v13.6.0 release notes

Unpaper or its underlying libraries don't seem to accept pnms with an
odd integer width. Although it's not clear if this is the issue at all.

In any case, keeping the image a PNG works around the issue. unpaper
only accepted PNM input in the past, which is why we send it PNM.
Since it now accepts PNG, we might as well use PNG.

Unpaper can write PNG as output too, but this added a few seconds to
the test suite was not committed.

Related issues:

https://github.com/ocrmypdf/OCRmyPDF/issues/887

https://github.com/ocrmypdf/OCRmyPDF/issues/665

https://github.com/unpaper/unpaper/issues/82

.docker/Dockerfile

@@ -1,17 +1,28 @@
 # OCRmyPDF
 #
-FROM ubuntu:20.04 as base
+
+FROM ubuntu:22.04 as base
+
+ENV LANG=C.UTF-8
+ENV TZ=UTC
+RUN echo 'debconf debconf/frontend select Noninteractive' | debconf-set-selections
+
+RUN apt-get update && apt-get install -y --no-install-recommends \
+  python3 \
+  libqpdf-dev \
+  zlib1g \
+  liblept5
 
 FROM base as builder
 
-ENV LANG=C.UTF-8
-
+# Note we need leptonica here to build jbig2
 RUN apt-get update && apt-get install -y --no-install-recommends \
   build-essential autoconf automake libtool \
   libleptonica-dev \
   zlib1g-dev \
-  python3 \
+  python3-dev \
   python3-distutils \
+  libffi-dev \
   ca-certificates \
   curl \
   git
@@ -35,26 +46,15 @@ COPY . /app
 
 WORKDIR /app
 
-RUN pip3 install --no-cache-dir \
-  -r requirements/main.txt \
-  -r requirements/webservice.txt \
-  -r requirements/test.txt \
-  -r requirements/watcher.txt \
-  .
+RUN pip3 install --no-cache-dir .[test,webservice,watcher]
 
 FROM base
 
-ENV LANG=C.UTF-8
-
 RUN apt-get update && apt-get install -y --no-install-recommends \
   ghostscript \
   img2pdf \
-  liblept5 \
   libsm6 libxext6 libxrender-dev \
-  zlib1g \
   pngquant \
-  python3 \
-  qpdf \
   tesseract-ocr \
   tesseract-ocr-chi-sim \
   tesseract-ocr-deu \
@@ -62,7 +62,8 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
   tesseract-ocr-fra \
   tesseract-ocr-por \
   tesseract-ocr-spa \
-  unpaper
+  unpaper \
+  && rm -rf /var/lib/apt/lists/*
 
 WORKDIR /app
 
@@ -74,8 +75,6 @@ COPY --from=builder /app/misc/watcher.py /app/
 
 # Copy minimal project files to get the test suite.
 COPY --from=builder /app/setup.cfg /app/setup.py /app/README.md /app/
-COPY --from=builder /app/requirements /app/requirements
 COPY --from=builder /app/tests /app/tests
-COPY --from=builder /app/src /app/src
 
 ENTRYPOINT ["/usr/local/bin/ocrmypdf"]

.github/ISSUE_TEMPLATE/2-problem-with-a-specific-input-file.md

@@ -22,7 +22,7 @@ Run with verbosity or higher `-v1` to see more detailed logging. This informatio
 **Example file**
 If your issue is a problem that affects only certain files, and we will require an input file (PDF or image) that demonstrates your issue.
 
-Please provide an input file with no personal or confidential information. At your option you may [GPG-encrypt the file](https://github.com/jbarlow83/OCRmyPDF/wiki) for OCRmyPDF's author only.
+Please provide an input file with no personal or confidential information. At your option you may [GPG-encrypt the file](https://github.com/ocrmypdf/OCRmyPDF/wiki) for OCRmyPDF's author only.
 
 Links to files hosted elsewhere are perfectly acceptable. You could also look in ``tests/resources`` and see if any of those files reproduce your issue.
 

.github/ISSUE_TEMPLATE/3-feature_request.md

@@ -14,7 +14,14 @@ A clear and concise description of what the problem is. Ex. I'm always frustrate
 A clear and concise description of what you want to happen.
 
 **Describe alternatives you've considered**
-A clear and concise description of any alternative solutions or features you've considered.
+A clear and concise description of any alternative solutions or features you've considered. Please include the versions of OCRmyPDF and other supporting programs (Tesseract OCR, Ghostscript) - maybe an alternative already exists in a newer version.
+
+**Example file**
+If your issue concerns how OCRmyPDF processes certain files, and please provide an example file that helps illustrate how OCRmyPDF's output could be improve.
+
+Please provide an input file with no personal or confidential information. At your option you may [GPG-encrypt the file](https://github.com/ocrmypdf/OCRmyPDF/wiki) for OCRmyPDF's author only.
+
+Links to files hosted elsewhere are perfectly acceptable. You could also look in ``tests/resources`` and see if any of those files reproduce your issue.
 
 **Additional context**
 Add any other context or screenshots about the feature request here.

.github/workflows/build.yml

@@ -0,0 +1,303 @@
+name: Test and deploy
+
+on:
+  push:
+    branches:
+      - master
+      - ci
+      - release/*
+      - feature/*
+    tags:
+      - v*
+    paths-ignore:
+      - README*
+  pull_request:
+
+jobs:
+  test_linux:
+    name: Test ${{ matrix.os }} with Python ${{ matrix.python }}
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        include:
+          - os: ubuntu-18.04
+            python: "3.7"
+          - os: ubuntu-20.04
+            python: "3.8"
+          - os: ubuntu-20.04
+            python: "3.9"
+          - os: ubuntu-20.04
+            python: "3.10"
+          - os: ubuntu-latest
+            python: "3.9"
+          - os: ubuntu-latest
+            python: "pypy-3.8"
+          - os: ubuntu-latest
+            python: "3.9"
+            tesseract5: true
+
+    env:
+      OS: ${{ matrix.os }}
+      PYTHON: ${{ matrix.python }}
+
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: "0" # 0=all, needed for setuptools-scm to resolve version tags
+
+      - uses: actions/setup-python@v2
+        name: Install Python
+        with:
+          python-version: ${{ matrix.python }}
+
+      - name: Install Tesseract 5
+        if: matrix.tesseract5
+        run: |
+          sudo add-apt-repository ppa:alex-p/tesseract-ocr-devel
+
+      - name: Install common packages
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y --no-install-recommends \
+            curl \
+            ghostscript \
+            img2pdf \
+            libffi-dev \
+            libsm6 libxext6 libxrender-dev \
+            pngquant \
+            poppler-utils \
+            tesseract-ocr \
+            tesseract-ocr-deu \
+            tesseract-ocr-eng \
+            tesseract-ocr-osd \
+            unpaper \
+            zlib1g
+
+      - name: Install Ubuntu 18.04 packages
+        if: matrix.os == 'ubuntu-18.04'
+        run: |
+          sudo apt-get install -y --no-install-recommends \
+            libexempi3
+
+      - name: Install Ubuntu 20.04 packages
+        if: matrix.os == 'ubuntu-20.04' || matrix.os == 'ubuntu-latest'
+        run: |
+          sudo apt-get install -y --no-install-recommends \
+            libexempi8
+
+      - name: Install Ubuntu packages for PyPy
+        if: startsWith(matrix.python, 'pypy')
+        run: |
+          sudo apt-get install -y --no-install-recommends \
+            libxml2-dev \
+            libxslt1-dev \
+            pypy3-dev
+
+      - name: Install Python packages
+        run: |
+          python -m pip install --upgrade pip wheel
+          python -m pip install .[test]
+
+      - name: Report versions
+        run: |
+          tesseract --version
+          gs --version
+          pngquant --version
+          unpaper --version
+          img2pdf --version
+
+      - name: Test
+        run: |
+          python -m pytest --cov-report xml --cov=ocrmypdf --cov=tests/ -n0 tests/
+
+      - name: Upload coverage to Codecov
+        uses: codecov/codecov-action@v1
+        with:
+          files: ./coverage.xml
+          env_vars: OS,PYTHON
+
+  test_macos:
+    name: Test macOS
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os: [macos-latest]
+        python: ["3.9", "3.10"]
+
+    env:
+      OS: ${{ matrix.os }}
+      PYTHON: ${{ matrix.python }}
+
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: "0" # 0=all, needed for setuptools-scm to resolve version tags
+
+      - uses: actions/setup-python@v2
+        name: Install Python
+        with:
+          python-version: ${{ matrix.python }}
+
+      - name: Install Homebrew deps
+        run: |
+          brew update
+          brew install \
+            exempi \
+            ghostscript \
+            jbig2enc \
+            openjpeg \
+            pngquant \
+            tesseract
+
+      - name: Install Python packages
+        run: |
+          python -m pip install --upgrade pip wheel
+          python -m pip install .[test]
+
+      - name: Report versions
+        run: |
+          tesseract --version
+          gs --version
+          pngquant --version
+          img2pdf --version
+
+      - name: Test
+        run: |
+          python -m pytest --cov-report xml --cov=ocrmypdf --cov=tests/ -n0 tests/
+
+      - name: Upload coverage to Codecov
+        uses: codecov/codecov-action@v1
+        with:
+          files: ./coverage.xml
+          env_vars: OS,PYTHON
+
+  test_windows:
+    name: Test Windows
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os: [windows-latest]
+        python: ["3.9", "3.10"]
+
+    env:
+      OS: ${{ matrix.os }}
+      PYTHON: ${{ matrix.python }}
+
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: "0" # 0=all, needed for setuptools-scm to resolve version tags
+
+      - uses: actions/setup-python@v2
+        name: Install Python
+        with:
+          python-version: ${{ matrix.python }}
+
+      - name: Install system packages
+        run: |
+          choco install --yes --no-progress --pre tesseract
+          choco install --yes --no-progress --ignore-checksums ghostscript
+
+      - name: Install Python packages
+        run: |
+          python -m pip install --upgrade pip wheel
+          python -m pip install .[test]
+
+      - name: Test
+        run: |
+          python -m pytest --cov-report xml --cov=ocrmypdf --cov=tests/ -n0 tests/
+
+      - name: Upload coverage to Codecov
+        uses: codecov/codecov-action@v1
+        with:
+          files: ./coverage.xml
+          env_vars: OS,PYTHON
+
+  wheel_sdist_linux:
+    name: Build sdist and wheels
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: "0" # 0=all, needed for setuptools-scm to resolve version tags
+
+      - uses: actions/setup-python@v2
+        name: Install Python
+        with:
+          python-version: "3.7"
+
+      - name: Make wheels and sdist
+        run: |
+          python -m pip install --upgrade pip wheel
+          python setup.py sdist
+          python setup.py bdist_wheel
+
+      - uses: actions/upload-artifact@v2
+        with:
+          path: |
+            ./dist/*.whl
+            ./dist/*.tar.gz
+
+  upload_pypi:
+    name: Deploy artifacts to PyPI
+    needs: [wheel_sdist_linux, test_linux, test_macos, test_windows]
+    runs-on: ubuntu-latest
+    if: github.event_name == 'push' && startsWith(github.event.ref, 'refs/tags/v')
+    steps:
+      - uses: actions/download-artifact@v2
+        with:
+          name: artifact
+          path: dist
+
+      - uses: pypa/gh-action-pypi-publish@master
+        with:
+          user: __token__
+          password: ${{ secrets.TOKEN_PYPI }}
+          # repository_url: https://test.pypi.org/legacy/
+
+  docker:
+    name: Build Docker images
+    needs: [wheel_sdist_linux, test_linux, test_macos, test_windows]
+    runs-on: ubuntu-latest
+    if: github.event_name != 'pull_request'
+    steps:
+      - name: Set image tag to release or branch
+        run: echo "DOCKER_IMAGE_TAG=${GITHUB_REF##*/}" >> $GITHUB_ENV
+
+      - name: If master, set to latest
+        run: echo 'DOCKER_IMAGE_TAG=latest' >> $GITHUB_ENV
+        if: env.DOCKER_IMAGE_TAG == 'master'
+
+      - name: Set Docker Hub repository to username
+        run: echo "DOCKER_REPOSITORY=jbarlow83" >> $GITHUB_ENV
+
+      - name: Set image name
+        run: echo "DOCKER_IMAGE_NAME=ocrmypdf" >> $GITHUB_ENV
+
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: "0" # 0=all, needed for setuptools-scm to resolve version tags
+
+      - name: Login to Docker Hub
+        uses: docker/login-action@v1
+        with:
+          username: jbarlow83
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
+
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v1
+
+      - name: Set up Docker Buildx
+        id: buildx
+        uses: docker/setup-buildx-action@v1
+
+      - name: Print image tag
+        run: echo "Building image ${DOCKER_REPOSITORY}/${DOCKER_IMAGE_NAME}:${DOCKER_IMAGE_TAG}"
+
+      - name: Build
+        run: |
+          docker buildx build \
+            --push \
+            --platform linux/arm64/v8,linux/amd64  \
+            --tag "${DOCKER_REPOSITORY}/${DOCKER_IMAGE_NAME}:${DOCKER_IMAGE_TAG}" \
+            --file .docker/Dockerfile .

.gitignore

@@ -6,8 +6,9 @@
 !.gitattributes
 !.gitignore
 !.pre-commit-config.yaml
-!.readthedocs.yml
+!.readthedocs.yaml
 !.github/
+!.docker/
 
 # Dev scratch
 *.ipynb
@@ -24,6 +25,7 @@ venv*/
 /debug_tests.py
 *.traineddata
 /private
+/coverage.xml
 
 # Package building
 *.egg-info/

.pre-commit-config.yaml

@@ -1,23 +1,37 @@
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v3.4.0
+    rev: v4.2.0
     hooks:
       - id: check-case-conflict
       - id: check-merge-conflict
       - id: check-toml
       - id: check-yaml
       - id: debug-statements
-  - repo: https://github.com/asottile/seed-isort-config
-    rev: v2.2.0
-    hooks:
-      - id: seed-isort-config
-  - repo: https://github.com/pre-commit/mirrors-isort
-    rev: v5.7.0 # pick the isort version you'd like to use from https://github.com/pre-commit/mirrors-isort/releases
+  - repo: https://github.com/pycqa/isort
+    rev: 5.10.1
     hooks:
       - id: isort
+        args: ["--profile", "black"]
   - repo: https://github.com/psf/black
-    rev: 20.8b1
+    rev: 22.3.0
     hooks:
       - id: black
         language_version: python
-        exclude: ^src/ocrmypdf/lib/_leptonica.py
+  - repo: https://github.com/asottile/setup-cfg-fmt
+    rev: v1.20.1
+    hooks:
+      - id: setup-cfg-fmt
+  - repo: https://github.com/asottile/pyupgrade
+    rev: v2.32.1
+    hooks:
+      - id: pyupgrade
+        args: ["--py37-plus"]
+  - repo: https://github.com/pre-commit/mirrors-mypy
+    rev: v0.950
+    hooks:
+      - id: mypy
+        additional_dependencies:
+          - types-toml
+          - types-setuptools
+          - types-requests
+          - types-Pillow

.readthedocs.yaml

@@ -0,0 +1,22 @@
+# Read the Docs configuration file
+# See https://docs.readthedocs.io/en/stable/config-file/v2.html for details
+
+# Required
+version: 2
+
+# Build documentation in the docs/ directory with Sphinx
+sphinx:
+  configuration: docs/conf.py
+
+# Optionally build your docs in additional formats such as PDF
+formats:
+  - pdf
+
+# Optionally set the version of Python and requirements required to build your docs
+python:
+  version: "3.7"
+  install:
+    - method: pip
+      path: .
+      extra_requirements:
+        - docs

.readthedocs.yml

@@ -1,10 +0,0 @@
-build:
-  image: latest
-
-python:
-  version: 3.6
-
-formats:
-  - pdf
-
-requirements_file: requirements/main.txt

README.md

@@ -1,9 +1,7 @@
 <img src="docs/images/logo.svg" width="240" alt="OCRmyPDF">
 
-[![Build Status][azure]](https://dev.azure.com/jim0585/ocrmypdf/_build/latest?definitionId=2&branchName=master) [![PyPI version][pypi]](https://pypi.org/project/ocrmypdf/) ![Homebrew version][homebrew] ![ReadTheDocs][docs] ![Python versions][pyversions]
+[![Build Status](https://github.com/ocrmypdf/OCRmyPDF/actions/workflows/build.yml/badge.svg)](https://github.com/ocrmypdf/OCRmyPDF/actions/workflows/build.yml) [![PyPI version][pypi]](https://pypi.org/project/ocrmypdf/) ![Homebrew version][homebrew] ![ReadTheDocs][docs] ![Python versions][pyversions]
 
-[azure]: https://dev.azure.com/jim0585/ocrmypdf/_apis/build/status/jbarlow83.OCRmyPDF?branchName=master
-[travis]: https://travis-ci.org/jbarlow83/OCRmyPDF.svg?branch=master "Travis build status"
 [pypi]: https://img.shields.io/pypi/v/ocrmypdf.svg "PyPI version"
 [homebrew]: https://img.shields.io/homebrew/v/ocrmypdf.svg "Homebrew version"
 [docs]: https://readthedocs.org/projects/ocrmypdf/badge/?version=latest "RTD"
@@ -57,25 +55,18 @@ I searched the web for a free command line tool to OCR PDF files: I found many,
 
 ## Installation
 
-Linux, Windows, macOS and FreeBSD are supported. Docker images are also available.
+Linux, Windows, macOS and FreeBSD are supported. Docker images are also available, for both x64 and ARM.
 
-Users of Debian 9 or later or Ubuntu 16.10 or later may simply
-
-```bash
-apt-get install ocrmypdf
-```
-
-and users of Fedora 29 or later may simply
-
-```bash
-dnf install ocrmypdf
-```
-
-and Homebrew users (macOS, Linux, Windows Subsystem for Linux) may simply
-
-```bash
-brew install ocrmypdf
-```
+| Operating system              | Install command               |
+| ----------------------------- | ------------------------------|
+| Debian, Ubuntu                | ``apt install ocrmypdf``      |
+| Windows Subsystem for Linux   | ``apt install ocrmypdf``      |
+| Fedora                        | ``dnf install ocrmypdf``      |
+| macOS (Homebrew)              | ``brew install ocrmypdf``     |
+| macOS (nix)                   | ``nix-env -i  ocrmypdf``      |
+| LinuxBrew                     | ``brew install ocrmypdf``     |
+| FreeBSD                       | ``pkg install py37-ocrmypdf`` |
+| Conda                         | ``conda install ocrmypdf``    |
 
 For everyone else, [see our documentation](https://ocrmypdf.readthedocs.io/en/latest/installation.html) for installation steps.
 
@@ -99,6 +90,11 @@ brew install tesseract-lang
 
 You can then pass the `-l LANG` argument to OCRmyPDF to give a hint as to what languages it should search for. Multiple languages can be requested.
 
+OCRmyPDF supports Tesseract 4.0 and the beta versions of Tesseract 5.0. It will
+automatically use whichever version it finds first on the `PATH` environment
+variable. On Windows, if `PATH` does not provide a Tesseract binary, we use
+the highest version number that is installed according to the Windows Registry.
+
 ## Documentation and support
 
 Once OCRmyPDF is installed, the built-in help which explains the command syntax and options can be accessed via:
@@ -109,11 +105,11 @@ ocrmypdf --help
 
 Our [documentation is served on Read the Docs](https://ocrmypdf.readthedocs.io/en/latest/index.html).
 
-Please report issues on our [GitHub issues](https://github.com/jbarlow83/OCRmyPDF/issues) page, and follow the issue template for quick response.
+Please report issues on our [GitHub issues](https://github.com/ocrmypdf/OCRmyPDF/issues) page, and follow the issue template for quick response.
 
 ## Requirements
 
-In addition to the required Python version (3.6+), OCRmyPDF requires external program installations of Ghostscript, Tesseract OCR, QPDF, and Leptonica. OCRmyPDF is pure Python, but uses CFFI to portably generate library bindings. OCRmyPDF works on pretty much everything: Linux, macOS, Windows and FreeBSD.
+In addition to the required Python version (3.7+), OCRmyPDF requires external program installations of Ghostscript and Tesseract OCR. OCRmyPDF is pure Python, and runs on pretty much everything: Linux, macOS, Windows and FreeBSD.
 
 ## Press & Media
 
@@ -123,6 +119,7 @@ In addition to the required Python version (3.6+), OCRmyPDF requires external pr
 - [heise Open Source, 09/2014: Texterkennung mit OCRmyPDF](https://heise.de/-2356670)
 - [heise Durchsuchbare PDF-Dokumente mit OCRmyPDF erstellen](https://www.heise.de/ratgeber/Durchsuchbare-PDF-Dokumente-mit-OCRmyPDF-erstellen-4607592.html)
 - [Excellent Utilities: OCRmyPDF](https://www.linuxlinks.com/excellent-utilities-ocrmypdf-add-ocr-text-layer-scanned-pdfs/)
+- [LinuxUser Texterkennung mit OCRmyPDF und Scanbd automatisieren](https://www.linux-community.de/ausgaben/linuxuser/2021/06/texterkennung-mit-ocrmypdf-und-scanbd-automatisieren/)
 
 ## Business enquiries
 

azure-pipelines.yml

@@ -1,252 +0,0 @@
-trigger:
-  tags:
-    include:
-      - v*
-  branches:
-    include:
-      - "*"
-    exclude:
-      - "travis"
-
-stages:
-  - stage: "Test"
-    jobs:
-      - job: Windows
-        pool:
-          vmImage: "vs2017-win2016"
-        strategy:
-          matrix:
-            Python36:
-              python.version: "3.6"
-            Python37:
-              python.version: "3.7"
-            Python38:
-              python.version: "3.8"
-            Python39:
-              python.version: "3.9"
-        steps:
-          - task: UsePythonVersion@0
-            inputs:
-              versionSpec: "$(python.version)"
-          - pwsh: |
-              choco install --yes --no-progress --pre tesseract
-              choco install --yes --no-progress python3
-              choco install --yes --no-progress ghostscript
-              # choco install --yes --no-progress pngquant
-            displayName: "Install system packages"
-          - pwsh: |
-              refreshenv
-              python -m pip install --upgrade pip wheel
-              python -m pip install -r requirements/main.txt -r requirements/test.txt .
-            displayName: "Install Python packages"
-          - pwsh: |
-              refreshenv
-              $env:pathext += ';.py'
-              # -n auto helps Windows
-              python -m pytest -n auto --junitxml=test.xml --cov=ocrmypdf --cov-report=xml
-            displayName: "Test"
-          - task: PublishTestResults@2
-            inputs:
-              testResultsFiles: "test.xml"
-              testRunTitle: "$(Agent.OS) - $(Build.DefinitionName) - Python $(python.version)"
-            condition: succeededOrFailed()
-      - job: "Ubuntu_1804"
-        pool:
-          vmImage: "ubuntu-18.04"
-        strategy:
-          matrix:
-            Python36:
-              python.version: "3.6"
-            Python37:
-              python.version: "3.7"
-            Python38:
-              python.version: "3.8"
-            Python39:
-              python.version: "3.9"
-        steps:
-          - task: UsePythonVersion@0
-            inputs:
-              versionSpec: "$(python.version)"
-          - bash: |
-              sudo apt-get update
-              sudo apt-get install -y --no-install-recommends \
-                python3-software-properties \
-                curl \
-                ghostscript \
-                img2pdf \
-                libexempi3 \
-                libffi-dev \
-                liblept5 \
-                libsm6 libxext6 libxrender-dev \
-                pngquant \
-                poppler-utils \
-                tesseract-ocr \
-                tesseract-ocr-deu \
-                tesseract-ocr-eng \
-                unpaper \
-                zlib1g
-            displayName: "Install system packages"
-          - bash: |
-              curl https://bootstrap.pypa.io/get-pip.py | python3
-              pip3 install -r requirements/main.txt -r requirements/test.txt .
-            displayName: "Install Python packages"
-          - bash: |
-              tesseract --version
-            displayName: "Record versions"
-          - bash: |
-              # -n auto is slower on Linux and breaks on Python 3.8
-              pytest -n0 --junitxml=test.xml --cov=ocrmypdf --cov-report=xml
-            displayName: "Test"
-          - task: PublishTestResults@2
-            inputs:
-              testResultsFiles: "test.xml"
-              testRunTitle: "$(Agent.OS) - $(Build.DefinitionName) - Python $(python.version)"
-            condition: succeededOrFailed()
-      - job: "Ubuntu_1604"
-        pool:
-          vmImage: "ubuntu-16.04"
-        strategy:
-          matrix:
-            Python36:
-              python.version: "3.6"
-        steps:
-          - task: UsePythonVersion@0
-            inputs:
-              versionSpec: "$(python.version)"
-          - bash: |
-              sudo apt-get update
-              sudo apt-get install -y --no-install-recommends \
-                software-properties-common
-              sudo add-apt-repository -y ppa:alex-p/tesseract-ocr
-              sudo apt-get update
-              sudo apt-get install -y --no-install-recommends \
-                ghostscript \
-                img2pdf \
-                libexempi3 \
-                libffi-dev \
-                liblept5 \
-                libsm6 libxext6 libxrender-dev \
-                pngquant \
-                poppler-utils \
-                tesseract-ocr \
-                tesseract-ocr-deu \
-                tesseract-ocr-eng \
-                unpaper \
-                zlib1g
-            displayName: "Install system packages"
-          - bash: |
-              curl https://bootstrap.pypa.io/get-pip.py | python3
-              pip3 install -r requirements/main.txt -r requirements/test.txt .
-            displayName: "Install Python packages"
-          - bash: |
-              tesseract --version
-            displayName: "Record versions"
-          - bash: |
-              # -n auto is slower on Linux and breaks on Python 3.8
-              pytest -n0 --junitxml=test.xml --cov=ocrmypdf --cov-report=xml
-            displayName: "Test"
-          - task: PublishTestResults@2
-            inputs:
-              testResultsFiles: "test.xml"
-              testRunTitle: "$(Agent.OS) - $(Build.DefinitionName) - Python $(python.version)"
-            condition: succeededOrFailed()
-      - job: "macOS_Mojave"
-        pool:
-          vmImage: "macos-10.14"
-        steps:
-          # https://github.com/actions/virtual-environments/issues/664
-          # - task: UsePythonVersion@0
-          #   inputs:
-          #     versionSpec: "$(python.version)"
-          - bash: |
-              brew update
-              brew upgrade python
-              echo "Using `python3 --version`"
-            displayName: "Update brew and Python"
-          - bash: |
-              brew install \
-                exempi \
-                ghostscript \
-                jbig2enc \
-                leptonica \
-                openjpeg \
-                pngquant \
-                tesseract
-            displayName: "Install system packages"
-          - bash: |
-              pip3 install --upgrade pip
-              pip3 install -r requirements/main.txt -r requirements/test.txt .
-            displayName: "Install Python packages"
-          - bash: |
-              tesseract --version
-            displayName: "Record versions"
-          - bash: pytest -nauto --junitxml=test.xml --cov=ocrmypdf --cov-report=xml
-            displayName: "Test"
-          - task: PublishTestResults@2
-            inputs:
-              testResultsFiles: "test.xml"
-              testRunTitle: "$(Agent.OS) - $(Build.DefinitionName) - Python $(python.version)"
-            condition: succeededOrFailed()
-          - task: PublishCodeCoverageResults@1
-            inputs:
-              codeCoverageTool: Cobertura
-              summaryFileLocation: "$(System.DefaultWorkingDirectory)/**/coverage.xml"
-
-  - stage: "Artifacts"
-    jobs:
-      - job: "sdist_wheel"
-        pool:
-          vmImage: "ubuntu-18.04"
-        steps:
-          - task: UsePythonVersion@0
-            inputs:
-              versionSpec: "3.7"
-          - bash: |
-              python -m pip install --upgrade pip wheel
-              python setup.py sdist bdist_wheel
-          - publish: dist
-            artifact: sdist_wheel
-
-  - stage: "Deploy"
-    jobs:
-      - deployment: "PyPI"
-        pool:
-          vmImage: "ubuntu-18.04"
-        environment: "deploy"
-        strategy:
-          runOnce:
-            deploy:
-              steps:
-                - download: current
-                  artifact: sdist_wheel
-                - script: |
-                    mkdir -p dist
-                    mv $(Pipeline.Workspace)/sdist_wheel/* dist
-                  displayName: "Move dist files"
-                - task: UsePythonVersion@0
-                  inputs:
-                    versionSpec: "3.8"
-                    architecture: x64
-                - script: |
-                    pip install --upgrade twine
-                  displayName: "Generate artifacts"
-                - script: |
-                    cat <<FILE >.pypirc
-                    [distutils]
-                    index-servers =
-                        pypi
-
-                    [pypi]
-                    username: __token__
-                    password: $(TOKEN_PYPI)
-
-                    FILE
-                  displayName: "Generate PyPI auth file"
-                - script: |
-                    python -m twine upload --config-file .pypirc dist/*
-                  displayName: "Upload to PyPI"
-                  condition: and(succeeded(), startsWith(variables['Build.SourceBranch'], 'refs/tags/'))
-                - script: |
-                    curl -X POST -d "token=$(TOKEN_RTD)" https://readthedocs.org/api/v2/webhook/pikepdf/39557/
-                  displayName: "Trigger ReadTheDocs"
-                  condition: and(succeeded(), or(startsWith(variables['Build.SourceBranch'], 'refs/tags/'), startsWith(variables['Build.SourceBranch'], 'refs/heads/master')))

debian/copyright

@@ -1,7 +1,7 @@
 Format: https://www.debian.org/doc/packaging-manuals/copyright-format/1.0/
 Upstream-Name: OCRmyPDF
 Upstream-Contact: James R. Barlow <barlow.jim@gmail.com>
-Source: https://github.com/jbarlow83/OCRmyPDF
+Source: https://github.com/ocrmypdf/OCRmyPDF
 
 Files: *
 Copyright:
@@ -60,11 +60,6 @@ Copyright: (C) 2010 Jonathan Brinley <jonathanbrinley@gmail.com>
  (C) 2015-16 James R. Barlow
 License: Expat
 
-Files: src/ocrmypdf/_unicodefun.py
-Copyright: (C) 2014 Armin Ronacher
- (C) 2017 James R. Barlow
-License: BSD-3-clause
-
 Files: tests/plugins/*
 Copyright: (C) 2016, 2017, 2016-2018 James R. Barlow
 License: Expat

docs/advanced.rst

@@ -67,11 +67,11 @@ without modifying the PDF. This is to ensure that PDFs that were
 previously OCRed or were "born digital" rather than scanned are not
 processed.
 
-If ``--skip-text`` is issued, then no OCR will be performed on pages
-that already have text. The page will be copied to the output. This may
-be useful for documents that contain both "born digital" and scanned
-content, or to use OCRmyPDF to normalize and convert to PDF/A regardless
-of their contents.
+If ``--skip-text`` is issued, then no image processing or OCR will be
+performed on pages that already have text. The page will be copied to
+the output. This may be useful for documents that contain both "born
+digital" and scanned content, or to use OCRmyPDF to normalize and
+convert to PDF/A regardless of their contents.
 
 If ``--redo-ocr`` is issued, then a detailed text analysis is performed.
 Text is categorized as either visible or invisible. Invisible text (OCR)
@@ -228,20 +228,14 @@ preprocessing is specified, then the image layer is a new PDF.
 Unlike ``sandwich`` this renderer is implemented within OCRmyPDF; anyone
 looking to customize how OCR is presented should look here. A major
 disadvantage of this renderer is it not capable of correctly handling
-text outside the Latin alphabet. Pull requests to improve the situation
-are welcome.
+text outside the Latin alphabet (specifically, it supports the ISO 8859-1
+character). Pull requests to improve the situation are welcome.
 
 Currently, this renderer has the best compatibility with Mozilla's
 PDF.js viewer.
 
 This works in all versions of Tesseract.
 
-The ``tesseract`` renderer
---------------------------
-
-The ``tesseract`` renderer was removed. OCRmyPDF's new approach to text
-layer grafting makes it functionally equivalent to ``sandwich``.
-
 Return code policy
 ==================
 

docs/api.rst

@@ -12,7 +12,7 @@ subprocess call anyway, as this provides isolation of its activities.
 Example
 =======
 
-OCRmyPDF one high-level function to run its main engine from an
+OCRmyPDF provides one high-level function to run its main engine from an
 application. The parameters are symmetric to the command line arguments
 and largely have the same functions.
 
@@ -23,7 +23,7 @@ and largely have the same functions.
     if __name__ == '__main__':  # To ensure correct behavior on Windows and macOS
         ocrmypdf.ocr('input.pdf', 'output.pdf', deskew=True)
 
-With a few exceptions, all of the command line arguments are available
+With some exceptions, all of the command line arguments are available
 and may be passed as equivalent keywords.
 
 A few differences are that ``verbose`` and ``quiet`` are not available.
@@ -41,33 +41,41 @@ execution. To do this, it will:
 - manage the signal flags of its worker processes
 - execute other subprocesses (forking and executing other programs)
 
-The Python process that calls ``ocrmypdf.ocr()`` must be sufficiently
+The Python process that calls :func:`ocrmypdf.ocr()` must be sufficiently
 privileged to perform these actions.
 
-There is no currently no option to manage how jobs are scheduled other
+There currently is no option to manage how jobs are scheduled other
 than the argument ``jobs=`` which will limit the number of worker
 processes.
 
-Creating a child process to call ``ocrmypdf.ocr()`` is suggested. That
+Creating a child process to call :func:`ocrmypdf.ocr()` is suggested. That
 way your application will survive and remain interactive even if
 OCRmyPDF fails for any reason.
 
-Programs that call ``ocrmypdf.ocr()`` should also install a SIGBUS signal
+Programs that call :func:`ocrmypdf.ocr()` should also install a SIGBUS signal
 handler (except on Windows), to raise an exception if access to a memory
 mapped file fails. OCRmyPDF may use memory mapping.
 
-``ocrmypdf.ocr()`` will take a threading lock to prevent multiple runs of itself
+:func:`ocrmypdf.ocr()` will take a threading lock to prevent multiple runs of itself
 in the same Python interpreter process. This is not thread-safe, because of how
 OCRmyPDF's plugins and Python's library import system work. If you need to parallelize
 OCRmyPDF, use processes.
 
 .. warning::
 
-    On Windows and macOS, the script that calls ``ocrmypdf.ocr()`` must be
+    On Windows and macOS, the script that calls :func:`ocrmypdf.ocr()` must be
     protected by an "ifmain" guard (``if __name__ == '__main__'``). If you do
     not take at least one of these steps, process semantics will prevent
     OCRmyPDF from working correctly.
 
+.. warning::
+
+    On macOS with Python 3.7, you must call
+    :func:`multiprocessing.set_start_method("spawn")`. Without this, multiprocessing
+    will be unstable. From the command line, OCRmyPDF does this automatically,
+    but as an API user you must do this. See Python bpo-33725 for details.
+    Python 3.8+ also resolve this automatically.
+
 Logging
 -------
 
@@ -96,7 +104,7 @@ Exceptions
 
 OCRmyPDF may throw standard Python exceptions, ``ocrmypdf.exceptions.*``
 exceptions, some exceptions related to multiprocessing, and
-``KeyboardInterrupt``. The parent process should provide an exception
+:exc:`KeyboardInterrupt`. The parent process should provide an exception
 handler. OCRmyPDF will clean up its temporary files and worker processes
 automatically when an exception occurs.
 

docs/batch.rst

@@ -36,18 +36,21 @@ Directory trees
 ===============
 
 This will walk through a directory tree and run OCR on all files in
-place, printing the output in a way that makes
+place, and printing each filename in between runs:
 
 .. code-block:: bash
 
-   find . -printf '%p' -name '*.pdf' -exec ocrmypdf '{}' '{}' \;
+   find . -printf '%p\n' -name '*.pdf' -exec ocrmypdf '{}' '{}' \;
 
-Alternatively, with a docker container (mounts a volume to the container
-where the PDFs are stored):
+Alternatively, with a Docker container and streaming the file through
+standard input and output:
 
 .. code-block:: bash
 
-   find . -printf '%p' -name '*.pdf' -exec docker run --rm -v <host dir>:<container dir> jbarlow83/ocrmypdf '<container dir>/{}' '<container dir>/{}' \;
+   find . -name '*.pdf' -print0 | xargs -0 | while read pdf; do
+       pdfout=$(mktemp)
+       docker run --rm -i jbarlow83/ocrmypdf - - <$pdf >$pdfout && cp $pdfout $pdf
+   done
 
 This only runs one ``ocrmypdf`` process at a time. This variation uses
 ``find`` to create a directory list and ``parallel`` to parallelize runs
@@ -111,7 +114,7 @@ Users may need to customize the script to meet their requirements.
 
 .. code-block:: bash
 
-    pip3 install -r requirements/watcher.txt
+    pip3 install ocrmypdf[watcher]
 
     env OCR_INPUT_DIRECTORY=/mnt/input-pdfs \
         OCR_OUTPUT_DIRECTORY=/mnt/output-pdfs \
@@ -124,7 +127,9 @@ Users may need to customize the script to meet their requirements.
 
     "OCR_INPUT_DIRECTORY", "Set input directory to monitor (recursive)"
     "OCR_OUTPUT_DIRECTORY", "Set output directory (should not be under input)"
+    "OCR_ARCHIVE_DIRECTORY", "Set archive directory for processed originals (should not be under input, requires ``OCR_ON_SUCCESS_ARCHIVE`` to be set)"
     "OCR_ON_SUCCESS_DELETE", "This will delete the input file if the exit code is 0 (OK)"
+    "OCR_ON_SUCCESS_ARCHIVE", "This will move the processed orignal file to ``OCR_ARCHIVE_DIRECTORY`` if the exit code is 0 (OK). Note that ``OCR_ON_SUCCESS_DELETE`` takes precedence over this option, i.e. if both options are set, the input file will be deleted."
     "OCR_OUTPUT_DIRECTORY_YEAR_MONTH", "This will place files in the output in ``{output}/{year}/{month}/{filename}``"
     "OCR_DESKEW", "Apply deskew to crooked input PDFs"
     "OCR_JSON_SETTINGS", "A JSON string specifying any other arguments for ``ocrmypdf.ocr``, e.g. ``'OCR_JSON_SETTINGS={""rotate_pages"": true}'``."
@@ -144,16 +149,18 @@ The watcher service is included in the OCRmyPDF Docker image. To run it:
     docker run \
         -v <path to files to convert>:/input \
         -v <path to store results>:/output \
+        -v <path to store processed originals>:/archive \
         -e OCR_OUTPUT_DIRECTORY_YEAR_MONTH=1 \
-        -e OCR_ON_SUCCESS_DELETE=1 \
+        -e OCR_ON_SUCCESS_ARCHIVE=1 \
         -e OCR_DESKEW=1 \
         -e PYTHONUNBUFFERED=1 \
         -it --entrypoint python3 \
         jbarlow83/ocrmypdf \
         watcher.py
 
-This service will watch for a file that matches ``/input/\*.pdf`` and will
-convert it to a OCRed PDF in ``/output/``. The parameters to this image are:
+This service will watch for a file that matches ``/input/\*.pdf``,
+convert it to a OCRed PDF in ``/output/``, and move the processed
+original to ``/archive``. The parameters to this image are:
 
 .. csv-table:: watcher.py parameters for Docker
     :header: "Parameter", "Description"
@@ -161,10 +168,11 @@ convert it to a OCRed PDF in ``/output/``. The parameters to this image are:
 
     "``-v <path to files to convert>:/input``", "Files placed in this location will be OCRed"
     "``-v <path to store results>:/output``", "This is where OCRed files will be stored"
-    "``-e OCR_OUTPUT_DIRECTORY_YEAR_MONTH=1``", "Define environment variable OCR_OUTPUT_DIRECTORY_YEAR_MONTH=1"
-    "``-e OCR_ON_SUCCESS_DELETE=1``", "Define environment variable"
-    "``-e OCR_DESKEW=1``", "Define environment variable"
-    "``-e PYTHONBUFFERED=1``", "This will force STDOUT to be unbuffered and allow you to see messages in docker logs"
+    "``-v <path to store processed originals>:/archive``", "Archive processed originals here"
+    "``-e OCR_OUTPUT_DIRECTORY_YEAR_MONTH=1``", "Define environment variable ``OCR_OUTPUT_DIRECTORY_YEAR_MONTH=1`` to place files in the output in ``{output}/{year}/{month}/{filename}``"
+    "``-e OCR_ON_SUCCESS_ARCHIVE=1``", "Define environment variable ``OCR_ON_SUCCESS_ARCHIVE`` to move processed originals"
+    "``-e OCR_DESKEW=1``", "Define environment variable ``OCR_DESKEW``  to apply deskew to crooked input PDFs"
+    "``-e PYTHONBUFFERED=1``", "This will force ``STDOUT`` to be unbuffered and allow you to see messages in docker logs"
 
 This service relies on polling to check for changes to the filesystem. It
 may not be suitable for some environments, such as filesystems shared on a
@@ -202,13 +210,6 @@ Alternatives
 -  `Watchman <https://facebook.github.io/watchman/>`__ is a more
    powerful alternative to ``watchmedo``.
 
-AWS Lambda is not viable
-------------------------
-
-AWS Lambda and its equivalents have low limits on execution time and payload
-size, relative to OCRmyPDF's needs. As of this writing, the request/response
-payload for AWS Lambda was 6 MB, which means many PDFs will not fit.
-
 macOS Automator
 ===============
 

docs/conf.py

@@ -31,9 +31,18 @@
 # Add any Sphinx extension module names here, as strings. They can be
 # extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
 # ones.
-extensions = ['sphinx.ext.napoleon']
+extensions = [
+    'sphinx.ext.autodoc',
+    'sphinx.ext.intersphinx',
+    'sphinx.ext.autosummary',
+    'sphinx.ext.napoleon',
+    'sphinx_issues',
+]
 
+# Extension settings
+intersphinx_mapping = {'https://docs.python.org/': None}
 napoleon_use_rtype = False
+issues_github_path = "ocrmypdf/OCRmyPDF"
 
 # Add any paths that contain templates here, relative to this directory.
 templates_path = ['_templates']
@@ -54,7 +63,7 @@ master_doc = 'index'
 # General information about the project.
 project = 'ocrmypdf'
 copyright = (
-    '2020, James R. Barlow. Licensed under Creative Commons Attribution-ShareAlike 4.0.'
+    '2022, James R. Barlow. Licensed under Creative Commons Attribution-ShareAlike 4.0.'
 )
 author = 'James R. Barlow'
 
@@ -82,15 +91,16 @@ if on_rtd:
         'pikepdf',
         'pikepdf.models',
         'pikepdf.models.metadata',
-        'ocrmypdf.leptonica',
     ]
     sys.modules.update((mod_name, Mock()) for mod_name in MOCK_MODULES)
 
-
-from pkg_resources import get_distribution, DistributionNotFound
+try:
+    from importlib_metadata import version as package_version
+except ModuleNotFoundError:
+    from importlib.metadata import version as package_version
 
 # The full version, including alpha/beta/rc tags.
-release = get_distribution('ocrmypdf').version
+release = package_version('ocrmypdf')
 version = '.'.join(release.split('.')[:2])
 
 
@@ -273,7 +283,7 @@ htmlhelp_basename = 'ocrmypdfdoc'
 
 # -- Options for LaTeX output ---------------------------------------------
 
-latex_elements = {
+latex_elements = {  # type: ignore
     # The paper size ('letterpaper' or 'a4paper').
     #
     # 'papersize': 'letterpaper',

docs/contributing.rst

@@ -46,17 +46,6 @@ Style guide: Is it OCRmyPDF or ocrmypdf?
 
 The program/project is OCRmyPDF and the name of the executable or library is ocrmypdf.
 
-Known ports/packagers
-=====================
-
-OCRmyPDF has been ported to many platforms already. If you are interesting in
-porting to a new platform, check with
-`Repology <https://repology.org/projects/?search=ocrmypdf>`__ to see the status
-of that platform.
-
-Packager maintainers, please ensure that the command line completion scripts in
-``misc/`` are installed.
-
 Copyright and license
 =====================
 

docs/cookbook.rst

@@ -58,9 +58,11 @@ portrait pages.
 You can increase (decrease) the parameter ``--rotate-pages-threshold``
 to make page rotation more (less) aggressive. The threshold number is the ratio
 of how confidence the OCR engine is that the document image should be changed,
-compared to kept the same. A value of ``15.0`` is the default, and is fairly
-conservative. A value of ``2.0`` will produce more rotations, and more false
-positives.
+compared to kept the same. The default value is quite conservative; on some files
+it may not attempt rotations at all unless it is very confident that the current
+rotation is wrong. A lower value of ``2.0`` will produce more rotations, and
+more false positives. Run with ``-v1`` to see the confidence level for each
+page to see if there may be a better value for your files.
 
 If the page is "just a little off horizontal", like a crooked picture,
 then you want ``--deskew``. ``--rotate-pages`` is for when the cardinal
@@ -102,6 +104,9 @@ This produces a file named "output.pdf" and a companion text file named
     because of options like ``--skip-big`` or ``--tesseract-timeout``, those pages
     will not be in the sidecar.
 
+    If you don't want to generate the output PDF, use ``--output-type=none`` to
+    avoid generating one. Set the output filename to ``-`` (i.e. redirect to stdout).
+
     To extract all text from a PDF, whether generated from OCR or otherwise,
     use a program like Poppler's ``pdftotext`` or ``pdfgrep``.
 
@@ -179,10 +184,7 @@ might remove desirable content, especially from poor quality scans.
    ignored. This should not be used on documents that contain color
    photos as it may remove them.
 -  ``--deskew`` will correct pages were scanned at a skewed angle by
-   rotating them back into place. Skew determination and correction is
-   performed using `Postl's variance of line
-   sums <http://www.leptonica.org/skew-measurement.html>`__ algorithm as
-   implemented in `Leptonica <http://www.leptonica.org/index.html>`__.
+   rotating them back into place.
 -  ``--clean`` uses
    `unpaper <https://www.flameeyes.eu/projects/unpaper>`__ to clean up
    pages before OCR, but does not alter the final output. This makes it
@@ -198,7 +200,7 @@ might remove desirable content, especially from poor quality scans.
 
 .. warning::
 
-   ``--clean-final`` and ``-remove-background`` may leave undesirable
+   ``--clean-final`` and ``--remove-background`` may leave undesirable
    visual artifacts in some images where their algorithms have
    shortcomings. Files should be visually reviewed after using these
    options.
@@ -241,10 +243,11 @@ You can also optimize all images without performing any OCR:
 
     ocrmypdf --tesseract-timeout=0 --optimize 3 --skip-text input.pdf output.pdf
 
-Perform OCR only certain pages
-------------------------------
+Process only certain pages
+--------------------------
 
-You can ask OCRmyPDF to only apply OCR to certain pages.
+You can ask OCRmyPDF to only apply `image processing <#image-processing>`__
+and OCR to certain pages.
 
 .. code-block:: bash
 
@@ -258,10 +261,10 @@ overlap pages. OCRmyPDF does not currently account for document page numbers,
 such as an introduction section of a book that uses Roman numerals. It simply
 counts the number of virtual pieces of paper since the start.
 
-Regardless of the argument to ``--pages``, OCRmyPDF will optimize all pages in
-the file and convert it to PDF/A, unless you disable those options. In this
-example, we want to OCR only the title and otherwise change the PDF as little
-as possible:
+Regardless of the argument to ``--pages``, OCRmyPDF will optimize all pages/images
+in the file and convert it to PDF/A, unless you disable those options. Both of these
+steps are "whole file" operations. In this example, we want to OCR only the title
+and otherwise change the PDF as little as possible:
 
 .. code-block:: bash
 

docs/docker.rst

@@ -59,7 +59,7 @@ Using the Docker image on the command line
 ==========================================
 
 **Unlike typical Docker containers**, in this section the OCRmyPDF Docker
-container is emphemeral – it runs for one OCR job and terminates, just like a
+container is ephemeral – it runs for one OCR job and terminates, just like a
 command line program. We are using Docker to deliver an application (as opposed
 to the more conventional case, where a Docker container runs as a server).
 
@@ -103,16 +103,41 @@ Adding languages to the Docker image
 By default the Docker image includes English, German, Simplified Chinese,
 French, Portuguese and Spanish, the most popular languages for OCRmyPDF
 users based on feedback. You may add other languages by creating a new
-Dockerfile based on the public one:
+Dockerfile based on the public one.
 
 .. code-block:: dockerfile
 
    FROM jbarlow83/ocrmypdf
 
-   # Add French
-   RUN apt install tesseract-ocr-fra
+   # Example: add Italian
+   RUN apt install tesseract-ocr-ita
 
-You can also copy training data to ``/usr/share/tesseract-ocr/<tesseract version>/tessdata``.
+To install language packs (training data) such as the 
+`tessdata_best <https://github.com/tesseract-ocr/tessdata_best>`_ suite or
+custom data, you first need to determine the version of Tesseract data files, which
+may differ from the Tesseract program version. Use this command to determine the data
+file version:
+
+.. code-block:: bash
+
+   docker run -i --rm --entrypoint /bin/ls jbarlow83/ocrmypdf /usr/share/tesseract-ocr
+
+As of 2021, the data file version is probably ``4.00``.
+
+You can then add new data with either a Dockerfile:
+
+.. code-block:: dockerfile
+
+   FROM jbarlow83/ocrmypdf
+
+   # Example: add a tessdata_best file
+   COPY chi_tra_vert.traineddata /usr/share/tesseract-ocr/<data version>/tessdata/
+
+Alternately, you can copy training data into a Docker container as follows:
+
+.. code-block:: bash
+
+   docker cp mycustomtraining.traineddata name_of_container:/usr/share/tesseract-ocr/<tesseract version>/tessdata/
 
 Executing the test suite
 ========================

docs/images/logo-square-256.svg

@@ -0,0 +1,239 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg
+   width="256"
+   height="256"
+   viewBox="0 0 256 256.00001"
+   version="1.1"
+   xml:space="preserve"
+   style="clip-rule:evenodd;fill-rule:evenodd;stroke-linecap:round;stroke-linejoin:round;stroke-miterlimit:1.5"
+   id="svg270"
+   sodipodi:docname="logo-square-256.svg"
+   inkscape:export-filename="/home/jb/src/ocrmypdf/docs/images/logo-square.png"
+   inkscape:export-xdpi="96"
+   inkscape:export-ydpi="96"
+   inkscape:version="1.1.2 (0a00cf5339, 2022-02-04)"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:serif="http://www.serif.com/"><metadata
+   id="metadata276"><rdf:RDF><cc:Work
+       rdf:about=""><dc:format>image/svg+xml</dc:format><dc:type
+         rdf:resource="http://purl.org/dc/dcmitype/StillImage" /></cc:Work></rdf:RDF></metadata><defs
+   id="defs274" /><sodipodi:namedview
+   pagecolor="#ffffff"
+   bordercolor="#666666"
+   borderopacity="1"
+   objecttolerance="10"
+   gridtolerance="10"
+   guidetolerance="10"
+   inkscape:pageopacity="0"
+   inkscape:pageshadow="2"
+   inkscape:window-width="2396"
+   inkscape:window-height="1691"
+   id="namedview272"
+   showgrid="false"
+   lock-margins="false"
+   inkscape:zoom="2.0079523"
+   inkscape:cx="189.74554"
+   inkscape:cy="54.533168"
+   inkscape:window-x="26"
+   inkscape:window-y="23"
+   inkscape:window-maximized="0"
+   inkscape:current-layer="svg270"
+   inkscape:pagecheckerboard="0"
+   width="256px"
+   fit-margin-top="0"
+   fit-margin-left="0"
+   fit-margin-right="0"
+   fit-margin-bottom="0" />
+    <g
+   id="svg"
+   transform="matrix(0.48534351,0,0,0.4057699,1.8106874,71.192214)">
+        <rect
+   x="0"
+   y="0"
+   width="520"
+   height="280"
+   style="fill:#ffffff"
+   id="rect188" />
+        <g
+   transform="matrix(1.03522,0,0,1.23823,-69.7528,-83.422)"
+   id="g267">
+            <g
+   transform="translate(243.977,20.0703)"
+   id="g218">
+                <g
+   id="Page">
+                    <g
+   transform="matrix(0.961773,0,0,1.05962,6.19811,-3.01071)"
+   id="g192">
+                        <path
+   d="m 328.5,97.682 c 0,-1.217 -0.517,-2.386 -1.444,-3.264 -7.03,-6.66 -37.614,-35.638 -44.828,-42.474 -0.977,-0.925 -2.327,-1.448 -3.738,-1.448 -13.997,0 -90.407,0 -111.151,0 -2.871,0 -5.198,2.113 -5.198,4.718 0,27.837 0,170.351 0,198.186 0,2.605 2.327,4.717 5.197,4.717 24.904,0 131.821,0 156.2,0 2.74,0 4.962,-2.016 4.962,-4.504 0,-24.345 0,-139.717 0,-155.931 z"
+   style="fill:#fdfdfd;stroke:#333333;stroke-width:3.95px"
+   id="path190" />
+                    </g>
+                    <g
+   id="Dog-ear"
+   serif:id="Dog ear"
+   transform="translate(-4,2)">
+                        <path
+   d="m 277.072,48.496 v 45.352 c 0,1.324 0.526,2.593 1.462,3.529 0.936,0.936 2.205,1.462 3.529,1.462 12.485,0 44.078,0 44.078,0"
+   style="fill:#f5f5f5;stroke:#333333;stroke-width:4px"
+   id="path194" />
+                    </g>
+                </g>
+                <g
+   transform="translate(-29.6816,-0.395178)"
+   id="g216">
+                    <g
+   transform="matrix(1.00243,0,0,1.11818,-144.72,-8.80181)"
+   id="g200">
+                        <path
+   d="m 465.73,119.654 c 0,-2.049 -1.856,-3.713 -4.142,-3.713 H 310.259 c -2.286,0 -4.142,1.664 -4.142,3.713 v 63.454 c 0,2.049 1.856,3.713 4.142,3.713 h 151.329 c 2.286,0 4.142,-1.664 4.142,-3.713 z"
+   style="fill:#f80000;stroke:#ffffff;stroke-width:3.77px"
+   id="path198" />
+                    </g>
+                    <g
+   transform="matrix(1.24571,0,0,1.35864,116.812,84.3924)"
+   id="g214">
+                        <g
+   transform="matrix(64,0,0,64,42.1437,77.6203)"
+   id="g204">
+                            <path
+   d="m 0.084,0 v -0.68 h 0.213 c 0.074,0 0.137,0.017 0.19,0.05 0.053,0.034 0.079,0.09 0.079,0.168 0,0.077 -0.028,0.134 -0.085,0.17 -0.057,0.037 -0.121,0.055 -0.193,0.055 H 0.213 V 0 Z m 0.209,-0.572 h -0.08 v 0.228 h 0.082 c 0.039,0 0.07,-0.009 0.094,-0.027 0.024,-0.017 0.037,-0.045 0.04,-0.083 0,-0.044 -0.012,-0.075 -0.036,-0.092 -0.024,-0.017 -0.057,-0.026 -0.1,-0.026 z"
+   style="fill:#ffffff;fill-rule:nonzero"
+   id="path202" />
+                        </g>
+                        <g
+   transform="matrix(64,0,0,64,79.7117,77.6203)"
+   id="g208">
+                            <path
+   d="M 0.332,0 H 0.084 v -0.68 h 0.252 c 0.105,0 0.182,0.032 0.233,0.095 0.051,0.063 0.076,0.144 0.076,0.241 0,0.105 -0.027,0.189 -0.082,0.251 C 0.508,-0.031 0.431,0 0.332,0 Z M 0.337,-0.57 H 0.213 v 0.461 H 0.33 c 0.055,0 0.099,-0.018 0.132,-0.054 C 0.495,-0.199 0.511,-0.259 0.511,-0.344 0.511,-0.415 0.497,-0.47 0.469,-0.51 0.441,-0.55 0.397,-0.57 0.337,-0.57 Z"
+   style="fill:#ffffff;fill-rule:nonzero"
+   id="path206" />
+                        </g>
+                        <g
+   transform="matrix(64,0,0,64,123.424,77.6203)"
+   id="g212">
+                            <path
+   d="M 0.405,-0.288 H 0.213 V 0 H 0.084 v -0.68 h 0.385 l 0.02,0.102 H 0.213 v 0.189 h 0.173 z"
+   style="fill:#ffffff;fill-rule:nonzero"
+   id="path210" />
+                        </g>
+                    </g>
+                </g>
+            </g>
+            <g
+   transform="matrix(1,0,0,1.52217,67.3796,10.7507)"
+   id="g222">
+                <rect
+   x="23.500999"
+   y="81.300003"
+   width="162.30499"
+   height="61.77"
+   style="fill:#b4d5ff"
+   id="rect220" />
+            </g>
+            <g
+   transform="matrix(0.967536,0,0,0.961535,5.90498,47.9703)"
+   id="g236">
+                <g
+   transform="matrix(90.4804,0,0,90.4804,82.6698,167.705)"
+   id="g226">
+                    <path
+   d="m 0.057,-0.337 c 0,-0.105 0.027,-0.19 0.082,-0.257 0.055,-0.066 0.132,-0.1 0.231,-0.102 0.107,0 0.186,0.034 0.237,0.103 0.051,0.069 0.077,0.152 0.077,0.249 0,0.105 -0.027,0.191 -0.082,0.258 -0.055,0.067 -0.133,0.1 -0.232,0.1 C 0.264,0.014 0.185,-0.02 0.134,-0.089 0.083,-0.157 0.057,-0.24 0.057,-0.337 Z m 0.135,-0.001 c 0,0.071 0.014,0.13 0.043,0.175 0.029,0.045 0.073,0.068 0.134,0.068 0.055,0 0.098,-0.02 0.131,-0.061 0.033,-0.041 0.049,-0.103 0.049,-0.188 0,-0.071 -0.014,-0.129 -0.043,-0.174 -0.029,-0.045 -0.073,-0.068 -0.134,-0.068 -0.053,0 -0.097,0.022 -0.13,0.067 -0.033,0.045 -0.05,0.105 -0.05,0.181 z"
+   style="fill:#333333;fill-rule:nonzero"
+   id="path224" />
+                </g>
+                <g
+   transform="matrix(90.4804,0,0,90.4804,147.906,167.705)"
+   id="g230">
+                    <path
+   d="M 0.505,-0.557 C 0.473,-0.567 0.448,-0.574 0.429,-0.579 0.41,-0.583 0.388,-0.585 0.361,-0.585 c -0.054,0 -0.096,0.022 -0.125,0.066 -0.029,0.044 -0.044,0.104 -0.044,0.181 0,0.066 0.012,0.123 0.037,0.171 0.025,0.048 0.066,0.072 0.124,0.072 0.029,0 0.056,-0.003 0.081,-0.009 0.025,-0.006 0.047,-0.013 0.068,-0.022 L 0.551,-0.03 C 0.525,-0.017 0.494,-0.006 0.457,0.002 0.42,0.01 0.388,0.014 0.36,0.014 0.254,0.014 0.177,-0.02 0.129,-0.088 0.081,-0.156 0.057,-0.239 0.057,-0.337 c 0,-0.105 0.027,-0.19 0.08,-0.257 0.053,-0.067 0.129,-0.1 0.228,-0.1 0.02,0 0.048,0.003 0.083,0.01 0.035,0.007 0.068,0.018 0.097,0.034 z"
+   style="fill:#333333;fill-rule:nonzero"
+   id="path228" />
+                </g>
+                <g
+   transform="matrix(90.4804,0,0,90.4804,199.751,167.705)"
+   id="g234">
+                    <path
+   d="m 0.293,-0.572 h -0.08 v 0.208 h 0.082 c 0.039,0 0.071,-0.008 0.096,-0.024 0.025,-0.015 0.038,-0.041 0.038,-0.077 0,-0.038 -0.012,-0.065 -0.036,-0.082 -0.024,-0.017 -0.057,-0.025 -0.1,-0.025 z M 0.479,0 0.335,-0.26 C 0.328,-0.259 0.32,-0.259 0.312,-0.259 0.304,-0.258 0.296,-0.258 0.288,-0.258 H 0.213 V 0 H 0.084 v -0.68 h 0.213 c 0.074,0 0.137,0.017 0.19,0.051 0.053,0.034 0.079,0.087 0.079,0.158 0,0.042 -0.011,0.078 -0.032,0.108 -0.022,0.031 -0.05,0.054 -0.084,0.071 L 0.617,0 Z"
+   style="fill:#333333;fill-rule:nonzero"
+   id="path232" />
+                </g>
+            </g>
+            <g
+   transform="matrix(0.916882,0,0,1,121.475,-32.6535)"
+   id="g246">
+                <g
+   transform="matrix(86.953,0,0,86.953,152.996,241.878)"
+   id="g240">
+                    <path
+   d="M 0.479,-0.428 C 0.5,-0.451 0.527,-0.47 0.562,-0.484 c 0.034,-0.013 0.065,-0.02 0.092,-0.02 0.066,0 0.113,0.019 0.141,0.058 0.027,0.039 0.041,0.086 0.041,0.142 V 0 H 0.705 v -0.298 c 0,-0.031 -0.007,-0.054 -0.022,-0.071 -0.015,-0.016 -0.036,-0.024 -0.064,-0.024 -0.019,0 -0.038,0.005 -0.059,0.015 -0.021,0.01 -0.039,0.021 -0.056,0.034 0.001,0.007 0.001,0.013 0.002,0.02 0.001,0.007 0.001,0.013 0.001,0.02 V 0 H 0.376 v -0.298 c 0,-0.031 -0.007,-0.054 -0.022,-0.071 -0.015,-0.016 -0.036,-0.024 -0.063,-0.024 -0.017,0 -0.033,0.003 -0.05,0.01 -0.017,0.007 -0.034,0.016 -0.049,0.027 V 0 H 0.062 V -0.485 H 0.13 l 0.032,0.044 c 0.022,-0.02 0.049,-0.035 0.08,-0.047 0.031,-0.011 0.058,-0.016 0.083,-0.016 0.038,0 0.07,0.007 0.095,0.02 0.025,0.014 0.045,0.033 0.059,0.056 z"
+   style="fill:#333333;fill-rule:nonzero"
+   id="path238" />
+                </g>
+                <g
+   transform="matrix(86.953,0,0,86.953,228.906,241.878)"
+   id="g244">
+                    <path
+   d="M 0.156,0.023 0.179,-0.034 0.006,-0.467 0.14,-0.485 0.252,-0.191 0.358,-0.485 H 0.495 L 0.278,0.064 C 0.263,0.103 0.236,0.137 0.197,0.165 0.158,0.193 0.118,0.212 0.075,0.222 L 0.029,0.115 C 0.052,0.105 0.077,0.093 0.104,0.079 0.13,0.064 0.147,0.046 0.156,0.023 Z"
+   style="fill:#333333;fill-rule:nonzero"
+   id="path242" />
+                </g>
+            </g>
+            <g
+   id="Selectors"
+   transform="matrix(0.965977,0,0,0.807602,67.3796,67.3718)">
+                <g
+   id="Right-selector"
+   serif:id="Right selector">
+                    <g
+   transform="matrix(1.03522,0,0,1.23823,2.07044,0)"
+   id="g250">
+                        <path
+   d="M 185.806,161.156 V 67.132"
+   style="fill:none;stroke:#4c9fff;stroke-width:4px;stroke-linecap:butt"
+   id="path248" />
+                    </g>
+                    <g
+   transform="matrix(1.03522,0,0,1.23823,161.788,169.469)"
+   id="g254">
+                        <circle
+   cx="31.523001"
+   cy="34.313999"
+   r="10.021"
+   style="fill:#4c9fff;stroke:#4c9fff;stroke-width:4px;stroke-linecap:butt"
+   id="circle252" />
+                    </g>
+                </g>
+                <g
+   id="Left-selector"
+   serif:id="Left selector">
+                    <g
+   transform="matrix(1.03522,0,0,1.23823,-170.092,0)"
+   id="g259">
+                        <path
+   d="M 185.806,161.156 V 67.132"
+   style="fill:none;stroke:#4c9fff;stroke-width:4px;stroke-linecap:butt"
+   id="path257" />
+                    </g>
+                    <g
+   transform="matrix(1.03522,0,0,1.23823,-10.3742,28.2274)"
+   id="g263">
+                        <circle
+   cx="31.523001"
+   cy="34.313999"
+   r="10.021"
+   style="fill:#4c9fff;stroke:#4c9fff;stroke-width:4px;stroke-linecap:butt"
+   id="circle261" />
+                    </g>
+                </g>
+            </g>
+        </g>
+    </g>
+</svg>

docs/images/logo-square.svg

@@ -0,0 +1,233 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg
+   xmlns:serif="http://www.serif.com/"
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="503"
+   height="503"
+   viewBox="0 0 503 503"
+   version="1.1"
+   xml:space="preserve"
+   style="clip-rule:evenodd;fill-rule:evenodd;stroke-linecap:round;stroke-linejoin:round;stroke-miterlimit:1.5"
+   id="svg270"
+   sodipodi:docname="logo-square.svg"
+   inkscape:export-filename="/home/jb/src/ocrmypdf/docs/images/logo-square.png"
+   inkscape:export-xdpi="96"
+   inkscape:export-ydpi="96"
+   inkscape:version="1.0.2 (e86c870879, 2021-01-15)"><metadata
+   id="metadata276"><rdf:RDF><cc:Work
+       rdf:about=""><dc:format>image/svg+xml</dc:format><dc:type
+         rdf:resource="http://purl.org/dc/dcmitype/StillImage" /><dc:title></dc:title></cc:Work></rdf:RDF></metadata><defs
+   id="defs274" /><sodipodi:namedview
+   pagecolor="#ffffff"
+   bordercolor="#666666"
+   borderopacity="1"
+   objecttolerance="10"
+   gridtolerance="10"
+   guidetolerance="10"
+   inkscape:pageopacity="0"
+   inkscape:pageshadow="2"
+   inkscape:window-width="2396"
+   inkscape:window-height="1691"
+   id="namedview272"
+   showgrid="false"
+   lock-margins="false"
+   inkscape:zoom="2.0079523"
+   inkscape:cx="251.5"
+   inkscape:cy="193.18317"
+   inkscape:window-x="26"
+   inkscape:window-y="23"
+   inkscape:window-maximized="0"
+   inkscape:current-layer="svg270" />
+    <g
+   id="svg"
+   transform="matrix(0.965977,0,0,0.807602,0,138.43572)">
+        <rect
+   x="0"
+   y="0"
+   width="520"
+   height="280"
+   style="fill:#ffffff"
+   id="rect188" />
+        <g
+   transform="matrix(1.03522,0,0,1.23823,-69.7528,-83.422)"
+   id="g267">
+            <g
+   transform="translate(243.977,20.0703)"
+   id="g218">
+                <g
+   id="Page">
+                    <g
+   transform="matrix(0.961773,0,0,1.05962,6.19811,-3.01071)"
+   id="g192">
+                        <path
+   d="m 328.5,97.682 c 0,-1.217 -0.517,-2.386 -1.444,-3.264 -7.03,-6.66 -37.614,-35.638 -44.828,-42.474 -0.977,-0.925 -2.327,-1.448 -3.738,-1.448 -13.997,0 -90.407,0 -111.151,0 -2.871,0 -5.198,2.113 -5.198,4.718 0,27.837 0,170.351 0,198.186 0,2.605 2.327,4.717 5.197,4.717 24.904,0 131.821,0 156.2,0 2.74,0 4.962,-2.016 4.962,-4.504 0,-24.345 0,-139.717 0,-155.931 z"
+   style="fill:#fdfdfd;stroke:#333333;stroke-width:3.95px"
+   id="path190" />
+                    </g>
+                    <g
+   id="Dog-ear"
+   serif:id="Dog ear"
+   transform="translate(-4,2)">
+                        <path
+   d="m 277.072,48.496 v 45.352 c 0,1.324 0.526,2.593 1.462,3.529 0.936,0.936 2.205,1.462 3.529,1.462 12.485,0 44.078,0 44.078,0"
+   style="fill:#f5f5f5;stroke:#333333;stroke-width:4px"
+   id="path194" />
+                    </g>
+                </g>
+                <g
+   transform="translate(-29.6816,-0.395178)"
+   id="g216">
+                    <g
+   transform="matrix(1.00243,0,0,1.11818,-144.72,-8.80181)"
+   id="g200">
+                        <path
+   d="m 465.73,119.654 c 0,-2.049 -1.856,-3.713 -4.142,-3.713 H 310.259 c -2.286,0 -4.142,1.664 -4.142,3.713 v 63.454 c 0,2.049 1.856,3.713 4.142,3.713 h 151.329 c 2.286,0 4.142,-1.664 4.142,-3.713 z"
+   style="fill:#f80000;stroke:#ffffff;stroke-width:3.77px"
+   id="path198" />
+                    </g>
+                    <g
+   transform="matrix(1.24571,0,0,1.35864,116.812,84.3924)"
+   id="g214">
+                        <g
+   transform="matrix(64,0,0,64,42.1437,77.6203)"
+   id="g204">
+                            <path
+   d="m 0.084,0 v -0.68 h 0.213 c 0.074,0 0.137,0.017 0.19,0.05 0.053,0.034 0.079,0.09 0.079,0.168 0,0.077 -0.028,0.134 -0.085,0.17 -0.057,0.037 -0.121,0.055 -0.193,0.055 H 0.213 V 0 Z m 0.209,-0.572 h -0.08 v 0.228 h 0.082 c 0.039,0 0.07,-0.009 0.094,-0.027 0.024,-0.017 0.037,-0.045 0.04,-0.083 0,-0.044 -0.012,-0.075 -0.036,-0.092 -0.024,-0.017 -0.057,-0.026 -0.1,-0.026 z"
+   style="fill:#ffffff;fill-rule:nonzero"
+   id="path202" />
+                        </g>
+                        <g
+   transform="matrix(64,0,0,64,79.7117,77.6203)"
+   id="g208">
+                            <path
+   d="M 0.332,0 H 0.084 v -0.68 h 0.252 c 0.105,0 0.182,0.032 0.233,0.095 0.051,0.063 0.076,0.144 0.076,0.241 0,0.105 -0.027,0.189 -0.082,0.251 C 0.508,-0.031 0.431,0 0.332,0 Z M 0.337,-0.57 H 0.213 v 0.461 H 0.33 c 0.055,0 0.099,-0.018 0.132,-0.054 C 0.495,-0.199 0.511,-0.259 0.511,-0.344 0.511,-0.415 0.497,-0.47 0.469,-0.51 0.441,-0.55 0.397,-0.57 0.337,-0.57 Z"
+   style="fill:#ffffff;fill-rule:nonzero"
+   id="path206" />
+                        </g>
+                        <g
+   transform="matrix(64,0,0,64,123.424,77.6203)"
+   id="g212">
+                            <path
+   d="M 0.405,-0.288 H 0.213 V 0 H 0.084 v -0.68 h 0.385 l 0.02,0.102 H 0.213 v 0.189 h 0.173 z"
+   style="fill:#ffffff;fill-rule:nonzero"
+   id="path210" />
+                        </g>
+                    </g>
+                </g>
+            </g>
+            <g
+   transform="matrix(1,0,0,1.52217,67.3796,10.7507)"
+   id="g222">
+                <rect
+   x="23.500999"
+   y="81.300003"
+   width="162.30499"
+   height="61.77"
+   style="fill:#b4d5ff"
+   id="rect220" />
+            </g>
+            <g
+   transform="matrix(0.967536,0,0,0.961535,5.90498,47.9703)"
+   id="g236">
+                <g
+   transform="matrix(90.4804,0,0,90.4804,82.6698,167.705)"
+   id="g226">
+                    <path
+   d="m 0.057,-0.337 c 0,-0.105 0.027,-0.19 0.082,-0.257 0.055,-0.066 0.132,-0.1 0.231,-0.102 0.107,0 0.186,0.034 0.237,0.103 0.051,0.069 0.077,0.152 0.077,0.249 0,0.105 -0.027,0.191 -0.082,0.258 -0.055,0.067 -0.133,0.1 -0.232,0.1 C 0.264,0.014 0.185,-0.02 0.134,-0.089 0.083,-0.157 0.057,-0.24 0.057,-0.337 Z m 0.135,-0.001 c 0,0.071 0.014,0.13 0.043,0.175 0.029,0.045 0.073,0.068 0.134,0.068 0.055,0 0.098,-0.02 0.131,-0.061 0.033,-0.041 0.049,-0.103 0.049,-0.188 0,-0.071 -0.014,-0.129 -0.043,-0.174 -0.029,-0.045 -0.073,-0.068 -0.134,-0.068 -0.053,0 -0.097,0.022 -0.13,0.067 -0.033,0.045 -0.05,0.105 -0.05,0.181 z"
+   style="fill:#333333;fill-rule:nonzero"
+   id="path224" />
+                </g>
+                <g
+   transform="matrix(90.4804,0,0,90.4804,147.906,167.705)"
+   id="g230">
+                    <path
+   d="M 0.505,-0.557 C 0.473,-0.567 0.448,-0.574 0.429,-0.579 0.41,-0.583 0.388,-0.585 0.361,-0.585 c -0.054,0 -0.096,0.022 -0.125,0.066 -0.029,0.044 -0.044,0.104 -0.044,0.181 0,0.066 0.012,0.123 0.037,0.171 0.025,0.048 0.066,0.072 0.124,0.072 0.029,0 0.056,-0.003 0.081,-0.009 0.025,-0.006 0.047,-0.013 0.068,-0.022 L 0.551,-0.03 C 0.525,-0.017 0.494,-0.006 0.457,0.002 0.42,0.01 0.388,0.014 0.36,0.014 0.254,0.014 0.177,-0.02 0.129,-0.088 0.081,-0.156 0.057,-0.239 0.057,-0.337 c 0,-0.105 0.027,-0.19 0.08,-0.257 0.053,-0.067 0.129,-0.1 0.228,-0.1 0.02,0 0.048,0.003 0.083,0.01 0.035,0.007 0.068,0.018 0.097,0.034 z"
+   style="fill:#333333;fill-rule:nonzero"
+   id="path228" />
+                </g>
+                <g
+   transform="matrix(90.4804,0,0,90.4804,199.751,167.705)"
+   id="g234">
+                    <path
+   d="m 0.293,-0.572 h -0.08 v 0.208 h 0.082 c 0.039,0 0.071,-0.008 0.096,-0.024 0.025,-0.015 0.038,-0.041 0.038,-0.077 0,-0.038 -0.012,-0.065 -0.036,-0.082 -0.024,-0.017 -0.057,-0.025 -0.1,-0.025 z M 0.479,0 0.335,-0.26 C 0.328,-0.259 0.32,-0.259 0.312,-0.259 0.304,-0.258 0.296,-0.258 0.288,-0.258 H 0.213 V 0 H 0.084 v -0.68 h 0.213 c 0.074,0 0.137,0.017 0.19,0.051 0.053,0.034 0.079,0.087 0.079,0.158 0,0.042 -0.011,0.078 -0.032,0.108 -0.022,0.031 -0.05,0.054 -0.084,0.071 L 0.617,0 Z"
+   style="fill:#333333;fill-rule:nonzero"
+   id="path232" />
+                </g>
+            </g>
+            <g
+   transform="matrix(0.916882,0,0,1,121.475,-32.6535)"
+   id="g246">
+                <g
+   transform="matrix(86.953,0,0,86.953,152.996,241.878)"
+   id="g240">
+                    <path
+   d="M 0.479,-0.428 C 0.5,-0.451 0.527,-0.47 0.562,-0.484 c 0.034,-0.013 0.065,-0.02 0.092,-0.02 0.066,0 0.113,0.019 0.141,0.058 0.027,0.039 0.041,0.086 0.041,0.142 V 0 H 0.705 v -0.298 c 0,-0.031 -0.007,-0.054 -0.022,-0.071 -0.015,-0.016 -0.036,-0.024 -0.064,-0.024 -0.019,0 -0.038,0.005 -0.059,0.015 -0.021,0.01 -0.039,0.021 -0.056,0.034 0.001,0.007 0.001,0.013 0.002,0.02 0.001,0.007 0.001,0.013 0.001,0.02 V 0 H 0.376 v -0.298 c 0,-0.031 -0.007,-0.054 -0.022,-0.071 -0.015,-0.016 -0.036,-0.024 -0.063,-0.024 -0.017,0 -0.033,0.003 -0.05,0.01 -0.017,0.007 -0.034,0.016 -0.049,0.027 V 0 H 0.062 V -0.485 H 0.13 l 0.032,0.044 c 0.022,-0.02 0.049,-0.035 0.08,-0.047 0.031,-0.011 0.058,-0.016 0.083,-0.016 0.038,0 0.07,0.007 0.095,0.02 0.025,0.014 0.045,0.033 0.059,0.056 z"
+   style="fill:#333333;fill-rule:nonzero"
+   id="path238" />
+                </g>
+                <g
+   transform="matrix(86.953,0,0,86.953,228.906,241.878)"
+   id="g244">
+                    <path
+   d="M 0.156,0.023 0.179,-0.034 0.006,-0.467 0.14,-0.485 0.252,-0.191 0.358,-0.485 H 0.495 L 0.278,0.064 C 0.263,0.103 0.236,0.137 0.197,0.165 0.158,0.193 0.118,0.212 0.075,0.222 L 0.029,0.115 C 0.052,0.105 0.077,0.093 0.104,0.079 0.13,0.064 0.147,0.046 0.156,0.023 Z"
+   style="fill:#333333;fill-rule:nonzero"
+   id="path242" />
+                </g>
+            </g>
+            <g
+   id="Selectors"
+   transform="matrix(0.965977,0,0,0.807602,67.3796,67.3718)">
+                <g
+   id="Right-selector"
+   serif:id="Right selector">
+                    <g
+   transform="matrix(1.03522,0,0,1.23823,2.07044,0)"
+   id="g250">
+                        <path
+   d="M 185.806,161.156 V 67.132"
+   style="fill:none;stroke:#4c9fff;stroke-width:4px;stroke-linecap:butt"
+   id="path248" />
+                    </g>
+                    <g
+   transform="matrix(1.03522,0,0,1.23823,161.788,169.469)"
+   id="g254">
+                        <circle
+   cx="31.523001"
+   cy="34.313999"
+   r="10.021"
+   style="fill:#4c9fff;stroke:#4c9fff;stroke-width:4px;stroke-linecap:butt"
+   id="circle252" />
+                    </g>
+                </g>
+                <g
+   id="Left-selector"
+   serif:id="Left selector">
+                    <g
+   transform="matrix(1.03522,0,0,1.23823,-170.092,0)"
+   id="g259">
+                        <path
+   d="M 185.806,161.156 V 67.132"
+   style="fill:none;stroke:#4c9fff;stroke-width:4px;stroke-linecap:butt"
+   id="path257" />
+                    </g>
+                    <g
+   transform="matrix(1.03522,0,0,1.23823,-10.3742,28.2274)"
+   id="g263">
+                        <circle
+   cx="31.523001"
+   cy="34.313999"
+   r="10.021"
+   style="fill:#4c9fff;stroke:#4c9fff;stroke-width:4px;stroke-linecap:butt"
+   id="circle261" />
+                    </g>
+                </g>
+            </g>
+        </g>
+    </g>
+</svg>

docs/index.rst

@@ -1,6 +1,8 @@
 OCRmyPDF documentation
 ======================
 
+.. figure:: images/logo.svg
+
 OCRmyPDF adds an optical character recognition (OCR) text layer to scanned PDF
 files, allowing them to be searched.
 
@@ -38,6 +40,7 @@ image processing and OCR to existing PDFs.
    plugins
    apiref
    contributing
+   maintainers
 
 Indices and tables
 ==================

docs/installation.rst

@@ -12,19 +12,23 @@ system/platform. This version may be out of date, however.
 
 These platforms have one-liner installs:
 
-+-----------------------------+-------------------------------+
-| Debian, Ubuntu              | ``apt install ocrmypdf``      |
-+-----------------------------+-------------------------------+
-| Windows Subsystem for Linux | ``apt install ocrmypdf``      |
-+-----------------------------+-------------------------------+
-| Fedora                      | ``dnf install ocrmypdf``      |
-+-----------------------------+-------------------------------+
-| macOS                       | ``brew install ocrmypdf``     |
-+-----------------------------+-------------------------------+
-| LinuxBrew                   | ``brew install ocrmypdf``     |
-+-----------------------------+-------------------------------+
-| FreeBSD                     | ``pkg install py37-ocrmypdf`` |
-+-----------------------------+-------------------------------+
++-------------------------------+-----------------------------------------+
+| Debian, Ubuntu                | ``apt install ocrmypdf``                |
++-------------------------------+-----------------------------------------+
+| Windows Subsystem for Linux   | ``apt install ocrmypdf``                |
++-------------------------------+-----------------------------------------+
+| Fedora                        | ``dnf install ocrmypdf``                |
++-------------------------------+-----------------------------------------+
+| macOS                         | ``brew install ocrmypdf``               |
++-------------------------------+-----------------------------------------+
+| LinuxBrew                     | ``brew install ocrmypdf``               |
++-------------------------------+-----------------------------------------+
+| FreeBSD                       | ``pkg install textproc/py-ocrmypdf``    |
++-------------------------------+-----------------------------------------+
+| Conda (WSL, macOS, Linux)     | ``conda install ocrmypdf``              |
++-------------------------------+-----------------------------------------+
+| Snap (snapcraft packaging)    | ``snap install ocrmypdf``               |
++-------------------------------+-----------------------------------------+
 
 More detailed procedures are outlined below. If you want to do a manual
 install, or install a more recent version than your platform provides, read on.
@@ -40,10 +44,10 @@ Debian and Ubuntu 18.04 or newer
 --------------------------------
 
 .. |deb-stable| image:: https://repology.org/badge/version-for-repo/debian_stable/ocrmypdf.svg
-    :alt: Debian 9 stable ("stretch")
+    :alt: Debian stable
 
 .. |deb-testing| image:: https://repology.org/badge/version-for-repo/debian_testing/ocrmypdf.svg
-    :alt: Debian 10 testing ("buster")
+    :alt: Debian testing
 
 .. |deb-unstable| image:: https://repology.org/badge/version-for-repo/debian_unstable/ocrmypdf.svg
     :alt: Debian unstable
@@ -54,6 +58,9 @@ Debian and Ubuntu 18.04 or newer
 .. |ubu-2004| image:: https://repology.org/badge/version-for-repo/ubuntu_20_04/ocrmypdf.svg
     :alt: Ubuntu 20.04 LTS
 
+.. |ubu-2204| image:: https://repology.org/badge/version-for-repo/ubuntu_22_04/ocrmypdf.svg
+    :alt: Ubuntu 22.04 LTS
+
 +-----------------------------------------------+
 | **OCRmyPDF versions in Debian & Ubuntu**      |
 +-----------------------------------------------+
@@ -61,7 +68,7 @@ Debian and Ubuntu 18.04 or newer
 +-----------------------------------------------+
 | |deb-stable| |deb-testing| |deb-unstable|     |
 +-----------------------------------------------+
-| |ubu-1804| |ubu-2004|                         |
+| |ubu-1804| |ubu-2004| |ubu-2204|              |
 +-----------------------------------------------+
 
 Users of Debian 9 ("stretch") or later, or Ubuntu 18.04 or later, including users
@@ -75,8 +82,7 @@ As indicated in the table above, Debian and Ubuntu releases may lag
 behind the latest version. If the version available for your platform is
 out of date, you could opt to install the latest version from source.
 See `Installing HEAD revision from
-sources <#installing-head-revision-from-sources>`__. Ubuntu 16.10 to 17.10
-inclusive also had ocrmypdf, but these versions are end of life.
+sources <#installing-head-revision-from-sources>`__.
 
 For full details on version availability for your platform, check the
 `Debian Package Tracker <https://tracker.debian.org/pkg/ocrmypdf>`__ or
@@ -86,18 +92,18 @@ For full details on version availability for your platform, check the
 
    OCRmyPDF for Debian and Ubuntu currently omit the JBIG2 encoder.
    OCRmyPDF works fine without it but will produce larger output files.
-   If you build jbig2enc from source, ocrmypdf 7.0.0 and later will
+   If you build jbig2enc from source, ocrmypdf will
    automatically detect it (specifically the ``jbig2`` binary) on the
    ``PATH``. To add JBIG2 encoding, see :ref:`jbig2`.
 
-Fedora 29 or newer
-------------------
+Fedora
+------
 
-.. |fedora-31| image:: https://repology.org/badge/version-for-repo/fedora_31/ocrmypdf.svg
-    :alt: Fedora 31
+.. |fedora-35| image:: https://repology.org/badge/version-for-repo/fedora_35/ocrmypdf.svg
+    :alt: Fedora 35
 
-.. |fedora-32| image:: https://repology.org/badge/version-for-repo/fedora_32/ocrmypdf.svg
-    :alt: Fedora 32
+.. |fedora-36| image:: https://repology.org/badge/version-for-repo/fedora_36/ocrmypdf.svg
+    :alt: Fedora 36
 
 .. |fedora-rawhide| image:: https://repology.org/badge/version-for-repo/fedora_rawhide/ocrmypdf.svg
     :alt: Fedore Rawhide
@@ -107,7 +113,7 @@ Fedora 29 or newer
 +-----------------------------------------------+
 | |latest|                                      |
 +-----------------------------------------------+
-| |fedora-31| |fedora-32| |fedora-rawhide|      |
+| |fedora-35| |fedora-36| |fedora-rawhide|      |
 +-----------------------------------------------+
 
 Users of Fedora 29 or later may simply
@@ -133,9 +139,29 @@ from sources <#installing-head-revision-from-sources>`__.
 
 .. _ubuntu-lts-latest:
 
-Installing the latest version on Ubuntu 20.04 LTS
+Installing the latest version on Ubuntu 22.04 LTS
 -------------------------------------------------
 
+Ubuntu 22.04 includes ocrmypdf 13.4.0 - you can install that with 
+``apt install ocrmypdf``. To install a more recent version for the current
+user, follow these steps:
+
+.. code-block:: bash
+
+    sudo apt-get update
+    sudo apt-get -y install ocrmypdf python3-pip
+
+    pip install --user --upgrade ocrmypdf
+
+If you get the message ``WARNING: The script ocrmypdf is installed in 
+'/home/$USER/.local/bin' which is not on PATH.``, you may need to re-login
+or open a new shell, or manually add this to your user's PATH.
+
+To add JBIG2 encoding, see :ref:`jbig2`.
+
+Ubuntu 20.04 LTS
+----------------
+
 Ubuntu 20.04 includes ocrmypdf 9.6.0 - you can install that with ``apt``. To
 install a more recent version, uninstall the system-provided version of
 ocrmypdf, and install the following dependencies:
@@ -147,7 +173,6 @@ ocrmypdf, and install the following dependencies:
     sudo apt-get -y install \
         ghostscript \
         icc-profiles-free \
-        liblept5 \
         libxml2 \
         pngquant \
         python3-pip \
@@ -158,7 +183,7 @@ To install ocrmypdf for the system:
 
 .. code-block:: bash
 
-    sudo pip3 install ocrmypdf
+    pip3 install ocrmypdf
 
 To install for the current user only:
 
@@ -167,6 +192,8 @@ To install for the current user only:
     export PATH=$HOME/.local/bin:$PATH
     pip3 install --user ocrmypdf
 
+To add JBIG2 encoding, see :ref:`jbig2`.
+
 Ubuntu 18.04 LTS
 ----------------
 
@@ -181,10 +208,8 @@ of ocrmypdf, and install the following dependencies:
     sudo apt-get -y install \
         ghostscript \
         icc-profiles-free \
-        liblept5 \
         libxml2 \
         pngquant \
-        python3-cffi \
         python3-distutils \
         python3-pkg-resources \
         python3-reportlab \
@@ -209,69 +234,6 @@ user's ``PATH`` to check for the user's Python packages.
 
 To add JBIG2 encoding, see :ref:`jbig2`.
 
-Ubuntu 16.04 LTS
-----------------
-
-No package is available for Ubuntu 16.04. OCRmyPDF 8.0 and newer require
-Python 3.6. Ubuntu 16.04 ships Python 3.5, but you can install Python
-3.6 on it. Or, you can skip Python 3.6 and install OCRmyPDF 7.x or older
-- for that procedure, please see the installation documentation for the
-version of OCRmyPDF you plan to use.
-
-**Install system packages for OCRmyPDF**
-
-.. code-block:: bash
-
-    sudo apt-get update
-    sudo apt-get install -y software-properties-common python-software-properties
-    sudo add-apt-repository -y \
-        ppa:jonathonf/python-3.6 \
-        ppa:alex-p/tesseract-ocr
-    sudo apt-get update
-    sudo apt-get install -y \
-        ghostscript \
-        libexempi3 \
-        libffi6 \
-        pngquant \
-        python3.6 \
-        qpdf \
-        tesseract-ocr \
-        unpaper
-
-This will install a Python 3.6 binary at ``/usr/bin/python3.6``
-alongside the system's Python 3.5. Do not remove the system Python. This
-will also install Tesseract 4.0 from a PPA, since the version available
-in Ubuntu 16.04 is too old for OCRmyPDF.
-
-Now install pip for Python 3.6. This will install the Python 3.6 version
-of ``pip`` at ``/usr/local/bin/pip``.
-
-.. code-block:: bash
-
-    curl https://bootstrap.pypa.io/get-pip.py | sudo python3.6
-
-**Install OCRmyPDF**
-
-OCRmyPDF requires the locale to be set for UTF-8. **On some minimal
-Ubuntu installations**, such as the Ubuntu 16.04 Docker images it may be
-necessary to set the locale.
-
-.. code-block:: bash
-
-    # Optional: Only need to set these if they are not already set
-    export LC_ALL=C.UTF-8
-    export LANG=C.UTF-8
-
-Now install OCRmyPDF for the current user, and ensure that the ``PATH``
-environment variable contains ``$HOME/.local/bin``.
-
-.. code-block:: bash
-
-    export PATH=$HOME/.local/bin:$PATH
-    pip3.6 install --user ocrmypdf
-
-To add JBIG2 encoding, see :ref:`jbig2`.
-
 Arch Linux (AUR)
 ----------------
 
@@ -352,43 +314,6 @@ To install OCRmyPDF for Alpine Linux:
 
     apk add ocrmypdf
 
-Mageia 7
---------
-
-Install the following dependencies:
-
-.. code-block:: bash
-
-    # As root user
-    urpmi.update -a
-    urpmi \
-        ghostscript \
-        icc-profiles-openicc \
-        jbig2dec \
-        lib64leptonica5 \
-        pngquant \
-        python3-pip \
-        python3-cffi \
-        python3-distutils-extra \
-        python3-pkg-resources \
-        python3-reportlab \
-        qpdf \
-        tesseract \
-        tesseract-osd \
-        tesseract-eng \
-        tesseract-fra
-
-To install ocrmypdf for the system:
-
-    # As root user
-    pip3 install ocrmypdf
-    ldconfig
-
-Or, to install for the current user only:
-
-    export PATH=$HOME/.local/bin:$PATH
-    pip3 install --user ocrmypdf
-
 Other Linux packages
 --------------------
 
@@ -423,19 +348,6 @@ languages you can optionally install them all:
 
     brew install tesseract-lang  # Optional: Install all language packs
 
-.. note::
-
-   Users who previously installed OCRmyPDF on macOS using
-   ``pip install ocrmypdf`` should remove the pip version
-   (``pip3 uninstall ocrmypdf``) before switching to the Homebrew
-   version.
-
-.. note::
-
-   Users who previously installed OCRmyPDF from the private tap should
-   switch to the mainline version (``brew untap jbarlow83/ocrmypdf``)
-   and install from there.
-
 Manual installation on macOS
 ----------------------------
 
@@ -454,7 +366,7 @@ Update Homebrew:
 
 Install or upgrade the required Homebrew packages, if any are missing.
 To do this, use ``brew edit ocrmypdf`` to obtain a recent list of Homebrew
-dependencies. You could also check the ``azure-pipelines.yml``.
+dependencies. You could also check the ``.workflows/build.yml``.
 
 This will include the English, French, German and Spanish language
 packs. If you need other languages you can optionally install them all:
@@ -469,19 +381,19 @@ Update the homebrew pip:
 
 .. code-block:: bash
 
-    pip3 install --upgrade pip
+    pip install --upgrade pip
 
 You can then install OCRmyPDF from PyPI, for the current user:
 
 .. code-block:: bash
 
-    pip3 install --user ocrmypdf
+    pip install --user ocrmypdf
 
 or system-wide:
 
 .. code-block:: bash
 
-    pip3 install ocrmypdf
+    pip install ocrmypdf
 
 The command line program should now be available:
 
@@ -536,14 +448,15 @@ to change the PATH.
 .. warning::
 
     As of early 2021, users have reported problems with the Microsoft Store version of
-    Python affected most third party Python packages including OCRmyPDF. Please use
-    Python downloaded from Python.org or Chocolatey as recommended here.
+    Python and OCRmyPDF. These issues affect many other third party Python packages.
+    Please download Python from Python.org or Chocolatey instead, and do not use the
+    Microsoft Store version.
 
 Windows Subsystem for Linux
 ---------------------------
 
-#. Install Ubuntu 18.04 for Windows Subsystem for Linux, if not already installed.
-#. Follow the procedure to install :ref:`OCRmyPDF on Ubuntu 18.04 <ubuntu-lts-latest>`.
+#. Install Ubuntu 22.04 for Windows Subsystem for Linux, if not already installed.
+#. Follow the procedure to install :ref:`OCRmyPDF on Ubuntu 22.04 <ubuntu-lts-latest>`.
 #. Open the Windows command prompt and create a symlink:
 
 .. code-block:: powershell
@@ -564,7 +477,7 @@ Cygwin64
 
 First install the the following prerequisite Cygwin packages using ``setup-x86_64.exe``::
 
-    python36 (or later)
+    python37 (or later)
     python3?-devel
     python3?-pip
     python3?-lxml
@@ -615,16 +528,13 @@ your command prompt can run the docker "hello world" container.
 Installing on FreeBSD
 =====================
 
-.. image:: https://repology.org/badge/version-for-repo/freebsd/python:ocrmypdf.svg
+.. image:: https://repology.org/badge/version-for-repo/freebsd/ocrmypdf.svg
     :alt: FreeBSD
-    :target: https://repology.org/project/python:ocrmypdf/versions
-
-FreeBSD 11.3, 12.0, 12.1-RELEASE and 13.0-CURRENT are supported. Other
-versions likely work but have not been tested.
+    :target: https://repology.org/project/ocrmypdf/versions
 
 .. code-block:: bash
 
-    pkg install py37-ocrmypdf
+    pkg install textproc/py-ocrmypdf
 
 To install a more recent version, you could attempt to first install the system
 version with ``pkg``, then use ``pip install --user ocrmypdf``.
@@ -645,6 +555,21 @@ the latest version. However, PyPI and ``pip`` cannot address the fact
 that ``ocrmypdf`` depends on certain non-Python system libraries and
 programs being installed.
 
+.. warning::
+
+    Debian and Ubuntu users: unfortunately, Debian and Ubuntu customize
+    Python in non-standard ways, and the nature of these customizations
+    varies from release to release. This can make for a frustrating
+    user experience. The instructions below work on almost all platforms that
+    have Python installed, except for Debian and Ubuntu, where you may need
+    to take additional steps. For best results on Debian and Ubuntu, use the
+    ``apt`` packages; or if these are too old, run
+    ``apt install python3-pip python3-venv``, create a virtual environment,
+    and install OCRmyPDF in that environment.
+
+    `See here for more inforation on Debian-Python issues
+    <https://gist.github.com/tiran/2dec9e03c6f901814f6d1e8dad09528e>`__.
+
 For best results, first install `your platform's
 version <https://repology.org/metapackage/ocrmypdf/versions>`__ of
 ``ocrmypdf``, using the instructions elsewhere in this document. Then
@@ -658,18 +583,18 @@ try:
 
 .. code-block:: bash
 
-    pip3 install --user ocrmypdf
+    pip install --user ocrmypdf
 
 You should then be able to run ``ocrmypdf --version`` and see that the
 latest version was located.
 
-Since ``pip3 install --user`` does not work correctly on some platforms,
+Since ``pip install --user`` does not work correctly on some platforms,
 notably Ubuntu 16.04 and older, and the Homebrew version of Python,
 instead use this for a system wide installation:
 
 .. code-block:: bash
 
-    pip3 install ocrmypdf
+    pip install ocrmypdf
 
 .. note::
 
@@ -685,16 +610,10 @@ OCRmyPDF currently requires these external programs and libraries to be
 installed, and must be satisfied using the operating system package
 manager. ``pip`` cannot provide them.
 
--  Python 3.6 or newer
--  Ghostscript 9.15 or newer
--  qpdf 8.1.0 or newer
--  Tesseract 4.0.0-beta or newer
+The following versions are required:
 
-As of ocrmypdf 7.2.1, the following versions are recommended:
-
--  Python 3.7 or 3.8
+-  Python 3.7 or newer
 -  Ghostscript 9.23 or newer
--  qpdf 8.2.1
 -  Tesseract 4.0.0 or newer
 -  jbig2enc 0.29 or newer
 -  pngquant 2.5 or newer
@@ -726,7 +645,7 @@ unfortunately, the ``pip install`` command cannot satisfy all of them.
 Installing HEAD revision from sources
 =====================================
 
-If you have ``git`` and Python 3.6 or newer installed, you can install
+If you have ``git`` and Python 3.7 or newer installed, you can install
 from source. When the ``pip`` installer runs, it will alert you if
 dependencies are missing.
 
@@ -740,7 +659,7 @@ environment:
 
 .. code-block:: bash
 
-    pip3 install git+https://github.com/jbarlow83/OCRmyPDF.git
+    pip install git+https://github.com/ocrmypdf/OCRmyPDF.git
 
 Or, to install in `development
 mode <https://pythonhosted.org/setuptools/setuptools.html#development-mode>`__,
@@ -748,18 +667,18 @@ allowing customization of OCRmyPDF, use the ``-e`` flag:
 
 .. code-block:: bash
 
-    pip3 install -e git+https://github.com/jbarlow83/OCRmyPDF.git
+    pip install -e git+https://github.com/ocrmypdf/OCRmyPDF.git
 
 You may find it easiest to install in a virtual environment, rather than
 system-wide:
 
 .. code-block:: bash
 
-    git clone -b master https://github.com/jbarlow83/OCRmyPDF.git
+    git clone -b master https://github.com/ocrmypdf/OCRmyPDF.git
     python3 -m venv
     source venv/bin/activate
     cd OCRmyPDF
-    pip3 install .
+    pip install .
 
 However, ``ocrmypdf`` will only be accessible on the system PATH when
 you activate the virtual environment.
@@ -782,12 +701,11 @@ To install all of the development and test requirements:
 
 .. code-block:: bash
 
-    git clone -b master https://github.com/jbarlow83/OCRmyPDF.git
-    python3 -m venv
+    git clone -b master https://github.com/ocrmypdf/OCRmyPDF.git
+    python -m venv
     source venv/bin/activate
     cd OCRmyPDF
-    pip install -e .
-    pip install -r requirements/dev.txt -r requirements/test.txt
+    pip install -e .[test]
 
 To add JBIG2 encoding, see :ref:`jbig2`.
 

docs/introduction.rst

@@ -2,7 +2,12 @@
 Introduction
 ============
 
-OCRmyPDF is a Python 3 application and library that adds OCR layers to PDFs.
+OCRmyPDF is an application and library that adds text "layers" to images
+in PDFs, making scanned image PDFs searchable. It uses OCR to guess what text
+is contained in images. It is written in Python. OCRmyPDF supports plugins
+that allow customization of its processing steps, and is very tolerant of
+PDFs that contain scanned images and "born digital" content that needs no
+text recognition.
 
 About OCR
 =========
@@ -26,7 +31,7 @@ exactly. They contain `vector
 graphics <http://vector-conversions.com/vectorizing/raster_vs_vector.html>`__
 that can contain raster objects such as scanned images. Because PDFs can
 contain multiple pages (unlike many image formats) and can contain fonts
-and text, it is a good formats for exchanging scanned documents.
+and text, it is a good format for exchanging scanned documents.
 
 |image|
 
@@ -35,9 +40,9 @@ have one image. Some scanners or scanning software will segment pages
 into monochromatic text and color regions for example, to improve the
 compression ratio and appearance of the page.
 
-Rasterizing a PDF is the process of generating an image suitable for
-display or analyzing with an OCR engine. OCR engines like Tesseract work
-with images, not vector objects.
+Rasterizing a PDF is the process of generating corresponding raster images.
+OCR engines like Tesseract work with images, not scalable vector graphics
+or mixed raster-vector-text graphics such as PDF.
 
 About PDF/A
 ===========
@@ -76,7 +81,7 @@ OCRmyPDF analyzes each page of a PDF to determine the colorspace and
 resolution (DPI) needed to capture all of the information on that page
 without losing content. It uses
 `Ghostscript <http://ghostscript.com/>`__ to rasterize the page, and
-then performs on OCR on the rasterized image to create an OCR "layer".
+then performs on OCR the rasterized image to create an OCR "layer".
 The layer is then grafted back onto the original PDF.
 
 While one can use a program like Ghostscript or ImageMagick to get an
@@ -84,9 +89,9 @@ image and put the image through Tesseract, that actually creates a new
 PDF and many details may be lost. OCRmyPDF can produce a minimally
 changed PDF as output.
 
-OCRmyPDF also some image processing options like deskew which improve
-the appearance of files and quality of OCR. When these are used, the OCR
-layer is grafted onto the processed image instead.
+OCRmyPDF also provides some image processing options, like deskew, which
+improves the appearance of files and quality of OCR. When these are used,
+the OCR layer is grafted onto the processed image instead.
 
 By default, OCRmyPDF produces archival PDFs – PDF/A, which are a
 stricter subset of PDF features designed for long term archives. If
@@ -139,7 +144,7 @@ Limitations
 OCRmyPDF is limited by the Tesseract OCR engine. As such it experiences
 these limitations, as do any other programs that rely on Tesseract:
 
--  The OCR is not as accurate as commercial solutions such as Abbyy.
+-  The OCR is not as accurate as commercial OCR solutions.
 -  It is not capable of recognizing handwriting.
 -  It may find gibberish and report this as OCR output.
 -  If a document contains languages outside of those given in the
@@ -207,8 +212,9 @@ consider one of these similar open source programs:
 
 -  pdf2pdfocr
 -  pdfsandwich
--  pypdfocr
--  pdfbeads
+
+Ghostscript recently added three "pdfocr" output devices. They work by
+rasterizing all content and converting all pages to a single colour space.
 
 Web front-ends
 ==============

docs/jbig2.rst

@@ -9,11 +9,11 @@ encoding was patented for a long time. All known JBIG2 US patents have
 expired as of 2017, but it is possible that unknown patents exist.
 
 JBIG2 encoding is recommended for OCRmyPDF and is used to losslessly
-create smaller PDFs. If JBIG2 encoding not available, lower quality
+create smaller PDFs. If JBIG2 encoding is not available, lower quality
 encodings will be used.
 
 JBIG2 decoding is not patented and is performed automatically by most
-PDF viewers. It is widely supported has been part of the PDF
+PDF viewers. It is widely supported and has been part of the PDF
 specification since 2001.
 
 On macOS, Homebrew packages jbig2enc and OCRmyPDF includes it by
@@ -32,12 +32,15 @@ For all other Linux, you must build a JBIG2 encoder from source:
 
 .. _jbig2-lossy:
 
+Dependencies include libtoolize and libleptonica, which on Ubuntu systems 
+are packaged as libtool and libleptonica-dev.
+
 Lossy mode JBIG2
 ================
 
 OCRmyPDF provides lossy mode JBIG2 as an advanced feature. Users should
 `review the technical concerns with JBIG2 in lossy
-mode <https://abbyy.technology/en:kb:tip:jbig2_compression_and_ocr>`__
+mode <https://en.wikipedia.org/wiki/JBIG2#Disadvantages>`__
 and decide if this feature is acceptable for their use case.
 
 JBIG2 lossy mode does achieve higher compression ratios than any other

docs/languages.rst

@@ -12,9 +12,9 @@ languages <https://github.com/tesseract-ocr/tesseract/blob/master/doc/tesseract.
 Languages are identified by standardized three-letter codes (called ISO 639-2 Alpha-3).
 Tesseract's documentation also lists the three-letter code for your language.
 Some are anglicized, e.g. Spanish is ``spa`` rather than ``esp``, while others
-are not, e.g. German is ``deu``.
+are not, e.g. German is ``deu`` and French is ``fra``.
 
-After you have installed a language pack, you can use it ``ocrmypdf -l <language>``,
+After you have installed a language pack, you can use it with ``ocrmypdf -l <language>``,
 for example ``ocrmypdf -l spa``. For multilingual documents, you can specify
 all languages to be expected, e.g. ``ocrmypdf -l eng+fra`` for English and French.
 English is assumed by default unless other language(s) are specified.
@@ -35,8 +35,8 @@ Debian and Ubuntu users
 
 You can then pass the ``-l LANG`` argument to OCRmyPDF to give a hint as
 to what languages it should search for. Multiple languages can be
-requested using either ``-l eng+fre`` (English and French) or
-``-l eng -l fre``.
+requested using either ``-l eng+fra`` (English and French) or
+``-l eng -l fra``.
 
 Fedora users
 ============
@@ -51,8 +51,35 @@ Fedora users
 
 You can then pass the ``-l LANG`` argument to OCRmyPDF to give a hint as
 to what languages it should search for. Multiple languages can be
-requested using either ``-l eng+fre`` (English and French) or
-``-l eng -l fre``.
+requested using either ``-l eng+fra`` (English and French) or
+``-l eng -l fra``.
+
+Gentoo users
+============
+
+On Gentoo the package ``app-text/tessdata_fast``, which ``app-text/tesseract`` depends on, handles Tesseract languages.
+It accepts USE flags to select what languages should be installed, these can be set in ``/etc/portage/package.use``.
+Alternatively one can globally set the `L10N use extension <https://wiki.gentoo.org/wiki/Localization/Guide#L10N>`__ in ``/etc/portage/make.conf``.
+This enables these languages for all packages (e.g. including aspell).
+
+.. code-block:: bash
+
+   # Display a list of all Tesseract language packs
+   equery uses app-text/tessdata_fast
+   
+   # Add English and German language support for Tesseract only
+   echo 'app-text/tessdata_fast l10n_de l10n_en' >> /etc/portage/package.use
+   
+   # Add global English and German language support (the `l10n_` from equery has to be omited)
+   echo L10N="de en" >> /etc/portage/make.conf
+   
+   # update system to reflect changed USE flags
+   emerge --update --deep --newuse @world
+
+You can then pass the ``-l LANG`` argument to OCRmyPDF to give a hint as
+to what languages it should search for. Multiple languages can be
+requested using either ``-l eng+fra`` (English and French) or
+``-l eng -l fra``.
 
 macOS users
 ===========
@@ -70,4 +97,7 @@ derived Docker image as
 Windows users
 =============
 
-The Tesseract installer provided by Chocolatey already includes 100 languages.
+The Tesseract installer provided by Chocolatey currently includes only English language. 
+To install other languages, download the respective language pack (``.traineddata`` file) 
+from https://github.com/tesseract-ocr/tessdata/ and place it in 
+``C:\\Program Files\\Tesseract-OCR\\tessdata`` (or wherever Tesseract OCR is installed).

docs/maintainers.rst

@@ -0,0 +1,60 @@
+================
+Maintainer notes
+================
+
+This is for those who package OCRmyPDF for downstream use.  (Thank you
+for your hard work.)
+
+Known ports/packagers
+=====================
+
+OCRmyPDF has been ported to many platforms already. If you are interesting in
+porting to a new platform, check with
+`Repology <https://repology.org/projects/?search=ocrmypdf>`__ to see the status
+of that platform.
+
+Make sure you can package pikepdf
+---------------------------------
+
+pikepdf, created by the same author, is a mixed Python and C++14 package with
+much stiffer build requirements. If you want to use OCRmyPDF on some novel platform
+or distribution, first make sure you can package pikepdf.
+
+Non-Python dependencies
+-----------------------
+
+Note that we have non-Python dependencies. In particular, OCRmyPDF requires
+Ghostscript and Tesseract OCR to be installed and needs to be able to locate their
+binaries on the system PATH. On Windows, OCRmyPDF will also check the registry
+for their locations.
+
+Tesseract OCR relies on SIMD for performance and only has proper support for this
+on ARM and x86_64. Performance may be poor on other processor architectures.
+
+Versioning scheme
+-----------------
+
+OCRmyPDF uses setuptools-scm for versioning, which derives the version from
+Git as a single source of truth. This may be unsuitable for some distributions, e.g.
+to indicate that your distribution modifies OCRmyPDF in some way.
+
+You can patch the ``__version__`` variable in ``src/ocrmypdf/_version.py`` if
+necessary.
+
+OCRmyPDF uses setuptools-scm-git-archive to ensure that tarballs downloaded from
+GitHub contain version information. Unfortunately, these tarballs are not always
+deterministic. See this
+`issue <https://github.com/ocrmypdf/OCRmyPDF/issues/841#issuecomment-936562696>`_.
+
+jbig2enc
+--------
+
+OCRmyPDF will use jbig2enc, a JBIG2 encoder, if one can be found. Some distributions
+have shied away from packaging JBIG2 because it contains patented algorithms, but
+all patents have expired since 2017. If possible, consider packaging it too to
+improve OCRmyPDF's compression.
+
+Command line completions
+------------------------
+
+Please ensure that command line completions are installed.

docs/pdfsecurity.rst

@@ -19,7 +19,7 @@ PDF is a rich, complex file format. The official PDF 1.7 specification,
 ISO 32000:2008, is hundreds of pages long and references several annexes
 each of which are similar in length. PDFs can contain video, audio, XML,
 JavaScript and other programming, and forms. In some cases, they can
-open internet connections to pre-selected URLs. All of these possible
+open internet connections to pre-selected URLs. All of these are possible
 attack vectors.
 
 In short, PDFs `may contain
@@ -31,7 +31,7 @@ describes a high-paranoia method which allows potentially hostile PDFs
 to be viewed and rasterized safely in a disposable virtual machine. A
 trusted PDF created in this manner is converted to images and loses all
 information making it searchable and losing all compression. OCRmyPDF
-could be used restore searchability.
+could be used to restore searchability.
 
 How OCRmyPDF processes PDFs
 ===========================
@@ -41,8 +41,8 @@ layer. First, it runs all PDFs through
 `pikepdf <https://github.com/pikepdf/pikepdf>`__, a library based on
 `qpdf <https://github.com/qpdf/qpdf>`__, a program that repairs PDFs
 with syntax errors. This is done because, in the author's experience, a
-significant number of PDFs in the wild especially those created by
-scanners are not well-formed files. qpdf makes it more likely that
+significant number of PDFs in the wild, especially those created by
+scanners, are not well-formed files. qpdf makes it more likely that
 OCRmyPDF will succeed, but offers no security guarantees. qpdf is also
 used to split the PDF into single page PDFs.
 
@@ -66,8 +66,8 @@ service. OCRmyPDF relies on Ghostscript, and therefore, if deployed
 online one should be prepared to comply with Ghostscript's Affero GPL
 license, and any other licenses.
 
-Setting aside these concerns, a side effect of OCRmyPDF is it may
-incidentally sanitize PDFs that contain certain types of malware. It
+Setting aside these concerns, a side effect of OCRmyPDF is that it may
+incidentally sanitize PDFs containing certain types of malware. It
 repairs the PDF with pikepdf/libqpdf, which could correct malformed PDF
 structures that are part of an attack. When PDF/A output is selected
 (the default), the input PDF is partially reconstructed by Ghostscript.
@@ -83,7 +83,7 @@ Limiting CPU usage
 OCRmyPDF will attempt to use all available CPUs and storage, so
 executing ``nice ocrmypdf`` or limiting the number of jobs with the
 ``-j`` argument may ensure the server remains available. Another option
-would be run OCRmyPDF jobs inside a Docker container, a virtual machine,
+would be to run OCRmyPDF jobs inside a Docker container, a virtual machine,
 or a cloud instance, which can impose its own limits on CPU usage and be
 terminated "from orbit" if it fails to complete.
 
@@ -128,8 +128,9 @@ Commercial alternatives
 The author also provides professional services that include OCR and
 building databases around PDFs, and is happy to provide consultation.
 
-Abbyy Cloud OCR is a viable commercial alternative with a web services
-API.
+Abbyy Cloud OCR is viable commercial alternative with a web services
+API. Amazon Textract, Google Cloud Vision, and Microsoft Azure 
+Computer Vision provide advanced OCR but have less PDF rendering capability.
 
 Password protection, digital signatures and certification
 =========================================================

docs/pipeline.svg

@@ -1,392 +0,0 @@
-<?xml version="1.0" encoding="UTF-8" standalone="no"?>
-<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN"
- "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
-<!-- Generated by graphviz version 2.40.1 (20161225.0304)
- -->
-<!-- Title: Pipeline: Pages: 1 -->
-<svg width="1484pt" height="1277pt"
- viewBox="0.00 0.00 1484.00 1277.20" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
-<g id="graph0" class="graph" transform="scale(1 1) rotate(0) translate(4 1273.1977)">
-<title>Pipeline:</title>
-<polygon fill="#ffffff" stroke="transparent" points="-4,4 -4,-1273.1977 1480,-1273.1977 1480,4 -4,4"/>
-<g id="clust1" class="cluster">
-<title>clustertasks</title>
-<polygon fill="none" stroke="#000000" points="8,-8 8,-1261.1977 1468,-1261.1977 1468,-8 8,-8"/>
-<text text-anchor="middle" x="738" y="-1233.1977" font-family="Times,serif" font-size="30.00" fill="#ff3232">Pipeline:</text>
-</g>
-<!-- t0 -->
-<g id="node1" class="node">
-<title>t0</title>
-<polygon fill="#efa03b" stroke="#000000" points="1215.7053,-1215.1977 1006.2947,-1215.1977 1002.2947,-1211.1977 1002.2947,-1179.1977 1211.7053,-1179.1977 1215.7053,-1183.1977 1215.7053,-1215.1977"/>
-<polyline fill="none" stroke="#000000" points="1211.7053,-1211.1977 1002.2947,-1211.1977 "/>
-<polyline fill="none" stroke="#000000" points="1211.7053,-1211.1977 1211.7053,-1179.1977 "/>
-<polyline fill="none" stroke="#000000" points="1211.7053,-1211.1977 1215.7053,-1215.1977 "/>
-<text text-anchor="middle" x="1109" y="-1191.1977" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.triage</text>
-</g>
-<!-- t1 -->
-<g id="node2" class="node">
-<title>t1</title>
-<polygon fill="#efa03b" stroke="#000000" points="1280.0433,-1157.1977 941.9567,-1157.1977 937.9567,-1153.1977 937.9567,-1121.1977 1276.0433,-1121.1977 1280.0433,-1125.1977 1280.0433,-1157.1977"/>
-<polyline fill="none" stroke="#000000" points="1276.0433,-1153.1977 937.9567,-1153.1977 "/>
-<polyline fill="none" stroke="#000000" points="1276.0433,-1153.1977 1276.0433,-1121.1977 "/>
-<polyline fill="none" stroke="#000000" points="1276.0433,-1153.1977 1280.0433,-1157.1977 "/>
-<text text-anchor="middle" x="1109" y="-1133.1977" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.repair_and_parse_pdf</text>
-</g>
-<!-- t0&#45;&gt;t1 -->
-<g id="edge1" class="edge">
-<title>t0&#45;&gt;t1</title>
-<path fill="none" stroke="#0044a0" d="M1109,-1179.1641C1109,-1175.4895 1109,-1171.5395 1109,-1167.604"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="1112.5001,-1167.4199 1109,-1157.4199 1105.5001,-1167.42 1112.5001,-1167.4199"/>
-</g>
-<!-- t2 -->
-<g id="node3" class="node">
-<title>t2</title>
-<polygon fill="#efa03b" stroke="#000000" points="1030.0926,-1078.2732 785,-1095.1731 539.9074,-1078.2732 540.1362,-1050.9285 1029.8638,-1050.9285 1030.0926,-1078.2732"/>
-<polygon fill="none" stroke="#000000" points="1034.1249,-1082.0021 785,-1099.1801 535.8751,-1082.0021 536.1685,-1046.931 1033.8315,-1046.931 1034.1249,-1082.0021"/>
-<text text-anchor="middle" x="785" y="-1064.7153" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.marker_pages</text>
-</g>
-<!-- t1&#45;&gt;t2 -->
-<g id="edge2" class="edge">
-<title>t1&#45;&gt;t2</title>
-<path fill="none" stroke="#0044a0" d="M1023.449,-1121.1152C984.5695,-1112.8974 938.0675,-1103.0685 896.4957,-1094.2816"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="897.1066,-1090.8335 886.5989,-1092.1898 895.6589,-1097.6822 897.1066,-1090.8335"/>
-</g>
-<!-- t16 -->
-<g id="node17" class="node">
-<title>t16</title>
-<polygon fill="#00cc66" stroke="#000000" points="1215.5059,-328.1861 942.4941,-328.1861 938.4941,-324.1861 938.4941,-292.1861 1211.5059,-292.1861 1215.5059,-296.1861 1215.5059,-328.1861"/>
-<polyline fill="none" stroke="#000000" points="1211.5059,-324.1861 938.4941,-324.1861 "/>
-<polyline fill="none" stroke="#000000" points="1211.5059,-324.1861 1211.5059,-292.1861 "/>
-<polyline fill="none" stroke="#000000" points="1211.5059,-324.1861 1215.5059,-328.1861 "/>
-<text text-anchor="middle" x="1077" y="-304.1861" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.weave_layers</text>
-</g>
-<!-- t1&#45;&gt;t16 -->
-<g id="edge24" class="edge">
-<title>t1&#45;&gt;t16</title>
-<path fill="none" stroke="#0044a0" d="M1097.2373,-1120.9576C1077.987,-1088.7184 1043.7317,-1019.7372 1063,-963.2656 1106.87,-834.6914 1482.8098,-573.3183 1409,-459.2656 1362.7595,-387.8137 1273.2613,-350.2799 1198.8305,-330.7528"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="1199.4555,-327.3005 1188.9015,-328.2301 1197.7316,-334.0849 1199.4555,-327.3005"/>
-</g>
-<!-- t17 -->
-<g id="node18" class="node">
-<title>t17</title>
-<polygon fill="#efa03b" stroke="#000000" points="1439.5156,-1009.7492 1076.4844,-1009.7492 1072.4844,-1005.7492 1072.4844,-973.7492 1435.5156,-973.7492 1439.5156,-977.7492 1439.5156,-1009.7492"/>
-<polyline fill="none" stroke="#000000" points="1435.5156,-1005.7492 1072.4844,-1005.7492 "/>
-<polyline fill="none" stroke="#000000" points="1435.5156,-1005.7492 1435.5156,-973.7492 "/>
-<polyline fill="none" stroke="#000000" points="1435.5156,-1005.7492 1439.5156,-1009.7492 "/>
-<text text-anchor="middle" x="1256" y="-985.7492" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.generate_postscript_stub</text>
-</g>
-<!-- t1&#45;&gt;t17 -->
-<g id="edge25" class="edge">
-<title>t1&#45;&gt;t17</title>
-<path fill="none" stroke="#0044a0" d="M1127.0251,-1121.1176C1153.0342,-1095.0292 1201.2007,-1046.7157 1230.7429,-1017.0834"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="1233.2458,-1019.5302 1237.8275,-1009.9772 1228.2885,-1014.5879 1233.2458,-1019.5302"/>
-</g>
-<!-- t18 -->
-<g id="node19" class="node">
-<title>t18</title>
-<polygon fill="#efa03b" stroke="#000000" points="1070.189,-214.4288 1263,-187.1099 1455.811,-214.4288 1455.631,-258.6317 1070.369,-258.6317 1070.189,-214.4288"/>
-<polygon fill="none" stroke="#000000" points="1066.1715,-210.9543 1263,-183.0662 1459.8285,-210.9543 1459.6181,-262.6353 1066.3819,-262.6353 1066.1715,-210.9543"/>
-<text text-anchor="middle" x="1263" y="-220.6461" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.metadata_fixup</text>
-</g>
-<!-- t1&#45;&gt;t18 -->
-<g id="edge28" class="edge">
-<title>t1&#45;&gt;t18</title>
-<path fill="none" stroke="#0044a0" d="M1241.1115,-1121.1192C1311.0076,-1105.2558 1393.5851,-1075.5895 1449,-1020.2329 1458.4922,-1010.7507 1459,-1005.1661 1459,-991.7492 1459,-991.7492 1459,-991.7492 1459,-393.7258 1459,-341.9727 1441.8377,-325.2216 1402,-292.1861 1390.9869,-283.0534 1378.4885,-274.871 1365.6104,-267.6415"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="1367.1079,-264.4723 1356.6476,-262.7888 1363.775,-270.6279 1367.1079,-264.4723"/>
-</g>
-<!-- t3 -->
-<g id="node4" class="node">
-<title>t3</title>
-<polygon fill="#efa03b" stroke="#000000" points="1015.5194,-999.3074 785,-1016.2081 554.4806,-999.3074 554.6958,-971.9616 1015.3042,-971.9616 1015.5194,-999.3074"/>
-<polygon fill="none" stroke="#000000" points="1019.5491,-1003.0204 785,-1020.2165 550.4509,-1003.0204 550.7267,-967.9639 1019.2733,-967.9639 1019.5491,-1003.0204"/>
-<text text-anchor="middle" x="785" y="-985.7492" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.ocr_or_skip</text>
-</g>
-<!-- t2&#45;&gt;t3 -->
-<g id="edge3" class="edge">
-<title>t2&#45;&gt;t3</title>
-<path fill="none" stroke="#0044a0" d="M785,-1046.5938C785,-1041.5635 785,-1036.139 785,-1030.7273"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="788.5001,-1030.4427 785,-1020.4427 781.5001,-1030.4428 788.5001,-1030.4427"/>
-</g>
-<!-- t4 -->
-<g id="node5" class="node">
-<title>t4</title>
-<polygon fill="#efa03b" stroke="#000000" points="889.3166,-941.2656 582.6834,-941.2656 578.6834,-937.2656 578.6834,-905.2656 885.3166,-905.2656 889.3166,-909.2656 889.3166,-941.2656"/>
-<polyline fill="none" stroke="#000000" points="885.3166,-937.2656 578.6834,-937.2656 "/>
-<polyline fill="none" stroke="#000000" points="885.3166,-937.2656 885.3166,-905.2656 "/>
-<polyline fill="none" stroke="#000000" points="885.3166,-937.2656 889.3166,-941.2656 "/>
-<text text-anchor="middle" x="734" y="-917.2656" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.rasterize_preview</text>
-</g>
-<!-- t3&#45;&gt;t4 -->
-<g id="edge4" class="edge">
-<title>t3&#45;&gt;t4</title>
-<path fill="none" stroke="#0044a0" d="M767.1627,-967.797C762.8396,-961.9919 758.1927,-955.7519 753.7858,-949.8343"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="756.3465,-947.4129 747.5666,-941.483 750.7323,-951.5939 756.3465,-947.4129"/>
-</g>
-<!-- t5 -->
-<g id="node6" class="node">
-<title>t5</title>
-<polygon fill="#efa03b" stroke="#000000" points="1028.0824,-883.2656 769.9176,-883.2656 765.9176,-879.2656 765.9176,-847.2656 1024.0824,-847.2656 1028.0824,-851.2656 1028.0824,-883.2656"/>
-<polyline fill="none" stroke="#000000" points="1024.0824,-879.2656 765.9176,-879.2656 "/>
-<polyline fill="none" stroke="#000000" points="1024.0824,-879.2656 1024.0824,-847.2656 "/>
-<polyline fill="none" stroke="#000000" points="1024.0824,-879.2656 1028.0824,-883.2656 "/>
-<text text-anchor="middle" x="897" y="-859.2656" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.orient_page</text>
-</g>
-<!-- t3&#45;&gt;t5 -->
-<g id="edge6" class="edge">
-<title>t3&#45;&gt;t5</title>
-<path fill="none" stroke="#0044a0" d="M869.1177,-967.8063C880.553,-961.0264 890.81,-952.3527 898,-941.2656 907.042,-927.3226 906.9457,-908.7053 904.5139,-893.387"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="907.914,-892.5383 902.5408,-883.4071 901.0469,-893.8961 907.914,-892.5383"/>
-</g>
-<!-- t4&#45;&gt;t5 -->
-<g id="edge5" class="edge">
-<title>t4&#45;&gt;t5</title>
-<path fill="none" stroke="#0044a0" d="M784.6807,-905.2319C801.0865,-899.3943 819.4462,-892.8614 836.4913,-886.7963"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="838.1026,-889.938 846.3506,-883.288 835.7559,-883.343 838.1026,-889.938"/>
-</g>
-<!-- t6 -->
-<g id="node7" class="node">
-<title>t6</title>
-<polygon fill="#efa03b" stroke="#000000" points="1064.7325,-825.2656 689.2675,-825.2656 685.2675,-821.2656 685.2675,-789.2656 1060.7325,-789.2656 1064.7325,-793.2656 1064.7325,-825.2656"/>
-<polyline fill="none" stroke="#000000" points="1060.7325,-821.2656 685.2675,-821.2656 "/>
-<polyline fill="none" stroke="#000000" points="1060.7325,-821.2656 1060.7325,-789.2656 "/>
-<polyline fill="none" stroke="#000000" points="1060.7325,-821.2656 1064.7325,-825.2656 "/>
-<text text-anchor="middle" x="875" y="-801.2656" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.rasterize_with_ghostscript</text>
-</g>
-<!-- t5&#45;&gt;t6 -->
-<g id="edge7" class="edge">
-<title>t5&#45;&gt;t6</title>
-<path fill="none" stroke="#0044a0" d="M890.1597,-847.2319C888.6839,-843.3412 887.0909,-839.1417 885.5115,-834.9776"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="888.731,-833.5965 881.9119,-825.4878 882.186,-836.0791 888.731,-833.5965"/>
-</g>
-<!-- t13 -->
-<g id="node16" class="node">
-<title>t13</title>
-<polygon fill="#00cc66" stroke="#000000" points="1077,-515.2656 754.4141,-487.2656 1077,-459.2656 1399.5859,-487.2656 1077,-515.2656"/>
-<text text-anchor="middle" x="1077" y="-481.2656" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.select_image_layer</text>
-</g>
-<!-- t5&#45;&gt;t13 -->
-<g id="edge20" class="edge">
-<title>t5&#45;&gt;t13</title>
-<path fill="none" stroke="#0044a0" d="M1000.2704,-847.241C1031.2359,-840.6873 1060.5415,-832.9862 1074,-825.2656 1114.0954,-802.2644 1145,-795.4899 1145,-749.2656 1145,-749.2656 1145,-749.2656 1145,-633.2656 1145,-590.368 1154.8687,-575.2844 1135,-537.2656 1131.571,-530.7041 1126.8893,-524.6314 1121.6622,-519.1451"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="1123.8989,-516.4405 1114.2626,-512.0367 1119.0494,-521.4885 1123.8989,-516.4405"/>
-</g>
-<!-- t7 -->
-<g id="node8" class="node">
-<title>t7</title>
-<polygon fill="#efa03b" stroke="#000000" points="856.0801,-767.2656 431.9199,-767.2656 427.9199,-763.2656 427.9199,-731.2656 852.0801,-731.2656 856.0801,-735.2656 856.0801,-767.2656"/>
-<polyline fill="none" stroke="#000000" points="852.0801,-763.2656 427.9199,-763.2656 "/>
-<polyline fill="none" stroke="#000000" points="852.0801,-763.2656 852.0801,-731.2656 "/>
-<polyline fill="none" stroke="#000000" points="852.0801,-763.2656 856.0801,-767.2656 "/>
-<text text-anchor="middle" x="642" y="-743.2656" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.preprocess_remove_background</text>
-</g>
-<!-- t6&#45;&gt;t7 -->
-<g id="edge8" class="edge">
-<title>t6&#45;&gt;t7</title>
-<path fill="none" stroke="#0044a0" d="M802.5546,-789.2319C777.7884,-783.067 749.9071,-776.1266 724.4138,-769.7806"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="724.95,-766.3073 714.4006,-767.288 723.259,-773.1 724.95,-766.3073"/>
-</g>
-<!-- t12 -->
-<g id="node15" class="node">
-<title>t12</title>
-<polygon fill="#efa03b" stroke="#000000" points="742,-593.2656 358.1874,-565.2656 742,-537.2656 1125.8126,-565.2656 742,-593.2656"/>
-<text text-anchor="middle" x="742" y="-559.2656" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.select_visible_page_image</text>
-</g>
-<!-- t6&#45;&gt;t12 -->
-<g id="edge18" class="edge">
-<title>t6&#45;&gt;t12</title>
-<path fill="none" stroke="#0044a0" d="M875.1443,-789.1129C874.6867,-773.2956 872.5831,-750.0189 865,-731.2656 844.2991,-680.0716 804.0126,-630.4165 775.1838,-598.9641"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="777.4937,-596.3085 768.1202,-591.3701 772.3682,-601.076 777.4937,-596.3085"/>
-</g>
-<!-- t8 -->
-<g id="node9" class="node">
-<title>t8</title>
-<polygon fill="#efa03b" stroke="#000000" points="732.6759,-709.2656 413.3241,-709.2656 409.3241,-705.2656 409.3241,-673.2656 728.6759,-673.2656 732.6759,-677.2656 732.6759,-709.2656"/>
-<polyline fill="none" stroke="#000000" points="728.6759,-705.2656 409.3241,-705.2656 "/>
-<polyline fill="none" stroke="#000000" points="728.6759,-705.2656 728.6759,-673.2656 "/>
-<polyline fill="none" stroke="#000000" points="728.6759,-705.2656 732.6759,-709.2656 "/>
-<text text-anchor="middle" x="571" y="-685.2656" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.preprocess_deskew</text>
-</g>
-<!-- t7&#45;&gt;t8 -->
-<g id="edge9" class="edge">
-<title>t7&#45;&gt;t8</title>
-<path fill="none" stroke="#0044a0" d="M619.9243,-731.2319C613.9709,-726.3685 607.4267,-721.0226 601.115,-715.8666"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="603.2653,-713.1037 593.3065,-709.4878 598.8367,-718.5248 603.2653,-713.1037"/>
-</g>
-<!-- t7&#45;&gt;t12 -->
-<g id="edge17" class="edge">
-<title>t7&#45;&gt;t12</title>
-<path fill="none" stroke="#0044a0" d="M717.0346,-731.1646C726.8757,-725.6884 735.7103,-718.5557 742,-709.2656 763.0291,-678.205 759.4083,-634.0888 752.8082,-602.8898"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="756.1481,-601.7976 750.487,-592.842 749.3278,-603.3733 756.1481,-601.7976"/>
-</g>
-<!-- t9 -->
-<g id="node10" class="node">
-<title>t9</title>
-<polygon fill="#efa03b" stroke="#000000" points="712.8828,-651.2656 411.1172,-651.2656 407.1172,-647.2656 407.1172,-615.2656 708.8828,-615.2656 712.8828,-619.2656 712.8828,-651.2656"/>
-<polyline fill="none" stroke="#000000" points="708.8828,-647.2656 407.1172,-647.2656 "/>
-<polyline fill="none" stroke="#000000" points="708.8828,-647.2656 708.8828,-615.2656 "/>
-<polyline fill="none" stroke="#000000" points="708.8828,-647.2656 712.8828,-651.2656 "/>
-<text text-anchor="middle" x="560" y="-627.2656" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.preprocess_clean</text>
-</g>
-<!-- t8&#45;&gt;t9 -->
-<g id="edge10" class="edge">
-<title>t8&#45;&gt;t9</title>
-<path fill="none" stroke="#0044a0" d="M567.5798,-673.2319C566.8624,-669.4493 566.0897,-665.3747 565.3216,-661.3247"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="568.7581,-660.6605 563.4559,-651.4878 561.8807,-661.9649 568.7581,-660.6605"/>
-</g>
-<!-- t8&#45;&gt;t12 -->
-<g id="edge16" class="edge">
-<title>t8&#45;&gt;t12</title>
-<path fill="none" stroke="#0044a0" d="M670.0045,-673.2618C693.2528,-667.2338 713.6919,-659.8307 722,-651.2656 734.1971,-638.6912 739.4901,-620.2051 741.6123,-603.5914"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="745.1041,-603.8427 742.5574,-593.5585 738.1349,-603.1862 745.1041,-603.8427"/>
-</g>
-<!-- t10 -->
-<g id="node11" class="node">
-<title>t10</title>
-<polygon fill="#efa03b" stroke="#000000" points="340.1094,-583.2656 35.8906,-583.2656 31.8906,-579.2656 31.8906,-547.2656 336.1094,-547.2656 340.1094,-551.2656 340.1094,-583.2656"/>
-<polyline fill="none" stroke="#000000" points="336.1094,-579.2656 31.8906,-579.2656 "/>
-<polyline fill="none" stroke="#000000" points="336.1094,-579.2656 336.1094,-547.2656 "/>
-<polyline fill="none" stroke="#000000" points="336.1094,-579.2656 340.1094,-583.2656 "/>
-<text text-anchor="middle" x="186" y="-559.2656" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.select_ocr_image</text>
-</g>
-<!-- t9&#45;&gt;t10 -->
-<g id="edge11" class="edge">
-<title>t9&#45;&gt;t10</title>
-<path fill="none" stroke="#0044a0" d="M460.7568,-615.2213C409.7279,-605.9434 347.4237,-594.6153 295.0537,-585.0935"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="295.5223,-581.6214 285.0575,-583.276 294.2701,-588.5085 295.5223,-581.6214"/>
-</g>
-<!-- t9&#45;&gt;t12 -->
-<g id="edge15" class="edge">
-<title>t9&#45;&gt;t12</title>
-<path fill="none" stroke="#0044a0" d="M608.2948,-615.2213C627.0226,-608.2242 648.8712,-600.0609 669.4134,-592.3859"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="670.9148,-595.5613 679.0573,-588.7826 668.4648,-589.004 670.9148,-595.5613"/>
-</g>
-<!-- t11 -->
-<g id="node12" class="node">
-<title>t11</title>
-<polygon fill="#00cc66" stroke="#000000" points="334.0996,-505.2656 19.9004,-505.2656 15.9004,-501.2656 15.9004,-469.2656 330.0996,-469.2656 334.0996,-473.2656 334.0996,-505.2656"/>
-<polyline fill="none" stroke="#000000" points="330.0996,-501.2656 15.9004,-501.2656 "/>
-<polyline fill="none" stroke="#000000" points="330.0996,-501.2656 330.0996,-469.2656 "/>
-<polyline fill="none" stroke="#000000" points="330.0996,-501.2656 334.0996,-505.2656 "/>
-<text text-anchor="middle" x="175" y="-481.2656" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.ocr_tesseract_hocr</text>
-</g>
-<!-- t10&#45;&gt;t11 -->
-<g id="edge12" class="edge">
-<title>t10&#45;&gt;t11</title>
-<path fill="none" stroke="#0044a0" d="M183.4487,-547.1748C182.1254,-537.7914 180.4788,-526.1154 178.9917,-515.5701"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="182.4378,-514.9417 177.5755,-505.5285 175.5064,-515.9193 182.4378,-514.9417"/>
-</g>
-<!-- t15 -->
-<g id="node14" class="node">
-<title>t15</title>
-<polygon fill="#ff69b4" stroke="#000000" points="736.2696,-505.2656 355.7304,-505.2656 351.7304,-501.2656 351.7304,-469.2656 732.2696,-469.2656 736.2696,-473.2656 736.2696,-505.2656"/>
-<polyline fill="none" stroke="#000000" points="732.2696,-501.2656 351.7304,-501.2656 "/>
-<polyline fill="none" stroke="#000000" points="732.2696,-501.2656 732.2696,-469.2656 "/>
-<polyline fill="none" stroke="#000000" points="732.2696,-501.2656 736.2696,-505.2656 "/>
-<text text-anchor="middle" x="544" y="-481.2656" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.ocr_tesseract_textonly_pdf</text>
-</g>
-<!-- t10&#45;&gt;t15 -->
-<g id="edge14" class="edge">
-<title>t10&#45;&gt;t15</title>
-<path fill="none" stroke="#0044a0" d="M269.0322,-547.1748C323.4947,-535.3086 394.7854,-519.776 451.1284,-507.5002"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="452.0829,-510.8744 461.1086,-505.3257 450.5927,-504.0348 452.0829,-510.8744"/>
-</g>
-<!-- t14 -->
-<g id="node13" class="node">
-<title>t14</title>
-<polygon fill="#00cc66" stroke="#000000" points="330.0433,-411.7258 21.9567,-411.7258 17.9567,-407.7258 17.9567,-375.7258 326.0433,-375.7258 330.0433,-379.7258 330.0433,-411.7258"/>
-<polyline fill="none" stroke="#000000" points="326.0433,-407.7258 17.9567,-407.7258 "/>
-<polyline fill="none" stroke="#000000" points="326.0433,-407.7258 326.0433,-375.7258 "/>
-<polyline fill="none" stroke="#000000" points="326.0433,-407.7258 330.0433,-411.7258 "/>
-<text text-anchor="middle" x="174" y="-387.7258" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.render_hocr_page</text>
-</g>
-<!-- t11&#45;&gt;t14 -->
-<g id="edge13" class="edge">
-<title>t11&#45;&gt;t14</title>
-<path fill="none" stroke="#0044a0" d="M174.8072,-469.2302C174.6642,-455.8584 174.4666,-437.3702 174.3027,-422.0357"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="177.8024,-421.987 174.1956,-412.025 170.8028,-422.0619 177.8024,-421.987"/>
-</g>
-<!-- t19 -->
-<g id="node22" class="node">
-<title>t19</title>
-<polygon fill="#efa03b" stroke="#000000" points="351.7205,-381.5085 545,-354.1897 738.2795,-381.5085 738.0991,-425.7112 351.9009,-425.7112 351.7205,-381.5085"/>
-<polygon fill="none" stroke="#000000" points="347.7033,-378.0329 545,-350.1463 742.2967,-378.0329 742.0858,-429.7148 347.9142,-429.7148 347.7033,-378.0329"/>
-<text text-anchor="middle" x="545" y="-387.7258" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.merge_sidecars</text>
-</g>
-<!-- t11&#45;&gt;t19 -->
-<g id="edge32" class="edge">
-<title>t11&#45;&gt;t19</title>
-<path fill="none" stroke="#0044a0" d="M246.3397,-469.2302C287.966,-458.7066 342.1269,-445.0142 392.8772,-432.184"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="393.8058,-435.5594 402.6429,-429.7151 392.0901,-428.773 393.8058,-435.5594"/>
-</g>
-<!-- t14&#45;&gt;t16 -->
-<g id="edge22" class="edge">
-<title>t14&#45;&gt;t16</title>
-<path fill="none" stroke="#0044a0" d="M230.6615,-375.7128C262.2207,-366.4581 302.4172,-355.9425 339,-350.1861 448.3957,-332.9725 745.7869,-320.7866 927.9215,-314.6743"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="928.248,-318.1655 938.1258,-314.3342 928.0148,-311.1693 928.248,-318.1655"/>
-</g>
-<!-- t15&#45;&gt;t16 -->
-<g id="edge21" class="edge">
-<title>t15&#45;&gt;t16</title>
-<path fill="none" stroke="#0044a0" d="M630.0636,-469.2083C667.5329,-460.6467 711.76,-449.5912 751,-437.2656 850.4985,-406.0121 963.1653,-359.5222 1026.586,-332.2894"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="1028.1566,-335.4238 1035.9553,-328.2523 1025.3866,-328.9951 1028.1566,-335.4238"/>
-</g>
-<!-- t15&#45;&gt;t19 -->
-<g id="edge31" class="edge">
-<title>t15&#45;&gt;t19</title>
-<path fill="none" stroke="#0044a0" d="M544.1928,-469.2302C544.2822,-460.8687 544.393,-450.5066 544.5035,-440.166"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="548.006,-439.9443 544.6132,-429.9075 541.0064,-439.8694 548.006,-439.9443"/>
-</g>
-<!-- t12&#45;&gt;t13 -->
-<g id="edge19" class="edge">
-<title>t12&#45;&gt;t13</title>
-<path fill="none" stroke="#0044a0" d="M833.9833,-543.8486C879.0975,-533.3444 933.4556,-520.6879 979.3116,-510.0109"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="980.3156,-513.3709 989.2614,-507.6943 978.7282,-506.5532 980.3156,-513.3709"/>
-</g>
-<!-- t13&#45;&gt;t16 -->
-<g id="edge23" class="edge">
-<title>t13&#45;&gt;t16</title>
-<path fill="none" stroke="#0044a0" d="M1077,-459.2068C1077,-426.2624 1077,-371.9187 1077,-338.7552"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="1080.5001,-338.3818 1077,-328.3819 1073.5001,-338.3819 1080.5001,-338.3818"/>
-</g>
-<!-- t16&#45;&gt;t18 -->
-<g id="edge27" class="edge">
-<title>t16&#45;&gt;t18</title>
-<path fill="none" stroke="#0044a0" d="M1117.3592,-292.0592C1133.8695,-284.6437 1153.6803,-275.7459 1173.2953,-266.9361"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="1174.8979,-270.0531 1182.586,-262.7632 1172.0299,-263.6676 1174.8979,-270.0531"/>
-</g>
-<!-- t17&#45;&gt;t18 -->
-<g id="edge26" class="edge">
-<title>t17&#45;&gt;t18</title>
-<path fill="none" stroke="#0044a0" d="M1326.2292,-973.7283C1377.8362,-955.768 1439,-922.4247 1439,-865.2656 1439,-865.2656 1439,-865.2656 1439,-393.7258 1439,-338.7092 1393.2056,-296.3464 1348.3356,-268.0433"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="1350.0403,-264.9831 1339.6842,-262.7465 1346.3852,-270.9531 1350.0403,-264.9831"/>
-</g>
-<!-- t20 -->
-<g id="node20" class="node">
-<title>t20</title>
-<polygon fill="#efa03b" stroke="#000000" points="1400.3184,-161.1062 1129.6816,-161.1062 1125.6816,-157.1062 1125.6816,-125.1062 1396.3184,-125.1062 1400.3184,-129.1062 1400.3184,-161.1062"/>
-<polyline fill="none" stroke="#000000" points="1396.3184,-157.1062 1125.6816,-157.1062 "/>
-<polyline fill="none" stroke="#000000" points="1396.3184,-157.1062 1396.3184,-125.1062 "/>
-<polyline fill="none" stroke="#000000" points="1396.3184,-157.1062 1400.3184,-161.1062 "/>
-<text text-anchor="middle" x="1263" y="-137.1062" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.optimize_pdf</text>
-</g>
-<!-- t18&#45;&gt;t20 -->
-<g id="edge29" class="edge">
-<title>t18&#45;&gt;t20</title>
-<path fill="none" stroke="#0044a0" d="M1263,-182.9188C1263,-179.129 1263,-175.3607 1263,-171.739"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="1266.5001,-171.4394 1263,-161.4394 1259.5001,-171.4394 1266.5001,-171.4394"/>
-</g>
-<!-- t21 -->
-<g id="node21" class="node">
-<title>t21</title>
-<polygon fill="#efa03b" stroke="#000000" points="1096.0652,-47.332 1263,-20.0049 1429.9348,-47.332 1429.779,-91.5483 1096.221,-91.5483 1096.0652,-47.332"/>
-<polygon fill="none" stroke="#000000" points="1092.0538,-43.9306 1263,-15.9468 1433.9462,-43.9306 1433.7643,-95.5532 1092.2357,-95.5532 1092.0538,-43.9306"/>
-<text text-anchor="middle" x="1263" y="-53.5531" font-family="Times,serif" font-size="20.00" fill="#000000">ocrmypdf.pipeline.copy_final</text>
-</g>
-<!-- t20&#45;&gt;t21 -->
-<g id="edge30" class="edge">
-<title>t20&#45;&gt;t21</title>
-<path fill="none" stroke="#0044a0" d="M1263,-124.9764C1263,-119.37 1263,-112.9163 1263,-106.2938"/>
-<polygon fill="#0044a0" stroke="#0044a0" points="1266.5001,-105.8456 1263,-95.8457 1259.5001,-105.8457 1266.5001,-105.8456"/>
-</g>
-</g>
-</svg>

docs/plugins.rst

@@ -61,9 +61,10 @@ similar to ``pytest`` packages such as ``pytest-cov`` (the package) and
 
 .. note::
 
-    We strongly recommend plugin authors name their plugins with the prefix
+    We recommend plugin authors name their plugins with the prefix
     ``ocrmypdf-`` (for the package name on PyPI) and ``ocrmypdf_`` (for the
-    module), just like pytest plugins.
+    module), just like pytest plugins. At the same time, please make it clear
+    that your package is not official.
 
 Setuptools plugins
 ==================
@@ -86,6 +87,13 @@ named ``ocrmypdf-exampleplugin``:
         entry_points={"ocrmypdf": ["exampleplugin = exampleplugin.pluginmodule"]},
     )
 
+.. code-block:: ini
+
+    # equivalent setup.cfg
+    [options.entry_points]
+    ocrmypdf =
+        exampleplugin = exampleplugin.pluginmodule
+
 Plugin requirements
 ===================
 
@@ -144,6 +152,16 @@ hooks. As such, you cannot "chain" a series of plugin filters together in this
 way. Instead, a single hook implementation should be responsible for any such
 chaining operations.
 
+Examples
+========
+
+* OCRmyPDF's test suite contains several plugins that are used to simulate certain
+  test conditions.
+* `ocrmypdf-papermerge <https://github.com/papermerge/OCRmyPDF_papermerge>`_ is
+  a production plugin that integrates OCRmyPDF and the Papermerge document
+  management system.
+
+
 Custom command line arguments
 -----------------------------
 
@@ -151,6 +169,18 @@ Custom command line arguments
 
 .. autofunction:: ocrmypdf.pluginspec.check_options
 
+Execution and progress reporting
+--------------------------------
+
+.. autoclass: ocrmypdf.pluginspec.Executor
+    :members:
+
+.. autofunction:: ocrmypdf.pluginspec.get_logging_console
+
+.. autofunction:: ocrmypdf.pluginspec.get_executor
+
+.. autofunction:: ocrmypdf.pluginspec.get_progressbar_class
+
 Applying special behavior before processing
 -------------------------------------------
 
@@ -168,6 +198,8 @@ Modifying intermediate images
 
 .. autofunction:: ocrmypdf.pluginspec.filter_page_image
 
+.. autofunction:: ocrmypdf.pluginspec.filter_pdf_page
+
 OCR engine
 ----------
 

misc/batch.py

@@ -24,45 +24,39 @@
 import logging
 import os
 import sys
+from pathlib import Path
 
 import ocrmypdf
 
 # pylint: disable=logging-format-interpolation
 # pylint: disable=logging-not-lazy
 
-script_dir = os.path.dirname(os.path.realpath(__file__))
-print(script_dir + '/batch.py: Start')
+script_dir = Path(__file__).parent
 
 if len(sys.argv) > 1:
-    start_dir = sys.argv[1]
+    start_dir = Path(sys.argv[1])
 else:
-    start_dir = '.'
+    start_dir = Path('.')
 
 if len(sys.argv) > 2:
-    log_file = sys.argv[2]
+    log_file = Path(sys.argv[2])
 else:
-    log_file = script_dir + '/ocr-tree.log'
+    log_file = script_dir.with_name('ocr-tree.log')
 
 logging.basicConfig(
     level=logging.INFO,
     format='%(asctime)s %(message)s',
     filename=log_file,
-    filemode='w',
+    filemode='a',
 )
 
 ocrmypdf.configure_logging(ocrmypdf.Verbosity.default)
 
-for dir_name, subdirs, file_list in os.walk(start_dir):
-    logging.info(dir_name + '\n')
-    os.chdir(dir_name)
-    for filename in file_list:
-        file_ext = os.path.splitext(filename)[1]
-        if file_ext == '.pdf':
-            full_path = dir_name + '/' + filename
-            print(full_path)
-            result = ocrmypdf.ocr(filename, filename, deskew=True)
-            if result == ocrmypdf.ExitCode.already_done_ocr:
-                print("Skipped document because it already contained text")
-            elif result == ocrmypdf.ExitCode.ok:
-                print("OCR complete")
-            logging.info(result)
+for filename in start_dir.glob("**/*.py"):
+    logging.info(f"Processing {filename}")
+    result = ocrmypdf.ocr(filename, filename, deskew=True)
+    if result == ocrmypdf.ExitCode.already_done_ocr:
+        logging.error("Skipped document because it already contained text")
+    elif result == ocrmypdf.ExitCode.ok:
+        logging.info("OCR complete")
+    logging.info(result)

misc/completion/ocrmypdf.bash

@@ -1,6 +1,6 @@
 # ocrmypdf completion                                     -*- shell-script -*-
 
-# Copyright 2019 Frank Pille
+# Copyright 2019, 2021 Frank Pille
 # Copyright 2020 Alex Willner
 #
 # Permission is hereby granted, free of charge, to any person obtaining a copy
@@ -23,95 +23,285 @@
 
 set -o errexit
 
-_ocrmypdf()
+__ocrmypdf_arguments()
 {
-    local cur prev cword words split
+    local arguments="--help                   (show help message)
+--language               (language(s) of the file to be OCRed)
+--image-dpi              (assume this DPI if input image DPI is unknown)
+--output-type            (select PDF output options)
+--sidecar                (write OCR to text file)
+--version                (print program version and exit)
+--jobs                   (how many worker processes to use)
+--quiet                  (suppress INFO messages)
+--verbose                (set verbosity level)
+--title                  (set metadata)
+--author                 (set metadata)
+--subject                (set metadata)
+--keywords               (set metadata)
+--rotate-pages           (rotate pages to correct orientation)
+--remove-background      (attempt to remove background from pages)
+--deskew                 (fix small horizontal alignment skew)
+--clean                  (clean document images before OCR)
+--clean-final            (clean document images and keep result)
+--unpaper-args           (a quoted string of arguments to pass to unpaper)
+--oversample             (oversample images to this DPI)
+--remove-vectors         (don\'t send vector objects to OCR)
+--threshold              (threshold images before OCR)
+--force-ocr              (OCR documents that already have printable text)
+--skip-text              (skip OCR on any pages that already contain text)
+--redo-ocr               (redo OCR on any pages that seem to have OCR already)
+--skip-big               (skip OCR on pages larger than this many MPixels)
+--optimize               (select optimization level)
+--jpeg-quality           (JPEG quality [0..100])
+--png-quality            (PNG quality [0..100])
+--jbig2-lossy            (enable lossy JBIG2 (see docs))
+--pages                  (apply OCR to only the specified pages)
+--max-image-mpixels      (image decompression bomb threshold)
+--pdf-renderer           (select PDF renderer options)
+--rotate-pages-threshold (page rotation confidence)
+--pdfa-image-compression (set PDF/A image compression options)
+--fast-web-view          (if file size if above this amount in MB linearize PDF)
+--plugin                 (name of plugin to import)
+--keep-temporary-files   (keep temporary files (debug)
+--tesseract-config       (set custom tesseract config file)
+--tesseract-pagesegmode  (set tesseract --psm)
+--tesseract-oem          (set tesseract --oem)
+--tesseract-thresholding (set tesseract image thresholding)
+--tesseract-timeout      (maximum number of seconds to wait for OCR)
+--user-words             (specify location of user words file)
+--user-patterns          (specify location of user patterns file)
+--no-progress-bar        (disable the progress bar)
+"
 
-    # Homebrew on Macs have version 1.3 of bash-completion which doesn't include - see #502
-    if declare -F _init_completions >/dev/null 2>&1; then
-        _init_completion -s || return
-    else
-        COMPREPLY=()
-        _get_comp_words_by_ref cur prev words cword
+    COMPREPLY=( $( compgen -W "$arguments" -- "$cur") )
+
+    # Remove description if only one completion exists
+    if [[ ${#COMPREPLY[*]} -eq 1 ]]; then
+        COMPREPLY=( ${COMPREPLY[0]%% *} )
     fi
+}
 
-    if [[ $cur == -* ]]; then
-        COMPREPLY=( $( compgen -W '--language --image-dpi --output-type
-            --sidecar --version --jobs --quiet --verbose --title --author
-            --subject --keywords --rotate-pages --remove-background --deskew
-            --clean --clean-final --unpaper-args --oversample --remove-vectors
-            --threshold --force-ocr --skip-text --redo-ocr
-            --skip-big --jpeg-quality --png-quality --jbig2-lossy
-            --max-image-mpixels --tesseract-config --tesseract-pagesegmode
-            --help --tesseract-oem --pdf-renderer --tesseract-timeout
-            --rotate-pages-threshold --pdfa-image-compression --user-words
-            --user-patterns --keep-temporary-files --output-type
-            --no-progress-bar --pages --fast-web-view' \
-            --  "$cur" ) )
-        return
-    else
-        _filedir
-        return
+__ocrmypdf_output-type()
+{
+    local choices="pdfa   (output a PDF/A (default))
+pdf    (output a standard PDF)
+pdfa-1 (output a PDF/A-1b)
+pdfa-2 (output a PDF/A-2b)
+pdfa-3 (output a PDF/A-3b)
+none   (do not produce an output PDF (for example, if you only care about --sidecar))"
+
+    COMPREPLY=( $( compgen -W "$choices" -- "$cur") )
+
+    # Remove description if only one completion exists
+    if [[ ${#COMPREPLY[*]} -eq 1 ]]; then
+        COMPREPLY=( ${COMPREPLY[0]%% *} )
     fi
+}
 
+__ocrmypdf_verbose()
+{
+    local choices="0  (standard output messages)
+1  (troubleshooting output messages)
+2  (debugging output messages)"
+
+    COMPREPLY=( $( compgen -W "$choices" -- "$cur") )
+
+    # Remove description if only one completion exists
+    if [[ ${#COMPREPLY[*]} -eq 1 ]]; then
+        COMPREPLY=( ${COMPREPLY[0]%% *} )
+    fi
+}
+
+__ocrmypdf_optimize()
+{
+    local choices="0  (do not optimize)
+1  (do safe, lossless optimizations (default))
+2  (do some lossy optimizations)
+3  (do aggressive lossy optimizations (including lossy JBIG2))"
+
+    COMPREPLY=( $( compgen -W "$choices" -- "$cur") )
+
+    # Remove description if only one completion exists
+    if [[ ${#COMPREPLY[*]} -eq 1 ]]; then
+        COMPREPLY=( ${COMPREPLY[0]%% *} )
+    fi
+}
+
+__ocrmypdf_pdf-renderer()
+{
+    local choices="auto      (auto select PDF renderer)
+hocr      (use hOCR renderer)
+hocrdebug (uses hOCR renderer in debug mode, showing recognized text)
+sandwich  (use sandwich renderer)"
+
+    COMPREPLY=( $( compgen -W "$choices" -- "$cur") )
+
+    # Remove description if only one completion exists
+    if [[ ${#COMPREPLY[*]} -eq 1 ]]; then
+        COMPREPLY=( ${COMPREPLY[0]%% *} )
+    fi
+}
+
+__ocrmypdf_pdfa-image-compression()
+{
+    local choices="auto     (let Ghostscript decide how to compress images)
+jpeg     (convert color and grayscale images to JPEG)
+lossless (convert color and grayscale images to lossless (PNG))"
+
+    COMPREPLY=( $( compgen -W "$choices" -- "$cur") )
+
+    # Remove description if only one completion exists
+    if [[ ${#COMPREPLY[*]} -eq 1 ]]; then
+        COMPREPLY=( ${COMPREPLY[0]%% *} )
+    fi
+}
+
+__ocrmypdf_tesseract-pagesegmode()
+{
+    local choices="0  (orientation and script detection (OSD) only)
+1  (automatic page segmentation with OSD)
+2  (automatic page segmentation, but no OSD, or OCR)
+3  (fully automatic page segmentation, but no OSD (default))
+4  (assume a single column of text of variable sizes)
+5  (assume a single uniform block of vertically aligned text)
+6  (assume a single uniform block of text)
+7  (treat the image as a single text line)
+8  (treat the image as a single word)
+9  (treat the image as a single word in a circle)
+10 (treat the image as a single character)
+11 (sparse text - find as much text as possible in no particular order)
+12 (sparse text with OSD)
+13 (raw line - treat the image as a single text line)"
+
+    COMPREPLY=( $( compgen -W "$choices" -- "$cur") )
+
+    # Remove description if only one completion exists
+    if [[ ${#COMPREPLY[*]} -eq 1 ]]; then
+        COMPREPLY=( ${COMPREPLY[0]%% *} )
+    fi
+}
+
+__ocrmypdf_tesseract-oem()
+{
+    local choices="0 (legacy engine only)
+1 (neural nets LSTM engine only)
+2 (legacy + LSTM engines)
+3 (default, based on what is available)"
+
+    COMPREPLY=( $( compgen -W "$choices" -- "$cur") )
+
+    # Remove description if only one completion exists
+    if [[ ${#COMPREPLY[*]} -eq 1 ]]; then
+        COMPREPLY=( ${COMPREPLY[0]%% *} )
+    fi
+}
+
+__ocrmypdf_tesseract-thresholding()
+{
+    local choices="auto          (let OCRmyPDF pick thresholding - current always uses otsu)
+otsu          (use hOCR renderer)
+adaptive-otsu (use adaptive Otsu thresholding)
+sauvola       (use Sauvola thresholding)"
+
+    COMPREPLY=( $( compgen -W "$choices" -- "$cur") )
+    # Remove description if only one completion exists
+    if [[ ${#COMPREPLY[*]} -eq 1 ]]; then
+        COMPREPLY=( ${COMPREPLY[0]%% *} )
+    fi
+}
+
+
+__ocrmypdf_check_previous()
+{
     case $prev in
-        --version|-h|--help)
-            return
-            ;;
-        --user-words|--user-patterns|--tesseract-config)
-            _filedir
-            return
-            ;;
-        --output-type)
-            COMPREPLY=( $( compgen -W 'pdfa pdf pdfa-1 pdfa-2 pdfa-3' -- \
-                "$cur" ) )
-            return
-            ;;
-        --pdf-renderer)
-            COMPREPLY=( $( compgen -W 'auto hocr sandwich' -- "$cur" ) )
-            return
-            ;;
-        --pdfa-image-compression)
-            COMPREPLY=( $( compgen -W 'auto jpeg lossless' -- "$cur" ) )
-            return
-            ;;
-        -O|--optimize|--tesseract-oem)
-            COMPREPLY=( $( compgen -W '{0..3}' -- "$cur" ) )
-            return
-            ;;
-        --jpeg-quality|--png-quality)
-            COMPREPLY=( $( compgen -W '{0..100}' -- "$cur" ) )
-            return
+        -h|--help|--version)
+            return 0
             ;;
         -l|--language)
             COMPREPLY=$( command tesseract --list-langs 2>/dev/null )
             COMPREPLY=( $( compgen -W '${COMPREPLY[@]##*:}' -- "$cur" ) )
-            return
+            return 0
             ;;
-        --image-dpi|--oversample|--skip-big|--max-image-mpixels|\
-        --tesseract-timeout|--rotate-pages-threshold)
-            COMPREPLY=( $( compgen -P "$cur" -W '{0..9}' ) )
-            return
+        --output-type)
+            __ocrmypdf_output-type
+            return 0
             ;;
         -j|--jobs)
             COMPREPLY=( $( compgen -W '{1..'$( _ncpus )'}' -- "$cur" ) )
-            return
+            return 0
             ;;
         -v|--verbose)
-            COMPREPLY=( $( compgen -W '{0..2}' -- "$cur" ) ) # max level ?
-            return
+            __ocrmypdf_verbose
+            return 0
+            ;;
+        -O|--optimize)
+            __ocrmypdf_optimize
+            return 0
+            ;;
+        --pdf-renderer)
+            __ocrmypdf_pdf-renderer
+            return 0
+            ;;
+        --pdfa-image-compression)
+            __ocrmypdf_pdfa-image-compression
+            return 0
             ;;
         --tesseract-pagesegmode)
-            COMPREPLY=( $( compgen -W '{1..13}' -- "$cur" ) )
-            return
+            __ocrmypdf_tesseract-pagesegmode
+            return 0
             ;;
-        --sidecar|--title|--author|--subject|--keywords|--unpaper-args|--pages|--fast-web-view)
+        --tesseract-oem)
+            __ocrmypdf_tesseract-oem
+            return 0
+            ;;
+        --tesseract-thresholding)
+            __ocrmypdf_tesseract-thresholding
+            return 0
+            ;;
+
+        --title|--author|--subject|--keywords|--unpaper-args|--pages|--plugin|\
+        --jpeg-quality|--png-quality|--image-dpi|--oversample|--skip-big|--max-image-mpixels|\
+        --tesseract-timeout|--rotate-pages-threshold|--fast-web-view)
             # argument required but no completions available
-            return
+            return 0
+            ;;
+        --tesseract-config|--user-words|--user-patterns|--sidecar)
+            _filedir
+            return 0
             ;;
     esac
 
-    $split && return
+    return 1
+}
+
+_ocrmypdf()
+{
+    local OLDIFS="$IFS"
+    local IFS=$'\n'
+
+    local cur prev
+
+    # Homebrew on Macs have version 1.3 of bash-completion which doesn't include - see #502
+    if declare -F _init_completion >/dev/null 2>&1; then
+      _init_completion  || return
+    else
+        COMPREPLY=()
+        _get_comp_words_by_ref cur prev
+    fi
+
+    if __ocrmypdf_check_previous -ne 0; then
+        return
+    fi
+
+    if [[ "$cur" == -* ]]; then
+        __ocrmypdf_arguments
+    else
+        _filedir
+    fi
+
+    IFS="$OLDIFS"
+
+    return
 } &&
 complete -F _ocrmypdf ocrmypdf
 

misc/completion/ocrmypdf.fish

@@ -29,7 +29,6 @@ complete -c ocrmypdf -s d -l deskew -d "fix small horizontal alignment skew"
 complete -c ocrmypdf -s c -l clean -d "clean document images before OCR"
 complete -c ocrmypdf -s i -l clean-final -d "clean document images and keep result"
 complete -c ocrmypdf -l remove-vectors -d "don't send vector objects to OCR"
-complete -c ocrmypdf -l threshold -d "threshold images before OCR"
 
 complete -c ocrmypdf -s f -l force-ocr -d "OCR documents that already have printable text"
 complete -c ocrmypdf -s s -l skip-ocr -d "skip OCR on pages that text, otherwise try OCR"
@@ -54,6 +53,7 @@ function __fish_ocrmypdf_output_type
     echo -e "pdfa-1\t"(_ "output a PDF/A-1b")
     echo -e "pdfa-2\t"(_ "output a PDF/A-2b")
     echo -e "pdfa-3\t"(_ "output a PDF/A-3b")
+    echo -e "none\t"(_ "do not produce an output PDF (for example, if you only care about --sidecar)")
 end
 complete -c ocrmypdf -x -l output-type -a '(__fish_ocrmypdf_output_type)' -d "select PDF output options"
 
@@ -105,20 +105,20 @@ complete -c ocrmypdf -x -l pages -d "apply OCR to only the specified pages"
 complete -c ocrmypdf -x -l tesseract-config -d "set custom tesseract config file"
 
 function __fish_ocrmypdf_tesseract_pagesegmode
-  echo -e "0\t"(_ "orientation and script detection (OSD) only")
-  echo -e "1\t"(_ "automatic page segmentation with OSD")
-  echo -e "2\t"(_ "automatic page segmentation, but no OSD, or OCR")
-  echo -e "3\t"(_ "fully automatic page segmentation, but no OSD (default)")
-  echo -e "4\t"(_ "assume a single column of text of variable sizes")
-  echo -e "5\t"(_ "assume a single uniform block of vertically aligned text")
-  echo -e "6\t"(_ "assume a single uniform block of text")
-  echo -e "7\t"(_ "treat the image as a single text line")
-  echo -e "8\t"(_ "treat the image as a single word")
-  echo -e "9\t"(_ "treat the image as a single word in a circle")
-  echo -e "10\t"(_ "treat the image as a single character")
-  echo -e "11\t"(_ "sparse text - find as much text as possible in no particular order")
-  echo -e "12\t"(_ "sparse text with OSD")
-  echo -e "13\t"(_ "raw line - treat the image as a single text line")
+    echo -e "0\t"(_ "orientation and script detection (OSD) only")
+    echo -e "1\t"(_ "automatic page segmentation with OSD")
+    echo -e "2\t"(_ "automatic page segmentation, but no OSD, or OCR")
+    echo -e "3\t"(_ "fully automatic page segmentation, but no OSD (default)")
+    echo -e "4\t"(_ "assume a single column of text of variable sizes")
+    echo -e "5\t"(_ "assume a single uniform block of vertically aligned text")
+    echo -e "6\t"(_ "assume a single uniform block of text")
+    echo -e "7\t"(_ "treat the image as a single text line")
+    echo -e "8\t"(_ "treat the image as a single word")
+    echo -e "9\t"(_ "treat the image as a single word in a circle")
+    echo -e "10\t"(_ "treat the image as a single character")
+    echo -e "11\t"(_ "sparse text - find as much text as possible in no particular order")
+    echo -e "12\t"(_ "sparse text with OSD")
+    echo -e "13\t"(_ "raw line - treat the image as a single text line")
 end
 complete -c ocrmypdf -x -l tesseract-pagesegmode -a '(__fish_ocrmypdf_tesseract_pagesegmode)' -d "set tesseract --psm"
 
@@ -129,6 +129,15 @@ function __fish_ocrmypdf_tesseract_oem
     echo -e "3\t"(_ "default, based on what is available")
 end
 complete -c ocrmypdf -x -l tesseract-oem -a '(__fish_ocrmypdf_tesseract_oem)' -d "set tesseract --oem"
+
+function __fish_ocrmypdf_tesseract_thresholding
+    echo -e "auto\t"(_ "let OCRmyPDF pick thresholding (current always uses otsu)")
+    echo -e "otsu\t"(_ "legacy Otsu thresholding")
+    echo -e "adaptive-otsu\t"(_ "use adaptive Otsu thresholding")
+    echo -e "sauvola\t"(_ "use Sauvola thresholding")
+end
+complete -c ocrmypdf -x -l tesseract-thresholding -a '(__fish_ocrmypdf_tesseract_thresholding)' -d "set tesseract thresholding method (needs Tesseract 5.x)"
+
 complete -c ocrmypdf -x -l tesseract-timeout -d "maximum number of seconds to wait for OCR"
 complete -c ocrmypdf -x -l rotate-pages-threshold -d "page rotation confidence"
 

misc/synology.py

@@ -46,7 +46,7 @@ if len(sys.argv) > 1:
 else:
     start_dir = '.'
 
-for dir_name, subdirs, file_list in os.walk(start_dir):
+for dir_name, _subdirs, file_list in os.walk(start_dir):
     logging.info(dir_name)
     os.chdir(dir_name)
     for filename in file_list:

misc/watcher.py

@@ -1,3 +1,4 @@
+#!/usr/bin/env python3
 # Copyright (C) 2019 Ian Alexander: https://github.com/ianalexander
 # Copyright (C) 2020 James R Barlow: https://github.com/jbarlow83
 #
@@ -22,6 +23,7 @@
 import json
 import logging
 import os
+import shutil
 import sys
 import time
 from datetime import datetime
@@ -36,14 +38,21 @@ import ocrmypdf
 
 # pylint: disable=logging-format-interpolation
 
+
+def getenv_bool(name: str, default: str = 'False'):
+    return os.getenv(name, default).lower() in ('true', 'yes', 'y', '1')
+
+
 INPUT_DIRECTORY = os.getenv('OCR_INPUT_DIRECTORY', '/input')
 OUTPUT_DIRECTORY = os.getenv('OCR_OUTPUT_DIRECTORY', '/output')
-OUTPUT_DIRECTORY_YEAR_MONTH = bool(os.getenv('OCR_OUTPUT_DIRECTORY_YEAR_MONTH', ''))
-ON_SUCCESS_DELETE = bool(os.getenv('OCR_ON_SUCCESS_DELETE', ''))
-DESKEW = bool(os.getenv('OCR_DESKEW', ''))
+ARCHIVE_DIRECTORY = os.getenv('OCR_ARCHIVE_DIRECTORY', '/processed')
+OUTPUT_DIRECTORY_YEAR_MONTH = getenv_bool('OCR_OUTPUT_DIRECTORY_YEAR_MONTH')
+ON_SUCCESS_DELETE = getenv_bool('OCR_ON_SUCCESS_DELETE')
+ON_SUCCESS_ARCHIVE = getenv_bool('OCR_ON_SUCCESS_ARCHIVE')
+DESKEW = getenv_bool('OCR_DESKEW')
 OCR_JSON_SETTINGS = json.loads(os.getenv('OCR_JSON_SETTINGS', '{}'))
 POLL_NEW_FILE_SECONDS = int(os.getenv('OCR_POLL_NEW_FILE_SECONDS', '1'))
-USE_POLLING = bool(os.getenv('OCR_USE_POLLING', ''))
+USE_POLLING = getenv_bool('OCR_USE_POLLING')
 LOGLEVEL = os.getenv('OCR_LOGLEVEL', 'INFO')
 PATTERNS = ['*.pdf', '*.PDF']
 
@@ -102,9 +111,13 @@ def execute_ocrmypdf(file_path):
         deskew=DESKEW,
         **OCR_JSON_SETTINGS,
     )
-    if exit_code == 0 and ON_SUCCESS_DELETE:
-        log.info(f'OCR is done. Deleting: {file_path}')
-        file_path.unlink()
+    if exit_code == 0:
+        if ON_SUCCESS_DELETE:
+            log.info(f'OCR is done. Deleting: {file_path}')
+            file_path.unlink()
+        elif ON_SUCCESS_ARCHIVE:
+            log.info(f'OCR is done. Archiving {file_path.name} to {ARCHIVE_DIRECTORY}')
+            shutil.move(file_path, f'{ARCHIVE_DIRECTORY}/{file_path.name}')
     else:
         log.info('OCR is done')
 
@@ -129,13 +142,16 @@ def main():
         f"Starting OCRmyPDF watcher with config:\n"
         f"Input Directory: {INPUT_DIRECTORY}\n"
         f"Output Directory: {OUTPUT_DIRECTORY}\n"
-        f"Output Directory Year & Month: {OUTPUT_DIRECTORY_YEAR_MONTH}"
+        f"Output Directory Year & Month: {OUTPUT_DIRECTORY_YEAR_MONTH}\n"
+        f"Archive Directory: {ARCHIVE_DIRECTORY}"
     )
     log.debug(
         f"INPUT_DIRECTORY: {INPUT_DIRECTORY}\n"
         f"OUTPUT_DIRECTORY: {OUTPUT_DIRECTORY}\n"
         f"OUTPUT_DIRECTORY_YEAR_MONTH: {OUTPUT_DIRECTORY_YEAR_MONTH}\n"
+        f"ARCHIVE_DIRECTORY: {ARCHIVE_DIRECTORY}\n"
         f"ON_SUCCESS_DELETE: {ON_SUCCESS_DELETE}\n"
+        f"ON_SUCCESS_ARCHIVE: {ON_SUCCESS_ARCHIVE}\n"
         f"DESKEW: {DESKEW}\n"
         f"ARGS: {OCR_JSON_SETTINGS}\n"
         f"POLL_NEW_FILE_SECONDS: {POLL_NEW_FILE_SECONDS}\n"

misc/webservice.py

@@ -1,3 +1,4 @@
+#!/usr/bin/env python3
 # webservice.py wrapper for OCRmyPDF
 # Copyright (C) 2019 James R. Barlow: github.com/jbarlow83
 #
@@ -28,16 +29,7 @@ import shlex
 from subprocess import PIPE, run
 from tempfile import TemporaryDirectory
 
-from flask import (
-    Flask,
-    Response,
-    abort,
-    flash,
-    redirect,
-    request,
-    send_from_directory,
-    url_for,
-)
+from flask import Flask, Response, request, send_from_directory
 from werkzeug.utils import secure_filename
 
 app = Flask(__name__)
@@ -45,7 +37,7 @@ app.secret_key = "secret"
 app.config['MAX_CONTENT_LENGTH'] = 50_000_000
 app.config.from_envvar("OCRMYPDF_WEBSERVICE_SETTINGS", silent=True)
 
-ALLOWED_EXTENSIONS = set(["pdf"])
+ALLOWED_EXTENSIONS = {"pdf"}
 
 
 def allowed_file(filename):
@@ -67,7 +59,7 @@ def do_ocrmypdf(file):
         return Response("--sidecar not supported", 501, mimetype='text/plain')
 
     ocrmypdf_args = ["ocrmypdf", *cmd_args, up_file, down_file]
-    proc = run(ocrmypdf_args, stdout=PIPE, stderr=PIPE, encoding="utf-8")
+    proc = run(ocrmypdf_args, capture_output=True, encoding="utf-8")
     if proc.returncode != 0:
         stderr = proc.stderr
         return Response(stderr, 400, mimetype='text/plain')

pyproject.toml

@@ -2,15 +2,17 @@
 requires = [
   "setuptools >= 30.3.0",
   "wheel",
-  "cffi",
-  "setuptools_scm",
+  "setuptools_scm[toml] >= 3.4",
   "setuptools_scm_git_archive"
 ]
 build-backend = "setuptools.build_meta"
 
+[tool.setuptools_scm]
+version_scheme = "post-release"
+
 [tool.black]
 line-length = 88
-target-version = ["py36", "py37", "py38"]
+target-version = ["py37", "py38"]
 skip-string-normalization = true
 include = '\.pyi?$'
 exclude = '''
@@ -28,6 +30,77 @@ exclude = '''
   | docs
   | misc
   | \.egg-info
-  | src/ocrmypdf/lib/_leptonica.py
 )/
 '''
+
+[tool.coverage.run]
+branch = true
+parallel = true
+concurrency = ["multiprocessing"]
+
+[tool.coverage.paths]
+source = ["src/ocrmypdf"]
+
+[tool.coverage.report]
+# Regexes for lines to exclude from consideration
+exclude_lines = [
+  # Have to re-enable the standard pragma
+  "pragma: no cover",
+  # Don't complain if tests don't hit defensive assertion code:
+  "raise AssertionError",
+  "raise NotImplementedError",
+  # Don't complain if non-runnable code isn't run:
+  "if 0:",
+  "if False:",
+  "if __name__ == .__main__.:",
+  "if TYPE_CHECKING:"
+]
+
+[tool.isort]
+profile = "black"
+known_first_party = "ocrmypdf"
+known_third_party = [
+  "PIL",
+  "flask",
+  "img2pdf",
+  "ocrmypdf",
+  "pdfminer",
+  "pikepdf",
+  "pkg_resources",
+  "pluggy",
+  "pytest",
+  "reportlab",
+  "setuptools",
+  "sphinx_rtd_theme",
+  "tqdm",
+  "watchdog",
+  "werkzeug"
+]
+
+[tool.pytest.ini_options]
+minversion = "6.0"
+norecursedirs = ["lib", ".pc", ".git", "venv", "output", "cache", "resources"]
+testpaths = ["tests"]
+addopts = "-n auto"
+markers = ["slow"]
+filterwarnings = ["ignore:.*XMLParser.*:DeprecationWarning"]
+
+[tool.mypy]
+
+[[tool.mypy.overrides]]
+module = [
+  'pluggy',
+  'tqdm',
+  'coloredlogs',
+  'img2pdf',
+  'pdfminer.*',
+  'reportlab.*',
+  'fitz',
+  'libxmp.utils'
+]
+ignore_missing_imports = true
+
+[tool.pylint.basic]
+good-names = ["i", "j", "k", "ex", "Run", "_", "e", "p", "im", "w", "h", "m", "x", "y", "a", "b", "fp", "n", "f", "s", "v", "q", "dx", "dy"]
+logging-format-style = "old"
+disable = ["raw-checker-failed", "bad-inline-option", "locally-disabled", "file-ignored", "suppressed-message", "useless-suppression", "deprecated-pragma", "use-symbolic-message-instead", "logging-fstring-interpolation", "missing-function-docstring", "too-few-public-methods"]

requirements/main.txt

@@ -1,12 +0,0 @@
-# requirements.txt can be used to replicate the developer's build environment
-# setup.py lists a separate set of requirements that are looser to simplify
-# installation
-cffi == 1.14.3
-coloredlogs == 14.0  # technically optional
-img2pdf == 0.4.0
-pdfminer.six == 20201018
-pikepdf == 2.0.0
-pluggy == 0.13.1
-Pillow == 8.0.1
-reportlab == 3.5.55
-tqdm == 4.51.0

requirements/test.txt

@@ -1,7 +0,0 @@
-pytest >= 5.0.0
-pytest-helpers-namespace >= 2019.1.8
-pytest-xdist >= 1.31.0
-pytest-cov >= 2.10.0
-python-xmp-toolkit == 2.0.1  # requires apt-get install libexempi3
-                     # or brew install exempi
-#PyMuPDF == 1.13.4    # optional

requirements/watcher.txt

@@ -1 +0,0 @@
-watchdog == 0.10.2

requirements/webservice.txt

@@ -1 +0,0 @@
-Flask >= 1, < 2

setup.cfg

@@ -1,58 +1,113 @@
+[metadata]
+name = ocrmypdf
+description = OCRmyPDF adds an OCR text layer to scanned PDF files, allowing them to be searched
+long_description = file: README.md
+long_description_content_type = text/markdown
+url = https://github.com/ocrmypdf/OCRmyPDF
+author = James R. Barlow
+author_email = james@purplerock.ca
+license = MPL-2.0
+license_file = LICENSE
+license_files =
+    LICENSE
+classifiers =
+    Development Status :: 5 - Production/Stable
+    Environment :: Console
+    Intended Audience :: End Users/Desktop
+    Intended Audience :: Science/Research
+    Intended Audience :: System Administrators
+    License :: OSI Approved :: Mozilla Public License 2.0 (MPL 2.0)
+    Operating System :: MacOS :: MacOS X
+    Operating System :: Microsoft :: Windows :: Windows 10
+    Operating System :: POSIX
+    Operating System :: POSIX :: BSD
+    Operating System :: POSIX :: Linux
+    Programming Language :: Python :: 3
+    Programming Language :: Python :: 3 :: Only
+    Programming Language :: Python :: 3.7
+    Programming Language :: Python :: 3.8
+    Programming Language :: Python :: 3.9
+    Programming Language :: Python :: 3.10
+    Topic :: Scientific/Engineering :: Image Recognition
+    Topic :: Text Processing :: Indexing
+    Topic :: Text Processing :: Linguistic
+keywords =
+    PDF
+    OCR
+    optical character recognition
+    PDF/A
+    scanning
+project_urls =
+    Documentation = https://ocrmypdf.readthedocs.io/
+    Source = https://github.com/ocrmypdf/OCRmyPDF
+    Tracker = https://github.com/ocrmypdf/OCRmyPDF/issues
+
+[options]
+packages = find:
+install_requires =
+    Pillow>=8.2.0
+    coloredlogs>=14.0  # strictly optional
+    img2pdf>=0.3.0  # pure Python
+    packaging>=20
+    pdfminer.six!=20200720,>=20191110
+    pikepdf!=5.0.0,>=4.0.0
+    pluggy>=0.13.0
+    reportlab>=3.5.66
+    tqdm>=4
+    importlib-metadata>=4;python_version<'3.8'  # until Python 3.8
+    importlib-resources>=5;python_version<'3.9'  # until Python 3.9
+python_requires = >=3.7
+include_package_data = True
+package_dir =
+    =src
+platforms = any
+setup_requires =
+    setuptools-scm
+    setuptools-scm-git-archive
+zip_safe = False
+
+[options.packages.find]
+where = src
+
+[options.entry_points]
+console_scripts =
+    ocrmypdf = ocrmypdf.__main__:run
+
+[options.extras_require]
+docs =
+    sphinx
+    sphinx-issues
+    sphinx-rtd-theme
+extended_test =
+    PyMuPDF==1.19.1
+test =
+    coverage[toml]>=5
+    pytest>=6.0.0
+    pytest-cov>=2.11.1
+    pytest-xdist>=2.2.0
+    python-xmp-toolkit==2.0.1  # also requires apt-get install libexempi3
+watcher =
+    watchdog>=1.0.2
+webservice =
+    Flask>=1
+
+[options.package_data]
+ocrmypdf =
+    data/sRGB.icc
+    py.typed
+
 [bdist_wheel]
-python-tag = py36
+python-tag = py37
 
 [aliases]
-test=pytest
+test = pytest
 
 [check-manifest]
 ignore =
-	.github
+    .github
 
-[tool:pytest]
-norecursedirs = lib .pc .git output cache resources
-testpaths = tests
-filterwarnings =
-	ignore:.*XMLParser.*:DeprecationWarning
-markers =
-	slow
-addopts =
-	-n auto
-
-[isort]
-multi_line_output=3
-include_trailing_comma=True
-force_grid_wrap=0
-use_parentheses=True
-line_length=88
-known_first_party = ocrmypdf
-known_third_party = PIL,_cffi_backend,cffi,flask,img2pdf,pdfminer,pikepdf,pkg_resources,pluggy,pytest,reportlab,setuptools,sphinx_rtd_theme,tqdm,watchdog,werkzeug
-
-[metadata]
-license_file = LICENSE
-
-[coverage:paths]
-source =
-	src/
-
-[coverage:run]
-branch = true
-parallel = true
-concurrency = multiprocessing
-source =
-	src/ocrmypdf
-
-[coverage:report]
-# Regexes for lines to exclude from consideration
-exclude_lines =
-    # Have to re-enable the standard pragma
-    pragma: no cover
-
-    # Don't complain if tests don't hit defensive assertion code:
-    raise AssertionError
-    raise NotImplementedError
-
-    # Don't complain if non-runnable code isn't run:
-    if 0:
-	if False:
-    if __name__ == .__main__.:
-	if TYPE_CHECKING:
+[flake8]
+ignore = D203,F401,W503,E501,E203,F841
+exclude = .git,__pycache__,docs/conf.py,build,dist,.venv,.venvpp,.eggs,tmp,src/ocrmypdf/lib/
+max-complexity = 10
+max-line-length = 100

setup.py

@@ -1,93 +1,18 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-# © 2015 James R. Barlow: github.com/jbarlow83
+# © 2021 James R. Barlow: github.com/jbarlow83
 #
 # This Source Code Form is subject to the terms of the Mozilla Public
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""setup.py to support older setuptools and pip."""
 
-from __future__ import print_function, unicode_literals
-
-import sys
-
-from setuptools import find_packages, setup
-
-if sys.version_info < (3, 6):
-    print("Python 3.6 or newer is required", file=sys.stderr)
-    sys.exit(1)
-
-if 'upload' in sys.argv[1:]:
-    print('Use twine to upload the package - setup.py upload is insecure')
-    sys.exit(1)
-
-tests_require = open('requirements/test.txt', encoding='utf-8').read().splitlines()
-
-
-def readme():
-    with open('README.md', encoding='utf-8') as f:
-        return f.read()
-
+from setuptools import setup
 
+# Minimal setup to support older setuptools/setuptools_scm
 setup(
-    name='ocrmypdf',
-    description='OCRmyPDF adds an OCR text layer to scanned PDF files, allowing them to be searched',
-    long_description=readme(),
-    long_description_content_type='text/markdown',
-    url='https://github.com/jbarlow83/OCRmyPDF',
-    author='James R. Barlow',
-    author_email='james@purplerock.ca',
-    packages=find_packages('src', exclude=["tests", "tests.*"]),
-    package_dir={'': 'src'},
-    keywords=['PDF', 'OCR', 'optical character recognition', 'PDF/A', 'scanning'],
-    classifiers=[
-        "Programming Language :: Python :: 3.6",
-        "Programming Language :: Python :: 3.7",
-        "Programming Language :: Python :: 3.8",
-        "Programming Language :: Python :: 3.9",
-        "Development Status :: 5 - Production/Stable",
-        "Environment :: Console",
-        "Intended Audience :: End Users/Desktop",
-        "Intended Audience :: Science/Research",
-        "Intended Audience :: System Administrators",
-        "License :: OSI Approved :: Mozilla Public License 2.0 (MPL 2.0)",
-        "Operating System :: MacOS :: MacOS X",
-        "Operating System :: Microsoft :: Windows :: Windows 10",
-        "Operating System :: POSIX",
-        "Operating System :: POSIX :: BSD",
-        "Operating System :: POSIX :: Linux",
-        "Topic :: Scientific/Engineering :: Image Recognition",
-        "Topic :: Text Processing :: Indexing",
-        "Topic :: Text Processing :: Linguistic",
-    ],
-    python_requires=' >= 3.6',
     setup_requires=[  # can be removed whenever we can drop pip 9 support
-        'cffi >= 1.9.1',  # to build the leptonica module
         'setuptools_scm',  # so that version will work
         'setuptools_scm_git_archive',  # enable version from github tarballs
     ],
     use_scm_version={'version_scheme': 'post-release'},
-    cffi_modules=['src/ocrmypdf/lib/compile_leptonica.py:ffibuilder'],
-    install_requires=[
-        'cffi >= 1.9.1',  # must be a setup and install requirement
-        'coloredlogs >= 14.0',  # strictly optional
-        'img2pdf >= 0.3.0, < 0.5',  # pure Python, so track HEAD closely
-        'pdfminer.six >= 20191110, != 20200720, <= 20201018',
-        "pikepdf >= 1.14.0, < 3 ; python_version < '3.9'",
-        "pikepdf >= 2.0.0 ; python_version >= '3.9'",
-        'Pillow >= 7.0.0',
-        'pluggy >= 0.13.0, < 1.0',
-        'reportlab >= 3.3.0',  # oldest released version with sane image handling
-        'tqdm >= 4',
-    ],
-    tests_require=tests_require,
-    entry_points={'console_scripts': ['ocrmypdf = ocrmypdf.__main__:run']},
-    package_data={'ocrmypdf': ['data/sRGB.icc', 'py.typed']},
-    include_package_data=True,
-    zip_safe=False,
-    project_urls={
-        'Documentation': 'https://ocrmypdf.readthedocs.io/',
-        'Source': 'https://github.com/jbarlow83/ocrmypdf',
-        'Tracker': 'https://github.com/jbarlow83/ocrmypdf/issues',
-    },
 )

snapcraft.yaml

@@ -0,0 +1,72 @@
+name: ocrmypdf
+title: OCRmyPDF
+base: core20
+version: git
+summary: OCRmyPDF adds optical character recognition (OCR) to PDFs
+description: OCRmyPDF packaged for snap
+grade: stable
+confinement: strict
+icon: docs/images/logo-square-256.svg
+license: MPL-2.0
+
+architectures: [amd64]
+
+environment:
+  TESSDATA_PREFIX: $SNAP/usr/share/tesseract-ocr/4.00/tessdata
+  GS_LIB: $SNAP/usr/share/ghostscript/9.50/Resource/Init
+  GS_FONTPATH: $SNAP/usr/share/ghostscript/9.50/Resource/Font
+  LD_LIBRARY_PATH: $SNAP/usr/lib/x86_64-linux-gnu
+
+apps:
+  ocrmypdf:
+    command: usr/bin/snapcraft-preload python3 -m ocrmypdf
+    plugs:
+      - desktop
+      - desktop-legacy
+      - wayland
+      - x11
+      - home
+      - removable-media
+
+parts:
+  snapcraft-preload:
+    source: https://github.com/sergiusens/snapcraft-preload.git
+    plugin: cmake
+    cmake-parameters:
+      - -DCMAKE_INSTALL_PREFIX=/usr -DLIBPATH=/usr/lib
+    build-packages:
+      - on amd64:
+          - gcc-multilib
+          - g++-multilib
+    stage-packages:
+      - lib32stdc++6
+
+  ocrmypdf:
+    plugin: python
+    source: https://github.com/ocrmypdf/OCRmyPDF.git
+
+    stage-packages:
+      - ghostscript
+      - icc-profiles-free
+      - liblept5
+      - libxml2
+      - pngquant
+      - tesseract-ocr-all
+      - unpaper
+      - qpdf
+      - zlib1g
+
+    python-packages:
+      - cffi
+      - pdfminer.six
+      - pikepdf
+      - Pillow
+      - pluggy
+      - reportlab
+      - setuptools
+      - tqdm
+      - pipe
+
+    override-build: |
+      snapcraftctl build
+      ln -sf ../usr/lib/libsnapcraft-preload.so $SNAPCRAFT_PART_INSTALL/lib/libsnapcraft-preload.so

src/ocrmypdf/RELEASE.md

@@ -0,0 +1,35 @@
+# Release checklist
+
+## Patch release
+
+- Check `pytest`
+
+- Update release notes
+
+## Minor release
+
+## Major release
+
+- Run `pre-commit autoupdate`
+
+- Check README.md
+
+- Check setup.py
+
+    - Are classifiers up to date?
+    - Is `python_requires` correct?
+    - Python 3.6 is EOL on December 2021-12. Could drop support then.
+    - Can we tighten any `install_requires` dependencies?
+
+- Search for old version shims we can remove
+
+    - "shim"
+    - ` pikepdf.__version__`
+
+- Search for deprecation: search all files for deprec*, etc.
+
+- Check requirements in setup.cfg
+
+- Delete `tests/cache`, do `pytest --runslow`, and update cache.
+
+- Do `pytest --cov-report html`

src/ocrmypdf/__init__.py

@@ -4,10 +4,12 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""Adds OCR layer to PDFs."""
 
 from pluggy import HookimplMarker as _HookimplMarker
 
-from ocrmypdf import helpers, hocrtransform, leptonica, pdfa, pdfinfo
+from ocrmypdf import helpers, hocrtransform, pdfa, pdfinfo
+from ocrmypdf._concurrent import Executor
 from ocrmypdf._jobcontext import PageContext, PdfContext
 from ocrmypdf._version import PROGRAM_NAME, __version__
 from ocrmypdf.api import Verbosity, configure_logging, ocr

src/ocrmypdf/__main__.py

@@ -5,17 +5,19 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""ocrmypdf command line entrypoint."""
 
 import logging
 import os
 import signal
 import sys
+from contextlib import suppress
 from multiprocessing import set_start_method
 
 from ocrmypdf import __version__
 from ocrmypdf._plugin_manager import get_parser_options_plugins
 from ocrmypdf._sync import run_pipeline
-from ocrmypdf._validation import check_closed_streams, check_options
+from ocrmypdf._validation import check_options
 from ocrmypdf.api import Verbosity, configure_logging
 from ocrmypdf.exceptions import (
     BadArgsError,
@@ -34,10 +36,7 @@ def sigbus(*args):
 def run(args=None):
     _parser, options, plugin_manager = get_parser_options_plugins(args=args)
 
-    if not check_closed_streams(options):
-        return ExitCode.bad_args
-
-    if hasattr(os, 'nice'):
+    with suppress(AttributeError, PermissionError):
         os.nice(5)
 
     verbosity = options.verbose
@@ -47,7 +46,10 @@ def run(args=None):
         verbosity = Verbosity.quiet
         options.progress_bar = False
     configure_logging(
-        verbosity, progress_bar_friendly=options.progress_bar, manage_root_logger=True
+        verbosity,
+        progress_bar_friendly=options.progress_bar,
+        manage_root_logger=True,
+        plugin_manager=plugin_manager,
     )
     log.debug('ocrmypdf %s', __version__)
     try:
@@ -62,7 +64,7 @@ def run(args=None):
         log.error(e)
         return ExitCode.missing_dependency
 
-    if hasattr(signal, 'SIGBUS'):
+    with suppress(AttributeError, OSError):
         signal.signal(signal.SIGBUS, sigbus)
 
     result = run_pipeline(options=options, plugin_manager=plugin_manager)

src/ocrmypdf/_concurrent.py

@@ -4,134 +4,136 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""OCRmyPDF concurrency abstractions."""
 
-import logging
-import logging.handlers
-import multiprocessing
-import os
-import signal
-import sys
 import threading
-from contextlib import suppress
-from multiprocessing import Pool as ProcessPool
-from multiprocessing.dummy import Pool as ThreadPool
+from abc import ABC, abstractmethod
 from typing import Callable, Iterable, Optional
 
-from tqdm import tqdm
 
-from ocrmypdf.exceptions import InputFileError
+def _task_noop(*_args, **_kwargs):
+    return
 
 
-def log_listener(queue):
-    """Listen to the worker processes and forward the messages to logging
+class NullProgressBar:
+    """Progress bar API that takes no actions."""
 
-    For simplicity this is a thread rather than a process. Only one process
-    should actually write to sys.stderr or whatever we're using, so if this is
-    made into a process the main application needs to be directed to it.
+    def __init__(self, **kwargs):
+        pass
 
-    See https://docs.python.org/3/howto/logging-cookbook.html#logging-to-a-single-file-from-multiple-processes
+    def __enter__(self):
+        return self
+
+    def __exit__(self, exc_type, exc_value, traceback):
+        return False
+
+    def update(self, _arg=None):
+        return
+
+
+class Executor(ABC):
+    """Abstract concurrent executor."""
+
+    pool_lock = threading.Lock()
+    pbar_class = NullProgressBar
+
+    def __init__(self, *, pbar_class=None):
+        if pbar_class:
+            self.pbar_class = pbar_class
+
+    def __call__(
+        self,
+        *,
+        use_threads: bool,
+        max_workers: int,
+        tqdm_kwargs: dict,
+        worker_initializer: Optional[Callable] = None,
+        task: Optional[Callable] = None,
+        task_arguments: Optional[Iterable] = None,
+        task_finished: Optional[Callable] = None,
+    ) -> None:
+        """
+        Set up parallel execution and progress reporting.
+
+        Args:
+            use_threads: If ``False``, the workload is the sort that will benefit from
+                running in a multiprocessing context (for example, it uses Python
+                heavily, and parallelizing it with threads is not expected to be
+                performant).
+            max_workers: The maximum number of workers that should be run.
+            tdqm_kwargs: Arguments to set up the progress bar.
+            worker_initializer: Called when a worker is initialized, in the worker's
+                execution context. If the child workers are processes, it must be
+                possible to marshall/pickle the worker initializer.
+                ``functools.partial`` can be used to bind parameters.
+            task: Called when the worker starts a new task, in the worker's execution
+                context. Must be possible to marshall to the worker.
+            task_finished: Called when a worker finishes a task, in the parent's
+                context.
+            task_arguments: An iterable that generates a group of parameters for each
+                task. This runs in the parent's context, but the parameters must be
+                marshallable to the worker.
+        """
+
+        if not task_arguments:
+            return  # Nothing to do!
+        if not worker_initializer:
+            worker_initializer = _task_noop
+        if not task_finished:
+            task_finished = _task_noop
+        if not task:
+            task = _task_noop
+
+        with self.pool_lock:
+            self._execute(
+                use_threads=use_threads,
+                max_workers=max_workers,
+                tqdm_kwargs=tqdm_kwargs,
+                worker_initializer=worker_initializer,
+                task=task,
+                task_arguments=task_arguments,
+                task_finished=task_finished,
+            )
+
+    @abstractmethod
+    def _execute(
+        self,
+        *,
+        use_threads: bool,
+        max_workers: int,
+        tqdm_kwargs: dict,
+        worker_initializer: Callable,
+        task: Callable,
+        task_arguments: Iterable,
+        task_finished: Callable,
+    ):
+        """Custom executors should override this method."""
+
+
+def setup_executor(plugin_manager) -> Executor:
+    pbar_class = plugin_manager.hook.get_progressbar_class()
+    return plugin_manager.hook.get_executor(progressbar_class=pbar_class)
+
+
+class SerialExecutor(Executor):
+    """Implements a purely sequential executor using the parallel protocol.
+
+    The current process/thread will be the worker that executes all tasks
+    in order. As such, ``worker_initializer`` will never be called.
     """
 
-    while True:
-        try:
-            record = queue.get()
-            if record is None:
-                break
-            logger = logging.getLogger(record.name)
-            logger.handle(record)
-        except Exception:  # pylint: disable=broad-except
-            import traceback  # pylint: disable=import-outside-toplevel
-
-            print("Logging problem", file=sys.stderr)
-            traceback.print_exc(file=sys.stderr)
-
-
-def process_sigbus(*args):
-    raise InputFileError("A worker process lost access to an input file")
-
-
-def process_init(queue, user_init, loglevel):
-    """Initialize a process pool worker"""
-
-    # Ignore SIGINT (our parent process will kill us gracefully)
-    signal.signal(signal.SIGINT, signal.SIG_IGN)
-
-    # Install SIGBUS handler (so our parent process can abort somewhat gracefully)
-    with suppress(AttributeError):  # Windows and Cygwin do not have SIGBUS
-        signal.signal(signal.SIGBUS, process_sigbus)
-
-    # Reconfigure the root logger for this process to send all messages to a queue
-    h = logging.handlers.QueueHandler(queue)
-    root = logging.getLogger()
-    root.setLevel(loglevel)
-    root.handlers = []
-    root.addHandler(h)
-
-    if user_init:
-        user_init()
-
-
-def thread_init(_queue, user_init, _loglevel):
-    # As a thread, block SIGBUS so the main thread deals with it...
-    with suppress(AttributeError):
-        # Windows and Cygwin do not have pthread_sigmask or SIGBUS
-        signal.pthread_sigmask(signal.SIG_BLOCK, {signal.SIGBUS})
-    if user_init:
-        user_init()
-
-
-def exec_progress_pool(
-    *,
-    use_threads: bool,
-    max_workers: int,
-    tqdm_kwargs: dict,
-    task_initializer: Optional[Callable] = None,
-    task: Optional[Callable] = None,
-    task_arguments: Optional[Iterable] = None,
-    task_finished: Optional[Callable] = None,
-):
-    log_queue: multiprocessing.Queue = multiprocessing.Queue(-1)
-    listener = threading.Thread(target=log_listener, args=(log_queue,))
-
-    if use_threads:
-        pool_class = ThreadPool
-        initializer = thread_init
-    else:
-        pool_class = ProcessPool
-        initializer = process_init
-    listener.start()
-
-    with tqdm(**tqdm_kwargs) as pbar:
-        pool = pool_class(
-            processes=max_workers,
-            initializer=initializer,
-            initargs=(log_queue, task_initializer, logging.getLogger("").level),
-        )
-        try:
-            results = pool.imap_unordered(task, task_arguments)
-            for result in results:
-                if task_finished:
-                    task_finished(result, pbar)
-                else:
-                    pbar.update()
-        except KeyboardInterrupt:
-            # Terminate pool so we exit instantly
-            pool.terminate()
-            # Don't try listener.join() here, will deadlock
-            raise
-        except Exception:
-            if not os.environ.get("PYTEST_CURRENT_TEST", ""):
-                # Unless inside pytest, exit immediately because no one wants
-                # to wait for child processes to finalize results that will be
-                # thrown away. Inside pytest, we want child processes to exit
-                # cleanly so that they output an error messages or coverage data
-                # we need from them.
-                pool.terminate()
-            raise
-        finally:
-            # Terminate log listener
-            log_queue.put_nowait(None)
-            pool.close()
-            pool.join()
-
-    listener.join()
+    def _execute(
+        self,
+        *,
+        use_threads: bool,
+        max_workers: int,
+        tqdm_kwargs: dict,
+        worker_initializer: Callable,
+        task: Callable,
+        task_arguments: Iterable,
+        task_finished: Callable,
+    ):  # pylint: disable=unused-argument
+        with self.pbar_class(**tqdm_kwargs) as pbar:
+            for args in task_arguments:
+                result = task(args)
+                task_finished(result, pbar)

src/ocrmypdf/_exec/ghostscript.py

@@ -10,43 +10,40 @@
 import logging
 import os
 import re
+import sys
 from io import BytesIO
 from os import fspath
 from pathlib import Path
-from shutil import which
 from subprocess import PIPE, CalledProcessError
 from typing import Optional
 
-from PIL import Image
+from PIL import Image, UnidentifiedImageError
 
-from ocrmypdf.exceptions import MissingDependencyError, SubprocessOutputError
+from ocrmypdf.exceptions import SubprocessOutputError
 from ocrmypdf.helpers import Resolution
 from ocrmypdf.subprocess import get_version, run, run_polling_stderr
 
+# Remove this workaround when we require Pillow >= 10
+try:
+    Transpose = Image.Transpose  # type: ignore
+except AttributeError:
+    # Pillow 9 shim
+    Transpose = Image  # type: ignore
+
 log = logging.getLogger(__name__)
 
-missing_gs_error = """
----------------------------------------------------------------------
-This error normally occurs when ocrmypdf find can't Ghostscript.
-Please ensure Ghostscript is installed and its location is added to
-the system PATH environment variable.
+# Most reliable what to get the bitness of Python interpreter, according to Python docs
+_IS_64BIT = sys.maxsize > 2**32
 
-For details see:
-    https://ocrmypdf.readthedocs.io/en/latest/installation.html
----------------------------------------------------------------------
-"""
-
-_gswin = None
+_GSWIN = None
 if os.name == 'nt':
-    _gswin = which('gswin64c')
-    if not _gswin:
-        _gswin = which('gswin32c')
-        if not _gswin:
-            raise MissingDependencyError(missing_gs_error)
-    _gswin = Path(_gswin).stem
+    if _IS_64BIT:
+        _GSWIN = 'gswin64c'
+    else:
+        _GSWIN = 'gswin32c'
 
-GS = _gswin if _gswin else 'gs'
-del _gswin
+GS = _GSWIN if _GSWIN else 'gs'
+del _GSWIN
 
 
 def version():
@@ -69,7 +66,8 @@ def jpeg_passthrough_available() -> bool:
 
 
 def _gs_error_reported(stream) -> bool:
-    return True if re.search(r'error', stream, flags=re.IGNORECASE) else False
+    match = re.search(r'error', stream, flags=re.IGNORECASE)
+    return bool(match)
 
 
 def rasterize_pdf(
@@ -95,6 +93,7 @@ def rasterize_pdf(
             '-dSAFER',
             '-dBATCH',
             '-dNOPAUSE',
+            '-dInterpolateControl=-1',
             f'-sDEVICE={raster_device}',
             f'-dFirstPage={pageno}',
             f'-dLastPage={pageno}',
@@ -104,7 +103,7 @@ def rasterize_pdf(
         + [
             '-o',
             '-',
-            '-sstdout=%stderr',
+            '-sstdout=%stderr',  # Literal %s, not string interpolation
             '-dAutoRotatePages=/None',  # Probably has no effect on raster
             '-f',
             fspath(input_file),
@@ -115,29 +114,38 @@ def rasterize_pdf(
         p = run(args_gs, stdout=PIPE, stderr=PIPE, check=True)
     except CalledProcessError as e:
         log.error(e.stderr.decode(errors='replace'))
-        raise SubprocessOutputError('Ghostscript rasterizing failed')
+        raise SubprocessOutputError('Ghostscript rasterizing failed') from e
     else:
         stderr = p.stderr.decode(errors='replace')
         if _gs_error_reported(stderr):
             log.error(stderr)
 
-    with Image.open(BytesIO(p.stdout)) as im:
-        if rotation is not None:
-            log.debug("Rotating output by %i", rotation)
-            # rotation is a clockwise angle and Image.ROTATE_* is
-            # counterclockwise so this cancels out the rotation
-            if rotation == 90:
-                im = im.transpose(Image.ROTATE_90)
-            elif rotation == 180:
-                im = im.transpose(Image.ROTATE_180)
-            elif rotation == 270:
-                im = im.transpose(Image.ROTATE_270)
-            if rotation % 180 == 90:
-                page_dpi = page_dpi.flip_axis()
-        im.save(fspath(output_file), dpi=page_dpi)
+    try:
+        with Image.open(BytesIO(p.stdout)) as im:
+            if rotation is not None:
+                log.debug("Rotating output by %i", rotation)
+                # rotation is a clockwise angle and Image.ROTATE_* is
+                # counterclockwise so this cancels out the rotation
+                if rotation == 90:
+                    im = im.transpose(Transpose.ROTATE_90)
+                elif rotation == 180:
+                    im = im.transpose(Transpose.ROTATE_180)
+                elif rotation == 270:
+                    im = im.transpose(Transpose.ROTATE_270)
+                if rotation % 180 == 90:
+                    page_dpi = page_dpi.flip_axis()
+            im.save(fspath(output_file), dpi=page_dpi)
+    except UnidentifiedImageError:
+        log.error(
+            f"Ghostscript (using {raster_device} at {raster_dpi} dpi) produced "
+            "an invalid page image file."
+        )
+        raise
 
 
 class GhostscriptFollower:
+    """Parses the output of Ghostscript and uses it to update the progress bar."""
+
     re_process = re.compile(r"Processing pages \d+ through (\d+).")
     re_page = re.compile(r"Page (\d+)")
 
@@ -158,14 +166,14 @@ class GhostscriptFollower:
                 )
                 return
         else:
-            m = self.re_page.match(line.strip())
-            if m:
+            if self.re_page.match(line.strip()):
                 self.progressbar.update()
 
 
 def generate_pdfa(
     pdf_pages,
     output_file: os.PathLike,
+    *,
     compression: str,
     pdf_version: str = '1.5',
     pdfa_part: str = '2',
@@ -195,17 +203,22 @@ def generate_pdfa(
             "-dAutoFilterGrayImages=true",
         ]
 
+    strategy = 'LeaveColorUnchanged'
     # Older versions of Ghostscript expect a leading slash in
     # sColorConversionStrategy, newer ones should not have it. See Ghostscript
     # git commit fe1c025d.
-    strategy = 'RGB' if version() >= '9.19' else '/RGB'
+    gs_version = version()
+    strategy = ('/' + strategy) if gs_version < '9.19' else strategy
 
-    if version() == '9.23':
+    if gs_version == '9.23':
         # 9.23: added JPEG passthrough as a new feature, but with a bug that
         # incorrectly formats some images. Fixed as of 9.24. So we disable this
         # feature for 9.23.
         # https://bugs.ghostscript.com/show_bug.cgi?id=699216
         compression_args.append('-dPassThroughJPEGImages=false')
+    elif gs_version == '9.56.0':
+        # 9.56.0 breaks our OCR...?
+        compression_args.append('-dNEWPDF=false')
 
     # nb no need to specify ProcessColorModel when ColorConversionStrategy
     # is set; see:
@@ -228,7 +241,7 @@ def generate_pdfa(
             "-dPDFACompatibilityPolicy=1",
             "-o",
             "-",
-            "-sstdout=%stderr",
+            "-sstdout=%stderr",  # Literal %s, not string interpolation
         ]
     )
     args_gs.extend(fspath(s) for s in pdf_pages)  # Stringify Path objs

src/ocrmypdf/_exec/jbig2enc.py

@@ -41,9 +41,17 @@ def convert_group(*, cwd, infiles, out_prefix):
     return proc
 
 
+def convert_group_mp(args):
+    return convert_group(cwd=args[0], infiles=args[1], out_prefix=args[2])
+
+
 def convert_single(*, cwd, infile, outfile):
     args = ['jbig2', '-p', infile]
     with open(outfile, 'wb') as fstdout:
         proc = run(args, cwd=cwd, stdout=fstdout, stderr=PIPE)
     proc.check_returncode()
     return proc
+
+
+def convert_single_mp(args):
+    return convert_single(cwd=args[0], infile=args[1], outfile=args[2])

src/ocrmypdf/_exec/pngquant.py

@@ -59,3 +59,7 @@ def quantize(input_file: Path, output_file: Path, quality_min: int, quality_max:
     if result.returncode == 0:
         # input_file could be the same as output_file, so we defer the write
         output_file.write_bytes(result.stdout)
+
+
+def quantize_mp(args):
+    return quantize(*args)

src/ocrmypdf/_exec/tesseract.py

@@ -8,14 +8,14 @@
 """Interface to Tesseract executable"""
 
 import logging
-import os
-import shutil
-from collections import namedtuple
+import re
+from math import pi
 from os import fspath
 from pathlib import Path
 from subprocess import PIPE, STDOUT, CalledProcessError, TimeoutExpired
-from typing import List, Optional
+from typing import Dict, List, Optional
 
+from packaging.version import Version
 from PIL import Image
 
 from ocrmypdf.exceptions import (
@@ -23,11 +23,11 @@ from ocrmypdf.exceptions import (
     SubprocessOutputError,
     TesseractConfigError,
 )
+from ocrmypdf.pluginspec import OrientationConfidence
 from ocrmypdf.subprocess import get_version, run
 
 log = logging.getLogger(__name__)
 
-OrientationConfidence = namedtuple('OrientationConfidence', ('angle', 'confidence'))
 
 HOCR_TEMPLATE = """<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
@@ -46,38 +46,77 @@ HOCR_TEMPLATE = """<?xml version="1.0" encoding="UTF-8"?>
 </html>
 """
 
+TESSERACT_THRESHOLDING_METHODS: Dict[str, int] = {
+    'auto': 0,
+    'otsu': 0,
+    'adaptive-otsu': 1,
+    'sauvola': 2,
+}
+
 
 class TesseractLoggerAdapter(logging.LoggerAdapter):
+    "Prepend [tesseract] to messages emitted from tesseract"
+
     def process(self, msg, kwargs):
         kwargs['extra'] = self.extra
-        return '[tesseract] %s' % (msg), kwargs
+        return f'[tesseract] {msg}', kwargs
 
 
-def version():
+TESSERACT_VERSION_PATTERN = r"""
+    v?
+    (?:
+        (?:(?P<epoch>[0-9]+)!)?                           # epoch
+        (?P<release>[0-9]+(?:\.[0-9]+)*)                  # release segment
+        (?P<pre>                                          # pre-release
+            [-_\.]?
+            (?P<pre_l>(a|b|c|rc|alpha|beta|pre|preview))
+            [-_\.]?
+            (?P<pre_n>[0-9]+)?
+        )?
+        (?P<post>                                         # post release
+            (?:-(?P<post_n1>[0-9]+))
+            |
+            (?:
+                [-_\.]?
+                (?P<post_l>post|rev|r)
+                [-_\.]?
+                (?P<post_n2>[0-9]+)?
+            )
+        )?
+        (?P<dev>                                          # dev release
+            [-_\.]?
+            (?P<dev_l>dev)
+            [-_\.]?
+            (?P<dev_n>[0-9]+)?
+        )?
+        (?P<date>
+            [-_\.]
+            (?:20[0-9][0-9] [0-1][0-9] [0-3][0-9])       # yyyy mm dd
+        )?
+        (?P<gitcount>
+            [-_\.]?
+            [0-9]+
+        )?
+        (?P<gitcommit>
+            [-_\.]?
+            g[0-9a-f]{2,10}
+        )?
+    )
+    (?:\+(?P<local>[a-z0-9]+(?:[-_\.][a-z0-9]+)*))?       # local version
+"""
+
+
+class TesseractVersion(Version):
+    "Modify standard packaging.Version regex to support Tesseract idiosyncracies."
+    _regex = re.compile(
+        r"^\s*" + TESSERACT_VERSION_PATTERN + r"\s*$", re.VERBOSE | re.IGNORECASE
+    )
+
+
+def version() -> str:
     return get_version('tesseract', regex=r'tesseract\s(.+)')
 
 
-def has_textonly_pdf(langs=None):
-    """Does Tesseract have textonly_pdf capability?
-
-    Available in v4.00.00alpha since January 2017. Best to
-    parse the parameter list.
-    """
-    args_tess = tess_base_args(langs, engine_mode=None) + ['--print-parameters', 'pdf']
-    params = ''
-    try:
-        proc = run(args_tess, check=True, stdout=PIPE, stderr=STDOUT)
-        params = proc.stdout
-    except CalledProcessError as e:
-        raise MissingDependencyError(
-            "Could not --print-parameters from tesseract. This can happen if the "
-            "TESSDATA_PREFIX environment is not set to a valid tessdata folder. "
-        ) from e
-    if b'textonly_pdf' in params:
-        return True
-    return False
-
-
 def has_user_words():
     """Does Tesseract have --user-words capability?
 
@@ -87,6 +126,11 @@ def has_user_words():
     return version() >= '4.1'
 
 
+def has_thresholding():
+    """Does Tesseract have -c thresholding method capability?"""
+    return version() >= '5.0'
+
+
 def get_languages():
     def lang_error(output):
         msg = (
@@ -115,7 +159,7 @@ def get_languages():
         if line.startswith('Error'):
             raise MissingDependencyError(lang_error(output))
     _header, *rest = output.splitlines()
-    return set(lang.strip() for lang in rest)
+    return {lang.strip() for lang in rest}
 
 
 def tess_base_args(langs: List[str], engine_mode: Optional[int]) -> List[str]:
@@ -127,7 +171,20 @@ def tess_base_args(langs: List[str], engine_mode: Optional[int]) -> List[str]:
     return args
 
 
-def get_orientation(input_file: Path, engine_mode: Optional[int], timeout: float):
+def _parse_tesseract_output(binary_output: bytes) -> Dict[str, str]:
+    def gen():
+        for line in binary_output.decode().splitlines():
+            line = line.strip()
+            parts = line.split(':', maxsplit=2)
+            if len(parts) == 2:
+                yield parts[0].strip(), parts[1].strip()
+
+    return dict(gen())
+
+
+def get_orientation(
+    input_file: Path, engine_mode: Optional[int], timeout: float
+) -> OrientationConfidence:
     args_tesseract = tess_base_args(['osd'], engine_mode) + [
         '--psm',
         '0',
@@ -137,7 +194,6 @@ def get_orientation(input_file: Path, engine_mode: Optional[int], timeout: float
 
     try:
         p = run(args_tesseract, stdout=PIPE, stderr=STDOUT, timeout=timeout, check=True)
-        stdout = p.stdout
     except TimeoutExpired:
         return OrientationConfidence(angle=0, confidence=0.0)
     except CalledProcessError as e:
@@ -149,19 +205,44 @@ def get_orientation(input_file: Path, engine_mode: Optional[int], timeout: float
         ):
             return OrientationConfidence(0, 0)
         raise SubprocessOutputError() from e
-    else:
-        osd = {}
-        for line in stdout.decode().splitlines():
-            line = line.strip()
-            parts = line.split(':', maxsplit=2)
-            if len(parts) == 2:
-                osd[parts[0].strip()] = parts[1].strip()
 
-        angle = int(osd.get('Orientation in degrees', 0))
-        oc = OrientationConfidence(
-            angle=angle, confidence=float(osd.get('Orientation confidence', 0))
-        )
-        return oc
+    osd = _parse_tesseract_output(p.stdout)
+    angle = int(osd.get('Orientation in degrees', 0))
+    orient_conf = OrientationConfidence(
+        angle=angle, confidence=float(osd.get('Orientation confidence', 0))
+    )
+    return orient_conf
+
+
+def get_deskew(
+    input_file: Path, languages: List[str], engine_mode: Optional[int], timeout: float
+) -> float:
+    """Gets angle to deskew this page, in degrees."""
+    args_tesseract = tess_base_args(languages, engine_mode) + [
+        '--psm',
+        '2',
+        fspath(input_file),
+        'stdout',
+    ]
+
+    try:
+        p = run(args_tesseract, stdout=PIPE, stderr=STDOUT, timeout=timeout, check=True)
+    except TimeoutExpired:
+        return 0.0
+    except CalledProcessError as e:
+        tesseract_log_output(e.stdout)
+        tesseract_log_output(e.stderr)
+        if b'Empty page!!' in e.output or (
+            e.output == b'' and e.returncode == 1
+        ):  # Not enough info for a skew angle - Tess 4 and 5 return different errors
+            return 0.0
+
+        raise SubprocessOutputError() from e
+
+    parsed = _parse_tesseract_output(p.stdout)
+    deskew_radians = float(parsed.get('Deskew angle', 0))
+    deskew_degrees = 180 / pi * deskew_radians
+    return deskew_degrees
 
 
 def tesseract_log_output(stream):
@@ -221,6 +302,7 @@ def _generate_null_hocr(output_hocr, output_text, image):
 
 
 def generate_hocr(
+    *,
     input_file: Path,
     output_hocr: Path,
     output_text: Path,
@@ -229,9 +311,11 @@ def generate_hocr(
     tessconfig: List[str],
     timeout: float,
     pagesegmode: int,
+    thresholding: int,
     user_words,
     user_patterns,
 ):
+    """Generate a hOCR file, which must be converted to PDF."""
     prefix = output_hocr.with_suffix('')
 
     args_tesseract = tess_base_args(languages, engine_mode)
@@ -239,6 +323,9 @@ def generate_hocr(
     if pagesegmode is not None:
         args_tesseract.extend(['--psm', str(pagesegmode)])
 
+    if thresholding != 0 and has_thresholding():
+        args_tesseract.extend(['-c', f'thresholding_method={thresholding}'])
+
     if user_words:
         args_tesseract.extend(['--user-words', user_words])
 
@@ -247,7 +334,8 @@ def generate_hocr(
 
     # Reminder: test suite tesseract test plugins will break after any changes
     # to the number of order parameters here
-    args_tesseract.extend([input_file, prefix, 'hocr', 'txt'] + tessconfig)
+    args_tesseract.extend([fspath(input_file), fspath(prefix), 'hocr', 'txt'])
+    args_tesseract.extend(tessconfig)
     try:
         p = run(args_tesseract, stdout=PIPE, stderr=STDOUT, timeout=timeout, check=True)
         stdout = p.stdout
@@ -259,7 +347,7 @@ def generate_hocr(
         _generate_null_hocr(output_hocr, output_text, input_file)
     except CalledProcessError as e:
         tesseract_log_output(e.output)
-        if b'Image too large' in e.output:
+        if b'Image too large' in e.output or b'Empty page!!' in e.output:
             _generate_null_hocr(output_hocr, output_text, input_file)
             return
 
@@ -269,7 +357,7 @@ def generate_hocr(
         # The sidecar text file will get the suffix .txt; rename it to
         # whatever caller wants it named
         if prefix.with_suffix('.txt').exists():
-            shutil.move(prefix.with_suffix('.txt'), output_text)
+            prefix.with_suffix('.txt').replace(output_text)
 
 
 def use_skip_page(output_pdf, output_text):
@@ -289,20 +377,15 @@ def generate_pdf(
     tessconfig: List[str],
     timeout: float,
     pagesegmode: int,
+    thresholding: int,
     user_words,
     user_patterns,
 ):
-    """Use Tesseract to render a PDF.
+    """Generate a PDF using Tesseract's internal PDF generator.
 
-    input_file -- image to analyze
-    output_pdf -- file to generate
-    output_text -- OCR text file
-    languages -- list of languages to consider
-    engine_mode -- engine mode argument for tess v4
-    tessconfig -- tesseract configuration
-    timeout -- timeout (seconds)
+    We specifically a text-only PDF which is more suitable for combining with
+    the input page.
     """
-
     args_tesseract = tess_base_args(languages, engine_mode)
 
     if pagesegmode is not None:
@@ -310,29 +393,33 @@ def generate_pdf(
 
     args_tesseract.extend(['-c', 'textonly_pdf=1'])
 
+    if thresholding != 0 and has_thresholding():
+        args_tesseract.extend(['-c', f'thresholding_method={thresholding}'])
+
     if user_words:
         args_tesseract.extend(['--user-words', user_words])
 
     if user_patterns:
         args_tesseract.extend(['--user-patterns', user_patterns])
 
-    prefix = os.path.splitext(output_pdf)[0]  # Tesseract appends suffixes
+    prefix = output_pdf.parent / Path(output_pdf.stem)
 
     # Reminder: test suite tesseract test plugins might break after any changes
     # to the number of order parameters here
 
-    args_tesseract.extend([input_file, prefix, 'pdf', 'txt'] + tessconfig)
+    args_tesseract.extend([fspath(input_file), fspath(prefix), 'pdf', 'txt'])
+    args_tesseract.extend(tessconfig)
     try:
         p = run(args_tesseract, stdout=PIPE, stderr=STDOUT, timeout=timeout, check=True)
         stdout = p.stdout
-        if os.path.exists(prefix + '.txt'):
-            shutil.move(prefix + '.txt', output_text)
+        if prefix.with_suffix('.txt').exists():
+            prefix.with_suffix('.txt').replace(output_text)
     except TimeoutExpired:
         page_timedout(timeout)
         use_skip_page(output_pdf, output_text)
     except CalledProcessError as e:
         tesseract_log_output(e.output)
-        if b'Image too large' in e.output:
+        if b'Image too large' in e.output or b'Empty page!!' in e.output:
             use_skip_page(output_pdf, output_text)
             return
         raise SubprocessOutputError() from e

src/ocrmypdf/_exec/unpaper.py

@@ -13,69 +13,119 @@
 import logging
 import os
 import shlex
+import sys
+from contextlib import contextmanager
 from decimal import Decimal
 from pathlib import Path
 from subprocess import PIPE, STDOUT
-from tempfile import TemporaryDirectory
-from typing import List, Optional, Tuple, Union
+from typing import Iterator, List, Optional, Tuple, Union
 
 from PIL import Image
 
 from ocrmypdf.exceptions import MissingDependencyError, SubprocessOutputError
-from ocrmypdf.subprocess import get_version
-from ocrmypdf.subprocess import run as external_run
+from ocrmypdf.subprocess import get_version, run
+
+if sys.version_info >= (3, 10):
+    from tempfile import TemporaryDirectory
+else:
+    from tempfile import TemporaryDirectory as _TemporaryDirectory
+
+    class TemporaryDirectory(_TemporaryDirectory):
+        """Shim to consume ignore_cleanup_errors kwarg on Python 3.9 and older.
+
+        The argument is consumed without action. If users are getting errors related
+        to temporary file cleanup, they should upgrade to Python 3.10 which properly
+        cleans up temporary directories on Windows.
+
+        See: https://github.com/python/cpython/pull/24793
+        """
+
+        def __init__(self, ignore_cleanup_errors=False, **kwargs):
+            super().__init__(**kwargs)
+
+    del _TemporaryDirectory
+
+
+UNPAPER_IMAGE_PIXEL_LIMIT = 256 * 1024 * 1024
 
 DecFloat = Union[Decimal, float]
 
 log = logging.getLogger(__name__)
 
 
+class UnpaperImageTooLargeError(Exception):
+    """To capture details when an image is too large for unpaper."""
+
+    def __init__(
+        self,
+        w,
+        h,
+        message="Image with size {}x{} is too large for cleaning with 'unpaper'.",
+    ):
+        self.w = w
+        self.h = h
+        self.message = message.format(w, h)
+        super().__init__(self.message)
+
+
 def version() -> str:
     return get_version('unpaper')
 
 
-def _setup_unpaper_io(tmpdir: Path, input_file: Path) -> Tuple[Path, Path]:
-    SUFFIXES = {'1': '.pbm', 'L': '.pgm', 'RGB': '.ppm'}
-    with Image.open(input_file) as im:
-        im_modified = False
-        if im.mode not in SUFFIXES:
-            log.info("Converting image to other colorspace")
-            try:
-                if im.mode == 'P' and len(im.getcolors()) == 2:
-                    im = im.convert(mode='1')
-                else:
-                    im = im.convert(mode='RGB')
-            except IOError as e:
-                raise MissingDependencyError(
-                    "Could not convert image with type " + im.mode
-                ) from e
-            else:
-                im_modified = True
+SUPPORTED_MODES = {'1', 'L', 'RGB'}
+
+
+def _convert_image(im: Image.Image) -> Tuple[Image.Image, bool]:
+    im_modified = False
+
+    if im.mode not in SUPPORTED_MODES:
+        log.info("Converting image to other colorspace")
         try:
-            suffix = SUFFIXES[im.mode]
-        except KeyError:
+            if im.mode == 'P' and len(im.getcolors()) == 2:
+                im = im.convert(mode='1')
+            else:
+                im = im.convert(mode='RGB')
+        except OSError as e:
+            raise MissingDependencyError(
+                "Could not convert image with type " + im.mode
+            ) from e
+        else:
+            im_modified = True
+        if im.mode not in SUPPORTED_MODES:
             raise MissingDependencyError(
                 "Failed to convert image to a supported format."
             ) from None
-
-        if im_modified or input_file.suffix != '.pnm':
-            input_pnm = tmpdir / 'input.pnm'
-            im.save(input_pnm, format='PPM')
-        else:
-            # No changes, PNG input, just use the file we already have
-            input_pnm = input_file
-        output_pnm = tmpdir / f'output{suffix}'
-    return input_pnm, output_pnm
+    return im, im_modified
 
 
-def run(
-    input_file: Path, output_file: Path, dpi: DecFloat, mode_args: List[str]
+@contextmanager
+def _setup_unpaper_io(input_file: Path) -> Iterator[Tuple[Path, Path, Path]]:
+    with Image.open(input_file) as im:
+        if im.width * im.height >= UNPAPER_IMAGE_PIXEL_LIMIT:
+            raise UnpaperImageTooLargeError(w=im.width, h=im.height)
+        im, im_modified = _convert_image(im)
+
+        with TemporaryDirectory(ignore_cleanup_errors=True) as tmpdir:
+            tmppath = Path(tmpdir)
+            if im_modified or input_file.suffix != '.png':
+                input_png = tmppath / 'input.png'
+                im.save(input_png, format='PNG')
+            else:
+                # No changes, PNG input, just use the file we already have
+                input_png = input_file
+
+            # unpaper can write .png too, but it seems to write them slowly
+            # adds a few seconds to test suite - so just use pnm
+            output_pnm = tmppath / 'output.pnm'
+            yield input_png, output_pnm, tmppath
+
+
+def run_unpaper(
+    input_file: Path, output_file: Path, *, dpi: DecFloat, mode_args: List[str]
 ) -> None:
     args_unpaper = ['unpaper', '-v', '--dpi', str(round(dpi, 6))] + mode_args
 
-    with TemporaryDirectory() as tmpdir:
-        input_pnm, output_pnm = _setup_unpaper_io(Path(tmpdir), input_file)
-
+    with _setup_unpaper_io(input_file) as (input_png, output_pnm, tmpdir):
         # To prevent any shenanigans from accepting arbitrary parameters in
         # --unpaper-args, we:
         # 1) run with cwd set to a tmpdir with only unpaper's files
@@ -83,8 +133,8 @@ def run(
         # 3) append absolute paths for the input and output file
         # This should ensure that a user cannot clobber some other file with
         # their unpaper arguments (whether intentionally or otherwise)
-        args_unpaper.extend([os.fspath(input_pnm), os.fspath(output_pnm)])
-        external_run(
+        args_unpaper.extend([os.fspath(input_png), os.fspath(output_pnm)])
+        run(
             args_unpaper,
             close_fds=True,
             check=True,
@@ -96,12 +146,12 @@ def run(
         try:
             with Image.open(output_pnm) as imout:
                 imout.save(output_file, dpi=(dpi, dpi))
-        except (FileNotFoundError, OSError):
+        except OSError as e:
             raise SubprocessOutputError(
                 "unpaper: failed to produce the expected output file. "
                 + " Called with: "
                 + str(args_unpaper)
-            ) from None
+            ) from e
 
 
 def validate_custom_args(args: str) -> List[str]:
@@ -114,9 +164,10 @@ def validate_custom_args(args: str) -> List[str]:
 def clean(
     input_file: Path,
     output_file: Path,
+    *,
     dpi: DecFloat,
     unpaper_args: Optional[List[str]] = None,
-):
+) -> Path:
     default_args = [
         '--layout',
         'none',
@@ -130,4 +181,9 @@ def clean(
     ]
     if not unpaper_args:
         unpaper_args = default_args
-    run(input_file, output_file, dpi, unpaper_args)
+    try:
+        run_unpaper(input_file, output_file, dpi=dpi, mode_args=unpaper_args)
+        return output_file
+    except UnpaperImageTooLargeError as e:
+        log.warning(str(e))
+        return input_file

src/ocrmypdf/_graft.py

@@ -4,35 +4,51 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""For grafting text-only PDF pages onto freeform PDF pages."""
 
 import logging
 from contextlib import suppress
 from pathlib import Path
 from typing import Optional
 
-import pikepdf
+from pikepdf import (
+    Dictionary,
+    Name,
+    Object,
+    Operator,
+    Pdf,
+    PdfError,
+    PdfMatrix,
+    Stream,
+    parse_content_stream,
+    unparse_content_stream,
+)
 
 log = logging.getLogger(__name__)
 MAX_REPLACE_PAGES = 100
 
 
-def _update_page_resources(*, page, font, font_key, procset):
-    """Update this page's fonts with a reference to the Glyphless font"""
+def _ensure_dictionary(obj, name):
+    if name not in obj:
+        obj[name] = Dictionary({})
+    return obj[name]
 
-    if '/Resources' not in page:
-        page['/Resources'] = pikepdf.Dictionary({})
-    resources = page['/Resources']
-    try:
-        fonts = resources['/Font']
-    except KeyError:
-        fonts = pikepdf.Dictionary({})
+
+def _update_resources(*, obj, font, font_key, procset):
+    """Update this obj's fonts with a reference to the Glyphless font.
+
+    obj can be a page or Form XObject.
+    """
+
+    resources = _ensure_dictionary(obj, Name.Resources)
+    fonts = _ensure_dictionary(resources, Name.Font)
     if font_key is not None and font_key not in fonts:
         fonts[font_key] = font
-    resources['/Font'] = fonts
 
     # Reassign /ProcSet to one that just lists everything - ProcSet is
     # obsolete and doesn't matter but recommended for old viewer support
-    resources['/ProcSet'] = procset
+    if procset:
+        resources['/ProcSet'] = procset
 
 
 def strip_invisible_text(pdf, page):
@@ -41,58 +57,43 @@ def strip_invisible_text(pdf, page):
     render_mode = 0
     text_objects = []
 
-    page.page_contents_coalesce()
-    for operands, operator in pikepdf.parse_content_stream(page, ''):
+    for operands, operator in parse_content_stream(page, ''):
         if not in_text_obj:
-            if operator == pikepdf.Operator('BT'):
+            if operator == Operator('BT'):
                 in_text_obj = True
                 render_mode = 0
                 text_objects.append((operands, operator))
             else:
                 stream.append((operands, operator))
         else:
-            if operator == pikepdf.Operator('Tr'):
+            if operator == Operator('Tr'):
                 render_mode = operands[0]
             text_objects.append((operands, operator))
-            if operator == pikepdf.Operator('ET'):
+            if operator == Operator('ET'):
                 in_text_obj = False
                 if render_mode != 3:
                     stream.extend(text_objects)
                 text_objects.clear()
 
-    def convert(op):
-        try:
-            return op.unparse()
-        except AttributeError:
-            return str(op).encode('ascii')
-
-    lines = []
-
-    for operands, operator in stream:
-        if operator == pikepdf.Operator('INLINE IMAGE'):
-            iim = operands[0]
-            line = iim.unparse()
-        else:
-            line = b' '.join(convert(op) for op in operands) + b' ' + operator.unparse()
-        lines.append(line)
-
-    content_stream = b'\n'.join(lines)
-    page.Contents = pikepdf.Stream(pdf, content_stream)
+    content_stream = unparse_content_stream(stream)
+    page.Contents = Stream(pdf, content_stream)
 
 
 class OcrGrafter:
+    """Manages grafting text-only PDFs onto regular PDFs."""
+
     def __init__(self, context):
         self.context = context
         self.path_base = context.origin
 
-        self.pdf_base = pikepdf.open(self.path_base)
+        self.pdf_base = Pdf.open(self.path_base)
         self.font, self.font_key = None, None
 
         self.pdfinfo = context.pdfinfo
         self.output_file = context.get_path('graft_layers.pdf')
 
         self.procset = self.pdf_base.make_indirect(
-            pikepdf.Object.parse(b'[ /PDF /Text /ImageB /ImageC /ImageI ]')
+            Object.parse(b'[ /PDF /Text /ImageB /ImageC /ImageI ]')
         )
 
         self.emplacements = 1
@@ -116,7 +117,7 @@ class OcrGrafter:
             # We are updating the old page with a rasterized PDF of the new
             # page (without changing objgen, to preserve references)
             log.debug("Emplacement update")
-            with pikepdf.open(image) as pdf_image:
+            with Pdf.open(path_image) as pdf_image:
                 self.emplacements += 1
                 foreign_image_page = pdf_image.pages[0]
                 self.pdf_base.pages.append(foreign_image_page)
@@ -169,13 +170,13 @@ class OcrGrafter:
         """
 
         page0 = self.pdf_base.pages[0]
-        _update_page_resources(
-            page=page0, font=self.font, font_key=self.font_key, procset=self.procset
+        _update_resources(
+            obj=page0, font=self.font, font_key=self.font_key, procset=self.procset
         )
 
         # We cannot read and write the same file, that will corrupt it
         # but we don't to keep more copies than we need to. Delete intermediates.
-        # {interim_count} is the opened file we were updateing
+        # {interim_count} is the opened file we were updating
         # {interim_count - 1} can be deleted
         # {interim_count + 1} is the new file will produce and open
         old_file = self.output_file.with_suffix(f'.working{self.interim_count - 1}.pdf')
@@ -189,7 +190,7 @@ class OcrGrafter:
         self.pdf_base.save(next_file)
         self.pdf_base.close()
 
-        self.pdf_base = pikepdf.open(next_file)
+        self.pdf_base = Pdf.open(next_file)
         self.procset = self.pdf_base.pages[0].Resources.ProcSet
         self.font, self.font_key = None, None  # Ensure we reacquire this information
         self.interim_count += 1
@@ -205,11 +206,12 @@ class OcrGrafter:
         font, font_key = None, None
         possible_font_names = ('/f-0-0', '/F1')
         try:
-            with pikepdf.open(text) as pdf_text:
+            with Pdf.open(text) as pdf_text:
                 try:
                     pdf_text_fonts = pdf_text.pages[0].Resources.get('/Font', {})
                 except (AttributeError, IndexError, KeyError):
                     return None, None
+                pdf_text_font = None
                 for f in possible_font_names:
                     pdf_text_font = pdf_text_fonts.get(f, None)
                     if pdf_text_font is not None:
@@ -218,7 +220,7 @@ class OcrGrafter:
                 if pdf_text_font:
                     font = self.pdf_base.copy_foreign(pdf_text_font)
                 return font, font_key
-        except (FileNotFoundError, pikepdf.PdfError):
+        except (FileNotFoundError, PdfError):
             # PdfError occurs if a 0-length file is written e.g. due to OCR timeout
             return None, None
 
@@ -227,20 +229,22 @@ class OcrGrafter:
         *,
         page_num: int,
         textpdf: Path,
-        font: pikepdf.Object,
-        font_key: pikepdf.Object,
-        procset: pikepdf.Object,
+        font: Object,
+        font_key: Object,
+        procset: Object,
         text_rotation: int,
         strip_old_text: bool,
     ):
         """Insert the text layer from text page 0 on to pdf_base at page_num"""
 
+        # pylint: disable=invalid-name
+
         log.debug("Grafting")
         if Path(textpdf).stat().st_size == 0:
             return
 
         # This is a pointer indicating a specific page in the base file
-        with pikepdf.open(textpdf) as pdf_text:
+        with Pdf.open(textpdf) as pdf_text:
             pdf_text_contents = pdf_text.pages[0].Contents.read_bytes()
 
             base_page = self.pdf_base.pages.p(page_num)
@@ -255,13 +259,13 @@ class OcrGrafter:
             mediabox = [float(base_page.MediaBox[v]) for v in range(4)]
             wp, hp = mediabox[2] - mediabox[0], mediabox[3] - mediabox[1]
 
-            translate = pikepdf.PdfMatrix().translated(-wt / 2, -ht / 2)
-            untranslate = pikepdf.PdfMatrix().translated(wp / 2, hp / 2)
-            corner = pikepdf.PdfMatrix().translated(mediabox[0], mediabox[1])
+            translate = PdfMatrix().translated(-wt / 2, -ht / 2)
+            untranslate = PdfMatrix().translated(wp / 2, hp / 2)
+            corner = PdfMatrix().translated(mediabox[0], mediabox[1])
             # -rotation because the input is a clockwise angle and this formula
             # uses CCW
             text_rotation = -text_rotation % 360
-            rotate = pikepdf.PdfMatrix().rotated(text_rotation)
+            rotate = PdfMatrix().rotated(text_rotation)
 
             # Because of rounding of DPI, we might get a text layer that is not
             # identically sized to the target page. Scale to adjust. Normally this
@@ -272,24 +276,36 @@ class OcrGrafter:
             scale_y = hp / ht
 
             # log.debug('%r', scale_x, scale_y)
-            scale = pikepdf.PdfMatrix().scaled(scale_x, scale_y)
+            scale = PdfMatrix().scaled(scale_x, scale_y)
 
             # Translate the text so it is centered at (0, 0), rotate it there, adjust
             # for a size different between initial and text PDF, then untranslate, and
             # finally move the lower left corner to match the mediabox
             ctm = translate @ rotate @ scale @ untranslate @ corner
 
-            pdf_text_contents = (
-                b'q %s cm\n' % ctm.encode() + pdf_text_contents + b'\nQ\n'
+            base_resources = _ensure_dictionary(base_page, Name.Resources)
+            base_xobjs = _ensure_dictionary(base_resources, Name.XObject)
+            text_xobj_name = Name.random(prefix="OCR-")
+            xobj = self.pdf_base.make_stream(pdf_text_contents)
+            base_xobjs[text_xobj_name] = xobj
+            xobj.Type = Name.XObject
+            xobj.Subtype = Name.Form
+            xobj.FormType = 1
+            xobj.BBox = mediabox
+            _update_resources(
+                obj=xobj, font=font, font_key=font_key, procset=[Name.PDF]
             )
 
-            new_text_layer = pikepdf.Stream(self.pdf_base, pdf_text_contents)
+            pdf_draw_xobj = (
+                (b'q %s cm\n' % ctm.encode()) + (b'%s Do\n' % text_xobj_name) + b'\nQ\n'
+            )
+            new_text_layer = Stream(self.pdf_base, pdf_draw_xobj)
 
             if strip_old_text:
                 strip_invisible_text(self.pdf_base, base_page)
 
-            base_page.page_contents_add(new_text_layer, prepend=True)
+            base_page.contents_add(new_text_layer, prepend=True)
 
-            _update_page_resources(
-                page=base_page, font=font, font_key=font_key, procset=procset
+            _update_resources(
+                obj=base_page, font=font, font_key=font_key, procset=procset
             )

src/ocrmypdf/_jobcontext.py

@@ -4,6 +4,7 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""Defines context objects that are passed to child processes/threads."""
 
 import os
 import shutil
@@ -83,7 +84,7 @@ class PageContext:
         The path will be based in a common temporary folder and have a prefix based
         on the page number.
         """
-        return self.work_folder / ("%06d_%s" % (self.pageno + 1, name))
+        return self.work_folder / f"{(self.pageno + 1):06d}_{name}"
 
     def __getstate__(self):
         state = self.__dict__.copy()

src/ocrmypdf/_logging.py

@@ -4,15 +4,17 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""Logging support classes."""
 
 import logging
-import sys
 from contextlib import suppress
 
 from tqdm import tqdm
 
 
 class PageNumberFilter(logging.Filter):
+    """Insert PDF page number that emitted log message to log record."""
+
     def filter(self, record):
         pageno = getattr(record, 'pageno', None)
         if isinstance(pageno, int):
@@ -28,22 +30,14 @@ class TqdmConsole:
     This routes log messages through tqdm so that it can print them above the
     progress bar, and then refresh the progress bar, rather than overwriting
     it which looks messy.
-
-    For some reason Python 3.6 prints extra empty messages from time to time,
-    so we suppress those.
     """
 
     def __init__(self, file):
         self.file = file
-        self.py36 = sys.version_info[0:2] == (3, 6)
 
     def write(self, msg):
         # When no progress bar is active, tqdm.write() routes to print()
-        if self.py36:
-            if msg.strip() != '':
-                tqdm.write(msg.rstrip(), end='\n', file=self.file)
-        else:
-            tqdm.write(msg.rstrip(), end='\n', file=self.file)
+        tqdm.write(msg.rstrip(), end='\n', file=self.file)
 
     def flush(self):
         with suppress(AttributeError):

src/ocrmypdf/_pipeline.py

@@ -4,6 +4,7 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""OCRmyPDF page processing pipeline functions."""
 
 import logging
 import os
@@ -13,15 +14,14 @@ from contextlib import suppress
 from datetime import datetime, timezone
 from pathlib import Path
 from shutil import copyfileobj
-from typing import BinaryIO, Dict, Iterable, Optional, Union, cast
+from typing import Dict, Iterable, Optional
 
 import img2pdf
 import pikepdf
 from pikepdf.models.metadata import encode_pdf_date
 from PIL import Image, ImageColor, ImageDraw
-from tqdm import tqdm
 
-from ocrmypdf import leptonica
+from ocrmypdf._concurrent import Executor
 from ocrmypdf._exec import unpaper
 from ocrmypdf._jobcontext import PageContext, PdfContext
 from ocrmypdf._version import PROGRAM_NAME
@@ -33,12 +33,18 @@ from ocrmypdf.exceptions import (
     PriorOcrFoundError,
     UnsupportedImageFormatError,
 )
-from ocrmypdf.helpers import Resolution, safe_symlink
+from ocrmypdf.helpers import IMG2PDF_KWARGS, Resolution, safe_symlink
 from ocrmypdf.hocrtransform import HocrTransform
-from ocrmypdf.optimize import optimize
 from ocrmypdf.pdfa import generate_pdfa_ps
 from ocrmypdf.pdfinfo import Colorspace, Encoding, PdfInfo
 
+# Remove this workaround when we require Pillow >= 10
+try:
+    BICUBIC = Image.Resampling.BICUBIC  # type: ignore
+except AttributeError:
+    # Pillow 9 shim
+    BICUBIC = Image.BICUBIC  # type: ignore
+
 log = logging.getLogger(__name__)
 
 VECTOR_PAGE_DPI = 400
@@ -48,7 +54,7 @@ def triage_image_file(input_file, output_file, options):
     log.info("Input file is not a PDF, checking if it is an image...")
     try:
         im = Image.open(input_file)
-    except EnvironmentError as e:
+    except OSError as e:
         # Recover the original filename
         log.error(str(e).replace(str(input_file), str(options.input_file)))
         raise UnsupportedImageFormatError() from e
@@ -99,8 +105,8 @@ def triage_image_file(input_file, output_file, options):
             img2pdf.convert(
                 os.fspath(input_file),
                 layout_fun=layout_fun,
-                with_pdfrw=False,
                 outputstream=outf,
+                **IMG2PDF_KWARGS,
             )
         log.info("Successfully converted to PDF, processing...")
     except img2pdf.ImageOpenError as e:
@@ -135,7 +141,7 @@ def triage(original_filename, input_file, output_file, options):
             # Origin file is a pdf create a symlink with pdf extension
             safe_symlink(input_file, output_file)
             return output_file
-    except EnvironmentError as e:
+    except OSError as e:
         log.debug(f"Temporary file was at: {input_file}")
         msg = str(e).replace(str(input_file), original_filename)
         raise InputFileError(msg) from e
@@ -146,6 +152,8 @@ def triage(original_filename, input_file, output_file, options):
 
 def get_pdfinfo(
     input_file,
+    *,
+    executor: Executor,
     detailed_analysis=False,
     progbar=False,
     max_workers=None,
@@ -158,11 +166,12 @@ def get_pdfinfo(
             progbar=progbar,
             max_workers=max_workers,
             check_pages=check_pages,
+            executor=executor,
         )
-    except pikepdf.PasswordError:
-        raise EncryptedPdfError()
-    except pikepdf.PdfError:
-        raise InputFileError()
+    except pikepdf.PasswordError as e:
+        raise EncryptedPdfError() from e
+    except pikepdf.PdfError as e:
+        raise InputFileError() from e
 
 
 def validate_pdfinfo_options(context: PdfContext):
@@ -323,7 +332,8 @@ def is_ocr_required(page_context: PageContext):
             ocr_required = False
             log.warning(
                 "page too big, skipping OCR "
-                f"({(pixel_count / 1_000_000):.1f} MPixels > {options.skip_big:.1f} MPixels --skip-big)"
+                f"({(pixel_count / 1_000_000):.1f} MPixels > "
+                f"{options.skip_big:.1f} MPixels --skip-big)"
             )
     return ocr_required
 
@@ -421,8 +431,8 @@ def rasterize(
     output_file = page_context.get_path(f'rasterize{output_tag}.png')
     pageinfo = page_context.pageinfo
 
-    def at_least(cs):
-        return max(device_idx, colorspaces.index(cs))
+    def at_least(colorspace):
+        return max(device_idx, colorspaces.index(colorspace))
 
     for image in pageinfo.images:
         if image.type_ != 'image':
@@ -462,9 +472,10 @@ def rasterize(
 
 def preprocess_remove_background(input_file: Path, page_context: PageContext):
     if any(image.bpc > 1 for image in page_context.pageinfo.images):
-        output_file = page_context.get_path('pp_rm_bg.png')
-        leptonica.remove_background(input_file, output_file)
-        return output_file
+        raise NotImplementedError("--remove-background is temporarily not implemented")
+        # output_file = page_context.get_path('pp_rm_bg.png')
+        # leptonica.remove_background(input_file, output_file)
+        # return output_file
     else:
         log.info("background removal skipped on mono page")
         return input_file
@@ -473,15 +484,32 @@ def preprocess_remove_background(input_file: Path, page_context: PageContext):
 def preprocess_deskew(input_file: Path, page_context: PageContext):
     output_file = page_context.get_path('pp_deskew.png')
     dpi = get_page_square_dpi(page_context.pageinfo, page_context.options)
-    leptonica.deskew(input_file, output_file, dpi.x)
+
+    ocr_engine = page_context.plugin_manager.hook.get_ocr_engine()
+    deskew_angle_degrees = ocr_engine.get_deskew(input_file, page_context.options)
+
+    with Image.open(input_file) as im:
+        # According to Pillow docs, .rotate() will automatically use Image.NEAREST
+        # resampling if image is mode '1' or 'P'
+        deskewed = im.rotate(
+            deskew_angle_degrees,
+            resample=BICUBIC,
+            fillcolor=ImageColor.getcolor('white', mode=im.mode),
+        )
+        deskewed.save(output_file, dpi=dpi)
+
     return output_file
 
 
 def preprocess_clean(input_file: Path, page_context: PageContext):
     output_file = page_context.get_path('pp_clean.png')
     dpi = get_page_square_dpi(page_context.pageinfo, page_context.options)
-    unpaper.clean(input_file, output_file, dpi.x, page_context.options.unpaper_args)
-    return output_file
+    return unpaper.clean(
+        input_file,
+        output_file,
+        dpi=dpi.x,
+        unpaper_args=page_context.options.unpaper_args,
+    )
 
 
 def create_ocr_image(image: Path, page_context: PageContext):
@@ -492,10 +520,6 @@ def create_ocr_image(image: Path, page_context: PageContext):
     output_file = page_context.get_path('ocr.png')
     options = page_context.options
     with Image.open(image) as im:
-        white = ImageColor.getcolor('#ffffff', im.mode)
-        # pink = ImageColor.getcolor('#ff0080', im.mode)
-        draw = ImageDraw.ImageDraw(im)
-
         log.debug('resolution %r', im.info['dpi'])
 
         if not options.force_ocr:
@@ -505,6 +529,7 @@ def create_ocr_image(image: Path, page_context: PageContext):
             if options.redo_ocr:
                 mask = True  # Mask visible text, but not invisible text
 
+            draw = ImageDraw.ImageDraw(im)
             for textarea in page_context.pageinfo.get_textareas(
                 visible=mask, corrupt=None
             ):
@@ -513,25 +538,15 @@ def create_ocr_image(image: Path, page_context: PageContext):
                 # be None)
                 bbox = [float(v) for v in textarea]
                 xyscale = tuple(float(coord) / 72.0 for coord in im.info['dpi'])
-                pixcoords = [
+                pixcoords = (
                     bbox[0] * xyscale[0],
                     im.height - bbox[3] * xyscale[1],
                     bbox[2] * xyscale[0],
                     im.height - bbox[1] * xyscale[1],
-                ]
-                pixcoords = [int(round(c)) for c in pixcoords]
+                )
                 log.debug('blanking %r', pixcoords)
-                draw.rectangle(pixcoords, fill=white)
-                # draw.rectangle(pixcoords, outline=pink)
-
-        if options.threshold:
-            pix = leptonica.Pix.frompil(im)
-            pix = pix.masked_threshold_on_background_norm()
-            im_pix = pix.topil()
-            im_pix.info['dpi'] = im.info['dpi']
-            im = im_pix
-
-        del draw
+                draw.rectangle(pixcoords, fill='white')
+                # draw.rectangle(pixcoords, outline='pink')
 
         filter_im = page_context.plugin_manager.hook.filter_ocr_image(
             page=page_context, image=im
@@ -607,10 +622,14 @@ def create_pdf_page_from_image(
 
         layout_fun = img2pdf.get_layout_fun(pagesize)
         img2pdf.convert(
-            imfile, with_pdfrw=False, layout_fun=layout_fun, outputstream=pdf
+            imfile, layout_fun=layout_fun, outputstream=pdf, **IMG2PDF_KWARGS
         )
         log.debug('convert done')
 
+    output_file = page_context.plugin_manager.hook.filter_pdf_page(
+        page=page_context, image_filename=image, output_pdf=output_file
+    )
+
     return output_file
 
 
@@ -620,9 +639,9 @@ def render_hocr_page(hocr: Path, page_context: PageContext):
     dpi = get_page_square_dpi(page_context.pageinfo, options)
     debug_mode = options.pdf_renderer == 'hocrdebug'
 
-    hocrtransform = HocrTransform(hocr, dpi.x)  # square
+    hocrtransform = HocrTransform(hocr_filename=hocr, dpi=dpi.x)  # square
     hocrtransform.to_pdf(
-        output_file,
+        out_filename=output_file,
         image_filename=None,
         show_bounding_boxes=False if not debug_mode else True,
         invisible_text=True if not debug_mode else False,
@@ -674,11 +693,6 @@ def get_docinfo(base_pdf: pikepdf.Pdf, context: PdfContext) -> Dict[str, str]:
 
     pdfmark['/Creator'] = f'{PROGRAM_NAME} {VERSION} / {creator_tag}'
     pdfmark['/Producer'] = f'pikepdf {pikepdf.__version__}'
-    if 'OCRMYPDF_CREATOR' in os.environ:
-        pdfmark['/Creator'] = os.environ['OCRMYPDF_CREATOR']
-    if 'OCRMYPDF_PRODUCER' in os.environ:
-        pdfmark['/Producer'] = os.environ['OCRMYPDF_PRODUCER']
-
     pdfmark['/ModDate'] = encode_pdf_date(datetime.now(timezone.utc))
     return pdfmark
 
@@ -726,7 +740,11 @@ def convert_to_pdfa(input_pdf: Path, input_ps_stub: Path, context: PdfContext):
         output_file=output_file,
         compression=options.pdfa_image_compression,
         pdfa_part=options.output_type[-1],  # is pdfa-1, pdfa-2, or pdfa-3
-        progressbar_class=tqdm if options.progress_bar else None,
+        progressbar_class=(
+            context.plugin_manager.hook.get_progressbar_class()
+            if options.progress_bar
+            else None
+        ),
     )
 
     return output_file
@@ -739,6 +757,24 @@ def should_linearize(working_file: Path, context: PdfContext):
     return False
 
 
+def get_pdf_save_settings(output_type: str):
+    if output_type == 'pdfa-1':
+        # Trigger recompression to ensure object streams are removed, because
+        # Acrobat complains about them in PDF/A-1b validation.
+        return dict(
+            preserve_pdfa=True,
+            compress_streams=True,
+            stream_decode_level=pikepdf.StreamDecodeLevel.generalized,
+            object_stream_mode=pikepdf.ObjectStreamMode.disable,
+        )
+    else:
+        return dict(
+            preserve_pdfa=True,
+            compress_streams=True,
+            object_stream_mode=(pikepdf.ObjectStreamMode.generate),
+        )
+
+
 def metadata_fixup(working_file: Path, context: PdfContext):
     output_file = context.get_path('metafix.pdf')
     options = context.options
@@ -781,41 +817,64 @@ def metadata_fixup(working_file: Path, context: PdfContext):
                 missing = set(meta_original.keys()) - set(meta.keys())
                 report_on_metadata(missing)
 
+        optimizing = context.plugin_manager.hook.is_optimization_enabled(
+            context=context
+        )
         pdf.save(
             output_file,
-            compress_streams=True,
-            preserve_pdfa=True,
-            object_stream_mode=pikepdf.ObjectStreamMode.generate,
+            **get_pdf_save_settings(options.output_type),
             linearize=(  # Don't linearize if optimize() will be linearizing too
-                should_linearize(working_file, context)
-                if options.optimize == 0
-                else False
+                not optimizing and should_linearize(working_file, context)
             ),
         )
 
     return output_file
 
 
-def optimize_pdf(input_file: Path, context: PdfContext):
+def optimize_pdf(input_file: Path, context: PdfContext, executor: Executor):
     output_file = context.get_path('optimize.pdf')
-    save_settings = dict(
-        compress_streams=True,
-        preserve_pdfa=True,
-        object_stream_mode=pikepdf.ObjectStreamMode.generate,
+    output_pdf, messages = context.plugin_manager.hook.optimize_pdf(
+        input_pdf=input_file,
+        output_pdf=output_file,
+        context=context,
+        executor=executor,
         linearize=should_linearize(input_file, context),
     )
-    optimize(input_file, output_file, context, save_settings)
-    return output_file
+
+    input_size = input_file.stat().st_size
+    output_size = output_file.stat().st_size
+    if output_size > 0:
+        ratio = input_size / output_size
+        savings = 1 - output_size / input_size
+        log.info(f"Optimize ratio: {ratio:.2f} savings: {(savings):.1%}")
+
+    return output_pdf, messages
+
+
+def enumerate_compress_ranges(iterable):
+    skipped_from, index = None, None
+    for index, txt_file in enumerate(iterable):
+        index += 1
+        if txt_file:
+            if skipped_from is not None:
+                yield (skipped_from, index - 1), None
+                skipped_from = None
+            yield (index, index), txt_file
+        else:
+            if skipped_from is None:
+                skipped_from = index
+    if skipped_from is not None:
+        yield (skipped_from, index), None
 
 
 def merge_sidecars(txt_files: Iterable[Optional[Path]], context: PdfContext):
     output_file = context.get_path('sidecar.txt')
     with open(output_file, 'w', encoding="utf-8") as stream:
-        for page_num, txt_file in enumerate(txt_files):
-            if page_num != 0:
+        for (from_, to_), txt_file in enumerate_compress_ranges(txt_files):
+            if from_ != 1:
                 stream.write('\f')  # Form feed between pages
             if txt_file:
-                with open(txt_file, 'r', encoding="utf-8") as in_:
+                with open(txt_file, encoding="utf-8") as in_:
                     txt = in_.read()
                     # Some OCR engines (e.g. Tesseract v4 alpha) add form feeds
                     # between pages, and some do not. For consistency, we ignore
@@ -825,7 +884,11 @@ def merge_sidecars(txt_files: Iterable[Optional[Path]], context: PdfContext):
                     else:
                         stream.write(txt)
             else:
-                stream.write(f'[OCR skipped on page {(page_num + 1)}]')
+                if from_ != to_:
+                    pages = f'{from_}-{to_}'
+                else:
+                    pages = f'{from_}'
+                stream.write(f'[OCR skipped on page(s) {pages}]')
     return output_file
 
 

src/ocrmypdf/_plugin_manager.py

@@ -4,6 +4,7 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""Plugin manager using pluggy."""
 
 import argparse
 import importlib
@@ -11,7 +12,7 @@ import importlib.util
 import pkgutil
 import sys
 from pathlib import Path
-from typing import List, Tuple, Union
+from typing import List, Sequence, Tuple, Union
 
 import pluggy
 
@@ -66,15 +67,26 @@ class OcrmypdfPluginManager(pluggy.PluginManager):
 
         # 1. Register builtins
         if self.__builtins:
-            for module in pkgutil.iter_modules(ocrmypdf.builtin_plugins.__path__):
+            for module in sorted(
+                pkgutil.iter_modules(ocrmypdf.builtin_plugins.__path__)
+            ):
                 name = f'ocrmypdf.builtin_plugins.{module.name}'
                 module = importlib.import_module(name)
                 self.register(module)
 
-        # 2. Register setuptools plugins
+        # 2. Install semfree if needed
+        try:
+            # pylint: disable=import-outside-toplevel
+            from multiprocessing.synchronize import SemLock
+
+            del SemLock
+        except ImportError:
+            self.register(importlib.import_module('ocrmypdf.extra_plugins.semfree'))
+
+        # 3. Register setuptools plugins
         self.load_setuptools_entrypoints('ocrmypdf')
 
-        # 3. Register plugins specified on command line
+        # 4. Register plugins specified on command line
         for name in self.__plugins:
             if isinstance(name, Path) or name.endswith('.py'):
                 # Import by filename
@@ -90,22 +102,27 @@ class OcrmypdfPluginManager(pluggy.PluginManager):
 
 
 def get_plugin_manager(plugins: List[Union[str, Path]], builtins=True):
-    pm = OcrmypdfPluginManager(
+    return OcrmypdfPluginManager(
         project_name='ocrmypdf',
         plugins=plugins,
         builtins=builtins,
     )
-    return pm
 
 
 def get_parser_options_plugins(
-    args,
+    args: Sequence[str],
 ) -> Tuple[argparse.ArgumentParser, argparse.Namespace, pluggy.PluginManager]:
     pre_options, _unused = plugins_only_parser.parse_known_args(args=args)
     plugin_manager = get_plugin_manager(pre_options.plugins)
 
     parser = get_parser()
+    plugin_manager.hook.initialize(  # pylint: disable=no-member
+        plugin_manager=plugin_manager
+    )
     plugin_manager.hook.add_options(parser=parser)  # pylint: disable=no-member
 
     options = parser.parse_args(args=args)
     return parser, options, plugin_manager
+
+
+__all__ = ['OcrmypdfPluginManager', 'get_plugin_manager', 'get_parser_options_plugins']

src/ocrmypdf/_sync.py

@@ -4,21 +4,25 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""Implements the concurrent and page synchronous parts of the pipeline."""
 
+
+import argparse
 import logging
 import logging.handlers
 import os
 import sys
 import threading
+from concurrent.futures.process import BrokenProcessPool
+from concurrent.futures.thread import BrokenThreadPool
 from functools import partial
 from pathlib import Path
 from tempfile import mkdtemp
-from typing import List, NamedTuple, Optional, Tuple
+from typing import List, NamedTuple, Optional, Sequence, Tuple, cast
 
-import pikepdf
 import PIL
 
-from ocrmypdf._concurrent import exec_progress_pool
+from ocrmypdf._concurrent import Executor, setup_executor
 from ocrmypdf._graft import OcrGrafter
 from ocrmypdf._jobcontext import PageContext, PdfContext, cleanup_working_files
 from ocrmypdf._logging import PageNumberFilter
@@ -47,7 +51,7 @@ from ocrmypdf._pipeline import (
     triage,
     validate_pdfinfo_options,
 )
-from ocrmypdf._plugin_manager import get_plugin_manager
+from ocrmypdf._plugin_manager import OcrmypdfPluginManager, get_plugin_manager
 from ocrmypdf._validation import (
     check_requested_output_file,
     create_input_file,
@@ -67,6 +71,8 @@ log = logging.getLogger(__name__)
 
 
 class PageResult(NamedTuple):
+    """Result when a page is finished processing."""
+
     pageno: int
     pdf_page_from_image: Optional[Path]
     ocr: Optional[Path]
@@ -166,7 +172,7 @@ def make_intermediate_images(
     return ocr_image, preprocess_out
 
 
-def exec_page_sync(page_context: PageContext):
+def exec_page_sync(page_context: PageContext) -> PageResult:
     options = page_context.options
     tls.pageno = page_context.pageno + 1
 
@@ -224,17 +230,19 @@ def exec_page_sync(page_context: PageContext):
     )
 
 
-def post_process(pdf_file, context: PdfContext):
+def post_process(
+    pdf_file: Path, context: PdfContext, executor: Executor
+) -> Tuple[Path, Sequence[str]]:
     pdf_out = pdf_file
     if context.options.output_type.startswith('pdfa'):
         ps_stub_out = generate_postscript_stub(context)
         pdf_out = convert_to_pdfa(pdf_out, ps_stub_out, context)
 
     pdf_out = metadata_fixup(pdf_out, context)
-    return optimize_pdf(pdf_out, context)
+    return optimize_pdf(pdf_out, context, executor)
 
 
-def worker_init(max_pixels: int):
+def worker_init(max_pixels: int) -> None:
     # In Windows, child process will not inherit our change to this value in
     # the parent process, so ensure workers get it set. Not needed when running
     # threaded, but harmless to set again.
@@ -242,7 +250,7 @@ def worker_init(max_pixels: int):
     pikepdf_enable_mmap()
 
 
-def exec_concurrent(context: PdfContext):
+def exec_concurrent(context: PdfContext, executor: Executor) -> Sequence[str]:
     """Execute the pipeline concurrently"""
 
     # Run exec_page_sync on every page context
@@ -269,7 +277,7 @@ def exec_concurrent(context: PdfContext):
         finally:
             tls.pageno = None
 
-    exec_progress_pool(
+    executor(
         use_threads=options.use_threads,
         max_workers=max_workers,
         tqdm_kwargs=dict(
@@ -279,7 +287,7 @@ def exec_concurrent(context: PdfContext):
             unit_scale=0.5,
             disable=not options.progress_bar,
         ),
-        task_initializer=partial(worker_init, PIL.Image.MAX_IMAGE_PIXELS),
+        worker_initializer=partial(worker_init, PIL.Image.MAX_IMAGE_PIXELS),
         task=exec_page_sync,
         task_arguments=context.get_page_contexts(),
         task_finished=update_page,
@@ -294,15 +302,20 @@ def exec_concurrent(context: PdfContext):
     # Merge layers to one single pdf
     pdf = ocrgraft.finalize()
 
-    # PDF/A and metadata
-    log.info("Postprocessing...")
-    pdf = post_process(pdf, context)
+    messages: List[str] = []
+    if options.output_type != 'none':
+        # PDF/A and metadata
+        log.info("Postprocessing...")
+        pdf, messages = post_process(pdf, context, executor)
 
-    # Copy PDF file to destination
-    copy_final(pdf, options.output_file, context)
+        # Copy PDF file to destination
+        copy_final(pdf, options.output_file, context)
+    return messages
 
 
-def configure_debug_logging(log_filename: Path, prefix: str = ''):
+def configure_debug_logging(
+    log_filename: Path, prefix: str = ''
+) -> logging.FileHandler:
     """
     Create a debug log file at a specified location.
 
@@ -321,7 +334,12 @@ def configure_debug_logging(log_filename: Path, prefix: str = ''):
     return log_file_handler
 
 
-def run_pipeline(options, *, plugin_manager, api=False):
+def run_pipeline(
+    options: argparse.Namespace,
+    *,
+    plugin_manager: Optional[OcrmypdfPluginManager],
+    api: bool = False,
+) -> ExitCode:
     # Any changes to options will not take effect for options that are already
     # bound to function parameters in the pipeline. (For example
     # options.input_file, options.pdf_renderer are already bound.)
@@ -346,6 +364,7 @@ def run_pipeline(options, *, plugin_manager, api=False):
 
     pikepdf_enable_mmap()
 
+    executor = setup_executor(plugin_manager)
     try:
         check_requested_output_file(options)
         start_input_file, original_filename = create_input_file(options, work_folder)
@@ -358,6 +377,7 @@ def run_pipeline(options, *, plugin_manager, api=False):
         # Gather pdfinfo and create context
         pdfinfo = get_pdfinfo(
             origin_pdf,
+            executor=executor,
             detailed_analysis=options.redo_ocr,
             progbar=options.progress_bar,
             max_workers=options.jobs if not options.use_threads else 1,  # To help debug
@@ -370,7 +390,7 @@ def run_pipeline(options, *, plugin_manager, api=False):
         validate_pdfinfo_options(context)
 
         # Execute the pipeline
-        exec_concurrent(context)
+        optimize_messages = exec_concurrent(context, executor)
 
         if options.output_file == '-':
             log.info("Output sent to stdout")
@@ -378,7 +398,7 @@ def run_pipeline(options, *, plugin_manager, api=False):
             hasattr(options.output_file, 'writable') and options.output_file.writable()
         ):
             log.info("Output written to stream")
-        elif samefile(options.output_file, os.devnull):
+        elif samefile(options.output_file, Path(os.devnull)):
             pass  # Say nothing when sending to dev null
         else:
             if options.output_type.startswith('pdfa'):
@@ -396,21 +416,44 @@ def run_pipeline(options, *, plugin_manager, api=False):
             if not check_pdf(options.output_file):
                 log.warning('Output file: The generated PDF is INVALID')
                 return ExitCode.invalid_output_pdf
-            report_output_file_size(options, start_input_file, options.output_file)
+            report_output_file_size(
+                options, start_input_file, options.output_file, optimize_messages
+            )
 
-    except (KeyboardInterrupt if not api else NeverRaise) as e:
+    except (KeyboardInterrupt if not api else NeverRaise):
         if options.verbose >= 1:
             log.exception("KeyboardInterrupt")
         else:
             log.error("KeyboardInterrupt")
         return ExitCode.ctrl_c
     except (ExitCodeException if not api else NeverRaise) as e:
-        if str(e):
+        e = cast(ExitCodeException, e)
+        if options.verbose >= 1:
+            log.exception("ExitCodeException")
+        elif str(e):
             log.error("%s: %s", type(e).__name__, str(e))
         else:
             log.error(type(e).__name__)
         return e.exit_code
-    except (Exception if not api else NeverRaise) as e:  # pylint: disable=broad-except
+    except (PIL.Image.DecompressionBombError if not api else NeverRaise):
+        log.exception(
+            "A decompression bomb error was encountered while executing the "
+            "pipeline. Use the argument --max-image-mpixels to raise the maximum "
+            "image pixel limit."
+        )
+        return ExitCode.other_error
+    except (
+        BrokenProcessPool if not api else NeverRaise,
+        BrokenThreadPool if not api else NeverRaise,
+    ):
+        log.exception(
+            "A worker process was terminated unexpectedly. This is known to occur if "
+            "processing your file takes all available swap space and RAM. It may "
+            "help to try again with a smaller number of jobs, using the --jobs "
+            "argument."
+        )
+        return ExitCode.child_process_error
+    except (Exception if not api else NeverRaise):  # pylint: disable=broad-except
         log.exception("An exception occurred while executing the pipeline")
         return ExitCode.other_error
     finally:
@@ -418,7 +461,7 @@ def run_pipeline(options, *, plugin_manager, api=False):
             try:
                 debug_log_handler.close()
                 log.removeHandler(debug_log_handler)
-            except EnvironmentError as e:
+            except OSError as e:
                 print(e, file=sys.stderr)
         cleanup_working_files(work_folder, options)
 

src/ocrmypdf/_unicodefun.py

@@ -1,118 +0,0 @@
-# Copyright (c) 2014, Armin Ronacher
-#
-# Copyright (c) 2017, James R Barlow
-#
-# Some rights reserved.
-#
-# Redistribution and use in source and binary forms, with or without
-# modification, are permitted provided that the following conditions are
-# met:
-#
-#     * Redistributions of source code must retain the above copyright
-#       notice, this list of conditions and the following disclaimer.
-#
-#     * Redistributions in binary form must reproduce the above
-#       copyright notice, this list of conditions and the following
-#       disclaimer in the documentation and/or other materials provided
-#       with the distribution.
-#
-#     * The names of the contributors may not be used to endorse or
-#       promote products derived from this software without specific
-#       prior written permission.
-#
-# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
-# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
-# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
-# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
-# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
-# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
-# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
-# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
-# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
-# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
-
-import codecs
-import os
-import sys
-
-
-def verify_python3_env():  # pragma: no cover
-    """Ensures that the environment is good for unicode on Python 3."""
-
-    # PEP 538 changes in Python 3.7 should make this wrangling unnecessary
-    if sys.version_info[0:3] >= (3, 7, 0):
-        return
-
-    try:
-        import locale
-
-        fs_enc = codecs.lookup(locale.getpreferredencoding()).name
-    except Exception:
-        fs_enc = 'ascii'
-    if fs_enc != 'ascii':
-        return
-
-    extra = ''
-    if os.name == 'posix':
-        import subprocess
-
-        rv = subprocess.run(
-            ['locale', '-a'], stdout=subprocess.PIPE, stderr=subprocess.PIPE
-        ).stdout
-        good_locales = set()
-        has_c_utf8 = False
-
-        # Make sure we're operating on text here.
-        if isinstance(rv, bytes):
-            rv = rv.decode('ascii', 'replace')
-
-        for line in rv.splitlines():
-            locale = line.strip()
-            if locale.lower().endswith(('.utf-8', '.utf8')):
-                good_locales.add(locale)
-                if locale.lower() in ('c.utf8', 'c.utf-8'):
-                    has_c_utf8 = True
-
-        extra += '\n\n'
-        if not good_locales:
-            extra += (
-                'Additional information: on this system no suitable UTF-8\n'
-                'locales were discovered.  This most likely requires resolving\n'
-                'by reconfiguring the locale system.'
-            )
-        elif has_c_utf8:
-            extra += (
-                'This system supports the C.UTF-8 locale which is recommended.\n'
-                'You might be able to resolve your issue by exporting the\n'
-                'following environment variables:\n\n'
-                '    export LC_ALL=C.UTF-8\n'
-                '    export LANG=C.UTF-8'
-            )
-        else:
-            extra += (
-                'This system lists a couple of UTF-8 supporting locales that\n'
-                'you can pick from.  The following suitable locales were\n'
-                'discovered: %s'
-            ) % ', '.join(sorted(good_locales))
-
-        bad_locale = None
-        for locale in os.environ.get('LC_ALL'), os.environ.get('LANG'):
-            if locale and locale.lower().endswith(('.utf-8', '.utf8')):
-                bad_locale = locale
-            if locale is not None:
-                break
-        if bad_locale is not None:
-            extra += (
-                '\nocrmypdf discovered that you exported a UTF-8 locale\n'
-                'but the locale system could not pick up from it because\n'
-                'it does not exist.  The exported locale is "%s" but it\n'
-                'is not supported'
-            ) % bad_locale
-
-    raise RuntimeError(
-        'ocrmypdf will abort further execution because Python 3 '
-        'was configured to use ASCII as encoding for the '
-        'environment.' + extra
-    )

src/ocrmypdf/_validation.py

@@ -5,6 +5,7 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""Validate a work order from API or command line."""
 
 import locale
 import logging
@@ -13,43 +14,35 @@ import sys
 import unicodedata
 from pathlib import Path
 from shutil import copyfileobj
-from typing import List, Set, Tuple, Union
+from typing import List, Optional, Sequence, Set, Tuple
 
 import pikepdf
 import PIL
 
-from ocrmypdf._exec import jbig2enc, pngquant, unpaper
-from ocrmypdf._unicodefun import verify_python3_env
+from ocrmypdf._exec import unpaper
 from ocrmypdf.exceptions import (
     BadArgsError,
     InputFileError,
     MissingDependencyError,
     OutputFileAccessError,
 )
-from ocrmypdf.helpers import (
-    is_file_writable,
-    is_iterable_notstr,
-    monotonic,
-    safe_symlink,
-)
+from ocrmypdf.helpers import is_file_writable, monotonic, safe_symlink
+from ocrmypdf.hocrtransform import HOCR_OK_LANGS
 from ocrmypdf.subprocess import check_external_program
 
 # -------------
 # External dependencies
 
-HOCR_OK_LANGS = frozenset(['eng', 'deu', 'spa', 'ita', 'por'])
 DEFAULT_LANGUAGE = 'eng'  # Enforce English hegemony
 
 log = logging.getLogger(__name__)
 
 
 # --------
-# Critical environment tests
-verify_python3_env()
 
 
 def check_platform():
-    if os.name == 'nt' and sys.maxsize <= 2 ** 32:  # pragma: no cover
+    if os.name == 'nt' and sys.maxsize <= 2**32:  # pragma: no cover
         # 32-bit interpreter on Windows
         log.error(
             "You are running OCRmyPDF in a 32-bit (x86) Python interpreter."
@@ -65,13 +58,14 @@ def check_options_languages(options, ocr_engine_languages):
             log.debug("No language specified; assuming --language %s", DEFAULT_LANGUAGE)
     if not ocr_engine_languages:
         return
-    if not options.languages.issubset(ocr_engine_languages):
+    missing_languages = options.languages - ocr_engine_languages
+    if missing_languages:
         msg = (
-            f"OCR engine does not have language data for the following "
+            "OCR engine does not have language data for the following "
             "requested languages: \n"
         )
-        for lang in options.languages - ocr_engine_languages:
-            msg += lang + '\n'
+        msg += '\n'.join(lang for lang in missing_languages)
+        msg += '\nNote: most languages are identified by a 3-digit ISO 639-2 Code'
         raise MissingDependencyError(msg)
 
 
@@ -79,12 +73,18 @@ def check_options_output(options):
     is_latin = options.languages.issubset(HOCR_OK_LANGS)
 
     if options.pdf_renderer.startswith('hocr') and not is_latin:
-        msg = (
+        log.warning(
             "The 'hocr' PDF renderer is known to cause problems with one "
             "or more of the languages in your document.  Use "
-            "--pdf-renderer auto (the default) to avoid this issue."
+            "`--pdf-renderer auto` (the default) to avoid this issue."
+        )
+
+    if options.output_type == 'none' and options.output_file not in (os.devnull, '-'):
+        raise BadArgsError(
+            "Since you specified `--output-type none`, the output file "
+            f"{options.output_file} cannot be produced. Set the output file to "
+            f"`-` to suppress this message."
         )
-        log.warning(msg)
 
     lossless_reconstruction = False
     if not any(
@@ -111,7 +111,15 @@ def check_options_sidecar(options):
             raise BadArgsError(
                 "--sidecar filename must be specified when output file is stdout."
             )
+        elif options.output_file == os.devnull:
+            raise BadArgsError(
+                "--sidecar filename must be specified when output file is /dev/null or NUL."
+            )
         options.sidecar = options.output_file + '.txt'
+    if options.sidecar == options.input_file or options.sidecar == options.output_file:
+        raise BadArgsError(
+            "--sidecar file must be different from the input and output files"
+        )
 
 
 def check_options_preprocessing(options):
@@ -133,29 +141,29 @@ def check_options_preprocessing(options):
                     options.unpaper_args
                 )
         except Exception as e:
-            raise BadArgsError(str(e))
+            raise BadArgsError("--unpaper-args: " + str(e)) from e
 
 
 def _pages_from_ranges(ranges: str) -> Set[int]:
-    if is_iterable_notstr(ranges):
-        return set(ranges)
     pages: List[int] = []
     page_groups = ranges.replace(' ', '').split(',')
-    for g in page_groups:
-        if not g:
+    for group in page_groups:
+        if not group:
             continue
         try:
-            start, end = g.split('-')
+            start, end = group.split('-')
         except ValueError:
-            pages.append(int(g) - 1)
+            pages.append(int(group) - 1)
         else:
             try:
                 new_pages = list(range(int(start) - 1, int(end)))
                 if not new_pages:
-                    raise BadArgsError(f"invalid page subrange '{start}-{end}'")
+                    raise BadArgsError(
+                        f"invalid page subrange '{start}-{end}'"
+                    ) from None
                 pages.extend(new_pages)
             except ValueError:
-                raise BadArgsError("invalid page range") from None
+                raise BadArgsError(f"invalid page subrange '{group}'") from None
 
     if not pages:
         raise BadArgsError(
@@ -177,50 +185,15 @@ def _pages_from_ranges(ranges: str) -> Set[int]:
 
 def check_options_ocr_behavior(options):
     exclusive_options = sum(
-        [
-            (1 if opt else 0)
-            for opt in (options.force_ocr, options.skip_text, options.redo_ocr)
-        ]
+        (1 if opt else 0)
+        for opt in (options.force_ocr, options.skip_text, options.redo_ocr)
     )
     if exclusive_options >= 2:
         raise BadArgsError("Choose only one of --force-ocr, --skip-text, --redo-ocr.")
-    if options.pages and options.sidecar:
-        raise BadArgsError("--pages and --sidecar are mutually exclusive")
     if options.pages:
         options.pages = _pages_from_ranges(options.pages)
 
 
-def check_options_optimizing(options):
-    if options.optimize >= 2:
-        check_external_program(
-            program='pngquant',
-            package='pngquant',
-            version_checker=pngquant.version,
-            need_version='2.0.1',
-            required_for='--optimize {2,3}',
-        )
-
-    if options.optimize >= 2:
-        # Although we use JBIG2 for optimize=1, don't nag about it unless the
-        # user is asking for more optimization
-        check_external_program(
-            program='jbig2',
-            package='jbig2enc',
-            version_checker=jbig2enc.version,
-            need_version='0.28',
-            required_for='--optimize {2,3} | --jbig2-lossy',
-            recommended=True if not options.jbig2_lossy else False,
-        )
-
-    if options.optimize == 0 and any(
-        [options.jbig2_lossy, options.png_quality, options.jpeg_quality]
-    ):
-        log.warning(
-            "The arguments --jbig2-lossy, --png-quality, and --jpeg-quality "
-            "will be ignored because --optimize=0."
-        )
-
-
 def check_options_advanced(options):
     if options.pdfa_image_compression != 'auto' and not options.output_type.startswith(
         'pdfa'
@@ -234,13 +207,13 @@ def check_options_advanced(options):
 def check_options_metadata(options):
     docinfo = [options.title, options.author, options.keywords, options.subject]
     for s in (m for m in docinfo if m):
-        for c in s:
-            if unicodedata.category(c) == 'Co' or ord(c) >= 0x10000:
+        for char in s:
+            if unicodedata.category(char) == 'Co' or ord(char) >= 0x10000:
+                hexchar = hex(ord(char))[2:].upper()
                 raise ValueError(
                     "One of the metadata strings contains "
-                    "an unsupported Unicode character: '{}' (U+{})".format(
-                        c, hex(ord(c))[2:].upper()
-                    )
+                    "an unsupported Unicode character: "
+                    f"{char} (U+{hexchar})"
                 )
 
 
@@ -258,7 +231,6 @@ def _check_options(options, plugin_manager, ocr_engine_languages):
     check_options_sidecar(options)
     check_options_preprocessing(options)
     check_options_ocr_behavior(options)
-    check_options_optimizing(options)
     check_options_advanced(options)
     check_options_pillow(options)
     plugin_manager.hook.check_options(options=options)
@@ -269,54 +241,6 @@ def check_options(options, plugin_manager):
     _check_options(options, plugin_manager, ocr_engine_languages)
 
 
-def check_closed_streams(options):  # pragma: no cover
-    """Work around Python issue with multiprocessing forking on closed streams
-
-    https://bugs.python.org/issue28326
-
-    Attempting to a fork/exec a new Python process when any of std{in,out,err}
-    are closed or not flushable for some reason may raise an exception.
-    Fix this by opening devnull if the handle seems to be closed.  Do this
-    globally to avoid tracking places all places that fork.
-
-    Seems to be specific to multiprocessing.Process not all Python process
-    forkers.
-
-    The error actually occurs when the stream object is not flushable,
-    but replacing an open stream object that is not flushable with
-    /dev/null is a bad idea since it will create a silent failure.  Replacing
-    a closed handle with /dev/null seems safe.
-
-    """
-
-    if sys.version_info[0:3] >= (3, 6, 4):
-        return True  # Issued fixed in Python 3.6.4+
-
-    if sys.stderr is None:
-        sys.stderr = open(os.devnull, 'w')
-
-    if sys.stdin is None:
-        if options.input_file == '-':
-            log.error("Trying to read from stdin but stdin seems closed")
-            return False
-        sys.stdin = open(os.devnull, 'r')
-
-    if sys.stdout is None:
-        if options.output_file == '-':
-            # Can't replace stdout if the user is piping
-            # If this case can even happen, it must be some kind of weird
-            # stream.
-            log.error(
-                "Output was set to stdout '-' but the stream attached to "
-                "stdout does not support the flush() system call.  This "
-                "will fail."
-            )
-            return False
-        sys.stdout = open(os.devnull, 'w')
-
-    return True
-
-
 def create_input_file(options, work_folder: Path) -> Tuple[Path, str]:
     if options.input_file == '-':
         # stdin
@@ -338,8 +262,18 @@ def create_input_file(options, work_folder: Path) -> Tuple[Path, str]:
             target = work_folder / 'origin'
             safe_symlink(options.input_file, target)
             return target, os.fspath(options.input_file)
-        except FileNotFoundError:
-            raise InputFileError(f"File not found - {options.input_file}")
+        except FileNotFoundError as e:
+            msg = f"File not found - {options.input_file}"
+            if Path('/.dockerenv').exists():  # pragma: no cover
+                msg += (
+                    "\nDocker cannot your working directory unless you "
+                    "explicitly share it with the Docker container and set up"
+                    "permissions correctly.\n"
+                    "You may find it easier to use stdin/stdout:"
+                    "\n"
+                    "\tdocker run -i --rm jbarlow83/ocrmypdf - - <input.pdf >output.pdf\n"
+                )
+            raise InputFileError(msg) from e
 
 
 def check_requested_output_file(options):
@@ -359,7 +293,16 @@ def check_requested_output_file(options):
         )
 
 
-def report_output_file_size(options, input_file, output_file):
+def report_output_file_size(
+    options,
+    input_file: Path,
+    output_file: Path,
+    optimize_messages: Optional[Sequence[str]] = None,
+    file_overhead: int = 4000,
+    page_overhead: int = 3000,
+):
+    if optimize_messages is None:
+        optimize_messages = []
     try:
         output_size = Path(output_file).stat().st_size
         input_size = Path(input_file).stat().st_size
@@ -368,9 +311,7 @@ def report_output_file_size(options, input_file, output_file):
     with pikepdf.open(output_file) as p:
         # Overhead constants obtained by estimating amount of data added by OCR
         # PDF/A conversion, and possible XMP metadata addition, with compression
-        FILE_OVERHEAD = 4000
-        OCR_PER_PAGE_OVERHEAD = 3000
-        reasonable_overhead = FILE_OVERHEAD + OCR_PER_PAGE_OVERHEAD * len(p.pages)
+        reasonable_overhead = file_overhead + page_overhead * len(p.pages)
     ratio = output_size / input_size
     reasonable_ratio = output_size / (input_size + reasonable_overhead)
     if reasonable_ratio < 1.35 or input_size < 25000:
@@ -390,19 +331,8 @@ def report_output_file_size(options, input_file, output_file):
                 f"The argument --{arg.replace('_', '-')} was issued, causing transcoding."
             )
 
-    if options.optimize == 0:
-        reasons.append("Optimization was disabled.")
-    else:
-        image_optimizers = {
-            'jbig2': jbig2enc.available(),
-            'pngquant': pngquant.available(),
-        }
-        for name, available in image_optimizers.items():
-            if not available:
-                reasons.append(
-                    f"The optional dependency '{name}' was not found, so some image "
-                    f"optimizations could not be attempted."
-                )
+    reasons.extend(optimize_messages)
+
     if options.output_type.startswith('pdfa'):
         reasons.append("PDF/A conversion was enabled. (Try `--output-type pdf`.)")
     if options.plugins:

src/ocrmypdf/_version.py

@@ -4,10 +4,17 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""Get version by introspecting package information.
 
-import pkg_resources
+OCRmyPDF uses setuptools_scm to derive version from git tags.
+"""
+
+try:
+    from importlib_metadata import version as _package_version
+except ImportError:
+    from importlib.metadata import version as _package_version
 
 PROGRAM_NAME = 'ocrmypdf'
 
 # Official PEP 396
-__version__ = pkg_resources.get_distribution('ocrmypdf').version
+__version__ = _package_version('ocrmypdf')

src/ocrmypdf/api.py

@@ -4,6 +4,7 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""Functions for using ocrmypdf as an API."""
 
 import logging
 import os
@@ -25,10 +26,13 @@ from ocrmypdf.helpers import is_iterable_notstr
 try:
     import coloredlogs
 except ModuleNotFoundError:
-    coloredlogs = None
+    coloredlogs = None  # pylint: disable=invalid-name
+
+if coloredlogs:
+    from humanfriendly.terminal import enable_ansi_support
 
 
-StrPath = Union[os.PathLike, AnyStr]
+StrPath = Union[Path, AnyStr]
 PathOrIO = Union[BinaryIO, StrPath]
 
 _api_lock = threading.Lock()
@@ -37,6 +41,7 @@ _api_lock = threading.Lock()
 class Verbosity(IntEnum):
     """Verbosity level for configure_logging."""
 
+    # pylint: disable=invalid-name
     quiet = -1  #: Suppress most messages
     default = 0  #: Default level of logging
     debug = 1  #: Output ocrmypdf debug messages
@@ -45,18 +50,23 @@ class Verbosity(IntEnum):
 
 def configure_logging(
     verbosity: Verbosity,
+    *,
     progress_bar_friendly: bool = True,
     manage_root_logger: bool = False,
+    plugin_manager=None,
 ):
     """Set up logging.
 
     Before calling :func:`ocrmypdf.ocr()`, you can use this function to
-    configure logging, if you want ocrmypdf's output to look like the ocrmypdf
+    configure logging if you want ocrmypdf's output to look like the ocrmypdf
     command line interface. It will register log handlers, log filters, and
     formatters, configure color logging to standard error, and adjust the log
     levels of third party libraries. Details of this are fine-tuned and subject
     to change. The ``verbosity`` argument is equivalent to the argument
-    ``--verbose`` and applies those settings.
+    ``--verbose`` and applies those settings. If you have a wrapper
+    script for ocrmypdf and you want it to be very similar to ocrmypdf, use this
+    function; if you are using ocrmypdf as part of an application that manages
+    its own logging, you probably do not want this function.
 
     If this function is not called, ocrmypdf will not configure logging, and it
     is up to the caller of ``ocrmypdf.ocr()`` to set up logging as it wishes using
@@ -74,12 +84,11 @@ def configure_logging(
     their own debug logging.
 
     Args:
-        verbosity (Verbosity): Verbosity level.
-        progress_bar_friendly (bool): Install the TqdmConsole log handler, which is
-            compatible with the tqdm progress bar; without this log messages will
-            overwrite the progress bar
-        manage_root_logger (bool): Configure the process's root logger, to ensure
-            all log output is sent through
+        verbosity: Verbosity level.
+        progress_bar_friendly: If True (the default), install a custom log handler
+            that is compatible with progress bars and colored output.
+        manage_root_logger: Configure the process's root logger.
+        plugin_manager: The plugin manager, used for obtaining the custom log handler.
 
     Returns:
         The toplevel logger for ocrmypdf (or the root logger, if we are managing it).
@@ -90,9 +99,11 @@ def configure_logging(
     log = logging.getLogger(prefix)
     log.setLevel(logging.DEBUG)
 
-    if progress_bar_friendly:
-        console = logging.StreamHandler(stream=TqdmConsole(sys.stderr))
-    else:
+    console = None
+    if plugin_manager and progress_bar_friendly:
+        console = plugin_manager.hook.get_logging_console()
+
+    if not console:
         console = logging.StreamHandler(stream=sys.stderr)
 
     if verbosity < 0:
@@ -110,16 +121,15 @@ def configure_logging(
         fmt = '%(pageno)s%(message)s'
 
     use_colors = progress_bar_friendly
-    if not coloredlogs:
-        use_colors = False
-    if use_colors:
-        if os.name == 'nt':
-            use_colors = coloredlogs.enable_ansi_support()
+    formatter = None
+    if coloredlogs and use_colors:
+        use_colors = enable_ansi_support()
         if use_colors:
             use_colors = coloredlogs.terminal_supports_colors()
-    if use_colors:
-        formatter = coloredlogs.ColoredFormatter(fmt=fmt)
-    else:
+        if use_colors:
+            formatter = coloredlogs.ColoredFormatter(fmt=fmt)
+
+    if not formatter:
         formatter = logging.Formatter(fmt=fmt)
 
     console.setFormatter(formatter)
@@ -187,7 +197,7 @@ def create_options(
     else:
         cmdline.append(os.fspath(output_file))
 
-    parser._api_mode = True
+    parser.enable_api_mode()
     options = parser.parse_args(cmdline)
     for keyword, val in deferred:
         setattr(options, keyword, val)
@@ -222,7 +232,6 @@ def ocr(  # pylint: disable=unused-argument
     unpaper_args: str = None,
     oversample: int = None,
     remove_vectors: bool = None,
-    threshold: bool = None,
     force_ocr: bool = None,
     skip_text: bool = None,
     redo_ocr: bool = None,
@@ -237,6 +246,7 @@ def ocr(  # pylint: disable=unused-argument
     tesseract_config: Iterable[str] = None,
     tesseract_pagesegmode: int = None,
     tesseract_oem: int = None,
+    tesseract_thresholding: int = None,
     pdf_renderer=None,
     tesseract_timeout: float = None,
     rotate_pages_threshold: float = None,
@@ -245,6 +255,7 @@ def ocr(  # pylint: disable=unused-argument
     user_patterns: os.PathLike = None,
     fast_web_view: float = None,
     plugins: Iterable[StrPath] = None,
+    plugin_manager=None,
     keep_temporary_files: bool = None,
     progress_bar: bool = None,
     **kwargs,
@@ -288,7 +299,7 @@ def ocr(  # pylint: disable=unused-argument
             text already, and settings did not tell us to proceed.
         ocrmypdf.InputFileError: Any other problem with the input file.
         ocrmypdf.SubprocessOutputError: Any error related to executing a subprocess.
-        ocrmypdf.EncryptedPdfERror: If the input PDF is encrypted (password protected).
+        ocrmypdf.EncryptedPdfError: If the input PDF is encrypted (password protected).
             OCRmyPDF does not remove passwords.
         ocrmypdf.TesseractConfigError: If Tesseract reported its configuration was not
             valid.
@@ -296,6 +307,9 @@ def ocr(  # pylint: disable=unused-argument
     Returns:
         :class:`ocrmypdf.ExitCode`
     """
+    if plugins and plugin_manager:
+        raise ValueError("plugins= and plugin_manager are mutually exclusive")
+
     if not plugins:
         plugins = []
     elif isinstance(plugins, (str, Path)):
@@ -315,7 +329,8 @@ def ocr(  # pylint: disable=unused-argument
         # they might install different plugins, and generally speaking we have areas
         # of code that use global state.
 
-        plugin_manager = get_plugin_manager(plugins)
+        if not plugin_manager:
+            plugin_manager = get_plugin_manager(plugins)
         plugin_manager.hook.add_options(parser=parser)  # pylint: disable=no-member
 
         if 'verbose' in kwargs:
@@ -324,3 +339,17 @@ def ocr(  # pylint: disable=unused-argument
         options = create_options(**create_options_kwargs)
         check_options(options, plugin_manager)
         return run_pipeline(options=options, plugin_manager=plugin_manager, api=True)
+
+
+__all__ = [
+    'PageNumberFilter',
+    'TqdmConsole',
+    'Verbosity',
+    'check_options',
+    'configure_logging',
+    'create_options',
+    'get_parser',
+    'get_plugin_manager',
+    'ocr',
+    'run_pipeline',
+]

src/ocrmypdf/builtin_plugins/__init__.py

@@ -3,3 +3,7 @@
 # This Source Code Form is subject to the terms of the Mozilla Public
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
+
+# This file exists only mark builtin_plugins as a package.
+# The plugin manager will not load it, so anything defined here may not be
+# processed as a module.

src/ocrmypdf/builtin_plugins/concurrency.py

@@ -0,0 +1,177 @@
+# © 2020 James R. Barlow: github.com/jbarlow83
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+
+# © 2020 James R. Barlow: github.com/jbarlow83
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+
+
+"""OCRmyPDF's multiprocessing/multithreading abstraction layer."""
+
+import logging
+import logging.handlers
+import multiprocessing
+import os
+import queue
+import signal
+import sys
+import threading
+from concurrent.futures import ProcessPoolExecutor, ThreadPoolExecutor, as_completed
+from contextlib import suppress
+from typing import Callable, Iterable, Type, Union
+
+from tqdm import tqdm
+
+from ocrmypdf import Executor, hookimpl
+from ocrmypdf._logging import TqdmConsole
+from ocrmypdf.exceptions import InputFileError
+from ocrmypdf.helpers import remove_all_log_handlers
+
+FuturesExecutorClass = Union[Type[ThreadPoolExecutor], Type[ProcessPoolExecutor]]
+Queue = Union[multiprocessing.Queue, queue.Queue]
+UserInit = Callable[[], None]
+WorkerInit = Callable[[Queue, UserInit, int], None]
+
+
+def log_listener(q: Queue):
+    """Listen to the worker processes and forward the messages to logging
+
+    For simplicity this is a thread rather than a process. Only one process
+    should actually write to sys.stderr or whatever we're using, so if this is
+    made into a process the main application needs to be directed to it.
+
+    See:
+    https://docs.python.org/3/howto/logging-cookbook.html#logging-to-a-single-file-from-multiple-processes
+    """
+
+    while True:
+        try:
+            record = q.get()
+            if record is None:
+                break
+            logger = logging.getLogger(record.name)
+            logger.handle(record)
+        except Exception:  # pylint: disable=broad-except
+            import traceback  # pylint: disable=import-outside-toplevel
+
+            print("Logging problem", file=sys.stderr)
+            traceback.print_exc(file=sys.stderr)
+
+
+def process_sigbus(*args):
+    raise InputFileError("A worker process lost access to an input file")
+
+
+def process_init(q: Queue, user_init: UserInit, loglevel) -> None:
+    """Initialize a process pool worker"""
+
+    # Ignore SIGINT (our parent process will kill us gracefully)
+    signal.signal(signal.SIGINT, signal.SIG_IGN)
+
+    # Install SIGBUS handler (so our parent process can abort somewhat gracefully)
+    with suppress(AttributeError):  # Windows and Cygwin do not have SIGBUS
+        # Windows and Cygwin do not have pthread_sigmask or SIGBUS
+        signal.signal(signal.SIGBUS, process_sigbus)
+
+    # Remove any log handlers that belong to the parent process
+    root = logging.getLogger()
+    remove_all_log_handlers(root)
+
+    # Set up our single log handler to forward messages to the parent
+    root.setLevel(loglevel)
+    root.addHandler(logging.handlers.QueueHandler(q))
+
+    user_init()
+    return
+
+
+def thread_init(q: Queue, user_init: UserInit, loglevel) -> None:
+    del q  # unused but required argument
+    del loglevel  # unused but required argument
+    # As a thread, block SIGBUS so the main thread deals with it...
+    with suppress(AttributeError):
+        signal.pthread_sigmask(signal.SIG_BLOCK, {signal.SIGBUS})
+
+    user_init()
+    return
+
+
+class StandardExecutor(Executor):
+    """Standard OCRmyPDF concurrent task executor."""
+
+    def _execute(
+        self,
+        *,
+        use_threads: bool,
+        max_workers: int,
+        tqdm_kwargs: dict,
+        worker_initializer: Callable,
+        task: Callable,
+        task_arguments: Iterable,
+        task_finished: Callable,
+    ):
+        if use_threads:
+            log_queue: Queue = queue.Queue(-1)
+            executor_class: FuturesExecutorClass = ThreadPoolExecutor
+            initializer: WorkerInit = thread_init
+        else:
+            log_queue = multiprocessing.Queue(-1)
+            executor_class = ProcessPoolExecutor
+            initializer = process_init
+
+        # Regardless of whether we use_threads for worker processes, the log_listener
+        # must be a thread. Make sure we create the listener after the worker pool,
+        # so that it does not get forked into the workers.
+        listener = threading.Thread(target=log_listener, args=(log_queue,))
+        listener.start()
+
+        with self.pbar_class(**tqdm_kwargs) as pbar, executor_class(
+            max_workers=max_workers,
+            initializer=initializer,
+            initargs=(log_queue, worker_initializer, logging.getLogger("").level),
+        ) as executor:
+            futures = [executor.submit(task, args) for args in task_arguments]
+            try:
+                for future in as_completed(futures):
+                    result = future.result()
+                    task_finished(result, pbar)
+            except KeyboardInterrupt:
+                # Terminate pool so we exit instantly
+                executor.shutdown(wait=False, cancel_futures=True)
+                raise
+            except Exception:
+                if not os.environ.get("PYTEST_CURRENT_TEST", ""):
+                    # Normally we shutdown without waiting for other child workers
+                    # on error, because there is no point in waiting for them. Their
+                    # results will be discard. But if the condition above is True,
+                    # then we are running in pytest, and we want everything to exit
+                    # as cleanly as possible so that we get good error messages.
+                    executor.shutdown(wait=False, cancel_futures=True)
+                raise
+            finally:
+                # Terminate log listener
+                log_queue.put_nowait(None)
+
+        # When the above succeeds, wait for the listener thread to exit. (If
+        # an exception occurs, we don't try to join, in case it deadlocks.)
+        listener.join()
+
+
+@hookimpl
+def get_executor(progressbar_class):
+    return StandardExecutor(pbar_class=progressbar_class)
+
+
+@hookimpl
+def get_progressbar_class():
+    return tqdm
+
+
+@hookimpl
+def get_logging_console():
+    return logging.StreamHandler(stream=TqdmConsole(sys.stderr))

src/ocrmypdf/builtin_plugins/default_filters.py

@@ -0,0 +1,16 @@
+# © 2021 James R. Barlow: github.com/jbarlow83
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+
+"""OCRmyPDF automatically installs these filters as plugins."""
+
+from ocrmypdf import hookimpl
+
+
+@hookimpl
+def filter_pdf_page(
+    page, image_filename, output_pdf
+):  # pylint: disable=unused-argument
+    return output_pdf

src/ocrmypdf/builtin_plugins/ghostscript.py

@@ -5,6 +5,8 @@
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
 
+"""Built-in plugin to implement PDF page rasterization and PDF/A production."""
+
 import logging
 
 from ocrmypdf import hookimpl
@@ -18,13 +20,13 @@ log = logging.getLogger(__name__)
 
 @hookimpl
 def check_options(options):
-    gs_version = ghostscript.version()
     check_external_program(
         program='gs',
         package='ghostscript',
-        version_checker=gs_version,
+        version_checker=ghostscript.version,
         need_version='9.15',  # limited by Travis CI / Ubuntu 14.04 backports
     )
+    gs_version = ghostscript.version()
     if gs_version in ('9.24', '9.51'):
         raise MissingDependencyError(
             f"Ghostscript {gs_version} contains serious regressions and is not "

src/ocrmypdf/builtin_plugins/optimize.py

@@ -0,0 +1,158 @@
+# © 2022 James R. Barlow: github.com/jbarlow83
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+
+
+"""Built-in plugin to implement PDF page optimization."""
+
+import argparse
+import logging
+from pathlib import Path
+from typing import Sequence, Tuple
+
+from ocrmypdf import Executor, PdfContext, hookimpl
+from ocrmypdf._exec import jbig2enc, pngquant
+from ocrmypdf._pipeline import get_pdf_save_settings
+from ocrmypdf.cli import numeric
+from ocrmypdf.optimize import optimize
+from ocrmypdf.subprocess import check_external_program
+
+log = logging.getLogger(__name__)
+
+
+@hookimpl
+def add_options(parser):
+    optimizing = parser.add_argument_group(
+        "Optimization options", "Control how the PDF is optimized after OCR"
+    )
+    optimizing.add_argument(
+        '-O',
+        '--optimize',
+        type=int,
+        choices=range(0, 4),
+        default=1,
+        help=(
+            "Control how PDF is optimized after processing:"
+            "0 - do not optimize; "
+            "1 - do safe, lossless optimizations (default); "
+            "2 - do lossy JPEG and JPEG2000 optimizations; "
+            "3 - do more aggressive lossy JPEG and JPEG2000 optimizations. "
+            "To enable lossy JBIG2, see --jbig2-lossy."
+        ),
+    )
+    optimizing.add_argument(
+        '--jpeg-quality',
+        type=numeric(int, 0, 100),
+        default=0,
+        metavar='Q',
+        help=(
+            "Adjust JPEG quality level for JPEG optimization. "
+            "100 is best quality and largest output size; "
+            "1 is lowest quality and smallest output; "
+            "0 uses the default."
+        ),
+    )
+    optimizing.add_argument(
+        '--jpg-quality',
+        type=numeric(int, 0, 100),
+        default=0,
+        metavar='Q',
+        dest='jpeg_quality',
+        help=argparse.SUPPRESS,  # Alias for --jpeg-quality
+    )
+    optimizing.add_argument(
+        '--png-quality',
+        type=numeric(int, 0, 100),
+        default=0,
+        metavar='Q',
+        help=(
+            "Adjust PNG quality level to use when quantizing PNGs. "
+            "Values have same meaning as with --jpeg-quality"
+        ),
+    )
+    optimizing.add_argument(
+        '--jbig2-lossy',
+        action='store_true',
+        help=(
+            "Enable JBIG2 lossy mode (better compression, not suitable for some "
+            "use cases - see documentation). Only takes effect if --optimize 1 or "
+            "higher is also enabled."
+        ),
+    )
+    optimizing.add_argument(
+        '--jbig2-page-group-size',
+        type=numeric(int, 1, 10000),
+        default=0,
+        metavar='N',
+        # Adjust number of pages to consider at once for JBIG2 compression
+        help=argparse.SUPPRESS,
+    )
+
+
+@hookimpl
+def check_options(options):
+    if options.optimize >= 2:
+        check_external_program(
+            program='pngquant',
+            package='pngquant',
+            version_checker=pngquant.version,
+            need_version='2.0.1',
+            required_for='--optimize {2,3}',
+        )
+
+    if options.optimize >= 2:
+        # Although we use JBIG2 for optimize=1, don't nag about it unless the
+        # user is asking for more optimization
+        check_external_program(
+            program='jbig2',
+            package='jbig2enc',
+            version_checker=jbig2enc.version,
+            need_version='0.28',
+            required_for='--optimize {2,3} | --jbig2-lossy',
+            recommended=True if not options.jbig2_lossy else False,
+        )
+
+    if options.optimize == 0 and any(
+        [options.jbig2_lossy, options.png_quality, options.jpeg_quality]
+    ):
+        log.warning(
+            "The arguments --jbig2-lossy, --png-quality, and --jpeg-quality "
+            "will be ignored because --optimize=0."
+        )
+
+
+@hookimpl
+def optimize_pdf(
+    input_pdf: Path,
+    output_pdf: Path,
+    context: PdfContext,
+    executor: Executor,
+    linearize: bool,
+) -> Tuple[Path, Sequence[str]]:
+    save_settings = dict(
+        linearize=linearize,
+        **get_pdf_save_settings(context.options.output_type),
+    )
+    result_path = optimize(input_pdf, output_pdf, context, save_settings, executor)
+    messages = []
+    if context.options.optimize == 0:
+        messages.append("Optimization was disabled.")
+    else:
+        image_optimizers = {
+            'jbig2': jbig2enc.available(),
+            'pngquant': pngquant.available(),
+        }
+        for name, available in image_optimizers.items():
+            if not available:
+                messages.append(
+                    f"The optional dependency '{name}' was not found, so some image "
+                    f"optimizations could not be attempted."
+                )
+    return result_path, messages
+
+
+@hookimpl
+def is_optimization_enabled(context: PdfContext) -> bool:
+    return context.options.optimize != 0

src/ocrmypdf/builtin_plugins/tesseract_ocr.py

@@ -4,14 +4,15 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""Built-in plugin to implement OCR using Tesseract."""
+
 
 import logging
 import os
 
 from ocrmypdf import hookimpl
 from ocrmypdf._exec import tesseract
-from ocrmypdf.cli import numeric
-from ocrmypdf.exceptions import MissingDependencyError
+from ocrmypdf.cli import numeric, str_to_int
 from ocrmypdf.helpers import clamp
 from ocrmypdf.pluginspec import OcrEngine
 from ocrmypdf.subprocess import check_external_program
@@ -44,13 +45,27 @@ def add_options(parser):
         metavar='MODE',
         choices=range(0, 4),
         help=(
-            "Set Tesseract 4.0 OCR engine mode: "
+            "Set Tesseract 4.0+ OCR engine mode: "
             "0 - original Tesseract only; "
             "1 - neural nets LSTM only; "
             "2 - Tesseract + LSTM; "
             "3 - default."
         ),
     )
+    tess.add_argument(
+        '--tesseract-thresholding',
+        action='store',
+        type=str_to_int(tesseract.TESSERACT_THRESHOLDING_METHODS),
+        default='auto',
+        metavar='METHOD',
+        help=(
+            "Set Tesseract 5.0+ input image thresholding mode. This may improve OCR "
+            "results on low quality images or those that contain high constrast color. "
+            "legacy-otsu is the Tesseract default; adaptive-otsu is an improved Otsu "
+            "algorithm with improved sort for background color changes; sauvola is "
+            "based on local standard deviation."
+        ),
+    )
     tess.add_argument(
         '--tesseract-timeout',
         default=180.0,
@@ -80,24 +95,24 @@ def check_options(options):
         program='tesseract',
         package={'linux': 'tesseract-ocr'},
         version_checker=tesseract.version,
-        need_version='4.0.0',  # using backport for Travis CI
+        need_version='4.0.0-beta.1',  # using backport for Travis CI
+        version_parser=tesseract.TesseractVersion,
     )
 
     # Decide on what renderer to use
     if options.pdf_renderer == 'auto':
         options.pdf_renderer = 'sandwich'
 
-    if options.pdf_renderer == 'sandwich' and not tesseract.has_textonly_pdf(
-        set(options.languages)
-    ):
-        raise MissingDependencyError(
-            "You are using an alpha version of Tesseract 4.0 that does not support "
-            "the textonly_pdf parameter. We don't support versions this old."
-        )
     if not tesseract.has_user_words() and (options.user_words or options.user_patterns):
         log.warning(
-            "Tesseract 4.0 ignores --user-words and --user-patterns, so these "
-            "arguments have no effect."
+            "Tesseract 4.0 (which you have installed) ignores --user-words and "
+            "--user-patterns, so these arguments have no effect."
+        )
+    if not tesseract.has_thresholding() and options.tesseract_thresholding != 0:
+        log.warning(
+            "The installed version of Tesseract does not support changes to its "
+            "thresholding method. The --tesseract-threshold argument will be "
+            "ignored."
         )
     if options.tesseract_pagesegmode in (0, 2):
         log.warning(
@@ -125,6 +140,8 @@ def validate(pdfinfo, options):
 
 
 class TesseractOcrEngine(OcrEngine):
+    """Implements OCR with Tesseract."""
+
     @staticmethod
     def version():
         return tesseract.version()
@@ -149,6 +166,15 @@ class TesseractOcrEngine(OcrEngine):
             timeout=options.tesseract_timeout,
         )
 
+    @staticmethod
+    def get_deskew(input_file, options) -> float:
+        return tesseract.get_deskew(
+            input_file,
+            languages=options.languages,
+            engine_mode=options.tesseract_oem,
+            timeout=options.tesseract_timeout,
+        )
+
     @staticmethod
     def generate_hocr(input_file, output_hocr, output_text, options):
         tesseract.generate_hocr(
@@ -160,6 +186,7 @@ class TesseractOcrEngine(OcrEngine):
             tessconfig=options.tesseract_config,
             timeout=options.tesseract_timeout,
             pagesegmode=options.tesseract_pagesegmode,
+            thresholding=options.tesseract_thresholding,
             user_words=options.user_words,
             user_patterns=options.user_patterns,
         )
@@ -175,6 +202,7 @@ class TesseractOcrEngine(OcrEngine):
             tessconfig=options.tesseract_config,
             timeout=options.tesseract_timeout,
             pagesegmode=options.tesseract_pagesegmode,
+            thresholding=options.tesseract_thresholding,
             user_words=options.user_words,
             user_patterns=options.user_patterns,
         )

src/ocrmypdf/cli.py

@@ -4,42 +4,68 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""Command line interface customization and validation."""
 
 import argparse
-from typing import Optional, Type, TypeVar
+from typing import Any, Callable, Mapping, Optional, TypeVar
 
 from ocrmypdf._version import PROGRAM_NAME as _PROGRAM_NAME
 from ocrmypdf._version import __version__ as _VERSION
 
-T = TypeVar('T')
+T = TypeVar('T', int, float)
 
 
-def numeric(basetype: Type[T], min_: Optional[T] = None, max_: Optional[T] = None):
+def numeric(
+    basetype: Callable[[Any], T], min_: Optional[T] = None, max_: Optional[T] = None
+):
     """Validator for numeric params"""
     min_ = basetype(min_) if min_ is not None else None
     max_ = basetype(max_) if max_ is not None else None
 
-    def _numeric(string):
-        value = basetype(string)
+    def _numeric(s: str) -> T:
+        value = basetype(s)
         if (min_ is not None and value < min_) or (max_ is not None and value > max_):
-            msg = "%r not in valid range %r" % (string, (min_, max_))
-            raise argparse.ArgumentTypeError(msg)
+            raise argparse.ArgumentTypeError(
+                f"{s!r} not in valid range {(min_, max_)!r}"
+            )
         return value
 
     _numeric.__name__ = basetype.__name__
     return _numeric
 
 
+def str_to_int(mapping: Mapping[str, int]):
+    """Accept text on command line and convert to integer."""
+
+    def _str_to_int(s: str) -> int:
+        try:
+            return mapping[s]
+        except KeyError:
+            raise argparse.ArgumentTypeError(
+                f"{s!r} must be one of: {', '.join(mapping.keys())}"
+            ) from None
+
+    return _str_to_int
+
+
 class ArgumentParser(argparse.ArgumentParser):
     """Override parser's default behavior of calling sys.exit()
 
     https://stackoverflow.com/questions/5943249/python-argparse-and-controlling-overriding-the-exit-status-code
+
+    OCRmyPDF began as a CLI but eventually acquired an API. The API works inside out,
+    by synthesizing a command line argument. So we subclass the standard parser with
+    one that doesn't call sys.exit(). Obviously this is not the ideal way to do things
+    but it works for us.
     """
 
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
         self._api_mode = False
 
+    def enable_api_mode(self):
+        self._api_mode = True
+
     def error(self, message):
         if not self._api_mode:
             super().error(message)
@@ -48,6 +74,8 @@ class ArgumentParser(argparse.ArgumentParser):
 
 
 class LanguageSetAction(argparse.Action):
+    """Manages a list of languages."""
+
     def __init__(self, option_strings, dest, default=None, **kwargs):
         if default is None:
             default = set()
@@ -125,7 +153,7 @@ Online documentation is located at:
         'output_file',
         metavar="output_pdf",
         help="Output searchable PDF file (or '-' to write to standard output). "
-        "Existing files will be ovewritten. If same as input file, the "
+        "Existing files will be overwritten. If same as input file, the "
         "input file will be updated only if processing is successful.",
     )
     parser.add_argument(
@@ -145,7 +173,7 @@ Online documentation is located at:
     )
     parser.add_argument(
         '--output-type',
-        choices=['pdfa', 'pdf', 'pdfa-1', 'pdfa-2', 'pdfa-3'],
+        choices=['pdfa', 'pdf', 'pdfa-1', 'pdfa-2', 'pdfa-3', 'none'],
         default='pdfa',
         help="Choose output type. 'pdfa' creates a PDF/A-2b compliant file for "
         "long term archiving (default, recommended) but may not suitable "
@@ -153,7 +181,8 @@ Online documentation is located at:
         "also has problems with full Unicode text. 'pdf' attempts to "
         "preserve file contents as much as possible. 'pdf-a1' creates a "
         "PDF/A1-b file. 'pdf-a2' is equivalent to 'pdfa'. 'pdf-a3' creates a "
-        "PDF/A3-b file.",
+        "PDF/A3-b file. 'none' will produce no output, which may be helpful if "
+        "only the --sidecar is desired.",
     )
 
     # Use null string '\0' as sentinel to indicate the user supplied no argument,
@@ -167,7 +196,8 @@ Online documentation is located at:
         metavar='FILE',
         help="Generate sidecar text files that contain the same text recognized "
         "by Tesseract. This may be useful for building a OCR text database. "
-        "If FILE is omitted, the sidecar file be named {output_file}.txt "
+        "If FILE is omitted, the sidecar file be named {output_file}.txt; the next "
+        "argument must NOT be the name of the input PDF. "
         "If FILE is set to '-', the sidecar is written to stdout (a "
         "convenient way to preview OCR quality). The output file and sidecar "
         "may not both use stdout at the same time.",
@@ -209,7 +239,13 @@ Online documentation is located at:
         help=argparse.SUPPRESS,
     )
     jobcontrol.add_argument(
-        '--use-threads', action='store_true', help=argparse.SUPPRESS
+        '--use-threads', action='store_true', default=True, help=argparse.SUPPRESS
+    )
+    jobcontrol.add_argument(
+        '--no-use-threads',
+        action='store_false',
+        dest='use_threads',
+        help=argparse.SUPPRESS,
     )
 
     metadata = parser.add_argument_group(
@@ -283,14 +319,6 @@ Online documentation is located at:
         help="EXPERIMENTAL. Mask out any vector objects in the PDF so that they "
         "will not be included in OCR. This can eliminate false characters.",
     )
-    preprocessing.add_argument(
-        '--threshold',
-        action='store_true',
-        help=(
-            "EXPERIMENTAL. Threshold image to 1bpp before sending it to Tesseract "
-            "for OCR. Can improve OCR quality compared to Tesseract's thresholder."
-        ),
-    )
 
     ocrsettings = parser.add_argument_group("OCR options", "Control how OCR is applied")
     ocrsettings.add_argument(
@@ -324,70 +352,6 @@ Online documentation is located at:
         "but include skipped pages in final output",
     )
 
-    optimizing = parser.add_argument_group(
-        "Optimization options", "Control how the PDF is optimized after OCR"
-    )
-    optimizing.add_argument(
-        '-O',
-        '--optimize',
-        type=int,
-        choices=range(0, 4),
-        default=1,
-        help=(
-            "Control how PDF is optimized after processing:"
-            "0 - do not optimize; "
-            "1 - do safe, lossless optimizations (default); "
-            "2 - do some lossy optimizations; "
-            "3 - do aggressive lossy optimizations (including lossy JBIG2)"
-        ),
-    )
-    optimizing.add_argument(
-        '--jpeg-quality',
-        type=numeric(int, 0, 100),
-        default=0,
-        metavar='Q',
-        help=(
-            "Adjust JPEG quality level for JPEG optimization. "
-            "100 is best quality and largest output size; "
-            "1 is lowest quality and smallest output; "
-            "0 uses the default."
-        ),
-    )
-    optimizing.add_argument(
-        '--jpg-quality',
-        type=numeric(int, 0, 100),
-        default=0,
-        metavar='Q',
-        dest='jpeg_quality',
-        help=argparse.SUPPRESS,  # Alias for --jpeg-quality
-    )
-    optimizing.add_argument(
-        '--png-quality',
-        type=numeric(int, 0, 100),
-        default=0,
-        metavar='Q',
-        help=(
-            "Adjust PNG quality level to use when quantizing PNGs. "
-            "Values have same meaning as with --jpeg-quality"
-        ),
-    )
-    optimizing.add_argument(
-        '--jbig2-lossy',
-        action='store_true',
-        help=(
-            "Enable JBIG2 lossy mode (better compression, not suitable for some "
-            "use cases - see documentation)."
-        ),
-    )
-    optimizing.add_argument(
-        '--jbig2-page-group-size',
-        type=numeric(int, 1, 10000),
-        default=0,
-        metavar='N',
-        # Adjust number of pages to consider at once for JBIG2 compression
-        help=argparse.SUPPRESS,
-    )
-
     advanced = parser.add_argument_group(
         "Advanced", "Advanced options to control OCRmyPDF"
     )
@@ -406,7 +370,7 @@ Online documentation is located at:
         metavar='MPixels',
         help="Set maximum number of pixels to unpack before treating an image as a "
         "decompression bomb",
-        default=128.0,
+        default=250.0,
     )
     advanced.add_argument(
         '--pdf-renderer',

src/ocrmypdf/data/__init__.py

@@ -0,0 +1,8 @@
+# © 2021 James R. Barlow: github.com/jbarlow83
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+
+
+"""Data files used to generate certain PDFs."""

src/ocrmypdf/exceptions.py

@@ -4,12 +4,16 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""OCRmyPDF's exceptions."""
 
 from enum import IntEnum
 from textwrap import dedent
 
 
 class ExitCode(IntEnum):
+    """OCRmyPDF's exit codes."""
+
+    # pylint: disable=invalid-name
     ok = 0
     bad_args = 1
     input_file = 2
@@ -26,6 +30,8 @@ class ExitCode(IntEnum):
 
 
 class ExitCodeException(Exception):
+    """An exception which should return an exit code with sys.exit()."""
+
     exit_code = ExitCode.other_error
     message = ""
 
@@ -37,17 +43,24 @@ class ExitCodeException(Exception):
 
 
 class BadArgsError(ExitCodeException):
+    """Invalid arguments on the command line or API."""
+
     exit_code = ExitCode.bad_args
 
 
-class PdfMergeFailedError(ExitCodeException):
+class PdfMergeFailedError(ExitCodeException):  # deprecated
+    """An intermediate PDF can't be merged.
+
+    No longer in use.
+    """
+
     exit_code = ExitCode.input_file
     message = dedent(
         '''\
         Failed to merge PDF image layer with OCR layer
 
         Usually this happens because the input PDF file is malformed and
-        ocrmypdf cannot automatically correct the problem on its own.
+        ocrmypdf cannot correct the problem on its own.
 
         Try using
             ocrmypdf --pdf-renderer sandwich  [..other args..]
@@ -56,34 +69,50 @@ class PdfMergeFailedError(ExitCodeException):
 
 
 class MissingDependencyError(ExitCodeException):
+    """A third-party dependency is missing."""
+
     exit_code = ExitCode.missing_dependency
 
 
 class UnsupportedImageFormatError(ExitCodeException):
+    """The image format is not supported."""
+
     exit_code = ExitCode.input_file
 
 
 class DpiError(ExitCodeException):
+    """Missing information about input image DPI."""
+
     exit_code = ExitCode.input_file
 
 
 class OutputFileAccessError(ExitCodeException):
+    """Cannot access the intended output file path."""
+
     exit_code = ExitCode.file_access_error
 
 
 class PriorOcrFoundError(ExitCodeException):
+    """This file already has OCR."""
+
     exit_code = ExitCode.already_done_ocr
 
 
 class InputFileError(ExitCodeException):
+    """Something is wrong with the input file."""
+
     exit_code = ExitCode.input_file
 
 
 class SubprocessOutputError(ExitCodeException):
+    """A subprocess returned an unexpected error."""
+
     exit_code = ExitCode.child_process_error
 
 
 class EncryptedPdfError(ExitCodeException):
+    """Input PDF is encrypted."""
+
     exit_code = ExitCode.encrypted_pdf
     message = dedent(
         '''\
@@ -100,5 +129,7 @@ class EncryptedPdfError(ExitCodeException):
 
 
 class TesseractConfigError(ExitCodeException):
+    """Tesseract config can't be parsed."""
+
     exit_code = ExitCode.invalid_config
     message = "Error occurred while parsing a Tesseract configuration file"

src/ocrmypdf/extra_plugins/semfree.py

@@ -0,0 +1,201 @@
+# © 2021 James R. Barlow: github.com/jbarlow83
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+
+
+"""Semaphore-free alternate executor.
+
+There are two popular environments that do not fully support the standard Python
+multiprocessing module: AWS Lambda, and Termux (a terminal emulator for Android).
+
+This alternate executor divvies up work among worker processes before processing,
+rather than having each worker consume work from a shared queue when they finish
+their task. This means workers have no need to coordinate with each other. Each
+worker communicates only with the main process.
+
+This is not without drawbacks. If the tasks are not "even" in size, which cannot
+be guaranteed, some workers may end up with too much work while others are idle.
+It is less efficient than the standard implementation, so not th edefault.
+"""
+
+import logging
+import logging.handlers
+import signal
+from contextlib import suppress
+from enum import Enum, auto
+from itertools import islice, repeat, takewhile, zip_longest
+from multiprocessing import Pipe, Process
+from multiprocessing.connection import Connection, wait
+from typing import Callable, Iterable, Iterator, List
+
+from ocrmypdf import Executor, hookimpl
+from ocrmypdf._concurrent import NullProgressBar
+from ocrmypdf.exceptions import InputFileError
+from ocrmypdf.helpers import remove_all_log_handlers
+
+
+class MessageType(Enum):
+    """Implement basic IPC messaging."""
+
+    exception = auto()  # pylint: disable=invalid-name
+    result = auto()  # pylint: disable=invalid-name
+    complete = auto()  # pylint: disable=invalid-name
+
+
+def split_every(n: int, iterable: Iterable) -> Iterator:
+    """Split iterable into groups of n.
+
+    >>> list(split_every(4, range(10)))
+    [[0, 1, 2], [3, 4, 5], [6, 7, 8], [9]]
+
+    https://stackoverflow.com/a/22919323
+    """
+    iterator = iter(iterable)
+    return takewhile(bool, (list(islice(iterator, n)) for _ in repeat(None)))
+
+
+def process_sigbus(*args):
+    raise InputFileError("A worker process lost access to an input file")
+
+
+class ConnectionLogHandler(logging.handlers.QueueHandler):
+    """Handler used by child processes to forward log messages to parent."""
+
+    def __init__(self, conn: Connection) -> None:
+        # sets the parent's queue to None - parent only touches queue
+        # in enqueue() which we override
+        super().__init__(None)  # type: ignore
+        self.conn = conn
+
+    def enqueue(self, record):
+        self.conn.send(('log', record))
+
+
+def process_loop(
+    conn: Connection, user_init: Callable[[], None], loglevel, task, task_args
+):
+    """Initialize a process pool worker"""
+
+    # Install SIGBUS handler (so our parent process can abort somewhat gracefully)
+    with suppress(AttributeError):  # Windows and Cygwin do not have SIGBUS
+        # Windows and Cygwin do not have pthread_sigmask or SIGBUS
+        signal.signal(signal.SIGBUS, process_sigbus)
+
+    # Reconfigure the root logger for this process to send all messages to a queue
+    h = ConnectionLogHandler(conn)
+    root = logging.getLogger()
+    remove_all_log_handlers(root)
+    root.setLevel(loglevel)
+    root.addHandler(h)
+
+    user_init()
+
+    for args in task_args:
+        try:
+            result = task(args)
+        except Exception as e:  # pylint: disable=broad-except
+            conn.send((MessageType.exception, e))
+            break
+        else:
+            conn.send((MessageType.result, result))
+
+    conn.send((MessageType.complete, None))
+    conn.close()
+    return
+
+
+class LambdaExecutor(Executor):
+    """Executor for AWS Lambda or similar environments that lack semaphores."""
+
+    def _execute(
+        self,
+        *,
+        use_threads: bool,
+        max_workers: int,
+        tqdm_kwargs: dict,
+        worker_initializer: Callable,
+        task: Callable,
+        task_arguments: Iterable,
+        task_finished: Callable,
+    ):
+        if use_threads and max_workers == 1:
+            with self.pbar_class(**tqdm_kwargs) as pbar:
+                for args in task_arguments:
+                    result = task(args)
+                    task_finished(result, pbar)
+            return
+
+        task_arguments = list(task_arguments)
+        grouped_args = list(
+            zip_longest(*list(split_every(max_workers, task_arguments)))
+        )
+        if not grouped_args:
+            return
+
+        processes: List[Process] = []
+        connections: List[Connection] = []
+        for chunk in grouped_args:
+            parent_conn, child_conn = Pipe()
+
+            worker_args = [args for args in chunk if args is not None]
+            process = Process(
+                target=process_loop,
+                args=(
+                    child_conn,
+                    worker_initializer,
+                    logging.getLogger("").level,
+                    task,
+                    worker_args,
+                ),
+            )
+            process.daemon = True
+            processes.append(process)
+            connections.append(parent_conn)
+
+        for process in processes:
+            process.start()
+
+        with self.pbar_class(**tqdm_kwargs) as pbar:
+            while connections:
+                for result in wait(connections):
+                    if not isinstance(result, Connection):
+                        raise NotImplementedError("We only support Connection()")
+                    try:
+                        msg_type, msg = result.recv()
+                    except EOFError:
+                        connections.remove(result)
+                        continue
+
+                    if msg_type == MessageType.result:
+                        if task_finished:
+                            task_finished(msg, pbar)
+                    elif msg_type == 'log':
+                        record = msg
+                        logger = logging.getLogger(record.name)
+                        logger.handle(record)
+                    elif msg_type == MessageType.complete:
+                        connections.remove(result)
+                    elif msg_type == MessageType.exception:
+                        for process in processes:
+                            process.terminate()
+                        raise msg
+
+        for process in processes:
+            process.join()
+
+
+@hookimpl
+def get_executor(progressbar_class):
+    return LambdaExecutor(pbar_class=progressbar_class)
+
+
+@hookimpl
+def get_logging_console():
+    return logging.StreamHandler()
+
+
+@hookimpl
+def get_progressbar_class():
+    return NullProgressBar

src/ocrmypdf/helpers.py

@@ -4,6 +4,7 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""Support functions."""
 
 import logging
 import multiprocessing
@@ -15,29 +16,56 @@ from collections.abc import Iterable
 from contextlib import suppress
 from functools import wraps
 from io import StringIO
-from math import isclose
+from math import isclose, isfinite
 from pathlib import Path
 from typing import Any, Sequence
 
+import img2pdf
 import pikepdf
+from packaging.version import Version
 
 log = logging.getLogger(__name__)
 
+if Version(img2pdf.__version__) < Version('0.4.0'):
+    IMG2PDF_KWARGS = dict(without_pdfw=True)
+elif Version(img2pdf.__version__) < Version('0.4.3'):
+    IMG2PDF_KWARGS = dict(engine=img2pdf.Engine.pikepdf)
+else:
+    IMG2PDF_KWARGS = dict(
+        engine=img2pdf.Engine.pikepdf, rotation=img2pdf.Rotation.ifvalid
+    )
+
 
 class Resolution(namedtuple('Resolution', ('x', 'y'))):
-    """The number of pixels per inch in each 2D direction."""
+    """The number of pixels per inch in each 2D direction.
+
+    Resolution objects are considered "equal" for == purposes if they are
+    equal to a reasonable tolerance.
+    """
 
     __slots__ = ()
 
+    # rel_tol after converting from dpi to pixels per meter and saving
+    # as integer with rounding, as many file formats
+    CONVERSION_ERROR = 0.002
+
     def round(self, ndigits: int):
         return Resolution(round(self.x, ndigits), round(self.y, ndigits))
 
     def to_int(self):
         return Resolution(int(round(self.x)), int(round(self.y)))
 
+    @classmethod
+    def _isclose(cls, a, b):
+        return isclose(a, b, rel_tol=cls.CONVERSION_ERROR)
+
     @property
     def is_square(self) -> bool:
-        return isclose(self.x, self.y, rel_tol=1e-3)
+        return self._isclose(self.x, self.y)
+
+    @property
+    def is_finite(self) -> bool:
+        return isfinite(self.x) and isfinite(self.y)
 
     def take_max(self, vals, yvals=None):
         if yvals is not None:
@@ -54,9 +82,16 @@ class Resolution(namedtuple('Resolution', ('x', 'y'))):
     def __str__(self):
         return f"{self.x:f}x{self.y:f}"
 
-    def __repr__(self):
+    def __repr__(self):  # pragma: no cover
         return f"Resolution({self.x}x{self.y} dpi)"
 
+    def __eq__(self, other):
+        if isinstance(other, tuple) and len(other) == 2:
+            other = Resolution(*other)
+        if not isinstance(other, Resolution):
+            return NotImplemented
+        return self._isclose(self.x, other.x) and self._isclose(self.y, other.y)
+
 
 class NeverRaise(Exception):
     """An exception that is never raised"""
@@ -87,10 +122,7 @@ def safe_symlink(input_file: os.PathLike, soft_link_name: os.PathLike):
         # do not delete or overwrite real (non-soft link) file
         if not os.path.islink(soft_link_name):
             raise FileExistsError(f"{soft_link_name} exists and is not a link")
-        try:
-            os.unlink(soft_link_name)
-        except OSError:
-            log.debug("Can't unlink %s", soft_link_name)
+        os.unlink(soft_link_name)
 
     if not os.path.exists(input_file):
         raise FileNotFoundError(f"trying to create a broken symlink to {input_file}")
@@ -106,11 +138,11 @@ def safe_symlink(input_file: os.PathLike, soft_link_name: os.PathLike):
     os.symlink(os.path.abspath(input_file), soft_link_name)
 
 
-def samefile(f1: os.PathLike, f2: os.PathLike):
+def samefile(file1: os.PathLike, file2: os.PathLike):
     if os.name == 'nt':
-        return f1 == f2
+        return file1 == file2
     else:
-        return os.path.samefile(f1, f2)
+        return os.path.samefile(file1, file2)
 
 
 def is_iterable_notstr(thing: Any) -> bool:
@@ -118,9 +150,9 @@ def is_iterable_notstr(thing: Any) -> bool:
     return isinstance(thing, Iterable) and not isinstance(thing, str)
 
 
-def monotonic(L: Sequence) -> bool:
+def monotonic(seq: Sequence) -> bool:
     """Does this sequence increase monotonically?"""
-    return all(b > a for a, b in zip(L, L[1:]))
+    return all(b > a for a, b in zip(seq, seq[1:]))
 
 
 def page_number(input_file: os.PathLike) -> int:
@@ -135,7 +167,7 @@ def available_cpu_count() -> int:
     except NotImplementedError:
         pass
     warnings.warn(
-        "Could not get CPU count.  Assuming one (1) CPU." "Use -j N to set manually."
+        "Could not get CPU count. Assuming one (1) CPU. Use -j N to set manually."
     )
     return 1
 
@@ -159,17 +191,17 @@ def is_file_writable(test_file: os.PathLike) -> bool:
                 os.W_OK,
                 effective_ids=(os.access in os.supports_effective_ids),
             )
+
+        try:
+            fp = p.open('wb')
+        except OSError:
+            return False
         else:
-            try:
-                fp = p.open('wb')
-            except OSError:
-                return False
-            else:
-                fp.close()
-                with suppress(OSError):
-                    p.unlink()
-            return True
-    except (EnvironmentError, RuntimeError) as e:
+            fp.close()
+            with suppress(OSError):
+                p.unlink()
+        return True
+    except (OSError, RuntimeError) as e:
         log.debug(e)
         log.error(str(e))
         return False
@@ -181,45 +213,44 @@ def check_pdf(input_file: Path) -> bool:
     Checks for proper formatting and proper linearization. Uses pikepdf (which in
     turn, uses QPDF) to perform the checks.
     """
-    pdf = None
     try:
         pdf = pikepdf.open(input_file)
     except pikepdf.PdfError as e:
         log.error(e)
         return False
     else:
-        messages = pdf.check()
-        for msg in messages:
-            if 'error' in msg.lower():
-                log.error(msg)
+        with pdf:
+            messages = pdf.check()
+            success = True
+            for msg in messages:
+                if 'error' in msg.lower():
+                    log.error(msg)
+                    success = False
+                elif (
+                    "/DecodeParms: operation for dictionary attempted on object "
+                    "of type null" in msg
+                ):
+                    pass  # Ignore/spurious warning
+                else:
+                    log.warning(msg)
+                    success = False
+
+            sio = StringIO()
+            linearize_msgs = ''
+            try:
+                # If linearization is missing entirely, we do not complain. We do
+                # complain if linearization is present but incorrect.
+                pdf.check_linearization(sio)
+            except (RuntimeError, pikepdf.ForeignObjectError):
+                pass
             else:
-                log.warning(msg)
+                linearize_msgs = sio.getvalue()
+                if linearize_msgs:
+                    log.warning(linearize_msgs)
 
-        sio = StringIO()
-        linearize_msgs = ''
-        try:
-            # If linearization is missing entirely, we do not complain. We do
-            # complain if linearization is present but incorrect.
-            pdf.check_linearization(sio)
-        except RuntimeError:
-            pass
-        except (
-            getattr(pikepdf, 'ForeignObjectError')
-            if pikepdf.__version__ == '2.1.0'  # This version may throw wrong exception
-            else NeverRaise
-        ):
-            pass
-        else:
-            linearize_msgs = sio.getvalue()
-            if linearize_msgs:
-                log.warning(linearize_msgs)
-
-        if not messages and not linearize_msgs:
-            return True
-        return False
-    finally:
-        if pdf:
-            pdf.close()
+            if success and not linearize_msgs:
+                return True
+            return False
 
 
 def clamp(n, smallest, largest):  # mypy doesn't understand types for this
@@ -227,6 +258,13 @@ def clamp(n, smallest, largest):  # mypy doesn't understand types for this
     return max(smallest, min(n, largest))
 
 
+def remove_all_log_handlers(logger):
+    "Remove all log handlers, usually used in a child process."
+    for handler in logger.handlers[:]:
+        logger.removeHandler(handler)
+        handler.close()  # To ensure handlers with opened resources are released
+
+
 def pikepdf_enable_mmap():
     # try:
     #     if pikepdf._qpdf.set_access_default_mmap(True):
@@ -236,7 +274,7 @@ def pikepdf_enable_mmap():
     # We found a race condition probably related to pybind issue #2252 that can
     # cause a crash. For now, disable pikepdf mmap to be on the safe side.
     # Fix is not in pybind11 2.6.0
-    log.debug("pikepdf mmap disabled")
+    # log.debug("pikepdf mmap disabled")
     return
 
 
@@ -247,7 +285,7 @@ def deprecated(func):
     def new_func(*args, **kwargs):
         warnings.simplefilter('always', DeprecationWarning)  # turn off filter
         warnings.warn(
-            "Call to deprecated function {}.".format(func.__name__),
+            f"Call to deprecated function {func.__name__}.",
             category=DeprecationWarning,
             stacklevel=2,
         )

src/ocrmypdf/hocrtransform.py

@@ -28,10 +28,11 @@
 # TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
 # SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 
+"""Transform .hocr and page image to text PDF."""
+
 import argparse
 import os
 import re
-from itertools import chain
 from math import atan, cos, sin
 from pathlib import Path
 from typing import Any, NamedTuple, Optional, Tuple, Union
@@ -41,10 +42,66 @@ from reportlab.lib.colors import black, cyan, magenta, red
 from reportlab.lib.units import inch
 from reportlab.pdfgen.canvas import Canvas
 
+# According to Wikipedia these languages are supported in the ISO-8859-1 character
+# set, meaning reportlab can generate them and they are compatible with hocr,
+# assuming Tesseract has the necessary languages installed. Note that there may
+# not be language packs for them.
+HOCR_OK_LANGS = frozenset(
+    [
+        # Languages fully covered by Latin-1:
+        'afr',  # Afrikaans
+        'alb',  # Albanian
+        'ast',  # Leonese
+        'baq',  # Basque
+        'bre',  # Breton
+        'cos',  # Corsican
+        'eng',  # English
+        'eus',  # Basque
+        'fao',  # Faoese
+        'gla',  # Scottish Gaelic
+        'glg',  # Galician
+        'glv',  # Manx
+        'ice',  # Icelandic
+        'ind',  # Indonesian
+        'isl',  # Icelandic
+        'ita',  # Italian
+        'ltz',  # Luxembourgish
+        'mal',  # Malay Rumi
+        'mga',  # Irish
+        'nor',  # Norwegian
+        'oci',  # Occitan
+        'por',  # Portugeuse
+        'roh',  # Romansh
+        'sco',  # Scots
+        'sma',  # Sami
+        'spa',  # Spanish
+        'sqi',  # Albanian
+        'swa',  # Swahili
+        'swe',  # Swedish
+        'tgl',  # Tagalog
+        'wln',  # Walloon
+        # Languages supported by Latin-1 except for a few rare characters that OCR
+        # is probably not trained to recognize anyway:
+        'cat',  # Catalan
+        'cym',  # Welsh
+        'dan',  # Danish
+        'deu',  # German
+        'dut',  # Dutch
+        'est',  # Estonian
+        'fin',  # Finnish
+        'fra',  # French
+        'hun',  # Hungarian
+        'kur',  # Kurdish
+        'nld',  # Dutch
+        'wel',  # Welsh
+    ]
+)
+
+
 Element = ElementTree.Element
 
 
-class Rect(NamedTuple):  # pylint: disable=inherit-non-class
+class Rect(NamedTuple):
     """A rectangle for managing PDF coordinates."""
 
     x1: Any
@@ -54,7 +111,7 @@ class Rect(NamedTuple):  # pylint: disable=inherit-non-class
 
 
 class HocrTransformError(Exception):
-    pass
+    """Error while applying hOCR transform."""
 
 
 class HocrTransform:
@@ -77,7 +134,7 @@ class HocrTransform:
         {'ff': 'ff', 'ffi': 'f‌f‌i', 'ffl': 'f‌f‌l', 'fi': 'fi', 'fl': 'fl'}
     )
 
-    def __init__(self, hocr_filename: Union[str, Path], dpi: float):
+    def __init__(self, *, hocr_filename: Union[str, Path], dpi: float):
         self.dpi = dpi
         self.hocr = ElementTree.parse(os.fspath(hocr_filename))
 
@@ -182,6 +239,7 @@ class HocrTransform:
 
     def to_pdf(
         self,
+        *,
         out_filename: Path,
         image_filename: Optional[Path] = None,
         show_bounding_boxes: bool = False,
@@ -231,7 +289,7 @@ class HocrTransform:
                 continue
 
             pxl_coords = self.element_coordinates(elem)
-            pt = self.pt_from_pixel(pxl_coords)
+            pt = self.pt_from_pixel(pxl_coords)  # pylint: disable=invalid-name
 
             # draw the bbox border
             if show_bounding_boxes:  # pragma: no cover
@@ -240,13 +298,11 @@ class HocrTransform:
                 )
 
         found_lines = False
-        for line in sorted(
-            chain(
-                self.hocr.iterfind(self._child_xpath('span', 'ocr_header')),
-                self.hocr.iterfind(self._child_xpath('span', 'ocr_line')),
-                self.hocr.iterfind(self._child_xpath('span', 'ocr_textfloat')),
-            ),
-            key=self.topdown_position,
+        for line in (
+            element
+            for element in self.hocr.iterfind(self._child_xpath('span'))
+            if 'class' in element.attrib
+            and element.attrib['class'] in {'ocr_header', 'ocr_line', 'ocr_textfloat'}
         ):
             found_lines = True
             self._do_line(
@@ -295,7 +351,7 @@ class HocrTransform:
         interword_spaces: bool,
         show_bounding_boxes: bool,
     ):
-        if not line:
+        if line is None:
             return
         pxl_line_coords = self.element_coordinates(line)
         line_box = self.pt_from_pixel(pxl_line_coords)
@@ -433,10 +489,10 @@ if __name__ == "__main__":
     parser.add_argument('outputfile', help='Path to the PDF file to be generated')
     args = parser.parse_args()
 
-    hocr = HocrTransform(args.hocrfile, args.resolution)
+    hocr = HocrTransform(hocr_filename=args.hocrfile, dpi=args.resolution)
     hocr.to_pdf(
-        args.outputfile,
-        args.image,
-        args.boundingboxes,
+        out_filename=args.outputfile,
+        image_filename=args.image,
+        show_bounding_boxes=args.boundingboxes,
         interword_spaces=args.interword_spaces,
     )

src/ocrmypdf/lib/compile_leptonica.py

@@ -1,516 +0,0 @@
-#!/usr/bin/env python3
-# © 2017 James R. Barlow: github.com/jbarlow83
-#
-# This Source Code Form is subject to the terms of the Mozilla Public
-# License, v. 2.0. If a copy of the MPL was not distributed with this
-# file, You can obtain one at http://mozilla.org/MPL/2.0/.
-
-
-from pathlib import Path
-
-from cffi import FFI
-
-ffibuilder = FFI()
-ffibuilder.cdef(
-    """
-typedef signed char             l_int8;
-typedef unsigned char           l_uint8;
-typedef short                   l_int16;
-typedef unsigned short          l_uint16;
-typedef int                     l_int32;
-typedef unsigned int            l_uint32;
-typedef float                   l_float32;
-typedef double                  l_float64;
-typedef long long               l_int64;
-typedef unsigned long long      l_uint64;
-
-typedef int l_ok; /*!< return type 0 if OK, 1 on error */
-
-struct Pix
-{
-    l_uint32             w;           /* width in pixels                   */
-    l_uint32             h;           /* height in pixels                  */
-    l_uint32             d;           /* depth in bits (bpp)               */
-    l_uint32             spp;         /* number of samples per pixel       */
-    l_uint32             wpl;         /* 32-bit words/line                 */
-    l_uint32             refcount;    /* reference count (1 if no clones)  */
-    l_int32              xres;        /* image res (ppi) in x direction    */
-                                      /* (use 0 if unknown)                */
-    l_int32              yres;        /* image res (ppi) in y direction    */
-                                      /* (use 0 if unknown)                */
-    l_int32              informat;    /* input file format, IFF_*          */
-    l_int32              special;     /* special instructions for I/O, etc */
-    char                *text;        /* text string associated with pix   */
-    struct PixColormap  *colormap;    /* colormap (may be null)            */
-    l_uint32            *data;        /* the image data                    */
-};
-typedef struct Pix PIX;
-
-struct PixColormap
-{
-    void            *array;     /* colormap table (array of RGBA_QUAD)     */
-    l_int32          depth;     /* of pix (1, 2, 4 or 8 bpp)               */
-    l_int32          nalloc;    /* number of color entries allocated       */
-    l_int32          n;         /* number of color entries used            */
-};
-typedef struct PixColormap  PIXCMAP;
-
-/*! Array of pix */
-struct Pixa
-{
-    l_int32             n;          /*!< number of Pix in ptr array        */
-    l_int32             nalloc;     /*!< number of Pix ptrs allocated      */
-    l_uint32            refcount;   /*!< reference count (1 if no clones)  */
-    struct Pix        **pix;        /*!< the array of ptrs to pix          */
-    struct Boxa        *boxa;       /*!< array of boxes                    */
-};
-typedef struct Pixa PIXA;
-
-/*! Array of compressed pix */
-struct PixaComp
-{
-    l_int32              n;         /*!< number of PixComp in ptr array    */
-    l_int32              nalloc;    /*!< number of PixComp ptrs allocated  */
-    l_int32              offset;    /*!< indexing offset into ptr array    */
-    struct PixComp     **pixc;      /*!< the array of ptrs to PixComp      */
-    struct Boxa         *boxa;      /*!< array of boxes                    */
-};
-typedef struct PixaComp PIXAC;
-
-struct Box
-{
-    l_int32            x;
-    l_int32            y;
-    l_int32            w;
-    l_int32            h;
-    l_uint32           refcount;      /* reference count (1 if no clones)  */
-
-};
-typedef struct Box    BOX;
-
-/*! Array of Box */
-struct Boxa
-{
-    l_int32            n;           /*!< number of box in ptr array        */
-    l_int32            nalloc;      /*!< number of box ptrs allocated      */
-    l_uint32           refcount;    /*!< reference count (1 if no clones)  */
-    struct Box       **box;         /*!< box ptr array                     */
-};
-typedef struct Boxa BOXA;
-
-/*! String array: an array of C strings */
-struct Sarray
-{
-    l_int32          nalloc;    /*!< size of allocated ptr array         */
-    l_int32          n;         /*!< number of strings allocated         */
-    l_int32          refcount;  /*!< reference count (1 if no clones)    */
-    char           **array;     /*!< string array                        */
-};
-typedef struct Sarray SARRAY;
-
-/*! Pdf formatted encoding types */
-enum {
-    L_DEFAULT_ENCODE  = 0,  /*!< use default encoding based on image        */
-    L_JPEG_ENCODE     = 1,  /*!< use dct encoding: 8 and 32 bpp, no cmap    */
-    L_G4_ENCODE       = 2,  /*!< use ccitt g4 fax encoding: 1 bpp           */
-    L_FLATE_ENCODE    = 3,  /*!< use flate encoding: any depth, cmap ok     */
-    L_JP2K_ENCODE     = 4   /*!< use jp2k encoding: 8 and 32 bpp, no cmap   */
-};
-
-/*! Compressed image data */
-struct L_Compressed_Data
-{
-    l_int32            type;         /*!< encoding type: L_JPEG_ENCODE, etc   */
-    l_uint8           *datacomp;     /*!< gzipped raster data                 */
-    size_t             nbytescomp;   /*!< number of compressed bytes          */
-    char              *data85;       /*!< ascii85-encoded gzipped raster data */
-    size_t             nbytes85;     /*!< number of ascii85 encoded bytes     */
-    char              *cmapdata85;   /*!< ascii85-encoded uncompressed cmap   */
-    char              *cmapdatahex;  /*!< hex pdf array for the cmap          */
-    l_int32            ncolors;      /*!< number of colors in cmap            */
-    l_int32            w;            /*!< image width                         */
-    l_int32            h;            /*!< image height                        */
-    l_int32            bps;          /*!< bits/sample; typ. 1, 2, 4 or 8      */
-    l_int32            spp;          /*!< samples/pixel; typ. 1 or 3          */
-    l_int32            minisblack;   /*!< tiff g4 photometry                  */
-    l_int32            predictor;    /*!< flate data has PNG predictors       */
-    size_t             nbytes;       /*!< number of uncompressed raster bytes */
-    l_int32            res;          /*!< resolution (ppi)                    */
-};
-typedef struct L_Compressed_Data L_COMP_DATA;
-
-/*! Selection */
-struct Sel
-{
-    l_int32       sy;        /*!< sel height                               */
-    l_int32       sx;        /*!< sel width                                */
-    l_int32       cy;        /*!< y location of sel origin                 */
-    l_int32       cx;        /*!< x location of sel origin                 */
-    l_int32     **data;      /*!< {0,1,2}; data[i][j] in [row][col] order  */
-    char         *name;      /*!< used to find sel by name                 */
-};
-typedef struct Sel SEL;
-
-enum {
-    REMOVE_CMAP_TO_BINARY = 0,     /*!< remove colormap for conv to 1 bpp  */
-    REMOVE_CMAP_TO_GRAYSCALE = 1,  /*!< remove colormap for conv to 8 bpp  */
-    REMOVE_CMAP_TO_FULL_COLOR = 2, /*!< remove colormap for conv to 32 bpp */
-    REMOVE_CMAP_WITH_ALPHA = 3,    /*!< remove colormap and alpha          */
-    REMOVE_CMAP_BASED_ON_SRC = 4   /*!< remove depending on src format     */
-};
-
-/*! Access and storage flags */
-enum {
-    L_NOCOPY = 0,     /*!< do not copy the object; do not delete the ptr  */
-    L_INSERT = L_NOCOPY,    /*!< stuff it in; do not copy or clone        */
-    L_COPY = 1,       /*!< make/use a copy of the object                  */
-    L_CLONE = 2,      /*!< make/use clone (ref count) of the object       */
-    L_COPY_CLONE = 3  /*!< make a new array object (e.g., pixa) and fill  */
-                      /*!< the array with clones (e.g., pix)              */
-};
-
-/*! Flags for method of extracting barcode widths */
-enum {
-    L_USE_WIDTHS = 1,     /*!< use histogram of barcode widths           */
-    L_USE_WINDOWS = 2     /*!< find best window for decoding transitions */
-};
-
-/*! Flags for barcode formats */
-enum {
-    L_BF_UNKNOWN = 0,     /*!< unknown format                            */
-    L_BF_ANY = 1,         /*!< try decoding with all known formats       */
-    L_BF_CODE128 = 2,     /*!< decode with Code128 format                */
-    L_BF_EAN8 = 3,        /*!< decode with EAN8 format                   */
-    L_BF_EAN13 = 4,       /*!< decode with EAN13 format                  */
-    L_BF_CODE2OF5 = 5,    /*!< decode with Code 2 of 5 format            */
-    L_BF_CODEI2OF5 = 6,   /*!< decode with Interleaved 2 of 5 format     */
-    L_BF_CODE39 = 7,      /*!< decode with Code39 format                 */
-    L_BF_CODE93 = 8,      /*!< decode with Code93 format                 */
-    L_BF_CODABAR = 9,     /*!< decode with Code93 format                 */
-    L_BF_UPCA = 10        /*!< decode with UPC A format                  */
-};
-
-enum {
-    L_SEVERITY_EXTERNAL = 0,   /* Get the severity from the environment   */
-    L_SEVERITY_ALL      = 1,   /* Lowest severity: print all messages     */
-    L_SEVERITY_DEBUG    = 2,   /* Print debugging and higher messages     */
-    L_SEVERITY_INFO     = 3,   /* Print informational and higher messages */
-    L_SEVERITY_WARNING  = 4,   /* Print warning and higher messages       */
-    L_SEVERITY_ERROR    = 5,   /* Print error and higher messages         */
-    L_SEVERITY_NONE     = 6    /* Highest severity: print no messages     */
-};
-
-enum {
-    SEL_DONT_CARE  = 0,
-    SEL_HIT        = 1,
-    SEL_MISS       = 2
-};
-
-"""
-)
-
-ffibuilder.cdef(
-    """
-PIX * pixRead ( const char *filename );
-PIX * pixReadMem ( const l_uint8 *data, size_t size );
-PIX * pixReadStream ( FILE *fp, l_int32 hint );
-PIX * pixScale ( PIX *pixs, l_float32 scalex, l_float32 scaley );
-l_int32 pixFindSkew ( PIX *pixs, l_float32 *pangle, l_float32 *pconf );
-l_int32 pixWriteImpliedFormat ( const char *filename, PIX *pix, l_int32 quality, l_int32 progressive );
-l_int32 getImpliedFileFormat ( const char *filename );
-l_ok pixWriteStream ( FILE *fp, PIX *pix, l_int32 format );
-l_ok pixWriteStreamJpeg ( FILE *fp, PIX *pixs, l_int32 quality, l_int32 progressive );
-l_ok pixWriteMem ( l_uint8 **pdata, size_t *psize, PIX *pix, l_int32 format );
-l_ok pixWriteMemJpeg ( l_uint8 **pdata, size_t *psize, PIX *pix, l_int32 quality, l_int32 progressive );
-l_int32
-pixWriteMemPng(l_uint8  **pdata,
-               size_t    *psize,
-               PIX       *pix,
-               l_float32  gamma);
-
-void pixDestroy ( PIX **ppix );
-
-l_ok
-pixEqual(PIX      *pix1,
-         PIX      *pix2,
-         l_int32 *psame);
-
-PIX *
-pixEndianByteSwapNew(PIX  *pixs);
-
-PIX * pixDeskew ( PIX *pixs, l_int32 redsearch );
-char * getLeptonicaVersion (  );
-l_int32 pixCorrelationBinary(PIX *pix1, PIX *pix2, l_float32 *pval);
-PIX *pixRotate180(PIX *pixd, PIX *pixs);
-PIX *
-pixRotateOrth(PIX     *pixs,
-              l_int32  quads);
-
-l_int32 pixCountPixels ( PIX *pix, l_int32 *pcount, l_int32 *tab8 );
-PIX * pixAnd ( PIX *pixd, PIX *pixs1, PIX *pixs2 );
-l_int32 * makePixelSumTab8 ( void );
-
-PIX * pixDeserializeFromMemory ( const l_uint32 *data, size_t nbytes );
-l_int32 pixSerializeToMemory ( PIX *pixs, l_uint32 **pdata, size_t *pnbytes );
-
-PIX * pixConvertRGBToLuminance(PIX *pixs);
-
-PIX * pixConvertTo8(PIX     *pixs, l_int32  cmapflag);
-
-PIX * pixRemoveColormap(PIX *pixs, l_int32  type);
-
-l_int32
-pixOtsuAdaptiveThreshold(PIX       *pixs,
-                         l_int32    sx,
-                         l_int32    sy,
-                         l_int32    smoothx,
-                         l_int32    smoothy,
-                         l_float32  scorefract,
-                         PIX      **ppixth,
-                         PIX      **ppixd);
-
-PIX *
-pixOtsuThreshOnBackgroundNorm(PIX       *pixs,
-                              PIX       *pixim,
-                              l_int32    sx,
-                              l_int32    sy,
-                              l_int32    thresh,
-                              l_int32    mincount,
-                              l_int32    bgval,
-                              l_int32    smoothx,
-                              l_int32    smoothy,
-                              l_float32  scorefract,
-                              l_int32   *pthresh);
-
-PIX *
-pixMaskedThreshOnBackgroundNorm(PIX       *pixs,
-                                PIX       *pixim,
-                                l_int32    sx,
-                                l_int32    sy,
-                                l_int32    thresh,
-                                l_int32    mincount,
-                                l_int32    smoothx,
-                                l_int32    smoothy,
-                                l_float32  scorefract,
-                                l_int32   *pthresh);
-
-PIX *
-pixCleanBackgroundToWhite(PIX       *pixs,
-                          PIX       *pixim,
-                          PIX       *pixg,
-                          l_float32  gamma,
-                          l_int32    blackval,
-                          l_int32    whiteval);
-
-BOX *
-pixFindPageForeground ( PIX *pixs,
-                        l_int32 threshold,
-                        l_int32 mindist,
-                        l_int32 erasedist,
-                        l_int32 showmorph,
-                        PIXAC *pixac );
-
-PIX *
-pixClipRectangle(PIX   *pixs,
-                 BOX   *box,
-                 BOX  **pboxc);
-
-PIX *
-pixBackgroundNorm(PIX     *pixs,
-                  PIX     *pixim,
-                  PIX     *pixg,
-                  l_int32  sx,
-                  l_int32  sy,
-                  l_int32  thresh,
-                  l_int32  mincount,
-                  l_int32  bgval,
-                  l_int32  smoothx,
-                  l_int32  smoothy);
-
-PIX *
-pixGammaTRC(PIX       *pixd,
-            PIX       *pixs,
-            l_float32  gamma,
-            l_int32    minval,
-            l_int32    maxval);
-
-
-l_int32
-pixNumSignificantGrayColors(PIX       *pixs,
-                            l_int32    darkthresh,
-                            l_int32    lightthresh,
-                            l_float32  minfract,
-                            l_int32    factor,
-                            l_int32   *pncolors);
-
-l_int32
-pixColorFraction(PIX        *pixs,
-                 l_int32     darkthresh,
-                 l_int32     lightthresh,
-                 l_int32     diffthresh,
-                 l_int32     factor,
-                 l_float32  *ppixfract,
-                 l_float32  *pcolorfract);
-
-PIX *
-pixColorMagnitude(PIX     *pixs,
-                  l_int32  rwhite,
-                  l_int32  gwhite,
-                  l_int32  bwhite,
-                  l_int32  type);
-
-PIX *
-pixMaskOverColorPixels(PIX     *pixs,
-                       l_int32  threshdiff,
-                       l_int32  mindist);
-
-l_int32
-pixGetAverageMaskedRGB(PIX        *pixs,
-                       PIX        *pixm,
-                       l_int32     x,
-                       l_int32     y,
-                       l_int32     factor,
-                       l_int32     type,
-                       l_float32  *prval,
-                       l_float32  *pgval,
-                       l_float32  *pbval);
-
-PIX *
-pixGlobalNormRGB(PIX * 	pixd,
-                 PIX * 	pixs,
-                 l_int32 	rval,
-                 l_int32 	gval,
-                 l_int32 	bval,
-                 l_int32 	mapval);
-
-PIX *
-pixInvert(PIX * pixd,
-          PIX * pixs);
-
-PIX *
-pixRemoveColormapGeneral(PIX     *pixs,
-                         l_int32  type,
-                         l_int32  ifnocmap);
-
-l_int32
-pixGenerateCIData(PIX           *pixs,
-                  l_int32        type,
-                  l_int32        quality,
-                  l_int32        ascii85,
-                  L_COMP_DATA **pcid);
-
-SARRAY *
-pixProcessBarcodes(PIX      *pixs,
-                   l_int32   format,
-                   l_int32   method,
-                   SARRAY  **psaw,
-                   l_int32 debugflag);
-
-PIX *
-pixaGetPix(PIXA    *pixa,
-           l_int32  index,
-           l_int32 accesstype);
-
-BOX*
-pixaGetBox 	(PIXA *  	pixa,
-		    l_int32  	index,
-		    l_int32  	accesstype );
-
-PIXA *
-pixExtractBarcodes(PIX     *pixs,
-                   l_int32 debugflag);
-
-BOXA *
-pixLocateBarcodes ( PIX *pixs,
- l_int32 thresh,
- PIX **ppixb,
- PIX **ppixm );
-
-SARRAY *
-pixReadBarcodes(PIXA     *pixa,
-                l_int32   format,
-                l_int32   method,
-                SARRAY  **psaw,
-                l_int32 debugflag);
-
-PIX *
-pixGenHalftoneMask(PIX      *pixs,
-                        PIX     **ppixtext,
-                        l_int32  *phtfound,
-                        PIXA     *pixadb);
-
-l_int32
-l_generateCIDataForPdf(const char *fname,
-                       PIX *pix,
-                       l_int32 quality,
-                       L_COMP_DATA **pcid);
-
-
-BOX *
-boxClone ( BOX *box );
-
-BOX *
-boxaGetBox ( BOXA *boxa, l_int32 index, l_int32 accessflag );
-
-SEL *
-selCreateFromString ( const char *text, l_int32 h, l_int32 w, const char *name );
-
-SEL *
-selCreateBrick ( l_int32 h, l_int32 w, l_int32 cy, l_int32 cx, l_int32 type );
-
-char *
-selPrintToString(SEL  *sel);
-
-PIX *
-pixDilate ( PIX *pixd, PIX *pixs, SEL *sel );
-
-PIX *
-pixErode ( PIX *pixd, PIX *pixs, SEL *sel );
-
-PIX *
-pixHMT ( PIX *pixd, PIX *pixs, SEL *sel );
-
-PIX *
-pixSubtract ( PIX *pixd, PIX *pixs1, PIX *pixs2 );
-
-void
-boxDestroy(BOX  **pbox);
-
-void
-boxaDestroy ( BOXA **pboxa );
-
-void
-pixaDestroy(PIXA **ppixa);
-
-l_ok
-pixRenderBoxa ( PIX *pix, BOXA *boxa, l_int32 width, l_int32 op );
-
-void
-l_CIDataDestroy(L_COMP_DATA **pcid);
-
-void
-sarrayDestroy(SARRAY  **psa);
-
-void
-lept_free(void *ptr);
-
-void selDestroy ( SEL **psel );
-
-l_int32
-setMsgSeverity(l_int32 newsev);
-
-void
-leptSetStderrHandler(void (*handler)(const char *));
-"""
-)
-
-
-ffibuilder.set_source("ocrmypdf.lib._leptonica", None)
-
-if __name__ == '__main__':
-    ffibuilder.compile(verbose=True)
-    if Path('ocrmypdf/lib/_leptonica.py').exists() and Path('src/ocrmypdf').exists():
-        output = Path('ocrmypdf/lib/_leptonica.py')
-        output.rename('src/ocrmypdf/lib/_leptonica.py')
-        Path('ocrmypdf/lib').rmdir()
-        Path('ocrmypdf').rmdir()

src/ocrmypdf/optimize.py

@@ -4,12 +4,14 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""Post-processing image optimization of OCR PDFs."""
+
 
 import logging
 import sys
 import tempfile
+import threading
 from collections import defaultdict
-from functools import partial
 from os import fspath
 from pathlib import Path
 from typing import (
@@ -24,19 +26,27 @@ from typing import (
     Sequence,
     Tuple,
 )
+from zlib import compress
 
 import img2pdf
-import pikepdf
-from pikepdf import Dictionary, Name, Object, Pdf, PdfImage
+from pikepdf import (
+    Dictionary,
+    Name,
+    Object,
+    ObjectStreamMode,
+    Pdf,
+    PdfError,
+    PdfImage,
+    Stream,
+    UnsupportedImageTypeError,
+)
 from PIL import Image
-from tqdm import tqdm
 
-from ocrmypdf import leptonica
-from ocrmypdf._concurrent import exec_progress_pool
+from ocrmypdf._concurrent import Executor, SerialExecutor
 from ocrmypdf._exec import jbig2enc, pngquant
 from ocrmypdf._jobcontext import PdfContext
 from ocrmypdf.exceptions import OutputFileAccessError
-from ocrmypdf.helpers import deprecated, safe_symlink
+from ocrmypdf.helpers import IMG2PDF_KWARGS, safe_symlink
 
 log = logging.getLogger(__name__)
 
@@ -47,7 +57,9 @@ DEFAULT_PNG_QUALITY = 70
 Xref = NewType('Xref', int)
 
 
-class XrefExt(NamedTuple):  # pylint: disable=inherit-non-class
+class XrefExt(NamedTuple):
+    """A PDF xref and image extension pair."""
+
     xref: Xref
     ext: str
 
@@ -64,47 +76,68 @@ def jpg_name(root: Path, xref: Xref) -> Path:
     return img_name(root, xref, '.jpg')
 
 
-def tif_name(root: Path, xref: Xref) -> Path:
-    return img_name(root, xref, '.tif')
-
-
 def extract_image_filter(
-    pike: Pdf, root: Path, image: Object, xref: Xref
+    pike: Pdf, root: Path, image: Stream, xref: Xref
 ) -> Optional[Tuple[PdfImage, Tuple[Name, Object]]]:
+    del pike  # unused args
+    del root
+
     if image.Subtype != Name.Image:
         return None
     if image.Length < 100:
-        log.debug(f"Skipping small image, xref {xref}")
+        log.debug(f"xref {xref}: skipping image with small stream size")
         return None
     if image.Width < 8 or image.Height < 8:  # Issue 732
-        log.debug(f"Skipping oddly sized image, xref {xref}")
+        log.debug(f"xref {xref}: skipping image with unusually small dimensions")
         return None
 
     pim = PdfImage(image)
 
     if len(pim.filter_decodeparms) > 1:
-        log.debug(f"Skipping multiply filtered image, xref {xref}")
-        return None
-    filtdp = pim.filter_decodeparms[0]
+        first_filtdp = pim.filter_decodeparms[0]
+        second_filtdp = pim.filter_decodeparms[1]
+        if (
+            len(pim.filter_decodeparms) == 2
+            and first_filtdp[0] == Name.FlateDecode
+            and first_filtdp[1].get(Name.Predictor, 1) == 1
+            and second_filtdp[0] == Name.DCTDecode
+            and not second_filtdp[1]
+        ):
+            log.debug(
+                f"xref {xref}: found image compressed as /FlateDecode /DCTDecode, "
+                "marked for JPEG optimization"
+            )
+            filtdp = pim.filter_decodeparms[1]
+        else:
+            log.debug(f"xref {xref}: skipping image with multiple compression filters")
+            return None
+    else:
+        filtdp = pim.filter_decodeparms[0]
 
     if pim.bits_per_component > 8:
-        log.debug(f"Skipping wide gamut image, xref {xref}")
+        log.debug(f"xref {xref}: skipping wide gamut image")
         return None  # Don't mess with wide gamut images
 
     if filtdp[0] == Name.JPXDecode:
-        log.debug(f"Skipping JPEG2000 iamge, xref {xref}")
+        log.debug(f"xref {xref}: skipping JPEG2000 image")
         return None  # Don't do JPEG2000
 
+    if filtdp[0] == Name.CCITTFaxDecode and filtdp[1].get('/K', 0) >= 0:
+        log.debug(f"xref {xref}: skipping CCITT Group 3 image")
+        return None  # pikepdf doesn't support Group 3 yet
+
     if Name.Decode in image:
-        log.debug(f"Skipping image with Decode table, xref {xref}")
+        log.debug(f"xref {xref}: skipping image with Decode table")
         return None  # Don't mess with custom Decode tables
 
     return pim, filtdp
 
 
 def extract_image_jbig2(
-    *, pike: pikepdf.Pdf, root: Path, image: Object, xref: Xref, options
+    *, pike: Pdf, root: Path, image: Stream, xref: Xref, options
 ) -> Optional[XrefExt]:
+    del options  # unused arg
+
     result = extract_image_filter(pike, root, image, xref)
     if result is None:
         return None
@@ -120,16 +153,16 @@ def extract_image_jbig2(
         # Showing the palette or ICC to jbig2enc will cause it to perform
         # colorspace transform to 1bpp, which will conflict the palette or
         # ICC if it exists.
-        colorspace = pim.obj.get(pikepdf.Name.ColorSpace, None)
+        colorspace = pim.obj.get(Name.ColorSpace, None)
         if colorspace is not None or pim.image_mask:
             try:
                 # Set to DeviceGray temporarily; we already in 1 bpc.
-                pim.obj.ColorSpace = pikepdf.Name.DeviceGray
+                pim.obj.ColorSpace = Name.DeviceGray
                 imgname = root / f'{xref:08d}'
                 with imgname.open('wb') as f:
                     ext = pim.extract_to(stream=f)
                 imgname.rename(imgname.with_suffix(ext))
-            except pikepdf.UnsupportedImageTypeError:
+            except UnsupportedImageTypeError:
                 return None
             finally:
                 # Restore image colorspace after temporarily setting it to DeviceGray
@@ -142,7 +175,7 @@ def extract_image_jbig2(
 
 
 def extract_image_generic(
-    *, pike: Pdf, root: Path, image: PdfImage, xref: Xref, options
+    *, pike: Pdf, root: Path, image: Stream, xref: Xref, options
 ) -> Optional[XrefExt]:
     result = extract_image_filter(pike, root, image, xref)
     if result is None:
@@ -162,20 +195,12 @@ def extract_image_generic(
         # jpeg_quality_estimate = 117.0 * (bytes_per_pixel ** 0.213)
         # if jpeg_quality_estimate < 65:
         #     return None
-
-        # We could get the ICC profile here, but there's no need to look at it
-        # for quality transcoding
-        # if icc:
-        #     stream = BytesIO(raw_jpeg.read_raw_bytes())
-        #     iccbytes = icc.read_bytes()
-        #     with Image.open(stream) as im:
-        #         im.save(jpg_name(root, xref), icc_profile=iccbytes)
         try:
             imgname = root / f'{xref:08d}'
             with imgname.open('wb') as f:
                 ext = pim.extract_to(stream=f)
             imgname.rename(imgname.with_suffix(ext))
-        except pikepdf.UnsupportedImageTypeError:
+        except UnsupportedImageTypeError:
             return None
         return XrefExt(xref, ext)
     elif (
@@ -190,7 +215,11 @@ def extract_image_generic(
     elif not pim.indexed and pim.colorspace in pim.SIMPLE_COLORSPACES:
         # An optimization opportunity here, not currently taken, is directly
         # generating a PNG from compressed data
-        pim.as_pil_image().save(png_name(root, xref))
+        try:
+            pim.as_pil_image().save(png_name(root, xref))
+        except NotImplementedError:
+            log.warning("PDF contains an atypical image that cannot be optimized.")
+            return None
         return XrefExt(xref, '.png')
     elif (
         not pim.indexed
@@ -241,13 +270,13 @@ def extract_images(
             if image.objgen[1] != 0:
                 continue  # Ignore images in an incremental PDF
             xref = Xref(image.objgen[0])
-            if hasattr(image, 'SMask'):
+            if Name.SMask in image:
                 # Ignore soft masks
                 smask_xref = Xref(image.SMask.objgen[0])
                 exclude_xrefs.add(smask_xref)
-                log.debug(f"Skipping image {smask_xref} because it is an SMask")
+                log.debug(f"xref {smask_xref}: skipping image because it is an SMask")
             include_xrefs.add(xref)
-            log.debug(f"Treating {xref} as an optimization candidate")
+            log.debug(f"xref {xref}: treating as an optimization candidate")
             if xref not in pageno_for_xref:
                 pageno_for_xref[xref] = pageno
 
@@ -258,8 +287,10 @@ def extract_images(
             result = extract_fn(
                 pike=pike, root=root, image=image, xref=xref, options=options
             )
-        except Exception as e:  # pylint: disable=broad-except
-            log.debug("Image xref %s, error %s", xref, repr(e))
+        except Exception:  # pylint: disable=broad-except
+            log.exception(
+                f"xref {xref}: While extracting this image, an error occurred"
+            )
             errors += 1
         else:
             if result:
@@ -280,7 +311,7 @@ def extract_images_generic(
             pngs.append(xref_ext.xref)
         elif xref_ext.ext == '.jpg':
             jpegs.append(xref_ext.xref)
-    log.debug("Optimizable images: JPEGs: %s PNGs: %s", len(jpegs), len(pngs))
+    log.debug(f"Optimizable images: JPEGs: {len(jpegs)} PNGs: {len(pngs)}")
     return jpegs, pngs
 
 
@@ -292,22 +323,22 @@ def extract_images_jbig2(pike: Pdf, root: Path, options) -> Dict[int, List[XrefE
         group = pageno // options.jbig2_page_group_size
         jbig2_groups[group].append(xref_ext)
 
-    log.debug("Optimizable images: JBIG2 groups: %s", (len(jbig2_groups),))
+    log.debug(f"Optimizable images: JBIG2 groups: {len(jbig2_groups)}")
     return jbig2_groups
 
 
 def _produce_jbig2_images(
-    jbig2_groups: Dict[int, List[XrefExt]], root: Path, options
+    jbig2_groups: Dict[int, List[XrefExt]], root: Path, options, executor: Executor
 ) -> None:
     """Produce JBIG2 images from their groups"""
 
     def jbig2_group_args(root: Path, groups: Dict[int, List[XrefExt]]):
         for group, xref_exts in groups.items():
             prefix = f'group{group:08d}'
-            yield dict(
-                cwd=fspath(root),
-                infiles=(img_name(root, xref, ext) for xref, ext in xref_exts),
-                out_prefix=prefix,
+            yield (
+                fspath(root),  # =cwd
+                (img_name(root, xref, ext) for xref, ext in xref_exts),  # =infiles
+                prefix,  # =out_prefix
             )
 
     def jbig2_single_args(root, groups: Dict[int, List[XrefExt]]):
@@ -316,23 +347,20 @@ def _produce_jbig2_images(
             # Second loop is to ensure multiple images per page are unpacked
             for n, xref_ext in enumerate(xref_exts):
                 xref, ext = xref_ext
-                yield dict(
-                    cwd=fspath(root),
-                    infile=img_name(root, xref, ext),
-                    outfile=root / f'{prefix}.{n:04d}',
+                yield (
+                    fspath(root),
+                    img_name(root, xref, ext),
+                    root / f'{prefix}.{n:04d}',
                 )
 
-    def convert_generic(fn, kwargs_dict):
-        return fn(**kwargs_dict)
-
     if options.jbig2_page_group_size > 1:
         jbig2_args = jbig2_group_args
-        jbig2_convert = partial(convert_generic, jbig2enc.convert_group)
+        jbig2_convert = jbig2enc.convert_group_mp
     else:
         jbig2_args = jbig2_single_args
-        jbig2_convert = partial(convert_generic, jbig2enc.convert_single)
+        jbig2_convert = jbig2enc.convert_single_mp
 
-    exec_progress_pool(
+    executor(
         use_threads=True,
         max_workers=options.jobs,
         tqdm_kwargs=dict(
@@ -347,7 +375,11 @@ def _produce_jbig2_images(
 
 
 def convert_to_jbig2(
-    pike: Pdf, jbig2_groups: Dict[int, List[XrefExt]], root: Path, options
+    pike: Pdf,
+    jbig2_groups: Dict[int, List[XrefExt]],
+    root: Path,
+    options,
+    executor: Executor,
 ) -> None:
     """Convert images to JBIG2 and insert into PDF.
 
@@ -361,15 +393,16 @@ def convert_to_jbig2(
     When the JBIG2 symbolic coder is not used, each JBIG2 stands on its own
     and needs no dictionary. Currently this must be lossless JBIG2.
     """
+    jbig2_globals_dict: Optional[Dictionary]
 
-    _produce_jbig2_images(jbig2_groups, root, options)
+    _produce_jbig2_images(jbig2_groups, root, options, executor)
 
     for group, xref_exts in jbig2_groups.items():
         prefix = f'group{group:08d}'
         jbig2_symfile = root / (prefix + '.sym')
         if jbig2_symfile.exists():
             jbig2_globals_data = jbig2_symfile.read_bytes()
-            jbig2_globals = pikepdf.Stream(pike, jbig2_globals_data)
+            jbig2_globals = Stream(pike, jbig2_globals_data)
             jbig2_globals_dict = Dictionary(JBIG2Globals=jbig2_globals)
         elif options.jbig2_page_group_size == 1:
             jbig2_globals_dict = None
@@ -386,36 +419,125 @@ def convert_to_jbig2(
             )
 
 
-def transcode_jpegs(pike: Pdf, jpegs: Sequence[Xref], root: Path, options) -> None:
-    for xref in tqdm(
-        jpegs, desc="JPEGs", unit='image', disable=not options.progress_bar
-    ):
-        in_jpg = jpg_name(root, xref)
-        opt_jpg = in_jpg.with_suffix('.opt.jpg')
+def _optimize_jpeg(args: Tuple[Xref, Path, Path, int]) -> Tuple[Xref, Optional[Path]]:
+    xref, in_jpg, opt_jpg, jpeg_quality = args
 
-        # This produces a debug warning from PIL
-        # DEBUG:PIL.Image:Error closing: 'NoneType' object has no attribute
-        # 'close'.  Seems to be mostly harmless
-        # https://github.com/python-pillow/Pillow/issues/1144
-        with Image.open(in_jpg) as im:
-            im.save(opt_jpg, optimize=True, quality=options.jpeg_quality)
+    with Image.open(in_jpg) as im:
+        im.save(opt_jpg, optimize=True, quality=jpeg_quality)
 
-        if opt_jpg.stat().st_size > in_jpg.stat().st_size:
-            log.debug("xref %s, jpeg, made larger - skip", xref)
-            continue
+    if opt_jpg.stat().st_size > in_jpg.stat().st_size:
+        log.debug(f"xref {xref}, jpeg, made larger - skip")
+        opt_jpg.unlink()
+        return xref, None
+    return xref, opt_jpg
 
-        compdata = leptonica.CompressedData.open(opt_jpg)
-        im_obj = pike.get_object(xref, 0)
-        im_obj.write(compdata.read(), filter=Name.DCTDecode)
+
+def transcode_jpegs(
+    pike: Pdf, jpegs: Sequence[Xref], root: Path, options, executor: Executor
+) -> None:
+    def jpeg_args() -> Iterator[Tuple[Xref, Path, Path, int]]:
+        for xref in jpegs:
+            in_jpg = jpg_name(root, xref)
+            opt_jpg = in_jpg.with_suffix('.opt.jpg')
+            yield xref, in_jpg, opt_jpg, options.jpeg_quality
+
+    def finish_jpeg(result: Tuple[Xref, Optional[Path]], pbar):
+        xref, opt_jpg = result
+        if opt_jpg:
+            compdata = opt_jpg.read_bytes()  # JPEG can inserted into PDF as is
+            im_obj = pike.get_object(xref, 0)
+            im_obj.write(compdata, filter=Name.DCTDecode)
+        pbar.update()
+
+    executor(
+        use_threads=True,  # Processes are significantly slower at this task
+        max_workers=options.jobs,
+        tqdm_kwargs=dict(
+            desc="Recompressing JPEGs",
+            total=len(jpegs),
+            unit='image',
+            disable=not options.progress_bar,
+        ),
+        task=_optimize_jpeg,
+        task_arguments=jpeg_args(),
+        task_finished=finish_jpeg,
+    )
+
+
+def _find_deflatable_jpeg(
+    *, pike: Pdf, root: Path, image: Stream, xref: Xref, options
+) -> Optional[XrefExt]:
+    result = extract_image_filter(pike, root, image, xref)
+    if result is None:
+        return None
+    _pim, filtdp = result
+
+    if filtdp[0] == Name.DCTDecode and not filtdp[1] and options.optimize >= 1:
+        return XrefExt(xref, '.memory')
+
+    return None
+
+
+def _deflate_jpeg(args: Tuple[Pdf, threading.Lock, Xref, int]) -> Tuple[Xref, bytes]:
+    pike, lock, xref, complevel = args
+    with lock:
+        xobj = pike.get_object(xref, 0)
+        try:
+            data = xobj.read_raw_bytes()
+        except PdfError:
+            return xref, b''
+    compdata = compress(data, complevel)
+    if len(compdata) >= len(data):
+        return xref, b''
+    return xref, compdata
+
+
+def deflate_jpegs(pike: Pdf, root: Path, options, executor: Executor) -> None:
+    jpegs = []
+    for _pageno, xref_ext in extract_images(pike, root, options, _find_deflatable_jpeg):
+        xref = xref_ext.xref
+        log.debug(f'xref {xref}: marking this JPEG as deflatable')
+        jpegs.append(xref)
+
+    complevel = 9 if options.optimize == 3 else 6
+
+    # Our calls to xobj.write() in finish() need coordination
+    lock = threading.Lock()
+
+    def deflate_args() -> Iterator:
+        for xref in jpegs:
+            yield pike, lock, xref, complevel
+
+    def finish(result, pbar):
+        xref, compdata = result
+        if len(compdata) > 0:
+            with lock:
+                xobj = pike.get_object(xref, 0)
+                xobj.write(compdata, filter=[Name.FlateDecode, Name.DCTDecode])
+        pbar.update()
+
+    executor(
+        use_threads=True,  # We're sharing the pdf directly, must use threads
+        max_workers=options.jobs,
+        tqdm_kwargs=dict(
+            desc="Deflating JPEGs",
+            total=len(jpegs),
+            unit='image',
+            disable=not options.progress_bar,
+        ),
+        task=_deflate_jpeg,
+        task_arguments=deflate_args(),
+        task_finished=finish,
+    )
 
 
 def _transcode_png(pike: Pdf, filename: Path, xref: Xref) -> bool:
     output = filename.with_suffix('.png.pdf')
     with output.open('wb') as f:
-        img2pdf.convert(fspath(filename), outputstream=f)
+        img2pdf.convert(fspath(filename), outputstream=f, **IMG2PDF_KWARGS)
 
-    with pikepdf.open(output) as pdf_image:
-        foreign_image = next(pdf_image.pages[0].images.values())
+    with Pdf.open(output) as pdf_image:
+        foreign_image = next(iter(pdf_image.pages[0].images.values()))
         local_image = pike.copy_foreign(foreign_image)
 
         im_obj = pike.get_object(xref, 0)
@@ -456,6 +578,7 @@ def transcode_pngs(
     image_name_fn: Callable[[Path, Xref], Path],
     root: Path,
     options,
+    executor,
 ) -> None:
     modified: MutableSet[Xref] = set()
     if options.optimize >= 2:
@@ -475,10 +598,7 @@ def transcode_pngs(
                 )
                 modified.add(xref)
 
-        def pngquant_fn(args):
-            pngquant.quantize(*args)
-
-        exec_progress_pool(
+        executor(
             use_threads=True,
             max_workers=options.jobs,
             tqdm_kwargs=dict(
@@ -487,7 +607,7 @@ def transcode_pngs(
                 unit='image',
                 disable=not options.progress_bar,
             ),
-            task=pngquant_fn,
+            task=pngquant.quantize_mp,
             task_arguments=pngquant_args(),
         )
 
@@ -496,86 +616,20 @@ def transcode_pngs(
         _transcode_png(pike, filename, xref)
 
 
-@deprecated
-def rewrite_png_as_g4(pike: Pdf, im_obj: Object, compdata) -> None:  # pragma: no cover
-    im_obj.BitsPerComponent = 1
-    im_obj.Width = compdata.w
-    im_obj.Height = compdata.h
-
-    im_obj.write(compdata.read())
-
-    log.debug(f"PNG to G4 {im_obj.objgen}")
-    if Name.Predictor in im_obj:
-        del im_obj.Predictor
-    if Name.DecodeParms in im_obj:
-        del im_obj.DecodeParms
-    im_obj.DecodeParms = Dictionary(
-        K=-1, BlackIs1=bool(compdata.minisblack), Columns=compdata.w
-    )
-
-    im_obj.Filter = Name.CCITTFaxDecode
-    return
+DEFAULT_EXECUTOR = SerialExecutor()
 
 
-@deprecated
-def rewrite_png(pike: Pdf, im_obj: Object, compdata) -> None:  # pragma: no cover
-    # When a PNG is inserted into a PDF, we more or less copy the IDAT section from
-    # the PDF and transfer the rest of the PNG headers to PDF image metadata.
-    # One thing we have to do is tell the PDF reader whether a predictor was used
-    # on the image before Flate encoding. (Typically one is.)
-    # According to Leptonica source, PDF readers don't actually need us
-    # to specify the correct predictor, they just need a value of either:
-    #   1 - no predictor
-    #   10-14 - there is a predictor
-    # Leptonica's compdata->predictor only tells TRUE or FALSE
-    # 10-14 means the actual predictor is specified in the data, so for any
-    # number >= 10 the PDF reader will use whatever the PNG data specifies.
-    # In practice Leptonica should use Paeth, 14, but 15 seems to be the
-    # designated value for "optimal". So we will use 15.
-    # See:
-    #   - PDF RM 7.4.4.4 Table 10
-    #   - https://github.com/DanBloomberg/leptonica/blob/master/src/pdfio2.c#L757
-    predictor = 15 if compdata.predictor > 0 else 1
-    dparms = Dictionary(Predictor=predictor)
-    if predictor > 1:
-        dparms.BitsPerComponent = compdata.bps  # Yes, this is redundant
-        dparms.Colors = compdata.spp
-        dparms.Columns = compdata.w
-
-    im_obj.BitsPerComponent = compdata.bps
-    im_obj.Width = compdata.w
-    im_obj.Height = compdata.h
-
-    log.debug(
-        f"PNG {im_obj.objgen}: palette={compdata.ncolors} spp={compdata.spp} bps={compdata.bps}"
-    )
-    if compdata.ncolors > 0:
-        # .ncolors is the number of colors in the palette, not the number of
-        # colors used in a true color image. The palette string is always
-        # given as RGB tuples even when the image is grayscale; see
-        # https://github.com/DanBloomberg/leptonica/blob/master/src/colormap.c#L2067
-        palette_pdf_string = compdata.get_palette_pdf_string()
-        palette_data = pikepdf.Object.parse(palette_pdf_string)
-        palette_stream = pikepdf.Stream(pike, bytes(palette_data))
-        palette = [Name.Indexed, Name.DeviceRGB, compdata.ncolors - 1, palette_stream]
-        cs = palette
-    else:
-        # ncolors == 0 means we are using a colorspace without a palette
-        if compdata.spp == 1:
-            cs = Name.DeviceGray
-        elif compdata.spp == 4:
-            cs = Name.DeviceCMYK
-        else:  # spp == 3
-            cs = Name.DeviceRGB
-    im_obj.ColorSpace = cs
-    im_obj.write(compdata.read(), filter=Name.FlateDecode, decode_parms=dparms)
-
-
-def optimize(input_file: Path, output_file: Path, context, save_settings) -> None:
+def optimize(
+    input_file: Path,
+    output_file: Path,
+    context,
+    save_settings,
+    executor: Executor = DEFAULT_EXECUTOR,
+) -> Path:
     options = context.options
     if options.optimize == 0:
         safe_symlink(input_file, output_file)
-        return
+        return output_file
 
     if options.jpeg_quality == 0:
         options.jpeg_quality = DEFAULT_JPEG_QUALITY if options.optimize < 3 else 40
@@ -584,19 +638,20 @@ def optimize(input_file: Path, output_file: Path, context, save_settings) -> Non
     if options.jbig2_page_group_size == 0:
         options.jbig2_page_group_size = 10 if options.jbig2_lossy else 1
 
-    with pikepdf.Pdf.open(input_file) as pike:
+    with Pdf.open(input_file) as pike:
         root = output_file.parent / 'images'
         root.mkdir(exist_ok=True)
 
         jpegs, pngs = extract_images_generic(pike, root, options)
-        transcode_jpegs(pike, jpegs, root, options)
+        transcode_jpegs(pike, jpegs, root, options, executor)
+        deflate_jpegs(pike, root, options, executor)
         # if options.optimize >= 2:
         # Try pngifying the jpegs
         #    transcode_pngs(pike, jpegs, jpg_name, root, options)
-        transcode_pngs(pike, pngs, png_name, root, options)
+        transcode_pngs(pike, pngs, png_name, root, options, executor)
 
         jbig2_groups = extract_images_jbig2(pike, root, options)
-        convert_to_jbig2(pike, jbig2_groups, root, options)
+        convert_to_jbig2(pike, jbig2_groups, root, options, executor)
 
         target_file = output_file.with_suffix('.opt.pdf')
         pike.remove_unreferenced_resources()
@@ -609,19 +664,22 @@ def optimize(input_file: Path, output_file: Path, context, save_settings) -> Non
             f"Output file not created after optimizing. We probably ran "
             f"out of disk space in the temporary folder: {tempfile.gettempdir()}."
         )
-    ratio = input_size / output_size
     savings = 1 - output_size / input_size
-    log.info(f"Optimize ratio: {ratio:.2f} savings: {(savings):.1%}")
 
     if savings < 0:
-        log.info("Image optimization did not improve the file - discarded")
+        log.info(
+            "Image optimization did not improve the file - "
+            "optimizations will not be used"
+        )
         # We still need to save the file
-        with pikepdf.open(input_file) as pike:
+        with Pdf.open(input_file) as pike:
             pike.remove_unreferenced_resources()
             pike.save(output_file, **save_settings)
     else:
         safe_symlink(target_file, output_file)
 
+    return output_file
+
 
 def main(infile, outfile, level, jobs=1):
     from shutil import copy  # pylint: disable=import-outside-toplevel
@@ -653,9 +711,9 @@ def main(infile, outfile, level, jobs=1):
         jb2lossy=False,
     )
 
-    with TemporaryDirectory() as td:
-        context = PdfContext(options, td, infile, None, None)
-        tmpout = Path(td) / 'out.pdf'
+    with TemporaryDirectory() as tmpdir:
+        context = PdfContext(options, tmpdir, infile, None, None)
+        tmpout = Path(tmpdir) / 'out.pdf'
         optimize(
             infile,
             tmpout,
@@ -663,7 +721,7 @@ def main(infile, outfile, level, jobs=1):
             dict(
                 compress_streams=True,
                 preserve_pdfa=True,
-                object_stream_mode=pikepdf.ObjectStreamMode.generate,
+                object_stream_mode=ObjectStreamMode.generate,
             ),
         )
         copy(fspath(tmpout), fspath(outfile))

src/ocrmypdf/pdfa.py

@@ -13,12 +13,14 @@ import base64
 from pathlib import Path
 from typing import Dict, Iterator, Union
 
+try:
+    from importlib_resources import files as package_files
+except ImportError:
+    from importlib.resources import files as package_files
+
 import pikepdf
-import pkg_resources
 
-ICC_PROFILE_RELPATH = 'data/sRGB.icc'
-
-SRGB_ICC_PROFILE = pkg_resources.resource_filename('ocrmypdf', ICC_PROFILE_RELPATH)
+SRGB_ICC_PROFILE_NAME = 'sRGB.icc'
 
 
 def _postscript_objdef(
@@ -95,19 +97,20 @@ def generate_pdfa_ps(target_filename: Path, icc: str = 'sRGB'):
         target_filename: filename to save
         icc: ICC identifier such as 'sRGB'
     References:
-        Adobe PDFMARK Reference: https://www.adobe.com/content/dam/acom/en/devnet/acrobat/pdfs/pdfmark_reference.pdf
+        Adobe PDFMARK Reference:
+        https://www.adobe.com/content/dam/acom/en/devnet/acrobat/pdfs/pdfmark_reference.pdf
     """
-    if icc == 'sRGB':
-        icc_profile = SRGB_ICC_PROFILE
-    else:
+    if icc != 'sRGB':
         raise NotImplementedError("Only supporting sRGB")
 
-    bytes_icc_profile = Path(icc_profile).read_bytes()
-    ps = '\n'.join(_make_postscript(icc, bytes_icc_profile, 3))
+    bytes_icc_profile = (
+        package_files('ocrmypdf.data') / SRGB_ICC_PROFILE_NAME
+    ).read_bytes()
+    postscript = '\n'.join(_make_postscript(icc, bytes_icc_profile, 3))
 
     # We should have encoded everything to pure ASCII by this point, and
     # to be safe, only allow ASCII in PostScript
-    Path(target_filename).write_text(ps, encoding='ascii')
+    Path(target_filename).write_text(postscript, encoding='ascii')
     return target_filename
 
 

src/ocrmypdf/pdfinfo/__init__.py

@@ -6,4 +6,6 @@
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
 
+"""For extracting information about PDFs prior to OCR."""
+
 from ocrmypdf.pdfinfo.info import Colorspace, Encoding, PdfInfo

src/ocrmypdf/pdfinfo/info.py

@@ -6,35 +6,86 @@
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
 
+"""Extract information about the content of a PDF."""
+
+import atexit
 import logging
 import re
-from collections import defaultdict, namedtuple
+from collections import defaultdict
+from contextlib import ExitStack
 from decimal import Decimal
-from enum import Enum
+from enum import Enum, auto
 from functools import partial
-from math import hypot, isclose
+from math import hypot, inf, isclose
 from os import PathLike
 from pathlib import Path
-from typing import Container, Iterator, Optional, Tuple, Union
+from typing import (
+    Container,
+    Dict,
+    Iterable,
+    Iterator,
+    List,
+    Mapping,
+    NamedTuple,
+    Optional,
+    Sequence,
+    Tuple,
+    Union,
+)
 from warnings import warn
 
-import pikepdf
-from pikepdf import Object, Pdf, PdfMatrix
+from pikepdf import (
+    Object,
+    Pdf,
+    PdfImage,
+    PdfInlineImage,
+    PdfMatrix,
+    UnsupportedImageTypeError,
+    parse_content_stream,
+)
 
-from ocrmypdf._concurrent import exec_progress_pool
+from ocrmypdf._concurrent import Executor, SerialExecutor
 from ocrmypdf.exceptions import EncryptedPdfError, InputFileError
 from ocrmypdf.helpers import Resolution, available_cpu_count, pikepdf_enable_mmap
 from ocrmypdf.pdfinfo.layout import get_page_analysis, get_text_boxes
 
 logger = logging.getLogger()
 
-Colorspace = Enum('Colorspace', 'gray rgb cmyk lab icc index sep devn pattern jpeg2000')
 
-Encoding = Enum(
-    'Encoding', 'ccitt jpeg jpeg2000 jbig2 asciihex ascii85 lzw flate runlength'
-)
+class Colorspace(Enum):
+    """Description of common image colorspaces in a PDF."""
 
-FRIENDLY_COLORSPACE = {
+    # pylint: disable=invalid-name
+    gray = auto()
+    rgb = auto()
+    cmyk = auto()
+    lab = auto()
+    icc = auto()
+    index = auto()
+    sep = auto()
+    devn = auto()
+    pattern = auto()
+    jpeg2000 = auto()
+
+
+class Encoding(Enum):
+    """Description of common image encodings in a PDF."""
+
+    # pylint: disable=invalid-name
+    ccitt = auto()
+    jpeg = auto()
+    jpeg2000 = auto()
+    jbig2 = auto()
+    asciihex = auto()
+    ascii85 = auto()
+    lzw = auto()
+    flate = auto()
+    runlength = auto()
+
+
+FloatRect = Tuple[float, float, float, float]
+
+FRIENDLY_COLORSPACE: Dict[str, Colorspace] = {
     '/DeviceGray': Colorspace.gray,
     '/CalGray': Colorspace.gray,
     '/DeviceRGB': Colorspace.rgb,
@@ -52,7 +103,7 @@ FRIENDLY_COLORSPACE = {
     '/I': Colorspace.index,
 }
 
-FRIENDLY_ENCODING = {
+FRIENDLY_ENCODING: Dict[str, Encoding] = {
     '/CCITTFaxDecode': Encoding.ccitt,
     '/DCTDecode': Encoding.jpeg,
     '/JPXDecode': Encoding.jpeg2000,
@@ -66,7 +117,7 @@ FRIENDLY_ENCODING = {
     '/RL': Encoding.runlength,
 }
 
-FRIENDLY_COMP = {
+FRIENDLY_COMP: Dict[Colorspace, int] = {
     Colorspace.gray: 1,
     Colorspace.rgb: 3,
     Colorspace.cmyk: 4,
@@ -84,24 +135,46 @@ def _is_unit_square(shorthand):
     return all(isclose(a, b, rel_tol=1e-3) for a, b in pairwise)
 
 
-XobjectSettings = namedtuple('XobjectSettings', ['name', 'shorthand', 'stack_depth'])
+class XobjectSettings(NamedTuple):
+    """Info about an XObject found in a PDF."""
 
-InlineSettings = namedtuple('InlineSettings', ['iimage', 'shorthand', 'stack_depth'])
+    name: str
+    shorthand: Tuple[float, float, float, float, float, float]
+    stack_depth: int
 
-ContentsInfo = namedtuple(
-    'ContentsInfo',
-    ['xobject_settings', 'inline_images', 'found_vector', 'found_text', 'name_index'],
-)
 
-TextboxInfo = namedtuple('TextboxInfo', ['bbox', 'is_visible', 'is_corrupt'])
+class InlineSettings(NamedTuple):
+    """Info about an inline image found in a PDF."""
+
+    iimage: PdfInlineImage
+    shorthand: Tuple[float, float, float, float, float, float]
+    stack_depth: int
+
+
+class ContentsInfo(NamedTuple):
+    """Info about various objects found in a PDF."""
+
+    xobject_settings: List[XobjectSettings]
+    inline_images: List[InlineSettings]
+    found_vector: bool
+    found_text: bool
+    name_index: Mapping[str, List[XobjectSettings]]
+
+
+class TextboxInfo(NamedTuple):
+    """Info about a text box found in a PDF."""
+
+    bbox: Tuple[float, float, float, float]
+    is_visible: bool
+    is_corrupt: bool
 
 
 class VectorMarker:
-    pass
+    """Sentinel indicating vector drawing operations were found on a page."""
 
 
 class TextMarker:
-    pass
+    """Sentinel indicating text drawing operations were found on a page."""
 
 
 def _normalize_stack(graphobjs):
@@ -144,8 +217,8 @@ def _interpret_contents(contentstream: Object, initial_shorthand=UNIT_SQUARE):
 
     stack = []
     ctm = PdfMatrix(initial_shorthand)
-    xobject_settings = []
-    inline_images = []
+    xobject_settings: List[XobjectSettings] = []
+    inline_images: List[InlineSettings] = []
     name_index = defaultdict(lambda: [])
     found_vector = False
     found_text = False
@@ -155,9 +228,7 @@ def _interpret_contents(contentstream: Object, initial_shorthand=UNIT_SQUARE):
     operator_whitelist = ' '.join(vector_ops | text_showing_ops | image_ops)
 
     for n, graphobj in enumerate(
-        _normalize_stack(
-            pikepdf.parse_content_stream(contentstream, operator_whitelist)
-        )
+        _normalize_stack(parse_content_stream(contentstream, operator_whitelist))
     ):
         operands, operator = graphobj
         if operator == 'q':
@@ -165,7 +236,7 @@ def _interpret_contents(contentstream: Object, initial_shorthand=UNIT_SQUARE):
             if len(stack) > 32:  # See docstring
                 if len(stack) > 128:
                     raise RuntimeError(
-                        "PDF graphics stack overflowed hard limit, operator %i" % n
+                        f"PDF graphics stack overflowed hard limit at operator {n}"
                     )
                 warn("PDF graphics stack overflowed spec limit")
         elif operator == 'Q':
@@ -183,7 +254,7 @@ def _interpret_contents(contentstream: Object, initial_shorthand=UNIT_SQUARE):
                 name=image_name, shorthand=ctm.shorthand, stack_depth=len(stack)
             )
             xobject_settings.append(settings)
-            name_index[image_name].append(settings)
+            name_index[str(image_name)].append(settings)
         elif operator == 'INLINE IMAGE':  # BI/ID/EI are grouped into this
             iimage = operands[0]
             inline = InlineSettings(
@@ -251,43 +322,48 @@ def _get_dpi(ctm_shorthand, image_size) -> Resolution:
 
     """
 
-    a, b, c, d, _, _ = ctm_shorthand
+    a, b, c, d, _, _ = ctm_shorthand  # pylint: disable=invalid-name
 
     # Calculate the width and height of the image in PDF units
-    image_drawn_width = hypot(a, b)
-    image_drawn_height = hypot(c, d)
+    image_drawn = hypot(a, b), hypot(c, d)
 
-    # The scale of the image is pixels per unit of default user space (1/72")
-    scale_w = image_size[0] / image_drawn_width
-    scale_h = image_size[1] / image_drawn_height
-
-    # DPI = scale * 72
-    dpi_w = scale_w * 72.0
-    dpi_h = scale_h * 72.0
+    def calc(drawn, pixels, inches_per_pt=72.0):
+        # The scale of the image is pixels per unit of default user space (1/72")
+        scale = pixels / drawn if drawn != 0 else inf
+        dpi = scale * inches_per_pt
+        return dpi
 
+    dpi_w, dpi_h = (calc(image_drawn[n], image_size[n]) for n in range(2))
     return Resolution(dpi_w, dpi_h)
 
 
 class ImageInfo:
+    """Information about an image found in a PDF."""
+
     DPI_PREC = Decimal('1.000')
 
+    _comp: Optional[int]
+    _name: str
+
     def __init__(
         self,
         *,
         name='',
         pdfimage: Optional[Object] = None,
-        inline: Optional[Object] = None,
+        inline: Optional[PdfInlineImage] = None,
         shorthand=None,
     ):
         self._name = str(name)
         self._shorthand = shorthand
 
+        pim: Union[PdfInlineImage, PdfImage]
+
         if inline is not None:
             self._origin = 'inline'
-            pim = inline.iimage
+            pim = inline
         elif pdfimage is not None:
             self._origin = 'xobject'
-            pim = pikepdf.PdfImage(pdfimage)
+            pim = PdfImage(pdfimage)
         else:
             raise ValueError("Either pdfimage or inline must be set")
         self._width = pim.width
@@ -303,32 +379,42 @@ class ImageInfo:
 
         self._bpc = int(pim.bits_per_component)
         try:
-            self._enc = FRIENDLY_ENCODING.get(pim.filters[0], 'image')
+            self._enc = FRIENDLY_ENCODING.get(pim.filters[0])
         except IndexError:
-            self._enc = '?'
+            self._enc = None
 
         try:
-            self._color = FRIENDLY_COLORSPACE.get(pim.colorspace, '?')
+            self._color = FRIENDLY_COLORSPACE.get(pim.colorspace or '')
         except NotImplementedError:
-            self._color = '?'
+            self._color = None
         if self._enc == Encoding.jpeg2000:
             self._color = Colorspace.jpeg2000
 
         if self._color == Colorspace.icc:
             # Check the ICC profile to determine actual colorspace
-            pim_icc = pim.icc
-            if pim_icc.profile.xcolor_space == 'GRAY':
-                self._comp = 1
-            elif pim_icc.profile.xcolor_space == 'CMYK':
-                self._comp = 4
-            else:
-                self._comp = 3
+            try:
+                pim_icc = pim.icc
+                if pim_icc.profile.xcolor_space == 'GRAY':
+                    self._comp = 1
+                elif pim_icc.profile.xcolor_space == 'CMYK':
+                    self._comp = 4
+                else:
+                    self._comp = 3
+            except UnsupportedImageTypeError as ex:
+                self._comp = None
+                logger.warning(
+                    f"An image with a corrupt or unreadable ICC profile was found. "
+                    f"The output PDF may not match the input PDF visually: {ex}. {self}"
+                )
         else:
-            self._comp = FRIENDLY_COMP.get(self._color, '?')
+            if isinstance(self._color, Colorspace):
+                self._comp = FRIENDLY_COMP.get(self._color)
+            else:
+                self._comp = None
 
             # Bit of a hack... infer grayscale if component count is uncertain
             # but encoding only supports monochrome.
-            if self._comp == '?' and self._enc in (Encoding.ccitt, Encoding.jbig2):
+            if self._comp is None and self._enc in (Encoding.ccitt, Encoding.jbig2):
                 self._comp = FRIENDLY_COMP[Colorspace.gray]
 
     @property
@@ -353,30 +439,29 @@ class ImageInfo:
 
     @property
     def color(self):
-        return self._color
+        return self._color if self._color is not None else '?'
 
     @property
     def comp(self):
-        return self._comp
+        return self._comp if self._comp is not None else '?'
 
     @property
     def enc(self):
-        return self._enc
+        return self._enc if self._enc is not None else 'image'
+
+    @property
+    def renderable(self):
+        return self.dpi.is_finite and self.width >= 0 and self.height >= 0
 
     @property
     def dpi(self):
         return _get_dpi(self._shorthand, (self._width, self._height))
 
     def __repr__(self):
-        class_locals = {
-            attr: getattr(self, attr, None)
-            for attr in dir(self)
-            if not attr.startswith('_')
-        }
         return (
-            "<ImageInfo '{name}' {type_} {width}x{height} {color} "
-            "{comp} {bpc} {enc} {dpi}>"
-        ).format(**class_locals)
+            f"<ImageInfo '{self.name}' {self.type_} {self.width}x{self.height} "
+            f"{self.color} {self.comp} {self.bpc} {self.enc} {self.dpi}>"
+        )
 
 
 def _find_inline_images(contentsinfo: ContentsInfo) -> Iterator[ImageInfo]:
@@ -384,7 +469,7 @@ def _find_inline_images(contentsinfo: ContentsInfo) -> Iterator[ImageInfo]:
 
     for n, inline in enumerate(contentsinfo.inline_images):
         yield ImageInfo(
-            name='inline-%02d' % n, shorthand=inline.shorthand, inline=inline
+            name=f'inline-{n:02d}', shorthand=inline.shorthand, inline=inline.iimage
         )
 
 
@@ -409,7 +494,7 @@ def _image_xobjects(container) -> Iterator[Tuple[Object, str]]:
     xobjs = resources['/XObject'].as_dict()
     for xobj in xobjs:
         candidate: Object = xobjs[xobj]
-        if not '/Subtype' in candidate:
+        if '/Subtype' not in candidate:
             continue
         if candidate['/Subtype'] == '/Image':
             pdfimage = candidate
@@ -525,10 +610,10 @@ def _process_content_streams(
     yield from _find_form_xobject_images(pdf, container, contentsinfo)
 
 
-def _page_has_text(text_blocks, page_width, page_height) -> bool:
+def _page_has_text(text_blocks: Iterable[FloatRect], page_width, page_height) -> bool:
     """Smarter text detection that ignores text in margins"""
 
-    pw, ph = float(page_width), float(page_height)
+    pw, ph = float(page_width), float(page_height)  # pylint: disable=invalid-name
 
     margin_ratio = 0.125
     interior_bbox = (
@@ -538,7 +623,7 @@ def _page_has_text(text_blocks, page_width, page_height) -> bool:
         margin_ratio * ph,  # bottom  (first quadrant: bottom < top)
     )
 
-    def rects_intersect(a, b) -> bool:
+    def rects_intersect(a: FloatRect, b: FloatRect) -> bool:
         """
         Where (a,b) are 4-tuple rects (left-0, top-1, right-2, bottom-3)
         https://stackoverflow.com/questions/306316/determine-if-two-rectangles-overlap-each-other
@@ -560,41 +645,54 @@ def simplify_textboxes(miner, textbox_getter) -> Iterator[TextboxInfo]:
     We do this to save memory and ensure that our objects are pickleable.
     """
     for box in textbox_getter(miner):
-        first_line = box._objs[0]
-        first_char = first_line._objs[0]
+        first_line = box._objs[0]  # pylint: disable=protected-access
+        first_char = first_line._objs[0]  # pylint: disable=protected-access
 
         visible = first_char.rendermode != 3
         corrupt = first_char.get_text() == '\ufffd'
         yield TextboxInfo(box.bbox, visible, corrupt)
 
 
-worker_pdf = None
+worker_pdf = None  # pylint: disable=invalid-name
 
 
-def _pdf_pageinfo_sync_init(infile: Path, pdfminer_loglevel):
-    global worker_pdf  # pylint: disable=global-statement
+def _pdf_pageinfo_sync_init(pdf: Pdf, infile: Path, pdfminer_loglevel):
+    global worker_pdf  # pylint: disable=global-statement,invalid-name
     pikepdf_enable_mmap()
 
     logging.getLogger('pdfminer').setLevel(pdfminer_loglevel)
 
-    # If this function is called as a thread initializer, we need a messy hack
-    # to close worker_pdf. If called as a process, it will be released when the
-    # process is terminated.
-    worker_pdf = pikepdf.open(infile)
+    # If the pdf is not opened, open a copy for our worker process to use
+    if pdf is None:
+        worker_pdf = Pdf.open(infile)
+
+        def on_process_close():
+            worker_pdf.close()
+
+        # Close when this process exits
+        atexit.register(on_process_close)
 
 
 def _pdf_pageinfo_sync(args):
-    global worker_pdf  # pylint: disable=global-statement
-    pageno, infile, check_pages, detailed_analysis = args
-    page = PageInfo(worker_pdf, pageno, infile, check_pages, detailed_analysis)
-    return page
+    pageno, thread_pdf, infile, check_pages, detailed_analysis = args
+    pdf = thread_pdf if thread_pdf is not None else worker_pdf
+    with ExitStack() as stack:
+        if not pdf:  # When called with SerialExecutor
+            pdf = stack.enter_context(Pdf.open(infile))
+        page = PageInfo(pdf, pageno, infile, check_pages, detailed_analysis)
+        return page
 
 
 def _pdf_pageinfo_concurrent(
-    pdf, infile, progbar, max_workers, check_pages, detailed_analysis=False
-):
-    global worker_pdf  # pylint: disable=global-statement
-    pages = [None] * len(pdf.pages)
+    pdf,
+    executor: Executor,
+    infile,
+    progbar,
+    max_workers,
+    check_pages,
+    detailed_analysis=False,
+) -> Sequence[Optional['PageInfo']]:
+    pages: Sequence[Optional['PageInfo']] = [None] * len(pdf.pages)
 
     def update_pageinfo(result, pbar):
         page = result
@@ -607,7 +705,6 @@ def _pdf_pageinfo_concurrent(
         max_workers = available_cpu_count()
 
     total = len(pdf.pages)
-    contexts = ((n, infile, check_pages, detailed_analysis) for n in range(total))
 
     use_threads = False  # No performance gain if threaded due to GIL
     n_workers = min(1 + len(pages) // 4, max_workers)
@@ -616,29 +713,41 @@ def _pdf_pageinfo_concurrent(
         # a separate process.
         use_threads = True
 
-    try:
-        exec_progress_pool(
-            use_threads=use_threads,
-            max_workers=n_workers,
-            tqdm_kwargs=dict(
-                total=total, desc="Scanning contents", unit='page', disable=not progbar
-            ),
-            task_initializer=partial(
-                _pdf_pageinfo_sync_init, infile, logging.getLogger('pdfminer').level
-            ),
-            task=_pdf_pageinfo_sync,
-            task_arguments=contexts,
-            task_finished=update_pageinfo,
-        )
-    finally:
-        if worker_pdf and use_threads:
-            assert n_workers == 1, "Should have only one worker when threaded"
-            # This is messy, but if we ran in thread, close worker_pdf
-            worker_pdf.close()
+    # If we use a thread, we can pass the already-open Pdf for them to use
+    # If we use processes, we pass a None which tells the init function to open its
+    # own
+    initial_pdf = pdf if use_threads else None
+
+    contexts = (
+        (n, initial_pdf, infile, check_pages, detailed_analysis) for n in range(total)
+    )
+    assert n_workers == 1 if use_threads else n_workers >= 1, "Not multithreadable"
+    executor(
+        use_threads=use_threads,
+        max_workers=n_workers,
+        tqdm_kwargs=dict(
+            total=total, desc="Scanning contents", unit='page', disable=not progbar
+        ),
+        worker_initializer=partial(
+            _pdf_pageinfo_sync_init,
+            initial_pdf,
+            infile,
+            logging.getLogger('pdfminer').level,
+        ),
+        task=_pdf_pageinfo_sync,
+        task_arguments=contexts,
+        task_finished=update_pageinfo,
+    )
     return pages
 
 
 class PageInfo:
+    """Information about type of contents on each page in a PDF."""
+
+    _has_text: Optional[bool]
+    _has_vector: Optional[bool]
+    _images: List[ImageInfo]
+
     def __init__(
         self,
         pdf: Pdf,
@@ -696,25 +805,27 @@ class PageInfo:
             self._has_vector = False
             self._has_text = False
             self._images = []
-            for ci in _process_content_streams(
+            for info in _process_content_streams(
                 pdf=pdf, container=page, shorthand=userunit_shorthand
             ):
-                if isinstance(ci, VectorMarker):
+                if isinstance(info, VectorMarker):
                     self._has_vector = True
-                elif isinstance(ci, TextMarker):
+                elif isinstance(info, TextMarker):
                     self._has_text = True
-                elif isinstance(ci, ImageInfo):
-                    self._images.append(ci)
+                elif isinstance(info, ImageInfo):
+                    self._images.append(info)
                 else:
                     raise NotImplementedError()
         else:
             self._has_vector = None  # i.e. "no information"
             self._has_text = None
-            self._images = None
+            self._images = []
 
         self._dpi = None
         if self._images:
-            dpi = Resolution(0.0, 0.0).take_max(image.dpi for image in self._images)
+            dpi = Resolution(0.0, 0.0).take_max(
+                image.dpi for image in self._images if image.renderable
+            )
             self._dpi = dpi
             self._width_pixels = int(round(dpi.x * float(self._width_inches)))
             self._height_pixels = int(round(dpi.y * float(self._height_inches)))
@@ -725,7 +836,7 @@ class PageInfo:
 
     @property
     def has_text(self) -> bool:
-        return self._has_text
+        return bool(self._has_text)
 
     @property
     def has_corrupt_text(self) -> bool:
@@ -735,7 +846,7 @@ class PageInfo:
 
     @property
     def has_vector(self) -> bool:
-        return self._has_vector
+        return bool(self._has_vector)
 
     @property
     def width_inches(self) -> Decimal:
@@ -813,26 +924,32 @@ class PageInfo:
         )
 
 
+DEFAULT_EXECUTOR = SerialExecutor()
+
+
 class PdfInfo:
     """Get summary information about a PDF"""
 
     def __init__(
         self,
         infile,
+        *,
         detailed_analysis: bool = False,
         progbar: bool = False,
         max_workers: int = None,
         check_pages=None,
+        executor: Executor = DEFAULT_EXECUTOR,
     ):
         self._infile = infile
         if check_pages is None:
             check_pages = range(0, 1_000_000_000)
 
-        with pikepdf.open(infile) as pdf:
+        with Pdf.open(infile) as pdf:
             if pdf.is_encrypted:
                 raise EncryptedPdfError()  # Triggered by encryption with empty passwd
             self._pages = _pdf_pageinfo_concurrent(
                 pdf,
+                executor,
                 infile,
                 progbar,
                 max_workers,
@@ -848,17 +965,17 @@ class PdfInfo:
                     self._has_acroform = True
 
     @property
-    def pages(self):
+    def pages(self) -> Sequence[Optional[PageInfo]]:
         return self._pages
 
     @property
     def min_version(self) -> str:
         # The minimum PDF is the maximum version that any particular page needs
-        return max(page.min_version for page in self.pages)
+        return max(page.min_version for page in self.pages if page)
 
     @property
     def has_userunit(self) -> bool:
-        return any(page.userunit != 1.0 for page in self.pages)
+        return any(page.userunit != 1.0 for page in self.pages if page)
 
     @property
     def has_acroform(self) -> bool:

src/ocrmypdf/pdfinfo/layout.py

@@ -135,7 +135,7 @@ class LTStateAwareChar(LTChar):
         return self._text
 
     def __repr__(self):
-        return '<%s %s matrix=%s rendermode=%r font=%r adv=%s text=%r>' % (
+        return '<{} {} matrix={} rendermode={!r} font={!r} adv={} text={!r}>'.format(
             self.__class__.__name__,
             bbox2str(self.bbox),
             matrix2str(self.matrix),

src/ocrmypdf/pluginspec.py

@@ -4,28 +4,80 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+"""OCRmyPDF pluggy plugin specification."""
 
-from abc import ABC, abstractmethod, abstractstaticmethod
+from abc import ABC, abstractmethod
 from argparse import ArgumentParser, Namespace
-from collections import namedtuple
+from logging import Handler
 from pathlib import Path
-from typing import TYPE_CHECKING, AbstractSet, List, Optional
+from typing import (
+    TYPE_CHECKING,
+    AbstractSet,
+    List,
+    NamedTuple,
+    Optional,
+    Sequence,
+    Tuple,
+)
 
 import pluggy
 
+from ocrmypdf import Executor, PdfContext
 from ocrmypdf.helpers import Resolution
 
 if TYPE_CHECKING:
     from PIL import Image
 
+    # pylint: disable=ungrouped-imports
     from ocrmypdf._jobcontext import PageContext
     from ocrmypdf.pdfinfo import PdfInfo
 
+    # pylint: enable=ungrouped-imports
+
 hookspec = pluggy.HookspecMarker('ocrmypdf')
 
 # pylint: disable=unused-argument
 
 
+@hookspec(firstresult=True)
+def get_logging_console() -> Handler:
+    """Returns a custom logging handler.
+
+    Generally this is necessary when both logging output and a progress bar are both
+    outputting to ``sys.stderr``.
+
+    Note:
+        This is a :ref:`firstresult hook<firstresult>`.
+    """
+
+
+@hookspec
+def initialize(plugin_manager: pluggy.PluginManager):
+    """Called when this plugin is first loaded into OCRmyPDF.
+
+    The primary intended use of this is for plugins to check compatibility with other
+    plugins and possibly block other blocks, a plugin that wishes to block ocrmypdf's
+    built-in optimize plugin could do:
+
+    .. code-block::
+
+        plugin_manager.set_blocked('ocrmypdf.builtin_plugins.optimize')
+
+    It would also be reasonable for an plugin implementation to check if it is unable
+    to proceed, for example, because a required dependency is missing. (If the plugin's
+    ability to proceed depends on options and arguments, use ``validate`` instead.)
+
+    Raises:
+        ocrmypdf.exceptions.ExitCodeException: If options are not acceptable
+            and the application should terminate gracefully with an informative
+            message and error code.
+
+    Note:
+        This hook will be called from the main process, and may modify global state
+        before child worker processes are forked.
+    """
+
+
 @hookspec
 def add_options(parser: ArgumentParser) -> None:
     """Allows the plugin to add its own command line and API arguments.
@@ -62,6 +114,68 @@ def check_options(options: Namespace) -> None:
     """
 
 
+@hookspec(firstresult=True)
+def get_executor(progressbar_class) -> Executor:
+    """Called to obtain an object that manages parallel execution.
+
+    This may be used to replace OCRmyPDF's default parallel execution system
+    with a third party alternative. For example, you could make OCRmyPDF run in a
+    distributed environment.
+
+    OCRmyPDF's executors are analogous to the standard Python executors in
+    ``conconcurrent.futures``, but they do not work the same way. Executors may
+    be reused for different, unrelated batch operations, since all of the context
+    for a given job are passed to :meth:`Executor.__call__`.
+
+    Should be of type :class:`Executor` or otherwise conforming to the protocol
+    of that call.
+
+    Arguments:
+        progressbar_class: A progress bar class, which will be created when
+
+    Note:
+        This hook will be called from the main process, and may modify global state
+        before child worker processes are forked.
+    Note:
+        This is a :ref:`firstresult hook<firstresult>`.
+    """
+
+
+@hookspec(firstresult=True)
+def get_progressbar_class():
+    """Called to obtain a class that can be used to monitor progress.
+
+    A progress bar is assumed, but this could be used for any type of monitoring.
+
+    The class should follow a tqdm-like protocol. Calling the class should return
+    a new progress bar object, which is activated with ``__enter__`` and terminated
+    ``__exit__``. An update method is called whenever the progress bar is updated.
+    Progress bar objects will not be reused; a new one will be created for each
+    group of tasks.
+
+    The progress bar is held in the main process/thread and not updated by child
+    process/threads. When a child notifies the parent of completed work, the
+    parent updates the progress bar.
+
+    The arguments are the same as `tqdm <https://github.com/tqdm/tqdm>`_ accepts.
+
+    Progress bars should never write to ``sys.stdout``, or they will corrupt the
+    output if OCRmyPDF writes a PDF to standard output.
+
+    The type of events that OCRmyPDF reports to a progress bar may change in
+    minor releases.
+
+    Here is how OCRmyPDF will use the progress bar:
+
+    Example:
+
+        pbar_class = pm.hook.get_progressbar_class()
+        with pbar_class(**tqdm_kwargs) as pbar:
+            ...
+            pbar.update(1)
+    """
+
+
 @hookspec
 def validate(pdfinfo: 'PdfInfo', options: Namespace) -> None:
     """Called to give a plugin an opportunity to review *options* and *pdfinfo*.
@@ -119,7 +233,7 @@ def rasterize_pdf_page(
 
 
 @hookspec(firstresult=True)
-def filter_ocr_image(page: 'PageContext', image: 'Image') -> 'Image':
+def filter_ocr_image(page: 'PageContext', image: 'Image.Image') -> 'Image.Image':
     """Called to filter the image before it is sent to OCR.
 
     This is the image that OCR sees, not what the user sees when they view the
@@ -158,16 +272,16 @@ def filter_page_image(page: 'PageContext', image_filename: Path) -> Path:
     ``image_filename``. The hook may overwrite ``image_filename`` with a new file.
 
     The output image should preserve the same physical unit dimensions, that is
-    (width * dpi_x, height * dpi_y). That is, if the image is resized, the DPI
+    ``(width * dpi_x, height * dpi_y)``. That is, if the image is resized, the DPI
     must be adjusted by the reciprocal. If this is not preserved, the PDF page
-    will be resized and the OCR layer misaligned. OCRmyPDF does not nothing
+    will be resized and the OCR layer misaligned. OCRmyPDF does nothing
     to enforce these constraints; it is up to the plugin to do sensible things.
 
-    OCRmyPDF will create the PDF page based on the image format used. If you
-    convert the image to a JPEG, the output page will be created as a JPEG, etc.
-    If you change the colorspace, that change will be kept. Note that the
-    OCRmyPDF image optimization stage, if enabled, may ultimately chose a
-    different format.
+    OCRmyPDF will create the PDF page based on the image format used (unless the
+    hook is overriden). If you convert the image to a JPEG, the output page will
+    be created as a JPEG, etc. If you change the colorspace, that change will be
+    kept. Note that the OCRmyPDF image optimization stage, if enabled, may
+    ultimately chose a different format.
 
     If the return value is a file that does not exist, ``FileNotFoundError``
     will occur. The return value should be a path to a file in the same folder
@@ -185,15 +299,62 @@ def filter_page_image(page: 'PageContext', image_filename: Path) -> Path:
     """
 
 
-OrientationConfidence = namedtuple('OrientationConfidence', ('angle', 'confidence'))
-"""Expresses an OCR engine's confidence in page rotation.
+@hookspec(firstresult=True)
+def filter_pdf_page(
+    page: 'PageContext', image_filename: Path, output_pdf: Path
+) -> Path:
+    """Called to convert a filtered whole page image into a PDF.
 
-Attributes:
-    angle (int): The clockwise angle (0, 90, 180, 270) that the page should be
-        rotated. 0 means no rotation.
-    confidence (float): How confident the OCR engine is that this the correct
-        rotation. 0 is not confident, 15 is very confident. Arbitrary units.
-"""
+    A whole page image is only produced when preprocessing command line arguments
+    are issued or when ``--force-ocr`` is issued. If no whole page is image is
+    produced for a given page, this function will not be called. This is not
+    the image that will be shown to OCR. The whole page image is filtered in
+    the hook above, ``filter_page_image``, then this function is called for
+    PDF conversion.
+
+    This function will only be called when OCRmyPDF runs in a mode such as
+    "force OCR" mode where rasterizing of all content is performed.
+
+    Clever things could be done at this stage such as segmenting the page image into
+    color regions or vector equivalents.
+
+    The provider of the hook implementation is responsible for ensuring that the
+    OCR text layer is aligned with the PDF produced here, or text misalignment
+    will result.
+
+    Currently this function must produce a single page PDF or the pipeline will
+    fail.  If the intent is to remove the PDF, then create a single page empty
+    PDF.
+
+    Args:
+        page: Context for this page.
+        image_filename: Filename of the input image used to create output_pdf,
+            for "reference" if recreating the output_pdf entirely.
+        output_pdf: The previous created output_pdf.
+
+    Returns:
+        output_pdf
+
+    Note:
+        This hook will be called from child processes. Modifying global state
+        will not affect the main process or other child processes.
+    Note:
+        This is a :ref:`firstresult hook<firstresult>`.
+    """
+
+
+class OrientationConfidence(NamedTuple):
+    """Expresses an OCR engine's confidence in page rotation.
+
+    Attributes:
+        angle: The clockwise angle (0, 90, 180, 270) that the page should be
+            rotated. 0 means no rotation.
+        confidence: How confident the OCR engine is that this the correct
+            rotation. 0 is not confident, 15 is very confident. Arbitrary units.
+    """
+
+    angle: int
+    confidence: float
 
 
 class OcrEngine(ABC):
@@ -203,11 +364,13 @@ class OcrEngine(ABC):
     Tesseract OCR.
     """
 
-    @abstractstaticmethod
+    @staticmethod
+    @abstractmethod
     def version() -> str:
         """Returns the version of the OCR engine."""
 
-    @abstractstaticmethod
+    @staticmethod
+    @abstractmethod
     def creator_tag(options: Namespace) -> str:
         """Returns the creator tag to identify this software's role in creating the PDF.
 
@@ -227,24 +390,33 @@ class OcrEngine(ABC):
         to the user, usually in an error message.
         """
 
-    @abstractstaticmethod
+    @staticmethod
+    @abstractmethod
     def languages(options: Namespace) -> AbstractSet[str]:
         """Returns the set of all languages that are supported by the engine.
 
         Languages are typically given in 3-letter ISO 3166-1 codes, but actually
         can be any value understood by the OCR engine."""
 
-    @abstractstaticmethod
+    @staticmethod
+    @abstractmethod
     def get_orientation(input_file: Path, options: Namespace) -> OrientationConfidence:
         """Returns the orientation of the image."""
 
-    @abstractstaticmethod
+    @staticmethod
+    def get_deskew(input_file: Path, options: Namespace) -> float:
+        """Returns the deskew angle of the image, in degrees."""
+        return 0.0
+
+    @staticmethod
+    @abstractmethod
     def generate_hocr(
         input_file: Path, output_hocr: Path, output_text: Path, options: Namespace
     ) -> None:
         """Called to produce a hOCR file and sidecar text file."""
 
-    @abstractstaticmethod
+    @staticmethod
+    @abstractmethod
     def generate_pdf(
         input_file: Path, output_pdf: Path, output_text: Path, options: Namespace
     ) -> None:
@@ -264,8 +436,7 @@ def get_ocr_engine() -> OcrEngine:
     """Returns an OcrEngine to use for processing this file.
 
     The OcrEngine may be instantiated multiple times, by both the main process
-    and child process. As such, it must be obtain store any state in ``options``
-    or some common location.
+    and child process.
 
     Note:
         This is a :ref:`firstresult hook<firstresult>`.
@@ -321,3 +492,63 @@ def generate_pdfa(
     See also:
         https://github.com/tqdm/tqdm
     """
+
+
+@hookspec(firstresult=True)
+def optimize_pdf(
+    input_pdf: Path,
+    output_pdf: Path,
+    context: PdfContext,
+    executor: Executor,
+    linearize: bool,
+) -> Tuple[Path, Sequence[str]]:
+    """Optimize a PDF after image, OCR and metadata processing.
+
+    If the input_pdf is a PDF/A, the plugin should modify input_pdf in a way
+    that preserves the PDF/A status, or report to the user when this is not possible.
+
+    If the implementation fails to produce a smaller file than the input file, it
+    should return input_pdf instead.
+
+    A plugin that implements a new optimizer may need to suppress the built-in
+    optimizer by implementing an ``initialize`` hook.
+
+    Arguments:
+        input_pdf: The input PDF, which has OCR added.
+        output_pdf: The requested filename of the output PDF which should be created
+            by this optimization hook.
+        context: The current context.
+        executor: An initialized executor which may be used during optimization,
+            to distribute optimization tasks.
+        linearize: If True, OCRmyPDF requires ``optimize_pdf`` to return a linearized,
+            also known as fast web view PDF.
+
+    Returns:
+        Path: If optimization is successful, the hook should return ``output_file``.
+            If optimization does not produce a smaller file, the hook should return
+            ``input_file``.
+        Sequence[str]: Any comments that the plugin wishes to report to the user,
+            especially reasons it was not able to further optimize the file. For
+            example, the plugin could report that a required third party was not
+            installed, so a specific optimization was not attempted.
+
+    Note:
+        This is a :ref:`firstresult hook<firstresult>`.
+    """
+
+
+@hookspec(firstresult=True)
+def is_optimization_enabled(context: PdfContext) -> bool:
+    """For a given PdfContext, OCRmyPDF asks the plugin if optimization is enabled.
+
+    An optimization plugin might be installed and active but could be disabled by
+    user settings.
+
+    If this returns False, OCRmyPDF will take certain actions to finalize the PDF.
+
+    Returns:
+        True if the plugin's optimization is enabled.
+
+    Note:
+        This is a :ref:`firstresult hook<firstresult>`.
+    """

src/ocrmypdf/subprocess/__init__.py

@@ -13,18 +13,28 @@ import re
 import sys
 from collections.abc import Mapping
 from contextlib import suppress
-from distutils.version import LooseVersion
 from functools import lru_cache
-from pathlib import Path
 from subprocess import PIPE, STDOUT, CalledProcessError, CompletedProcess, Popen
 from subprocess import run as subprocess_run
+from typing import Callable, Optional, Type, Union
+
+from packaging.version import Version
 
 from ocrmypdf.exceptions import MissingDependencyError
 
+# pylint: disable=logging-format-interpolation
+
 log = logging.getLogger(__name__)
 
 
-def run(args, *, env=None, logs_errors_to_stdout=False, **kwargs):
+def run(
+    args,
+    *,
+    env=None,
+    logs_errors_to_stdout: bool = False,
+    check: bool = False,
+    **kwargs,
+) -> CompletedProcess:
     """Wrapper around :py:func:`subprocess.run`
 
     The main purpose of this wrapper is to log subprocess output in an orderly
@@ -45,7 +55,7 @@ def run(args, *, env=None, logs_errors_to_stdout=False, **kwargs):
     stderr = None
     stderr_name = 'stderr' if not logs_errors_to_stdout else 'stdout'
     try:
-        proc = subprocess_run(args, env=env, **kwargs)
+        proc = subprocess_run(args, env=env, check=check, **kwargs)
     except CalledProcessError as e:
         stderr = getattr(e, stderr_name, None)
         raise
@@ -62,7 +72,9 @@ def run(args, *, env=None, logs_errors_to_stdout=False, **kwargs):
     return proc
 
 
-def run_polling_stderr(args, *, callback, check=False, env=None, **kwargs):
+def run_polling_stderr(
+    args, *, callback: Callable[[str], None], check: bool = False, env=None, **kwargs
+) -> CompletedProcess:
     """Run a process like ``ocrmypdf.subprocess.run``, and poll stderr.
 
     Every line of produced by stderr will be forwarded to the callback function.
@@ -80,6 +92,8 @@ def run_polling_stderr(args, *, callback, check=False, env=None, **kwargs):
     with Popen(args, env=env, **kwargs) as proc:
         lines = []
         while proc.poll() is None:
+            if proc.stderr is None:
+                continue
             for msg in iter(proc.stderr.readline, ''):
                 if process_log.isEnabledFor(logging.DEBUG):
                     process_log.debug(msg.strip())
@@ -99,9 +113,10 @@ def _fix_process_args(args, env, kwargs):
         env = os.environ
 
     # Search in spoof path if necessary
-    program = args[0]
+    program = str(args[0])
 
     if os.name == 'nt':
+        # pylint: disable=import-outside-toplevel
         from ocrmypdf.subprocess._windows import fix_windows_args
 
         args = fix_windows_args(program, args, env)
@@ -109,22 +124,14 @@ def _fix_process_args(args, env, kwargs):
     log.debug("Running: %s", args)
     process_log = log.getChild(os.path.basename(program))
     text = kwargs.get('text', False)
-    if sys.version_info < (3, 7):
-        if os.name == 'nt':
-            # Can't use close_fds=True on Windows with Python 3.6 or older
-            # https://bugs.python.org/issue19575, etc.
-            kwargs['close_fds'] = False
-        if 'text' in kwargs:
-            # Convert run(...text=) to run(...universal_newlines=) for Python 3.6
-            kwargs['universal_newlines'] = kwargs['text']
-            del kwargs['text']
+
     return args, env, process_log, text
 
 
 @lru_cache(maxsize=None)
 def get_version(
     program: str, *, version_arg: str = '--version', regex=r'(\d+(\.\d+)*)', env=None
-):
+) -> str:
     """Get the version of the specified program
 
     Arguments:
@@ -145,7 +152,7 @@ def get_version(
             check=True,
             env=env,
         )
-        output = proc.stdout
+        output: str = proc.stdout
     except FileNotFoundError as e:
         raise MissingDependencyError(
             f"Could not find program '{program}' on the PATH"
@@ -170,42 +177,42 @@ def get_version(
     return version
 
 
-missing_program = '''
+MISSING_PROGRAM = '''
 The program '{program}' could not be executed or was not found on your
 system PATH.
 '''
 
-missing_optional_program = '''
+MISSING_OPTIONAL_PROGRAM = '''
 The program '{program}' could not be executed or was not found on your
 system PATH.  This program is required when you use the
 {required_for} arguments.  You could try omitting these arguments, or install
 the package.
 '''
 
-missing_recommend_program = '''
+MISSING_RECOMMEND_PROGRAM = '''
 The program '{program}' could not be executed or was not found on your
 system PATH.  This program is recommended when using the {required_for} arguments,
 but not required, so we will proceed.  For best results, install the program.
 '''
 
-old_version = '''
+OLD_VERSION = '''
 OCRmyPDF requires '{program}' {need_version} or higher.  Your system appears
 to have {found_version}.  Please update this program.
 '''
 
-old_version_required_for = '''
+OLD_VERSION_REQUIRED_FOR = '''
 OCRmyPDF requires '{program}' {need_version} or higher when run with the
 {required_for} arguments.  If you omit these arguments, OCRmyPDF may be able to
 proceed.  For best results, install the program.
 '''
 
-osx_install_advice = '''
+OSX_INSTALL_ADVICE = '''
 If you have homebrew installed, try these command to install the missing
 package:
     brew install {package}
 '''
 
-linux_install_advice = '''
+LINUX_INSTALL_ADVICE = '''
 On systems with the aptitude package manager (Debian, Ubuntu), try these
 commands:
     sudo apt-get update
@@ -215,7 +222,7 @@ On RPM-based systems (Red Hat, Fedora), search for instructions on
 installing the RPM for {program}.
 '''
 
-windows_install_advice = '''
+WINDOWS_INSTALL_ADVICE = '''
 If not already installed, install the Chocolatey package manager. Then use
 a command prompt to install the missing package:
     choco install {package}
@@ -233,66 +240,95 @@ def _get_platform():
 
 
 def _error_trailer(program, package, **kwargs):
+    del kwargs
     if isinstance(package, Mapping):
         package = package.get(_get_platform(), program)
 
     if _get_platform() == 'darwin':
-        log.info(osx_install_advice.format(**locals()))
+        log.info(OSX_INSTALL_ADVICE.format(**locals()))
     elif _get_platform() == 'linux':
-        log.info(linux_install_advice.format(**locals()))
+        log.info(LINUX_INSTALL_ADVICE.format(**locals()))
     elif _get_platform() == 'windows':
-        log.info(windows_install_advice.format(**locals()))
+        log.info(WINDOWS_INSTALL_ADVICE.format(**locals()))
 
 
 def _error_missing_program(program, package, required_for, recommended):
+    # pylint: disable=unused-argument
     if recommended:
-        log.warning(missing_recommend_program.format(**locals()))
+        log.warning(MISSING_RECOMMEND_PROGRAM.format(**locals()))
     elif required_for:
-        log.error(missing_optional_program.format(**locals()))
+        log.error(MISSING_OPTIONAL_PROGRAM.format(**locals()))
     else:
-        log.error(missing_program.format(**locals()))
+        log.error(MISSING_PROGRAM.format(**locals()))
     _error_trailer(**locals())
 
 
 def _error_old_version(program, package, need_version, found_version, required_for):
+    # pylint: disable=unused-argument
     if required_for:
-        log.error(old_version_required_for.format(**locals()))
+        log.error(OLD_VERSION_REQUIRED_FOR.format(**locals()))
     else:
-        log.error(old_version.format(**locals()))
+        log.error(OLD_VERSION.format(**locals()))
     _error_trailer(**locals())
 
 
+def _remove_leading_v(s):
+    if sys.version_info >= (3, 9):
+        return s.removeprefix('v')
+
+    if s.startswith('v'):
+        return s[1:]
+    return s
+
+
 def check_external_program(
     *,
-    program,
-    package,
-    version_checker,
-    need_version,
-    required_for=None,
+    program: str,
+    package: str,
+    version_checker: Callable,
+    need_version: str,
+    required_for: Optional[str] = None,
     recommended=False,
+    version_parser: Type[Version] = Version,
 ):
+    """Check for required version of external program and raise exception if not.
+
+    Args:
+        program: The name of the program to test.
+        package: The name of a software package that typically supplies this program.
+            Usually the same as program.
+        version_check: A callable without arguments that retrieves the installed
+            version of program.
+        need_version: The minimum required version.
+        required_for: The name of an argument of feature that requires this program.
+        recommended: If this external program is recommended, instead of raising
+            an exception, log a warning and allow execution to continue.
+        version_parser: A class that should be used to parse and compare version
+            numbers. Used when version numbers do not follow standard conventions.
+    """
+
     try:
         if callable(version_checker):
             found_version = version_checker()
-        else:
+        else:  # deprecated
             found_version = version_checker
-    except (CalledProcessError, FileNotFoundError, MissingDependencyError):
+    except (CalledProcessError, FileNotFoundError) as e:
         _error_missing_program(program, package, required_for, recommended)
         if not recommended:
-            raise MissingDependencyError()
+            raise MissingDependencyError(program) from e
+        return
+    except MissingDependencyError:
+        _error_missing_program(program, package, required_for, recommended)
+        if not recommended:
+            raise
         return
 
-    def remove_leading_v(s):
-        if s.startswith('v'):
-            return s[1:]
-        return s
+    found_version = _remove_leading_v(found_version)
+    need_version = _remove_leading_v(need_version)
 
-    found_version = remove_leading_v(found_version)
-    need_version = remove_leading_v(need_version)
-
-    if found_version and LooseVersion(found_version) < LooseVersion(need_version):
+    if found_version and version_parser(found_version) < version_parser(need_version):
         _error_old_version(program, package, need_version, found_version, required_for)
         if not recommended:
-            raise MissingDependencyError()
+            raise MissingDependencyError(program)
 
     log.debug('Found %s %s', program, found_version)

src/ocrmypdf/subprocess/_windows.py

@@ -4,37 +4,52 @@
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
+# type: ignore
+# Non-Windows mypy now breaks when trying to typecheck winreg
+
+"""Find Tesseract and Ghostscript binaries on Windows using the registry."""
+
 import logging
 import os
 import shutil
 import sys
-from distutils.version import LooseVersion
-from itertools import chain, filterfalse
+from itertools import chain
 from pathlib import Path
-from typing import Any, Callable, Iterator, Optional, Tuple, TypeVar, cast
+from typing import Any, Callable, Iterable, Iterator, Set, Tuple, TypeVar
 
 try:
     import winreg
-except ModuleNotFoundError as e:
-    raise ModuleNotFoundError("This module is for Windows only") from e
+except ModuleNotFoundError as _notfound_ex:
+    raise ModuleNotFoundError("This module is for Windows only") from _notfound_ex
 
 log = logging.getLogger(__name__)
 
 T = TypeVar('T')
 
 
+def ghostscript_version_key(s: str) -> Tuple[int, int, int]:
+    """Compare Ghostscript version numbers."""
+    try:
+        release = [int(elem) for elem in s.split('.', maxsplit=3)]
+        while len(release) < 3:
+            release.append(0)
+        return (release[0], release[1], release[2])
+    except ValueError:
+        return (0, 0, 0)
+
+
 def registry_enum(
     key: winreg.HKEYType, enum_fn: Callable[[winreg.HKEYType, int], T]
 ) -> Iterator[T]:
-    LIMIT = 999
+    limit = 999
     n = 0
-    while n < LIMIT:
+    while n < limit:
         try:
             yield enum_fn(key, n)
             n += 1
         except OSError:
             break
-    if n == LIMIT:
+    if n == limit:
         raise ValueError(f"Too many registry keys under {key}")
 
 
@@ -47,11 +62,14 @@ def registry_values(key: winreg.HKEYType) -> Iterator[Tuple[str, Any, int]]:
 
 
 def registry_path_ghostscript(env=None) -> Iterator[Path]:
+    del env  # unused (but needed for protocol)
     try:
         with winreg.OpenKey(
             winreg.HKEY_LOCAL_MACHINE, r"SOFTWARE\Artifex\GPL Ghostscript"
         ) as k:
-            latest_gs = max(registry_subkeys(k), key=LooseVersion)
+            latest_gs = max(
+                registry_subkeys(k), key=ghostscript_version_key, default=(0, 0, 0)
+            )
         with winreg.OpenKey(
             winreg.HKEY_LOCAL_MACHINE, fr"SOFTWARE\Artifex\GPL Ghostscript\{latest_gs}"
         ) as k:
@@ -62,6 +80,7 @@ def registry_path_ghostscript(env=None) -> Iterator[Path]:
 
 
 def registry_path_tesseract(env=None) -> Iterator[Path]:
+    del env  # unused (but needed for protocol)
     try:
         with winreg.OpenKey(winreg.HKEY_LOCAL_MACHINE, r"SOFTWARE\Tesseract-OCR") as k:
             for subkey, val, _valtype in registry_values(k):
@@ -113,7 +132,7 @@ SHIMS = [
 ]
 
 
-def fix_windows_args(program, args, env):
+def fix_windows_args(program: str, args, env):
     """Adjust our desired program and command line arguments for use on Windows"""
 
     if sys.version_info < (3, 8):
@@ -137,14 +156,12 @@ def fix_windows_args(program, args, env):
     return args
 
 
-def unique_everseen(iterable, key=None):
-    "List unique elements, preserving order. Remember all elements ever seen."
+def unique_everseen(iterable: Iterable[T], key: Callable[[T], T]) -> Iterator[T]:
+    "List unique elements, preserving order."
     # unique_everseen('AAAABBBCCDAABBB') --> A B C D
     # unique_everseen('ABBCcAD', str.lower) --> A B C D
-    seen = set()
+    seen: Set[T] = set()
     seen_add = seen.add
-    if key is None:
-        key = lambda x: x
     for element in iterable:
         k = key(element)
         if k not in seen:

tests/__init__.py

@@ -1,8 +1,7 @@
-# © 2017 James R. Barlow: github.com/jbarlow83
+# © 2021 James R. Barlow: github.com/jbarlow83
 #
 # This Source Code Form is subject to the terms of the Mozilla Public
 # License, v. 2.0. If a copy of the MPL was not distributed with this
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
-
-"""Bindings to external libraries"""
+# Empty __init__.py file

tests/cache/2400dpi/__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt/stderr.bin

@@ -1 +1 @@
-Tesseract Open Source OCR Engine v4.1.1 with Leptonica
+Tesseract Open Source OCR Engine v5.0.0-beta-20210916-12-g19cc9 with Leptonica

tests/cache/2400dpi/__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt/txt.bin

@@ -1,4 +1,6 @@
-a la Waterman
+i a la Waterman
+
+ 
 
 4 ons linzen
 
@@ -12,14 +14,14 @@ bloem, boter
 
 laurier, kruidnagel, kerrie, zout
 
-De linzgen wassen en in -l liter kokend wa-
+De linzgen wassen en in-l liter kokend wa-
 ter 1 dag laten weken, 2 liter water bij
 de linzen voegen, zonder het water waarin
-ze geweekt zijn af te gieten, De helft van
-de uien bakken met laurier en Kruicdnagel.
-Alle uien, kerrie en gout bij de linzen
-voegen, Alles aan de kook brengen,. Van de
+ze geweekt zijn af te gieten., De helft van
+de uien bakken met laurier en Kruidnagel.
+Alle uien, kerrie en zgout bij de linzen
+voegen, Alles aan de kook brengen, Van de
 bloem met boter en melk een papje maken en
 verder afmaken met de soep, Als de linzen
-gaar Zijn is de soep klaar.
+gfgaar Zijn is de soep klaar.
 

tests/cache/3small/__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt/stderr.bin

@@ -1 +1 @@
-Tesseract Open Source OCR Engine v4.1.1 with Leptonica
+Tesseract Open Source OCR Engine v5.0.0-beta-20210916-12-g19cc9 with Leptonica

tests/cache/3small/__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt/txt.bin

@@ -22,9 +22,9 @@ Mugerre
 
 Milafranga Komunikabideak
 
-BAIONA i zeettnansise —
+BAIONA  zeiteninsiie —
 
-1 Trenbideak -- ~~~
+7 Trenbideak -----
 
-t\ Basusarri — spmsans20141004 se: . a ~
+t\ Basusarri — spmeans:20141004 ae: . _ ~
 

tests/cache/3small/__-l__eng__000002_ocr.png__000002_ocr_tess__pdf__txt/stderr.bin

@@ -1 +1 @@
-Tesseract Open Source OCR Engine v4.1.1 with Leptonica
+Tesseract Open Source OCR Engine v5.0.0-beta-20210916-12-g19cc9 with Leptonica

tests/cache/3small/__-l__eng__000003_ocr.png__000003_ocr_tess__pdf__txt/stderr.bin

@@ -1 +1 @@
-Tesseract Open Source OCR Engine v4.1.1 with Leptonica
+Tesseract Open Source OCR Engine v5.0.0-beta-20210916-12-g19cc9 with Leptonica

tests/cache/3small/__-l__eng__000003_ocr.png__000003_ocr_tess__pdf__txt/txt.bin

@@ -14,14 +14,14 @@ bloem, boter
 
 laurier, kruidnagel, kerrie, zout
 
-De linzgen wassen en in -l liter kokend wa-
+De linzgen wassen en in-l liter kokend wa-
 ter 1 dag laten weken, 2 liter water bij
 de linzen voegen, zonder het water waarin
-ze geweekt zijn af te gieten, De helft van
-de uien bakken met laurier en Kruicdnagel.
-Alle uien, kerrie en gout bij de linzen
-voegen, Alles aan de kook brengen,. Van de
+ze geweekt zijn af te gieten., De helft van
+de uien bakken met laurier en Kruidnagel.
+Alle uien, kerrie en zgout bij de linzen
+voegen, Alles aan de kook brengen, Van de
 bloem met boter en melk een papje maken en
 verder afmaken met de soep, Als de linzen
-gaar Zijn is de soep klaar.
+gfgaar Zijn is de soep klaar.