -131,8 +135,8 @@ repo/tesseract/Makefile.in: repo/tesseract
 # phony to ensure this recipe is fired (as in empty directory after clone)
 .PHONY: repo/tesserocr repo/tesseract repo/assets
 repo/tesserocr repo/tesseract repo/assets:
-	git submodule sync $@
-	git submodule update --init $@
+	$(GIT_SUBMODULE) sync $@
+	$(GIT_SUBMODULE) update --init $@
 
 # Install this package
 install: deps
@@ -141,13 +145,16 @@ install: deps
 	ocrd resmgr download ocrd-tesserocr-recognize osd.traineddata
 	ocrd resmgr download ocrd-tesserocr-recognize equ.traineddata
 
+build:
+	$(PIP) install build
+	$(PYTHON) -m build .
+
+test test-cli coverage: export OCRD_MISSING_OUTPUT := ABORT
+
 # Run unit tests
 test: test/assets deps-test
 	@# declare -p HTTP_PROXY
-	#$(PYTHON) -m pytest -n auto --continue-on-collection-errors test $(PYTEST_ARGS)
-	# workaround for pytest-xdist not isolating setenv calls in click.CliRunner from each other:
-	$(PYTHON) -m pytest --continue-on-collection-errors test/test_cli.py $(PYTEST_ARGS)
-	$(PYTHON) -m pytest --continue-on-collection-errors test/test_{segment_{region,table,line,word},recognize}.py $(PYTEST_ARGS)
+	$(PYTHON) -m pytest test --durations=0 --continue-on-collection-errors $(PYTEST_ARGS)
 
 # Run unit tests and determine test coverage
 coverage:
@@ -166,7 +173,7 @@ test-cli: test/assets deps-test
 		ocrd-tesserocr-recognize      -l DEBUG -I OCR-D-SEG-LINE -O OCR-D-TESS-OCR -P model deu
 
 .PHONY: test test-cli install deps deps-ubuntu deps-test help
-.PHONY: install-tesseract install-tesserocr install-tesseract-training 
+.PHONY: install-tesseract install-tesserocr install-tesseract-training build
 
 #
 # Assets