Merge branch 'master' into scriptsender

Update pyproj
2021-02-18 01:48:00 -08:00 · 2021-02-18 01:08:49 -08:00
253 changed files with 11014 additions and 11058 deletions
@@ -1,28 +1,17 @@
 # OCRmyPDF
 #
-
-FROM ubuntu:21.04 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 ubuntu:20.04 as base

 FROM base as builder

-# Note we need leptonica here to build jbig2
+ENV LANG=C.UTF-8
+
 RUN apt-get update && apt-get install -y --no-install-recommends \
  build-essential autoconf automake libtool \
  libleptonica-dev \
  zlib1g-dev \
-  python3-dev \
+  python3 \
  python3-distutils \
-  libffi-dev \
  ca-certificates \
  curl \
  git
@@ -46,15 +35,26 @@ COPY . /app

 WORKDIR /app

-RUN pip3 install --no-cache-dir .[test,webservice,watcher]
+RUN pip3 install --no-cache-dir \
+  -r requirements/main.txt \
+  -r requirements/webservice.txt \
+  -r requirements/test.txt \
+  -r requirements/watcher.txt \
+  .

 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 \
@@ -62,8 +62,7 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
  tesseract-ocr-fra \
  tesseract-ocr-por \
  tesseract-ocr-spa \
-  unpaper \
-  && rm -rf /var/lib/apt/lists/*
+  unpaper

 WORKDIR /app

@@ -75,6 +74,8 @@ 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"]
@@ -14,14 +14,7 @@ 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. 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/jbarlow83/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.
+A clear and concise description of any alternative solutions or features you've considered.

 **Additional context**
 Add any other context or screenshots about the feature request here.
@@ -1,301 +0,0 @@
-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-latest
-            python: 3.9
-          - os: ubuntu-latest
-            python: "pypy-3.7"
-          - 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"]
-
-    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"]
-
-    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 pngquant
-
-      - 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 .
@@ -6,9 +6,8 @@
 !.gitattributes
 !.gitignore
 !.pre-commit-config.yaml
-!.readthedocs.yaml
+!.readthedocs.yml
 !.github/
-!.docker/

 # Dev scratch
 *.ipynb
@@ -25,7 +24,6 @@ venv*/
 /debug_tests.py
 *.traineddata
 /private
-/coverage.xml

 # Package building
 *.egg-info/
@@ -1,37 +1,23 @@
 repos:
  - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.0.1
+    rev: v3.4.0
    hooks:
      - id: check-case-conflict
      - id: check-merge-conflict
      - id: check-toml
      - id: check-yaml
      - id: debug-statements
-  - repo: https://github.com/pycqa/isort
-    rev: 5.9.3
+  - 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
    hooks:
      - id: isort
-        args: ["--profile", "black"]
  - repo: https://github.com/psf/black
-    rev: 21.9b0
+    rev: 20.8b1
    hooks:
      - id: black
        language_version: python
-  - repo: https://github.com/asottile/setup-cfg-fmt
-    rev: v1.19.0
-    hooks:
-      - id: setup-cfg-fmt
-  - repo: https://github.com/asottile/pyupgrade
-    rev: v2.29.0
-    hooks:
-      - id: pyupgrade
-        args: ["--py37-plus"]
-  - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v0.910-1
-    hooks:
-      - id: mypy
-        additional_dependencies:
-          - types-toml
-          - types-setuptools
-          - types-requests
-          - types-Pillow
+        exclude: ^src/ocrmypdf/lib/_leptonica.py
@@ -1,22 +0,0 @@
-# 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
@@ -0,0 +1,10 @@
+build:
+  image: latest
+
+python:
+  version: 3.6
+
+formats:
+  - pdf
+
+requirements_file: requirements/main.txt
@@ -1,6 +1,6 @@
 <img src="docs/images/logo.svg" width="240" alt="OCRmyPDF">

-[![Build Status](https://github.com/jbarlow83/OCRmyPDF/actions/workflows/build.yml/badge.svg)](https://github.com/jbarlow83/OCRmyPDF/actions/workflows/build.yml) [![PyPI version][pypi]](https://pypi.org/project/ocrmypdf/) ![Homebrew version][homebrew] ![ReadTheDocs][docs] ![Python versions][pyversions]
+[![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]

 [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"
@@ -57,17 +57,25 @@ 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, for both x64 and ARM.
+Linux, Windows, macOS and FreeBSD are supported. Docker images are also available.

-| Operating system              | Install command               |
-| ----------------------------- | ------------------------------|
-| 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`` |
-| Conda                         | ``conda install ocrmypdf``    |
+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
+```

 For everyone else, [see our documentation](https://ocrmypdf.readthedocs.io/en/latest/installation.html) for installation steps.

@@ -91,11 +99,6 @@ 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:
@@ -110,7 +113,7 @@ Please report issues on our [GitHub issues](https://github.com/jbarlow83/OCRmyPD

 ## Requirements

-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.
+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.

 ## Press & Media

@@ -120,7 +123,6 @@ In addition to the required Python version (3.7+), 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

@@ -0,0 +1,252 @@
+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')))
@@ -60,6 +60,11 @@ 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
@@ -228,14 +228,20 @@ 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 (specifically, it supports the ISO 8859-1
-character). Pull requests to improve the situation are welcome.
+text outside the Latin alphabet. 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
 ==================

@@ -12,7 +12,7 @@ subprocess call anyway, as this provides isolation of its activities.
 Example
 =======

-OCRmyPDF provides one high-level function to run its main engine from an
+OCRmyPDF 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 some exceptions, all of the command line arguments are available
+With a few 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.
@@ -41,29 +41,29 @@ 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 :func:`ocrmypdf.ocr()` must be sufficiently
+The Python process that calls ``ocrmypdf.ocr()`` must be sufficiently
 privileged to perform these actions.

-There currently is no option to manage how jobs are scheduled other
+There is no currently 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 :func:`ocrmypdf.ocr()` is suggested. That
+Creating a child process to call ``ocrmypdf.ocr()`` is suggested. That
 way your application will survive and remain interactive even if
 OCRmyPDF fails for any reason.

-Programs that call :func:`ocrmypdf.ocr()` should also install a SIGBUS signal
+Programs that call ``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.

-:func:`ocrmypdf.ocr()` will take a threading lock to prevent multiple runs of itself
+``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 :func:`ocrmypdf.ocr()` must be
+    On Windows and macOS, the script that calls ``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.
@@ -96,7 +96,7 @@ Exceptions

 OCRmyPDF may throw standard Python exceptions, ``ocrmypdf.exceptions.*``
 exceptions, some exceptions related to multiprocessing, and
-:exc:`KeyboardInterrupt`. The parent process should provide an exception
+``KeyboardInterrupt``. The parent process should provide an exception
 handler. OCRmyPDF will clean up its temporary files and worker processes
 automatically when an exception occurs.

@@ -111,7 +111,7 @@ Users may need to customize the script to meet their requirements.

 .. code-block:: bash

-    pip3 install ocrmypdf[watcher]
+    pip3 install -r requirements/watcher.txt

    env OCR_INPUT_DIRECTORY=/mnt/input-pdfs \
        OCR_OUTPUT_DIRECTORY=/mnt/output-pdfs \
@@ -202,6 +202,13 @@ 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
 ===============

@@ -31,18 +31,9 @@
 # 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.autodoc',
-    'sphinx.ext.intersphinx',
-    'sphinx.ext.autosummary',
-    'sphinx.ext.napoleon',
-    'sphinx_issues',
-]
+extensions = ['sphinx.ext.napoleon']

-# Extension settings
-intersphinx_mapping = {'https://docs.python.org/': None}
 napoleon_use_rtype = False
-issues_github_path = "jbarlow83/OCRmyPDF"

 # Add any paths that contain templates here, relative to this directory.
 templates_path = ['_templates']
@@ -63,7 +54,7 @@ master_doc = 'index'
 # General information about the project.
 project = 'ocrmypdf'
 copyright = (
-    '2021, James R. Barlow. Licensed under Creative Commons Attribution-ShareAlike 4.0.'
+    '2020, James R. Barlow. Licensed under Creative Commons Attribution-ShareAlike 4.0.'
 )
 author = 'James R. Barlow'

@@ -91,16 +82,15 @@ if on_rtd:
        'pikepdf',
        'pikepdf.models',
        'pikepdf.models.metadata',
+        'ocrmypdf.leptonica',
    ]
    sys.modules.update((mod_name, Mock()) for mod_name in MOCK_MODULES)

-try:
-    from importlib_metadata import version as package_version
-except ModuleNotFoundError:
-    from importlib.metadata import version as package_version
+
+from pkg_resources import get_distribution, DistributionNotFound

 # The full version, including alpha/beta/rc tags.
-release = package_version('ocrmypdf')
+release = get_distribution('ocrmypdf').version
 version = '.'.join(release.split('.')[:2])


@@ -283,7 +273,7 @@ htmlhelp_basename = 'ocrmypdfdoc'

 # -- Options for LaTeX output ---------------------------------------------

-latex_elements = {  # type: ignore
+latex_elements = {
    # The paper size ('letterpaper' or 'a4paper').
    #
    # 'papersize': 'letterpaper',
@@ -46,6 +46,17 @@ 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
 =====================

@@ -58,11 +58,9 @@ 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. 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.
+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.

 If the page is "just a little off horizontal", like a crooked picture,
 then you want ``--deskew``. ``--rotate-pages`` is for when the cardinal
@@ -104,9 +102,6 @@ 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``.

@@ -184,7 +179,10 @@ 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.
+   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>`__.
 -  ``--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
@@ -103,41 +103,16 @@ 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

-   # Example: add Italian
-   RUN apt install tesseract-ocr-ita
+   # Add French
+   RUN apt install tesseract-ocr-fra

-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/
+You can also copy training data to ``/usr/share/tesseract-ocr/<tesseract version>/tessdata``.

 Executing the test suite
 ========================
@@ -1,233 +0,0 @@
-<?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>
@@ -1,8 +1,6 @@
 OCRmyPDF documentation
 ======================

-.. figure:: images/logo.svg
-
 OCRmyPDF adds an optical character recognition (OCR) text layer to scanned PDF
 files, allowing them to be searched.

@@ -40,7 +38,6 @@ image processing and OCR to existing PDFs.
   plugins
   apiref
   contributing
-   maintainers

 Indices and tables
 ==================
@@ -12,21 +12,19 @@ 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 py38-ocrmypdf`` |
-+-------------------------------+-------------------------------+
-| Conda (WSL, macOS, Linux)     | ``conda install 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 py37-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.
@@ -56,9 +54,6 @@ 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-2110| image:: https://repology.org/badge/version-for-repo/ubuntu_21_10/ocrmypdf.svg
-    :alt: Ubuntu 21.10
-
 +-----------------------------------------------+
 | **OCRmyPDF versions in Debian & Ubuntu**      |
 +-----------------------------------------------+
@@ -66,7 +61,7 @@ Debian and Ubuntu 18.04 or newer
 +-----------------------------------------------+
 | |deb-stable| |deb-testing| |deb-unstable|     |
 +-----------------------------------------------+
-| |ubu-1804| |ubu-2004| |ubu-2110|              |
+| |ubu-1804| |ubu-2004|                         |
 +-----------------------------------------------+

 Users of Debian 9 ("stretch") or later, or Ubuntu 18.04 or later, including users
@@ -95,14 +90,14 @@ For full details on version availability for your platform, check the
   automatically detect it (specifically the ``jbig2`` binary) on the
   ``PATH``. To add JBIG2 encoding, see :ref:`jbig2`.

-Fedora
------
+Fedora 29 or newer
+------------------

-.. |fedora-34| image:: https://repology.org/badge/version-for-repo/fedora_34/ocrmypdf.svg
-    :alt: Fedora 34
+.. |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-rawhide| image:: https://repology.org/badge/version-for-repo/fedora_rawhide/ocrmypdf.svg
    :alt: Fedore Rawhide
@@ -112,7 +107,7 @@ Fedora
 +-----------------------------------------------+
 | |latest|                                      |
 +-----------------------------------------------+
-| |fedora-34| |fedora-35| |fedora-rawhide|      |
+| |fedora-31| |fedora-32| |fedora-rawhide|      |
 +-----------------------------------------------+

 Users of Fedora 29 or later may simply
@@ -152,6 +147,7 @@ ocrmypdf, and install the following dependencies:
    sudo apt-get -y install \
        ghostscript \
        icc-profiles-free \
+        liblept5 \
        libxml2 \
        pngquant \
        python3-pip \
@@ -162,7 +158,7 @@ To install ocrmypdf for the system:

 .. code-block:: bash

-    pip3 install ocrmypdf
+    sudo pip3 install ocrmypdf

 To install for the current user only:

@@ -185,8 +181,10 @@ 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 \
@@ -211,6 +209,69 @@ 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)
 ----------------

@@ -294,8 +355,7 @@ To install OCRmyPDF for Alpine Linux:
 Mageia 7
 --------

-There is no OS-level packaging available for Mageia, so you must install the
-dependencies:
+Install the following dependencies:

 .. code-block:: bash

@@ -305,8 +365,10 @@ dependencies:
        ghostscript \
        icc-profiles-openicc \
        jbig2dec \
+        lib64leptonica5 \
        pngquant \
        python3-pip \
+        python3-cffi \
        python3-distutils-extra \
        python3-pkg-resources \
        python3-reportlab \
@@ -318,16 +380,12 @@ dependencies:

 To install ocrmypdf for the system:

-.. code-block:: bash
-
    # As root user
    pip3 install ocrmypdf
    ldconfig

 Or, to install for the current user only:

-.. code-block:: bash
-
    export PATH=$HOME/.local/bin:$PATH
    pip3 install --user ocrmypdf

@@ -396,7 +454,7 @@ 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 ``.workflows/build.yml``.
+dependencies. You could also check the ``azure-pipelines.yml``.

 This will include the English, French, German and Spanish language
 packs. If you need other languages you can optionally install them all:
@@ -478,15 +536,14 @@ to change the PATH.
 .. warning::

    As of early 2021, users have reported problems with the Microsoft Store version of
-    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.
+    Python affected most third party Python packages including OCRmyPDF. Please use
+    Python downloaded from Python.org or Chocolatey as recommended here.

 Windows Subsystem for Linux
 ---------------------------

-#. Install Ubuntu 20.04 for Windows Subsystem for Linux, if not already installed.
-#. Follow the procedure to install :ref:`OCRmyPDF on Ubuntu 20.04 <ubuntu-lts-latest>`.
+#. 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>`.
 #. Open the Windows command prompt and create a symlink:

 .. code-block:: powershell
@@ -507,7 +564,7 @@ Cygwin64

 First install the the following prerequisite Cygwin packages using ``setup-x86_64.exe``::

-    python37 (or later)
+    python36 (or later)
    python3?-devel
    python3?-pip
    python3?-lxml
@@ -567,7 +624,7 @@ versions likely work but have not been tested.

 .. code-block:: bash

-    pkg install py38-ocrmypdf
+    pkg install py37-ocrmypdf

 To install a more recent version, you could attempt to first install the system
 version with ``pkg``, then use ``pip install --user ocrmypdf``.
@@ -588,21 +645,6 @@ 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
@@ -643,14 +685,16 @@ 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 3.7 or newer
+-  Python 3.6 or newer
 -  Ghostscript 9.15 or newer
+-  qpdf 8.1.0 or newer
 -  Tesseract 4.0.0-beta or newer

 As of ocrmypdf 7.2.1, the following versions are recommended:

-  Python 3.9 or newer
+-  Python 3.7 or 3.8
 -  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
@@ -682,7 +726,7 @@ unfortunately, the ``pip install`` command cannot satisfy all of them.
 Installing HEAD revision from sources
 =====================================

-If you have ``git`` and Python 3.7 or newer installed, you can install
+If you have ``git`` and Python 3.6 or newer installed, you can install
 from source. When the ``pip`` installer runs, it will alert you if
 dependencies are missing.

@@ -742,7 +786,8 @@ To install all of the development and test requirements:
    python3 -m venv
    source venv/bin/activate
    cd OCRmyPDF
-    pip install -e .[test]
+    pip install -e .
+    pip install -r requirements/dev.txt -r requirements/test.txt

 To add JBIG2 encoding, see :ref:`jbig2`.

@@ -2,12 +2,7 @@
 Introduction
 ============

-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.
+OCRmyPDF is a Python 3 application and library that adds OCR layers to PDFs.

 About OCR
 =========
@@ -31,7 +26,7 @@ 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 format for exchanging scanned documents.
+and text, it is a good formats for exchanging scanned documents.

 |image|

@@ -40,9 +35,9 @@ 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 corresponding raster images.
-OCR engines like Tesseract work with images, not scalable vector graphics
-or mixed raster-vector-text graphics such as PDF.
+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.

 About PDF/A
 ===========
@@ -81,7 +76,7 @@ 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 the rasterized image to create an OCR "layer".
+then performs on OCR on 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
@@ -89,9 +84,9 @@ 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 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.
+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.

 By default, OCRmyPDF produces archival PDFs – PDF/A, which are a
 stricter subset of PDF features designed for long term archives. If
@@ -144,7 +139,7 @@ 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 OCR solutions.
+-  The OCR is not as accurate as commercial solutions such as Abbyy.
 -  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
@@ -212,9 +207,8 @@ consider one of these similar open source programs:

 -  pdf2pdfocr
 -  pdfsandwich
-
-Ghostscript recently added three "pdfocr" output devices. They work by
-rasterizing all content and converting all pages to a single colour space.
+-  pypdfocr
+-  pdfbeads

 Web front-ends
 ==============
@@ -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 is not available, lower quality
+create smaller PDFs. If JBIG2 encoding 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 and has been part of the PDF
+PDF viewers. It is widely supported has been part of the PDF
 specification since 2001.

 On macOS, Homebrew packages jbig2enc and OCRmyPDF includes it by
@@ -37,7 +37,7 @@ Lossy mode JBIG2

 OCRmyPDF provides lossy mode JBIG2 as an advanced feature. Users should
 `review the technical concerns with JBIG2 in lossy
-mode <https://en.wikipedia.org/wiki/JBIG2#Disadvantages>`__
+mode <https://abbyy.technology/en:kb:tip:jbig2_compression_and_ocr>`__
 and decide if this feature is acceptable for their use case.

 JBIG2 lossy mode does achieve higher compression ratios than any other
@@ -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`` and French is ``fra``.
+are not, e.g. German is ``deu``.

-After you have installed a language pack, you can use it with ``ocrmypdf -l <language>``,
+After you have installed a language pack, you can use it ``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+fra`` (English and French) or
-``-l eng -l fra``.
+requested using either ``-l eng+fre`` (English and French) or
+``-l eng -l fre``.

 Fedora users
 ============
@@ -51,8 +51,8 @@ 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+fra`` (English and French) or
-``-l eng -l fra``.
+requested using either ``-l eng+fre`` (English and French) or
+``-l eng -l fre``.

 macOS users
 ===========
@@ -70,7 +70,4 @@ derived Docker image as
 Windows users
 =============

-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).
+The Tesseract installer provided by Chocolatey already includes 100 languages.
@@ -1,60 +0,0 @@
-================
-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/jbarlow83/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.
@@ -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 are possible
+open internet connections to pre-selected URLs. All of these 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 to restore searchability.
+could be used 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 that it may
-incidentally sanitize PDFs containing certain types of malware. It
+Setting aside these concerns, a side effect of OCRmyPDF is it may
+incidentally sanitize PDFs that contain 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 to run OCRmyPDF jobs inside a Docker container, a virtual machine,
+would be 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.

@@ -128,9 +128,8 @@ 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 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.
+Abbyy Cloud OCR is a viable commercial alternative with a web services
+API.

 Password protection, digital signatures and certification
 =========================================================
@@ -0,0 +1,392 @@
+<?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>
@@ -61,10 +61,9 @@ similar to ``pytest`` packages such as ``pytest-cov`` (the package) and

 .. note::

-    We recommend plugin authors name their plugins with the prefix
+    We strongly 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. At the same time, please make it clear
-    that your package is not official.
+    module), just like pytest plugins.

 Setuptools plugins
 ==================
@@ -87,13 +86,6 @@ named ``ocrmypdf-exampleplugin``:
        entry_points={"ocrmypdf": ["exampleplugin = exampleplugin.pluginmodule"]},
    )

-.. code-block:: ini
-
-    # equivalent setup.cfg
-    [options.entry_points]
-    ocrmypdf =
-        exampleplugin = exampleplugin.pluginmodule
-
 Plugin requirements
 ===================

@@ -159,18 +151,6 @@ 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
 -------------------------------------------

@@ -188,8 +168,6 @@ Modifying intermediate images

 .. autofunction:: ocrmypdf.pluginspec.filter_page_image

-.. autofunction:: ocrmypdf.pluginspec.filter_pdf_page
-
 OCR engine
 ----------

@@ -12,298 +12,17 @@ may be unreliable. Use the API to depend on precise behavior.
 The public API may be useful in scripts that launch OCRmyPDF processes or that
 wish to use some of its features for working with PDFs.

-.. note::
-
-   Python 3.6 reaches end of life on December 23, 2021. We have already ended support
-   for Python 3.6 but might release fixes for critical issues if necessary before that
-   date.
-
-v13.1.0
-=======
-
-  Changed to using Python concurrent.futures-based parallel execution instead of
-   pools, since futures have now exceed pools in features.
-  If a child worker is terminated (perhaps by the operating system or the user
-   killing it in a task manager), the parallel task will fail an error message.
-   Previously, the main ocrmypdf process would "hang" indefinitely, waiting for the
-   child to report.
-  Added new argument ``--tesseract-thresholding`` to provide control over Tesseract 5's
-   threshold parameter.
-  Documentation updates and changes. Better documentation for ``--output-type none``,
-   added a few releases ago. Removed some obsolete documentation.
-  Improved bash completions - thanks to @FPille.
-
-v13.0.0
-=======
-
-**Breaking changes**
-
-  The deprecated module ``ocrmypdf.leptonica`` has been removed.
-  We no longer depend on Leptonica (``liblept``) or CFFI (``libffi``,
-   ``python3-cffi``). (Note that Tesseract still requires Leptonica; OCRmyPDF no longer
-   directly uses this library.)
-  The argument ``--remove-background`` is temporarily disabled while we search for an
-   alternative to the Leptonica implementation of this feature.
-  The ``--threshold`` argument has been removed, since this also depended on Leptonica.
-   Tesseract 5.x has implemented improvements to thresholding, so this feature will be
-   redundant anyway.
-  ``--deskew`` was previous calculated by a Leptonica algorithm. We now use a feature
-   of Tesseract to term the appropriate the angle to deskew a page. The deskew angle
-   according to Tesseract may differ from Leptonica's algorithm. At least in theory,
-   Tesseract's deskew angle is informed by a more complex analysis than Leptonica,
-   so this should improve results in general. We also use Pillow to perform the
-   deskewing, which may affect the appearance of the image compared to Leptonica.
-  Support for Python 3.6 was dropped, since this release is approaching end of life.
-  We now require pikepdf 4.0 or newer. This, in turn, means that OCRmyPDF requires
-   a system compatible with the manylinux2014 specification. This change was "forced"
-   by Pillow not releasing manylinux2010 wheels anymore.
-  We no longer provide requirements.txt-style files. Use ``pip install ocrmypdf[...]``
-   instead.
-  Bumped required versions of several libraries.
-
-**Fixes**
-
-  Fixed an issue where OCRmyPDF failed to find Ghostscript on Windows even when
-   installed, and would exit with an error.
-  By removing Leptonica, we fixed all issues related to Leptonica on Apple
-   Silicon or Leptonica failing to import on Windows.
-
-v12.7.2
-=======
-
-  Fixed "invalid version number" error for Tesseract packaging with nonstandard
-   version "5.0.0-rc1.20211030".
-  Fixed use of deprecated ``importlib.resources.read_binary``.
-  Replace some uses of string paths with ``pathlib.Path``.
-  Fixed a leaked file handle when using ``--output-type none``.
-  Removed shims to support versions of pikepdf that are no longer supported.
-
-v12.7.1
-=======
-
-  Declare support for pdfminer.six v20211012.
-
-v12.7.0
-=======
-
-  Fixed test suite failure when using pikepdf 3.2.0 that was compiled with pybind11
-   2.8.0. :issue:`843`
-  Improve advice to user about using ``--max-image-mpixels`` if OCR fails for this
-   reason.
-  Minor documentation fixes. (Thanks to @mara004.)
-  Don't require importlib-metadata and importlib-resources backports on versions of
-   Python where the standard library implementation is sufficient.
-   (Thanks to Marco Genasci.)
-
-v12.6.0
-=======
-
-  Implemented ``--output-type=none`` to skip producing PDFs for applications that
-   only want sidecar files (:issue:`787`).
-  Fixed ambiguities in descriptions of behavior of ``--jbig2-lossy``.
-  Various improvements to documentation.
-
-v12.5.0
-=======
-
-  Fixed build failure for the combination of PyPy 3.6 and pikepdf 3.0. This
-   combination can work in a source build but does not work with wheels.
-  Accepted bot that wanted to upgrade our deprecated requirements.txt.
-  Documentation updates.
-  Replace pkg_resources and install dependency on setuptools with
-   importlib-metadata and importlib-resources.
-  Fixed regression in hocrtransform causing text to be omitted when this
-   renderer was used.
-  Fixed some typing errors.
-
-v12.4.0
-=======
-
-  When grafting text layers, use pikepdf's ``unparse_content_stream`` if available.
-  Confirmed support for pluggy 1.0. (Thanks @QuLogic.)
-  Fixed some typing issues, improved pre-commit settings, and fixed issues
-   flagged by linters.
-  PyPy 7.3.3 (=Python 3.6) is now supported. Note that PyPy does not necessarily
-   run faster, because the vast majority of OCRmyPDF's execution time is spent
-   running OCR or generally executing native code. However, PyPy may bring speed
-   improvements in some areas.
-
-v12.3.3
-=======
-
-  watcher.py: fixed interpretation of boolean env vars (:issue:`821`).
-  Adjust CI scripts to test Tesseract 5 betas.
-  Document our support for the Tesseract 5 betas.
-
-v12.3.2
-=======
-
-  Indicate support for flask 2.x, watcher 2.x (:issue:`815, 816`).
-
-v12.3.1
-=======
-
-  Fixed issue with selection of text when using the hOCR renderer (:issue:`813`).
-  Fixed build errors with the Docker image by upgrading to a newer Ubuntu.
-   Also set the timezone of this image to UTC.
-
-v12.3.0
-=======
-
-  Fixed a regression introduced in Pillow 8.3.0. Pillow no longer rounds DPI
-   for image resolutions. We now account for this (:issue:`802`).
-  We no longer use some API calls that are deprecated in the latest versions of
-   pikepdf.
-  Improved error message when a language is requested that doesn't look like a
-   typical ISO 639-2 code.
-  Fixed some tests that attempted to symlink on Windows, breaking tests on a
-   Windows desktop but not usually on CI.
-  Documentation fixes (thanks to @mara004)
-
-v12.2.0
-=======
-
-  Fixed invalid Tesseract version number on Windows (:issue:`795`).
-  Documentation tweaks. Documentation build now depends on sphinx-issues package.
-
-v12.1.0
-=======
-
-  For security reasons we now require Pillow >= 8.2.x. (Older versions will continue
-   to work if upgrading is not an option.)
-  The build system was reorganized to rely on ``setup.cfg`` instead of ``setup.py``.
-   All changes should work with previously supported versions of setuptools.
-  The files in ``requirements/*`` are now considered deprecated but will be retained for v12.
-   Instead use ``pip install ocrmypdf[test]`` instead of ``requirements/test.txt``, etc.
-   These files will be removed in v13.
-
-v12.0.3
-=======
-
-  Expand the list of languages supported by the hocr PDF renderer.
-   Several languages were previously considered not supported, particularly those
-   non-European languages that use the Latin alphabet.
-  Fixed a case where the exception stack trace was suppressed in verbose mode.
-  Improved documentation around commercial OCR.
-
-v12.0.2
-=======
-
-  Fixed exception thrown when using ``--remove-background`` on files containing small
-   images (:issue:`769`).
-  Improve documentation for description of adding language packs to the Docker image
-   and corrected name of French language pack.
-
-v12.0.1
-=======
-
-  Fixed "invalid version number" for untagged tesseract versions (:issue:`770`).
-
-v12.0.0
-=======
-
-**Breaking changes**
-
-  Due to recent security issues in pikepdf, Pillow and reportlab, we now require
-   newer versions of these libraries and some of their dependencies. (If necessary,
-   package maintainers may override these versions at their discretion; lower
-   versions will often work.)
-  We now use the "LeaveColorUnchanged" color conversion strategy when directing
-   Ghostscript to create a PDF/A. Generally this is faster than performing a
-   color conversion, which is not always necessary.
-  OCR text is now packaged in a Form XObject. This makes it easier to isolate
-   OCR from other document content. However, some poorly implemented PDF text
-   extraction algorithms may fail to detect the text.
-  Many API functions have stricter parameter checking or expect keyword arguments
-   were they previously did not.
-  Some deprecated functions in ``ocrmypdf.optimize`` were removed.
-  The ``ocrmypdf.leptonica`` module is now deprecated, due to difficulties with
-   the current strategy of ABI binding on newer platforms like Apple Silicon.
-   It will be removed and replaced, either by repackaging Leptonica as an
-   independent library using or using a different image processing library.
-  Continuous integration moved to GitHub Actions.
-  We no longer depend on ``pytest_helpers_namespace`` for testing.
-
-**New features**
-
-  New plugin hook: ``get_progressbar_class``, for progress reporting,
-   allowing developers to replace the standard console progress bar with some
-   other mechanism, such as updating a GUI progress bar.
-  New plugin hook: ``get_executor``, for replacing the concurrency model.
-   This is primarily to support execution on AWS Lambda, which does not support
-   standard Python ``multiprocessing`` due to its lack of shared memory.
-  New plugin hook: ``get_logging_console``, for replacing the standard
-   way OCRmyPDF outputs its messages.
-  New plugin hook: ``filter_pdf_page``, for modifying individual PDF
-   pages produced by OCRmyPDF.
-  OCRmyPDF now runs on nonstandard execution environments that do not have
-   interprocess semaphores, such as AWS Lambda and Android Termux. If the environment
-   does not have semaphores, OCRmyPDF will automatically select an alternate
-   process executor that does not use semaphores.
-  Continuous integration moved to GitHub Actions.
-  We now generate an ARM64-compatible Docker image alongside the x64 image.
-   Thanks to @andkrause for doing most of the work in a pull request several months
-   ago, which we were finally able to integrate now. Also thanks to @0x326 for
-   review comments.
-
-**Fixes**
-
-  Fixed a possible deadlock on attempting to flush ``sys.stderr`` when older
-   versions of Leptonica are in use.
-  Some worker processes inherited resources from their parents such as log
-   handlers that may have also lead to deadlocks. These resources are now released.
-  Improvements to test coverage.
-  Removed vestiges of support for Tesseract versions older than 4.0.0-beta1 (
-   which ships with Ubuntu 18.04).
-  OCRmyPDF can now parse all of Tesseract version numbers, since several
-   schemes have been in use.
-  Fixed an issue with parsing PDFs that contain images drawn at a scale of 0. (:issue:`761`)
-  Removed a frequently repeated message about disabling mmap.
-
-v11.7.3
-=======
-
-  Exclude CCITT Group 3 images from being optimized. Some libraries
-   OCRmyPDF uses do not seem to handle this obscure compression format properly.
-   You may get errors or possible corrupted output images without this fix.
-
-v11.7.2
-=======
-
-  Updated pinned versions in main.txt, primarily to upgrade Pillow to 8.1.2, due
-   to recently disclosed security vulnerabilities in that software.
-  The ``--sidecar`` parameter now causes an exception if set to the same file as
-   the input or output PDF.
-
-v11.7.1
-=======
-
-  Some exceptions while attempting image optimization were only logged at the debug
-   level, causing them to be suppressed. These errors are now logged appropriately.
-  Improved the error message related to ``--unpaper-args``.
-  Updated documentation to mention the new conda distribution.
-
-v11.7.0
-=======
-
-  We now support using ``--sidecar`` in conjunction with ``--pages``; these arguments
-   used to be mutually exclusive. (:issue:`735`)
-  Fixed a possible issue with PDF/A-1b generation. Acrobat complained that our PDFs use
-   object streams. More robust PDF/A validators like veraPDF don't consider this a
-   problem, but we'll honor Acrobat's objection from here on. This may increase file
-   size of PDF/A-1b files. PDF/A-2b files will not be affected.
-
 v11.6.2
 =======

 -  Fixed a regression where the wrong page orientation would be produced when using
-   arguments such as ``--deskew --rotate-pages`` (:issue:`730`).
+   arguments such as ``--deskew --rotate-pages`` (#730).

 v11.6.1
 =======

 -  Fixed an issue with attempting optimize unusually narrow-width images by excluding
-   these images from optimization (:issue:`732`).
+   these images from optimization (#732).
 -  Remove an obsolete compatibility shim for a version of pikepdf that is no longer
   supported.

@@ -341,7 +60,7 @@ v11.4.5
 v11.4.4
 =======

-  Fixed ``AttributeError: 'NoneType' object has no attribute 'userunit'`` (:issue:`700`),
+-  Fixed ``AttributeError: 'NoneType' object has no attribute 'userunit'``, issue #700,
   related to OCRmyPDF not properly forwarded an error message from pdfminer.six.
 -  Adjusted typing of some arguments.
 -  ``ocrmypdf.ocr`` now takes a ``threading.Lock`` for reasons outlined in the
@@ -384,7 +103,7 @@ v11.4.0
   ``com.github.ocrmypdf`` to ``ocrmypdf.io``. Scripts that chose to depend on this
   prefix may need to be adjusted. (This has always been an implementation detail so is
   not considered part of the semantic versioning "contract".)
-  Fixed :issue:`692`, where a particular file with malformed fonts would flood an
+-  Fixed issue #692, where a particular file with malformed fonts would flood an
   internal message cue by generating so many debug messages.
 -  Fixed an exception on processing hOCR files with no page record. Tesseract
   is not known to generate such files.
@@ -403,7 +122,7 @@ v11.3.3
 =======

 -  If unpaper outputs non-UTF-8 data, quietly fix this rather than choke on the
-   conversion. (Possibly addresses :issue:`671`.)
+   conversion. (Possibly addresses #671.)

 v11.3.2
 =======
@@ -416,7 +135,7 @@ v11.3.2
   as optimization candidates.
 -  On some systems, unpaper seems to be unable to process the PNGs we offer it
   as input. We now convert the input to PNM format, which unpaper always accepts.
-   Fixes :issue:`665` and :issue:`667`.
+   Fixes #665 and #667.
 -  DPI sent to unpaper is now rounded to a more reasonable number of decimal digits.
 -  Debug and error messages from unpaper were being suppressed.
 -  Some documentation tweaks.
@@ -425,7 +144,7 @@ v11.3.1
 =======

 -  Declare support for new versions: pdfminer.six 20201018 and pikepdf 2.x
-  Fixed warning related to ``--pdfa-image-compression`` that appears at the wrong
+-  Fix warning related to ``--pdfa-image-compression`` that appears at the wrong
   time.

 v11.3.0
@@ -445,7 +164,7 @@ v11.3.0
   macOS and Windows only where the parent process is not forked.
 -  Fixed the hookspec of rasterize_pdf_page to remove default parameters that
   were not handled in an expected way by pluggy.
-  Fixed another issue with automatic page rotation (:issue:`658`) due to the issue above.
+-  Fixed another issue with automatic page rotation (#658) due to the issue above.

 v11.2.1
 =======
@@ -467,7 +186,7 @@ v11.1.2
 -  Fixed hOCR renderer writing the text in roughly reverse order. This should not
   affect reasonably smart PDF readers that properly locate the position of all
   text, but may confuse those that rely on the order of objects in the content
-   stream. (:issue:`642`)
+   stream. (#642)

 v11.1.1
 =======
@@ -480,9 +199,9 @@ v11.1.1
 v11.1.0
 =======

-  Fixed page rotation issues: :issue:`634,589`.
+-  Fixed page rotation issues: #634, #589.
 -  Fixed some cases where optimization created an invalid image such as a
-   1-bit "RGB" image: :issue:`629,620`.
+   1-bit "RGB" image: #629, #620.
 -  Page numbers are now displayed in debug logs when pages are being grafted.
 -  ocrmypdf.optimize.rewrite_png and ocrmypdf.optimize.rewrite_png_as_g4 were
   marked deprecated. Strictly speaking these should have been internal APIs,
@@ -495,7 +214,7 @@ v11.1.0
 v11.0.2
 =======

-  Fixed :issue:`612`, TypeError exception. Fixed by eliminating unnecessary repair of
+-  Fixed issue #612, TypeError exception. Fixed by eliminating unnecessary repair of
   input PDF metadata in memory.

 v11.0.1
@@ -512,7 +231,7 @@ v11.0.0
 -  Project license changed to Mozilla Public License 2.0. Some miscellaneous
   code is now under MIT license and non-code content/media remains under
   CC-BY-SA 4.0. License changed with approval of all people who were found
-   to have contributed to GPLv3 licensed sections of the project. (:issue:`600`)
+   to have contributed to GPLv3 licensed sections of the project. (#600)
 -  Because the license changed, this is being treated as a major version number
   change; however, there are no known breaking changes in functional behavior
   or API compared to v10.x.
@@ -521,7 +240,7 @@ v10.3.3
 =======

 -  Fixed a "KeyError: 'dpi'" error message when using ``--threshold`` on an image.
-   (:issue:`607`)
+   (#607)

 v10.3.2
 =======
@@ -564,16 +283,16 @@ v10.2.0
 =======

 -  Update Docker image to use Ubuntu 20.04.
-  Fixed issue PDF/A acquires title "Untitled" after conversion. (:issue:`582`)
+-  Fixed issue PDF/A acquires title "Untitled" after conversion. (#582)
 -  Fixed a problem where, when using ``--pdf-renderer hocr``, some text would
   be missing from the output when using a more recent version of Tesseract.
   Tesseract began adding more detailed markup about the semantics of text
   that our HOCR transform did not recognize, so it ignored them. This option is
   not the default. If necessary ``--redo-ocr`` also redoing OCR to fix such issues.
 -  Fixed an error in Python 3.9 beta, due to removal of deprecated
-   ``Element.getchildren()``. (:issue:`584`)
+   ``Element.getchildren()``. (#584)
 -  Implemented support using the API with ``BytesIO`` and other file stream objects.
-   (:issue:`545`)
+   (#545)

 v10.1.1
 =======
@@ -666,7 +385,7 @@ v9.8.0

 -  Fixed issue where only the first PNG (FlateDecode) image in a file would be
   considered for optimization. File sizes should be improved from here on.
-  Fixed a startup crash when the chosen language was Japanese (:issue:`543`).
+-  Fixed a startup crash when the chosen language was Japanese (#543).
 -  Added options to configure polling and log level to watcher.py.

 v9.7.2
@@ -713,11 +432,11 @@ v9.6.1
        they can be copied out as whole files, and to ensure syntax checking
        is maintained.

-  Fixed an error that caused bash completions to fail on macOS. (:issue:`502,504`;
+-  Fixed an error that caused bash completions to fail on macOS. (#502, #504;
   @AlexanderWillner)
 -  Fixed a rare case where OCRmyPDF threw an exception while processing a PDF
   with the wrong object type in its ``/Trailer /Info``. The error is now logged
-   and incorrect object is ignored. (:issue:`497`)
+   and incorrect object is ignored. (#497)
 -  Removed potentially non-free file ``enron1.pdf`` and simplified the test that
   used it.
 -  Removed potentially non-free file ``misc/media/logo.afdesign``.
@@ -919,7 +638,7 @@ v8.3.1
 ======

 -  Fixed an issue where PDFs with malformed metadata would be rendered as
-   blank pages. :issue:`398`.
+   blank pages. `#398 <https://github.com/jbarlow83/OCRmyPDF/issues/398>`_.

 v8.3.0
 ======
@@ -1002,7 +721,7 @@ v8.2.0
   designed. However, quality would not be impacted. Lossless JBIG2 was
   entirely unaffected.
 -  Updated dependencies, including pikepdf to 1.1.0. This fixes
-   :issue:`358`.
+   `#358 <https://github.com/jbarlow83/OCRmyPDF/issues/358>`__.
 -  The install-time version checks for certain external programs have
   been removed from setup.py. These tests are now performed at
   run-time.
@@ -1023,7 +742,7 @@ v8.1.0
   (specifically, those with invalid destination objects)
 -  Fixed an issue when using ``--tesseract-timeout`` and image
   processing features on a file with more than 100 pages.
-   :issue:`347`
+   `#347 <https://github.com/jbarlow83/OCRmyPDF/issues/347>`__
 -  OCRmyPDF now always calls ``os.nice(5)`` to signal to operating
   systems that it is a background process.

@@ -1031,7 +750,7 @@ v8.0.1
 ======

 -  Fixed an exception when parsing PDFs that are missing a required
-   field. :issue:`325`
+   field. `#325 <https://github.com/jbarlow83/OCRmyPDF/issues/325>`__
 -  pikepdf 1.0.5 is now required, to address some other PDF parsing
   issues.

@@ -1052,7 +771,7 @@ older versions of certain dependencies.
 **Other changes**

 -  Fixed an unhandled exception when attempting to mask barcodes.
-   :issue:`322`
+   `#322 <https://github.com/jbarlow83/OCRmyPDF/issues/322>`__
 -  It is now possible to use ocrmypdf without pdfminer.six, to support
   distributions that do not have it or cannot currently use it (e.g.
   Homebrew). Downstream maintainers should include pdfminer.six if
@@ -1079,13 +798,13 @@ v7.4.0
 -  chardet >= 3.0.4 is temporarily listed as required. pdfminer.six
   depends on it, but the most recent release does not specify this
   requirement.
-   (:issue:`326`)
+   (`#326 <https://github.com/jbarlow83/OCRmyPDF/issues/326>`__)
 -  python-xmp-toolkit and libexempi are no longer required.
 -  A new Docker image is now being provided for users who wish to access
   OCRmyPDF over a simple HTTP interface, instead of the command line.
 -  Increase tolerance of PDFs that overflow or underflow the PDF
   graphics stack.
-   (:issue:`325`)
+   (`#325 <https://github.com/jbarlow83/OCRmyPDF/issues/325>`__)

 v7.3.1
 ======
@@ -1159,7 +878,7 @@ v7.3.0
 v7.2.1
 ======

-  Fixed compatibility with an API change in pikepdf 0.3.5.
+-  Fix compatibility with an API change in pikepdf 0.3.5.
 -  A kludge to support Leptonica versions older than 1.72 in the test
   suite was dropped. Older versions of Leptonica are likely still
   compatible. The only impact is that a portion of the test suite will
@@ -1196,7 +915,7 @@ Users who did not install an optional JBIG2 encoder are unaffected.
   will now attempt to further optimize that image as CCITT or JBIG2,
   instead of keeping it in the "flate" encoding which is not efficient
   for 1 bpp images.
-   (:issue:`297`)
+   (`#297 <https://github.com/jbarlow83/OCRmyPDF/issues/297>`__)
 -  Images in PDFs that are used as soft masks (i.e. transparency masks
   or alpha channels) are now excluded from optimization.
 -  Fixed handling of Tesseract 4.0-rc1 which now accepts invalid
@@ -1208,14 +927,15 @@ v7.1.0
 -  Improve the performance of initial text extraction, which is done to
   determine if a file contains existing text of some kind or not. On
   large files, this initial processing is now about 20x times faster.
-   (:issue:`299`)
+   (`#299 <https://github.com/jbarlow83/OCRmyPDF/issues/299>`__)
 -  pikepdf 0.3.3 is now required.
-  Fixed :issue:`231`, a
+-  Fixed issue
+   `#231 <https://github.com/jbarlow83/OCRmyPDF/issues/231>`__, a
   problem with JPEG2000 images where image metadata was only available
   inside the JPEG2000 file.
 -  Fixed some additional Ghostscript 9.25 compatibility issues.
 -  Improved handling of KeyboardInterrupt error messages.
-   (:issue:`301`)
+   (`#301 <https://github.com/jbarlow83/OCRmyPDF/issues/301>`__)
 -  README.md is now served in GitHub markdown instead of
   reStructuredText.

@@ -1247,34 +967,35 @@ v7.0.5
 v7.0.4
 ======

-  Fixed exception thrown when trying to optimize a certain type of PNG
+-  Fix exception thrown when trying to optimize a certain type of PNG
   embedded in a PDF with the ``-O2``
 -  Update to pikepdf 0.3.2, to gain support for optimizing some
   additional image types that were previously excluded from
   optimization (CMYK and grayscale). Fixes
-   :issue:`285`.
+   `#285 <https://github.com/jbarlow83/OCRmyPDF/issues/285>`__.

 v7.0.3
 ======

-  Fixed :issue:`284`, an error
+-  Fix issue
+   `#284 <https://github.com/jbarlow83/OCRmyPDF/issues/284>`__, an error
   when parsing inline images that have are also image masks, by
   upgrading pikepdf to 0.3.1

 v7.0.2
 ======

-  Fixed a regression with ``--rotate-pages`` on pages that already had
+-  Fix a regression with ``--rotate-pages`` on pages that already had
   rotations applied.
-   (:issue:`279`)
+   (`#279 <https://github.com/jbarlow83/OCRmyPDF/issues/279>`__)
 -  Improve quality of page rotation in some cases by rasterizing a
   higher quality preview image.
-   (:issue:`281`)
+   (`#281 <https://github.com/jbarlow83/OCRmyPDF/issues/281>`__)

 v7.0.1
 ======

-  Fixed compatibility with img2pdf >= 0.3.0 by rejecting input images
+-  Fix compatibility with img2pdf >= 0.3.0 by rejecting input images
   that have an alpha channel
 -  Add forward compatibility for pikepdf 0.3.0 (unrelated to img2pdf)
 -  Various documentation updates for v7.0.0 changes
@@ -1377,7 +1098,7 @@ v6.2.4
 v6.2.3
 ======

-  Fixed compatibility with img2pdf >= 0.3.0 by rejecting input images
+-  Fix compatibility with img2pdf >= 0.3.0 by rejecting input images
   that have an alpha channel
 -  This version will be included in Ubuntu 18.10

@@ -1394,8 +1115,9 @@ v6.2.2
 v6.2.1
 ======

-  Fixed recent versions of Tesseract (after 4.0.0-beta1) not being
-   detected as supporting the ``sandwich`` renderer (:issue:`271`).
+-  Fix recent versions of Tesseract (after 4.0.0-beta1) not being
+   detected as supporting the ``sandwich`` renderer
+   (`#271 <https://github.com/ppjbarlow83/OCRmyPDF/issues/271>`__).

 v6.2.0
 ======
@@ -1408,19 +1130,21 @@ v6.2.0
 -  Creation of PDF/A-3 is now supported. However, there is no ability to
   attach files to PDF/A-3.
 -  Lists more reasons why the file size might grow.
-  Fixed :issue:`262`,
+-  Fix issue
+   `#262 <https://github.com/ppjbarlow83/OCRmyPDF/issues/262>`__,
   ``--remove-background`` error on PDFs contained colormapped
   (paletted) images.
-  Fixed another XMP metadata validation issue, in cases where the input
+-  Fix another XMP metadata validation issue, in cases where the input
   file's creation date has no timezone and the creation date is not
   overridden.

 v6.1.5
 ======

-  Fixed :issue:`253`, a
+-  Fix issue
+   `#253 <https://github.com/jbarlow83/OCRmyPDF/issues/253>`__, a
   possible division by zero when using the ``hocr`` renderer.
-  Fixed incorrectly formatted ``<xmp:ModifyDate>`` field inside XMP
+-  Fix incorrectly formatted ``<xmp:ModifyDate>`` field inside XMP
   metadata for PDF/As. veraPDF flags this as a PDF/A validation
   failure. The error is caused the timezone and final digit of the
   seconds of modified time to be omitted, so at worst the modification
@@ -1429,7 +1153,7 @@ v6.1.5
 v6.1.4
 ======

-  Fixed :issue:`248`
+-  Fix issue `#248 <https://github.com/jbarlow83/OCRmyPDF/issues/248>`__
   ``--clean`` argument may remove OCR from left column of text on
   certain documents. We now set ``--layout none`` to suppress this.
 -  The test cache was updated to reflect the change above.
@@ -1454,7 +1178,8 @@ Notes
 v6.1.3
 ======

-  Fixed :issue:`247`,
+-  Fix issue
+   `#247 <https://github.com/jbarlow83/OCRmyPDF/issues/247>`__,
   ``/CreationDate`` metadata not copied from input to output.
 -  A warning is now issued when Python 3.5 is used on files with a large
   page count, as this case is known to regress to single core
@@ -1464,13 +1189,13 @@ v6.1.2
 ======

 -  Upgrade to PyMuPDF v1.12.5 which includes a more complete fix to
-   :issue:`239`.
+   `#239 <https://github.com/jbarlow83/OCRmyPDF/issues/239>`__.
 -  Add ``defusedxml`` dependency.

 v6.1.1
 ======

-  Fixed text being reported as found on all pages if PyMuPDF is not
+-  Fix text being reported as found on all pages if PyMuPDF is not
   installed.

 v6.1.0
@@ -1481,15 +1206,15 @@ v6.1.0
   PyMuPDF than the author anticipated. (For version 6.x only) install
   OCRmyPDF with ``pip install ocrmypdf[fitz]`` to use it to its full
   potential.
-  Fixed ``FileExistsError`` that could occur if OCR timed out while it
+-  Fix ``FileExistsError`` that could occur if OCR timed out while it
   was generating the output file.
-   (:issue:`218`)
-  Fixed table of contents/bookmarks all being redirected to page 1 when
+   (`#218 <https://github.com/jbarlow83/OCRmyPDF/issues/218>`__)
+-  Fix table of contents/bookmarks all being redirected to page 1 when
   generating a PDF/A (with PyMuPDF). (Without PyMuPDF the table of
   contents is removed in PDF/A mode.)
-  Fixed "RuntimeError: invalid key in dict" when table of
+-  Fix "RuntimeError: invalid key in dict" when table of
   contents/bookmarks titles contained the character ``)``.
-   (:issue:`239`)
+   (`#239 <https://github.com/jbarlow83/OCRmyPDF/issues/239>`__)
 -  Added a new argument ``--skip-repair`` to skip the initial PDF repair
   step if the PDF is already well-formed (because another program
   repaired it).
@@ -1516,29 +1241,35 @@ v6.0.0

 -  Fixed an issue where OCRmyPDF failed to detect existing text on
   pages, depending on how the text and fonts were encoded within the
-   PDF. (:issue:`233,232`)
+   PDF. (`#233 <https://github.com/jbarlow83/OCRmyPDF/issues/233>`__,
+   `#232 <https://github.com/jbarlow83/OCRmyPDF/issues/232>`__)
 -  Fixed an issue that caused dramatic inflation of file sizes when
   ``--skip-text --output-type pdf`` was used. OCRmyPDF now removes
   duplicate resources such as fonts, images and other objects that it
-   generates. (:issue:`237`)
+   generates.
+   (`#237 <https://github.com/jbarlow83/OCRmyPDF/issues/237>`__)
 -  Improved performance of the initial page splitting step. Originally
   this step was not believed to be expensive and ran in a process.
   Large file testing revealed it to be a bottleneck, so it is now
   parallelized. On a 700 page file with quad core machine, this change
-   saves about 2 minutes. (:issue:`234`)
+   saves about 2 minutes.
+   (`#234 <https://github.com/jbarlow83/OCRmyPDF/issues/234>`__)
 -  The test suite now includes a cache that can be used to speed up test
   runs across platforms. This also does not require computing
-   checksums, so it's faster. (:issue:`217`)
+   checksums, so it's faster.
+   (`#217 <https://github.com/jbarlow83/OCRmyPDF/issues/217>`__)

 v5.7.0
 ======

 -  Fixed an issue that caused poor CPU utilization on machines with more
-   than 4 cores when running Tesseract 4. (Related to :issue:`217`.)
+   than 4 cores when running Tesseract 4. (Related to issue
+   `#217 <https://github.com/jbarlow83/OCRmyPDF/issues/217>`__.)
 -  The 'hocr' renderer has been improved. The 'sandwich' and 'tesseract'
   renderers are still better for most use cases, but 'hocr' may be
   useful for people who work with the PDF.js renderer in English/ASCII
-   languages. (:issue:`225`)
+   languages.
+   (`#225 <https://github.com/jbarlow83/OCRmyPDF/issues/225>`__)

   -  It now formats text in a matter that is easier for certain PDF
      viewers to select and extract copy and paste text. This should
@@ -1566,10 +1297,11 @@ v5.6.2
 v5.6.1
 ======

-  Fixed :issue:`219`: change
+-  Fix issue
+   `#219 <https://github.com/jbarlow83/OCRmyPDF/issues/219>`__: change
   how the final output file is created to avoid triggering permission
   errors when the output is a special file such as ``/dev/null``
-  Fixed test suite failures due to a qpdf 8.0.0 regression and Python
+-  Fix test suite failures due to a qpdf 8.0.0 regression and Python
   3.5's handling of symlink
 -  The "encrypted PDF" error message was different depending on the type
   of PDF encryption. Now a single clear message appears for all types
@@ -1582,7 +1314,8 @@ v5.6.1
 v5.6.0
 ======

-  Fixed :issue:`216`: preserve
+-  Fix issue
+   `#216 <https://github.com/jbarlow83/OCRmyPDF/issues/216>`__: preserve
   "text as curves" PDFs without rasterizing file
 -  Related to the above, messages about rasterizing are more consistent
 -  For consistency versions minor releases will now get the trailing .0
@@ -1594,32 +1327,34 @@ v5.5
 -  Add new argument ``--max-image-mpixels``. Pillow 5.0 now raises an
   exception when images may be decompression bombs. This argument can
   be used to override the limit Pillow sets.
-  Fixed output page cropped when using the sandwich renderer and OCR is
+-  Fix output page cropped when using the sandwich renderer and OCR is
   skipped on a rotated and image-processed page
 -  A warning is now issued when old versions of Ghostscript are used in
   cases known to cause issues with non-Latin characters
-  Fixed a few parameter validation checks for ``-output-type pdfa-1`` and
+-  Fix a few parameter validation checks for ``-output-type pdfa-1`` and
   ``pdfa-2``

 v5.4.4
 ======

-  Fixed :issue:`181`: fix
+-  Fix issue
+   `#181 <https://github.com/jbarlow83/OCRmyPDF/issues/181>`__: fix
   final merge failure for PDFs with more pages than the system file
   handle limit (``ulimit -n``)
-  Fixed :issue:`200`: an
+-  Fix issue
+   `#200 <https://github.com/jbarlow83/OCRmyPDF/issues/200>`__: an
   uncommon syntax for formatting decimal numbers in a PDF would cause
   qpdf to issue a warning, which ocrmypdf treated as an error. Now this
   the warning is relayed.
-  Fixed an issue where intermediate PDFs would be created at version 1.3
+-  Fix an issue where intermediate PDFs would be created at version 1.3
   instead of the version of the original file. It's possible but
   unlikely this had side effects.
 -  A warning is now issued when older versions of qpdf are used since
   issues like
-   :issue:`200` cause
+   `#200 <https://github.com/jbarlow83/OCRmyPDF/issues/200>`__ cause
   qpdf to infinite-loop
 -  Address issue
-   :issue:`140`: if
+   `#140 <https://github.com/jbarlow83/OCRmyPDF/issues/140>`__: if
   Tesseract outputs invalid UTF-8, escape it and print its message
   instead of aborting with a Unicode error
 -  Adding previously unlisted setup requirement, pytest-runner
@@ -1685,13 +1420,13 @@ v5.3
   forwarded to Tesseract OCR as words and regular expressions
   respective to use to guide OCR. Supplying a list of subject-domain
   words should assist Tesseract with resolving words.
-   (:issue:`165`)
+   (`#165 <https://github.com/jbarlow83/OCRmyPDF/issues/165>`__)
 -  Using a non Latin-1 language with the "hocr" renderer now warns about
   possible OCR quality and recommends workarounds
-   (:issue:`176`)
+   (`#176 <https://github.com/jbarlow83/OCRmyPDF/issues/176>`__)
 -  Output file path added to error message when that location is not
   writable
-   (:issue:`175`)
+   (`#175 <https://github.com/jbarlow83/OCRmyPDF/issues/175>`__)
 -  Otherwise valid PDFs with leading whitespace at the beginning of the
   file are now accepted

@@ -1719,7 +1454,8 @@ v5.1
 v5.0.1
 ======

-  Fixed :issue:`169`,
+-  Fixed issue
+   `#169 <https://github.com/jbarlow83/OCRmyPDF/issues/169>`__,
   exception due to failure to create sidecar text files on some
   versions of Tesseract 3.04, including the jbarlow83/ocrmypdf Docker
   image
@@ -1737,17 +1473,19 @@ v5.0
 -  Add a new feature, ``--sidecar``, which allows creating "sidecar"
   text files which contain the OCR results in plain text. These OCR
   text is more reliable than extracting text from PDFs. Closes
-   :issue:`126`.
+   `#126 <https://github.com/jbarlow83/OCRmyPDF/issues/126>`__.

 -  New feature: ``--pdfa-image-compression``, which allows overriding
   Ghostscript's lossy-or-lossless image encoding heuristic and making
   all images JPEG encoded or lossless encoded as desired. Fixes
-   :issue:`163`.
+   `#163 <https://github.com/jbarlow83/OCRmyPDF/issues/163>`__.

-  Fixed :issue:`143`, added
+-  Fixed issue
+   `#143 <https://github.com/jbarlow83/OCRmyPDF/issues/143>`__, added
   ``--quiet`` to suppress "INFO" messages

-  Fixed :issue:`164`, a typo
+-  Fixed issue
+   `#164 <https://github.com/jbarlow83/OCRmyPDF/issues/164>`__, a typo

 -  Removed the command line parameters ``-n`` and ``--just-print`` since
   they have not worked for some time (reported as Ubuntu bug
@@ -1756,14 +1494,17 @@ v5.0
 v4.5.6
 ======

-  Fixed :issue:`156`,
+-  Fixed issue
+   `#156 <https://github.com/jbarlow83/OCRmyPDF/issues/156>`__,
   'NoneType' object has no attribute 'getObject' on pages with no
   optional /Contents record. This should resolve all issues related to
   pages with no /Contents record.
-  Fixed :issue:`158`, ocrmypdf
+-  Fixed issue
+   `#158 <https://github.com/jbarlow83/OCRmyPDF/issues/158>`__, ocrmypdf
   now stops and terminates if Ghostscript fails on an intermediate
   step, as it is not possible to proceed.
-  Fixed :issue:`160`,
+-  Fixed issue
+   `#160 <https://github.com/jbarlow83/OCRmyPDF/issues/160>`__,
   exception thrown on certain invalid arguments instead of error
   message

@@ -1771,19 +1512,20 @@ v4.5.5
 ======

 -  Automated update of macOS homebrew tap
-  Fixed :issue:`154`, KeyError
+-  Fixed issue
+   `#154 <https://github.com/jbarlow83/OCRmyPDF/issues/154>`__, KeyError
   '/Contents' when searching for text on blank pages that have no
   /Contents record. Note: incomplete fix for this issue.

 v4.5.4
 ======

-  Fixed ``--skip-big`` raising an exception if a page contains no images
-   (:issue:`152`) (thanks
+-  Fix ``--skip-big`` raising an exception if a page contains no images
+   (`#152 <https://github.com/jbarlow83/OCRmyPDF/issues/152>`__) (thanks
   to @TomRaz)
-  Fixed an issue where pages with no images might trigger "cannot write
+-  Fix an issue where pages with no images might trigger "cannot write
   mode P as JPEG"
-   (:issue:`151`)
+   (`#151 <https://github.com/jbarlow83/OCRmyPDF/issues/151>`__)

 v4.5.3
 ======
@@ -1802,7 +1544,8 @@ v4.5.3
 v4.5.2
 ======

-  Fixed :issue:`147`,
+-  Fix issue
+   `#147 <https://github.com/jbarlow83/OCRmyPDF/issues/147>`__.
   ``--pdf-renderer tess4 --clean`` will produce an oversized page
   containing the original image in the bottom left corner, due to loss
   DPI information.
@@ -1812,7 +1555,8 @@ v4.5.2
 v4.5.1
 ======

-  Fixed :issue:`137`,
+-  Fix issue
+   `#137 <https://github.com/jbarlow83/OCRmyPDF/issues/137>`__,
   proportions of images with a non-square pixel aspect ratio would be
   distorted in output for ``--force-ocr`` and some other combinations
   of flags
@@ -1821,7 +1565,7 @@ v4.5
 ====

 -  PDFs containing "Form XObjects" are now supported (issue
-   :issue:`134`; PDF
+   `#134 <https://github.com/jbarlow83/OCRmyPDF/issues/134>`__; PDF
   reference manual 8.10), and images they contain are taken into
   account when determining the resolution for rasterizing
 -  The Tesseract 4 Docker image no longer includes all languages,
@@ -1950,7 +1694,7 @@ v4.2.5
 ======

 -  Fixed an issue
-   (:issue:`100`) with
+   (`#100 <https://github.com/jbarlow83/OCRmyPDF/issues/100>`__) with
   PDFs that omit the optional /BitsPerComponent parameter on images
 -  Removed non-free file milk.pdf

@@ -1958,7 +1702,7 @@ v4.2.4
 ======

 -  Fixed an error
-   (:issue:`90`) caused by
+   (`#90 <https://github.com/jbarlow83/OCRmyPDF/issues/90>`__) caused by
   PDFs that use stencil masks properly
 -  Fixed handling of PDFs that try to draw images or stencil masks
   without properly setting up the graphics state (such images are now
@@ -1996,7 +1740,7 @@ v4.2

 -  ocrmypdf will now try to convert single image files to PDFs if they
   are provided as input
-   (:issue:`15`)
+   (`#15 <https://github.com/jbarlow83/OCRmyPDF/issues/15>`__)

   -  This is a basic convenience feature. It only supports a single
      image and always makes the image fill the whole page.
@@ -2024,11 +1768,11 @@ v4.2
   -  supports reinterpreting PDFs where text was rendered as curves for
      printing, and text needs to be recovered
   -  fixes issue
-      :issue:`82`
+      `#82 <https://github.com/jbarlow83/OCRmyPDF/issues/82>`__

 -  Fixes an issue where, with certain settings, monochrome images in
   PDFs would be converted to 8-bit grayscale, increasing file size
-   (:issue:`79`)
+   (`#79 <https://github.com/jbarlow83/OCRmyPDF/issues/79>`__)
 -  Support for Ubuntu 12.04 LTS "precise" has been dropped in favor of
   (roughly) Ubuntu 14.04 LTS "trusty"

@@ -2056,7 +1800,7 @@ v4.1.3

 -  More helpful error message for PDFs with version 4 security handler
 -  Update usage instructions for Windows/Docker users
-  Fixed order of operations for matrix multiplication (no effect on most
+-  Fix order of operations for matrix multiplication (no effect on most
   users)
 -  Add a few leptonica wrapper functions (no effect on most users)

@@ -2149,7 +1893,7 @@ New features
   dominant orientation of detectable text. This feature is fairly
   reliable but some false positives occur especially if there is not
   much text to work with.
-   (:issue:`4`)
+   (`#4 <https://github.com/jbarlow83/OCRmyPDF/issues/4>`__)
 -  Deskewing is now performed using Leptonica instead of unpaper.
   Leptonica is faster and more reliable at image deskewing than
   unpaper.
@@ -2162,13 +1906,13 @@ Fixes
 -  Fixed an issue where lossless reconstruction could misalign the
   graphics layer with respect to text layer if the page had been
   cropped such that its origin is not (0, 0)
-   (:issue:`49`)
+   (`#49 <https://github.com/jbarlow83/OCRmyPDF/issues/49>`__)

 Changes

 -  Logging output is now much easier to read
 -  ``--deskew`` is now performed by Leptonica instead of unpaper
-   (:issue:`25`)
+   (`#25 <https://github.com/jbarlow83/OCRmyPDF/issues/25>`__)
 -  libffi is now required
 -  Some changes were made to the Docker and Travis build environments to
   support libffi
@@ -2183,7 +1927,7 @@ v3.2.1

 Changes

-  Fixed :issue:`47`
+-  Fixed issue `#47 <https://github.com/jbarlow83/OCRmyPDF/issues/47>`__
   "convert() got and unexpected keyword argument 'dpi'" by upgrading to
   img2pdf 0.2
 -  Tweaked the Dockerfiles
@@ -2228,7 +1972,8 @@ Changes
 -  Python 3.5 and macOS El Capitan are now supported platforms - no
   changes were needed to implement support
 -  Improved some error messages related to missing input files
-  Fixed :issue:`20`: uppercase .PDF extension not accepted
+-  Fixed issue `#20 <https://github.com/jbarlow83/OCRmyPDF/issues/20>`__
+   - uppercase .PDF extension not accepted
 -  Fixed an issue where OCRmyPDF failed to text that certain pages
   contained previously OCR'ed text, such as OCR text produced by
   Tesseract 3.04
@@ -2305,19 +2050,19 @@ Release candidates^

 -  rc9:

-   -  Fix
-      :issue:`118`:
+   -  fix issue
+      `#118 <https://github.com/jbarlow83/OCRmyPDF/issues/118>`__:
      report error if ghostscript iccprofiles are missing
   -  fixed another issue related to
-      :issue:`111`: PDF
+      `#111 <https://github.com/jbarlow83/OCRmyPDF/issues/111>`__: PDF
      rasterized to palette file
   -  add support image files with a palette
   -  don't try to validate PDF file after an exception occurs

 -  rc8:

-   -  Fix
-      :issue:`111`:
+   -  fix issue
+      `#111 <https://github.com/jbarlow83/OCRmyPDF/issues/111>`__:
      exception thrown if PDF is missing DocumentInfo dictionary

 -  rc7:
@@ -24,39 +24,45 @@
 import logging
 import os
 import sys
-from pathlib import Path

 import ocrmypdf

 # pylint: disable=logging-format-interpolation
 # pylint: disable=logging-not-lazy

-script_dir = Path(__file__).parent
+script_dir = os.path.dirname(os.path.realpath(__file__))
+print(script_dir + '/batch.py: Start')

 if len(sys.argv) > 1:
-    start_dir = Path(sys.argv[1])
+    start_dir = sys.argv[1]
 else:
-    start_dir = Path('.')
+    start_dir = '.'

 if len(sys.argv) > 2:
-    log_file = Path(sys.argv[2])
+    log_file = sys.argv[2]
 else:
-    log_file = script_dir.with_name('ocr-tree.log')
+    log_file = script_dir + '/ocr-tree.log'

 logging.basicConfig(
    level=logging.INFO,
    format='%(asctime)s %(message)s',
    filename=log_file,
-    filemode='a',
+    filemode='w',
 )

 ocrmypdf.configure_logging(ocrmypdf.Verbosity.default)

-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)
+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)
@@ -1,6 +1,6 @@
 # ocrmypdf completion                                     -*- shell-script -*-

-# Copyright 2019, 2021 Frank Pille
+# Copyright 2019 Frank Pille
 # Copyright 2020 Alex Willner
 #
 # Permission is hereby granted, free of charge, to any person obtaining a copy
@@ -23,285 +23,95 @@

 set -o errexit

-__ocrmypdf_arguments()
+_ocrmypdf()
 {
-    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)
-"
+    local cur prev cword words split

-    COMPREPLY=( $( compgen -W "$arguments" -- "$cur") )
-
-    # Remove description if only one completion exists
-    if [[ ${#COMPREPLY[*]} -eq 1 ]]; then
-        COMPREPLY=( ${COMPREPLY[0]%% *} )
+    # 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
    fi
-}

-__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]%% *} )
+    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
    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
-        -h|--help|--version)
-            return 0
+        --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
            ;;
        -l|--language)
            COMPREPLY=$( command tesseract --list-langs 2>/dev/null )
            COMPREPLY=( $( compgen -W '${COMPREPLY[@]##*:}' -- "$cur" ) )
-            return 0
+            return
            ;;
-        --output-type)
-            __ocrmypdf_output-type
-            return 0
+        --image-dpi|--oversample|--skip-big|--max-image-mpixels|\
+        --tesseract-timeout|--rotate-pages-threshold)
+            COMPREPLY=( $( compgen -P "$cur" -W '{0..9}' ) )
+            return
            ;;
        -j|--jobs)
            COMPREPLY=( $( compgen -W '{1..'$( _ncpus )'}' -- "$cur" ) )
-            return 0
+            return
            ;;
        -v|--verbose)
-            __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
+            COMPREPLY=( $( compgen -W '{0..2}' -- "$cur" ) ) # max level ?
+            return
            ;;
        --tesseract-pagesegmode)
-            __ocrmypdf_tesseract-pagesegmode
-            return 0
+            COMPREPLY=( $( compgen -W '{1..13}' -- "$cur" ) )
+            return
            ;;
-        --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)
+        --sidecar|--title|--author|--subject|--keywords|--unpaper-args|--pages|--fast-web-view)
            # argument required but no completions available
-            return 0
-            ;;
-        --tesseract-config|--user-words|--user-patterns|--sidecar)
-            _filedir
-            return 0
+            return
            ;;
    esac

-    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
+    $split && return
 } &&
 complete -F _ocrmypdf ocrmypdf

@@ -29,6 +29,7 @@ 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"
@@ -53,7 +54,6 @@ 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"

@@ -129,15 +129,6 @@ 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"

@@ -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:
@@ -1,4 +1,3 @@
-#!/usr/bin/env python3
 # Copyright (C) 2019 Ian Alexander: https://github.com/ianalexander
 # Copyright (C) 2020 James R Barlow: https://github.com/jbarlow83
 #
@@ -37,19 +36,14 @@ 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 = getenv_bool('OCR_OUTPUT_DIRECTORY_YEAR_MONTH')
-ON_SUCCESS_DELETE = getenv_bool('OCR_ON_SUCCESS_DELETE')
-DESKEW = getenv_bool('OCR_DESKEW')
+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', ''))
 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 = getenv_bool('OCR_USE_POLLING')
+USE_POLLING = bool(os.getenv('OCR_USE_POLLING', ''))
 LOGLEVEL = os.getenv('OCR_LOGLEVEL', 'INFO')
 PATTERNS = ['*.pdf', '*.PDF']

@@ -1,4 +1,3 @@
-#!/usr/bin/env python3
 # webservice.py wrapper for OCRmyPDF
 # Copyright (C) 2019 James R. Barlow: github.com/jbarlow83
 #
@@ -29,7 +28,16 @@ import shlex
 from subprocess import PIPE, run
 from tempfile import TemporaryDirectory

-from flask import Flask, Response, request, send_from_directory
+from flask import (
+    Flask,
+    Response,
+    abort,
+    flash,
+    redirect,
+    request,
+    send_from_directory,
+    url_for,
+)
 from werkzeug.utils import secure_filename

 app = Flask(__name__)
@@ -37,7 +45,7 @@ app.secret_key = "secret"
 app.config['MAX_CONTENT_LENGTH'] = 50_000_000
 app.config.from_envvar("OCRMYPDF_WEBSERVICE_SETTINGS", silent=True)

-ALLOWED_EXTENSIONS = {"pdf"}
+ALLOWED_EXTENSIONS = set(["pdf"])


 def allowed_file(filename):
@@ -2,17 +2,15 @@
 requires = [
  "setuptools >= 30.3.0",
  "wheel",
-  "setuptools_scm[toml] >= 3.4",
+  "cffi",
+  "setuptools_scm",
  "setuptools_scm_git_archive"
 ]
 build-backend = "setuptools.build_meta"

-[tool.setuptools_scm]
-version_scheme = "post-release"
-
 [tool.black]
 line-length = 88
-target-version = ["py37", "py38"]
+target-version = ["py36", "py37", "py38", "py39"]
 skip-string-normalization = true
 include = '\.pyi?$'
 exclude = '''
@@ -30,72 +28,6 @@ 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
@@ -0,0 +1,12 @@
+# 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
@@ -0,0 +1,7 @@
+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
@@ -0,0 +1 @@
+watchdog == 0.10.2
@@ -0,0 +1 @@
+Flask >= 1, < 2
@@ -1,112 +1,58 @@
-[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/jbarlow83/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/jbarlow83/ocrmypdf
-    Tracker = https://github.com/jbarlow83/ocrmypdf/issues
-
-[options]
-packages = find:
-install_requires =
-    Pillow>=8.2.0
-    coloredlogs>=14.0  # strictly optional
-    img2pdf>=0.3.0,<0.5  # pure Python
-    pdfminer.six!=20200720,>=20191110,<=20211012
-    pikepdf>=4.0.0
-    pluggy>=0.13.0,<2
-    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,<3
-webservice =
-    Flask>=1,<3
-
-[options.package_data]
-ocrmypdf =
-    data/sRGB.icc
-    py.typed
-
 [bdist_wheel]
-python-tag = py37
+python-tag = py36

 [aliases]
-test = pytest
+test=pytest

 [check-manifest]
 ignore =
-    .github
+	.github

-[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
+[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:
@@ -1,17 +1,93 @@
-# © 2021 James R. Barlow: github.com/jbarlow83
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+# © 2015 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 setuptools import setup
+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()
+

-# 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',
+    },
 )
@@ -1,35 +0,0 @@
-# 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`
@@ -7,8 +7,7 @@

 from pluggy import HookimplMarker as _HookimplMarker

-from ocrmypdf import helpers, hocrtransform, pdfa, pdfinfo
-from ocrmypdf._concurrent import Executor
+from ocrmypdf import helpers, hocrtransform, leptonica, pdfa, pdfinfo
 from ocrmypdf._jobcontext import PageContext, PdfContext
 from ocrmypdf._version import PROGRAM_NAME, __version__
 from ocrmypdf.api import Verbosity, configure_logging, ocr
@@ -15,7 +15,7 @@ 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_options
+from ocrmypdf._validation import check_closed_streams, check_options
 from ocrmypdf.api import Verbosity, configure_logging
 from ocrmypdf.exceptions import (
    BadArgsError,
@@ -34,6 +34,9 @@ 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'):
        os.nice(5)

@@ -44,10 +47,7 @@ def run(args=None):
        verbosity = Verbosity.quiet
        options.progress_bar = False
    configure_logging(
-        verbosity,
-        progress_bar_friendly=options.progress_bar,
-        manage_root_logger=True,
-        plugin_manager=plugin_manager,
+        verbosity, progress_bar_friendly=options.progress_bar, manage_root_logger=True
    )
    log.debug('ocrmypdf %s', __version__)
    try:
@@ -4,130 +4,134 @@
 # 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/.

+
+import logging
+import logging.handlers
+import multiprocessing
+import os
+import signal
+import sys
 import threading
-from abc import ABC, abstractmethod
+from contextlib import suppress
+from multiprocessing import Pool as ProcessPool
+from multiprocessing.dummy import Pool as ThreadPool
 from typing import Callable, Iterable, Optional

+from tqdm import tqdm

-def _task_noop(*_args, **_kwargs):
-    return
+from ocrmypdf.exceptions import InputFileError


-class NullProgressBar:
-    def __init__(self, **kwargs):
-        pass
+def log_listener(queue):
+    """Listen to the worker processes and forward the messages to logging

-    def __enter__(self):
-        return self
+    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 __exit__(self, exc_type, exc_value, traceback):
-        return False
-
-    def update(self, _arg=None):
-        return
-
-
-class Executor(ABC):
-    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.
+    See https://docs.python.org/3/howto/logging-cookbook.html#logging-to-a-single-file-from-multiple-processes
    """

-    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)
+    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()
@@ -10,7 +10,6 @@
 import logging
 import os
 import re
-import sys
 from io import BytesIO
 from os import fspath
 from pathlib import Path
@@ -37,15 +36,14 @@ For details see:
 ---------------------------------------------------------------------
 """

-# Most reliable what to get the bitness of Python interpreter, according to Python docs
-_is_64bit = sys.maxsize > 2 ** 32
-
 _gswin = None
 if os.name == 'nt':
-    if _is_64bit:
-        _gswin = 'gswin64c'
-    else:
-        _gswin = 'gswin32c'
+    _gswin = which('gswin64c')
+    if not _gswin:
+        _gswin = which('gswin32c')
+        if not _gswin:
+            raise MissingDependencyError(missing_gs_error)
+    _gswin = Path(_gswin).stem

 GS = _gswin if _gswin else 'gs'
 del _gswin
@@ -97,7 +95,6 @@ def rasterize_pdf(
            '-dSAFER',
            '-dBATCH',
            '-dNOPAUSE',
-            '-dInterpolateControl=-1',
            f'-sDEVICE={raster_device}',
            f'-dFirstPage={pageno}',
            f'-dLastPage={pageno}',
@@ -169,7 +166,6 @@ class GhostscriptFollower:
 def generate_pdfa(
    pdf_pages,
    output_file: os.PathLike,
-    *,
    compression: str,
    pdf_version: str = '1.5',
    pdfa_part: str = '2',
@@ -199,11 +195,10 @@ 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 = ('/' + strategy) if version() < '9.19' else strategy
+    strategy = 'RGB' if version() >= '9.19' else '/RGB'

    if version() == '9.23':
        # 9.23: added JPEG passthrough as a new feature, but with a bug that
@@ -41,17 +41,9 @@ 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])
@@ -59,7 +59,3 @@ 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)
@@ -8,13 +8,13 @@
 """Interface to Tesseract executable"""

 import logging
-import re
-from distutils.version import StrictVersion
-from math import pi
+import os
+import shutil
+from collections import namedtuple
 from os import fspath
 from pathlib import Path
 from subprocess import PIPE, STDOUT, CalledProcessError, TimeoutExpired
-from typing import Dict, Iterator, List, Optional
+from typing import List, Optional

 from PIL import Image

@@ -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"
@@ -46,13 +46,6 @@ 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):
    def process(self, msg, kwargs):
@@ -60,31 +53,31 @@ class TesseractLoggerAdapter(logging.LoggerAdapter):
        return '[tesseract] %s' % (msg), kwargs


-class TesseractVersion(StrictVersion):
-
-    version_re = re.compile(
-        r'''
-        ^(\d+) \. (\d+) (\. (\d+))?  # groups: 1/major, 2/minor, 3/[skip], 4/patch
-        [-]?  # optional hyphen separator
-        (?: ((?:alpha|beta|rc|dev)\d*)? [.\-\ ]? (\d+)? )?  # 5/prerelease, 6/prerelease_num
-        (?:(?:-\d+)?-g[0-9a-f]+)?  # untagged git version
-        $
-        ''',
-        re.VERBOSE | re.ASCII,
-    )
-
-    def parse(self, vstring):
-        try:
-            super().parse(vstring)
-        except TypeError as e:
-            if 'int() argument must be a string' in str(e):
-                super().parse(vstring + '-0')
-
-
-def version() -> str:
+def version():
    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?

@@ -94,11 +87,6 @@ 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 = (
@@ -127,7 +115,7 @@ def get_languages():
        if line.startswith('Error'):
            raise MissingDependencyError(lang_error(output))
    _header, *rest = output.splitlines()
-    return {lang.strip() for lang in rest}
+    return set(lang.strip() for lang in rest)


 def tess_base_args(langs: List[str], engine_mode: Optional[int]) -> List[str]:
@@ -139,20 +127,7 @@ def tess_base_args(langs: List[str], engine_mode: Optional[int]) -> List[str]:
    return args


-def _parse_tesseract_output(binary_output: bytes) -> Dict[str, str]:
-    def g():
-        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 {k: v for k, v in g()}
-
-
-def get_orientation(
-    input_file: Path, engine_mode: Optional[int], timeout: float
-) -> OrientationConfidence:
+def get_orientation(input_file: Path, engine_mode: Optional[int], timeout: float):
    args_tesseract = tess_base_args(['osd'], engine_mode) + [
        '--psm',
        '0',
@@ -162,6 +137,7 @@ def get_orientation(

    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:
@@ -173,42 +149,19 @@ def get_orientation(
        ):
            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()

-    osd = _parse_tesseract_output(p.stdout)
-    angle = int(osd.get('Orientation in degrees', 0))
-    oc = OrientationConfidence(
-        angle=angle, confidence=float(osd.get('Orientation confidence', 0))
-    )
-    return oc
-
-
-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:  # Not enough info for a skew angle
-            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
+        angle = int(osd.get('Orientation in degrees', 0))
+        oc = OrientationConfidence(
+            angle=angle, confidence=float(osd.get('Orientation confidence', 0))
+        )
+        return oc


 def tesseract_log_output(stream):
@@ -268,7 +221,6 @@ def _generate_null_hocr(output_hocr, output_text, image):


 def generate_hocr(
-    *,
    input_file: Path,
    output_hocr: Path,
    output_text: Path,
@@ -277,11 +229,9 @@ 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)
@@ -289,9 +239,6 @@ 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])

@@ -300,8 +247,7 @@ def generate_hocr(

    # Reminder: test suite tesseract test plugins will break after any changes
    # to the number of order parameters here
-    args_tesseract.extend([fspath(input_file), fspath(prefix), 'hocr', 'txt'])
-    args_tesseract.extend(tessconfig)
+    args_tesseract.extend([input_file, prefix, 'hocr', 'txt'] + tessconfig)
    try:
        p = run(args_tesseract, stdout=PIPE, stderr=STDOUT, timeout=timeout, check=True)
        stdout = p.stdout
@@ -323,7 +269,7 @@ 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():
-            prefix.with_suffix('.txt').replace(output_text)
+            shutil.move(prefix.with_suffix('.txt'), output_text)


 def use_skip_page(output_pdf, output_text):
@@ -343,15 +289,20 @@ def generate_pdf(
    tessconfig: List[str],
    timeout: float,
    pagesegmode: int,
-    thresholding: int,
    user_words,
    user_patterns,
 ):
-    """Generate a PDF using Tesseract's internal PDF generator.
+    """Use Tesseract to render a PDF.

-    We specifically a text-only PDF which is more suitable for combining with
-    the input page.
+    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)
    """
+
    args_tesseract = tess_base_args(languages, engine_mode)

    if pagesegmode is not None:
@@ -359,27 +310,23 @@ 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 = output_pdf.parent / Path(output_pdf.stem)
+    prefix = os.path.splitext(output_pdf)[0]  # Tesseract appends suffixes

    # Reminder: test suite tesseract test plugins might break after any changes
    # to the number of order parameters here

-    args_tesseract.extend([fspath(input_file), fspath(prefix), 'pdf', 'txt'])
-    args_tesseract.extend(tessconfig)
+    args_tesseract.extend([input_file, prefix, 'pdf', 'txt'] + tessconfig)
    try:
        p = run(args_tesseract, stdout=PIPE, stderr=STDOUT, timeout=timeout, check=True)
        stdout = p.stdout
-        if prefix.with_suffix('.txt').exists():
-            prefix.with_suffix('.txt').replace(output_text)
+        if os.path.exists(prefix + '.txt'):
+            shutil.move(prefix + '.txt', output_text)
    except TimeoutExpired:
        page_timedout(timeout)
        use_skip_page(output_pdf, output_text)
@@ -45,7 +45,7 @@ def _setup_unpaper_io(tmpdir: Path, input_file: Path) -> Tuple[Path, Path]:
                    im = im.convert(mode='1')
                else:
                    im = im.convert(mode='RGB')
-            except OSError as e:
+            except IOError as e:
                raise MissingDependencyError(
                    "Could not convert image with type " + im.mode
                ) from e
@@ -69,7 +69,7 @@ def _setup_unpaper_io(tmpdir: Path, input_file: Path) -> Tuple[Path, Path]:


 def run(
-    input_file: Path, output_file: Path, *, dpi: DecFloat, mode_args: List[str]
+    input_file: Path, output_file: Path, dpi: DecFloat, mode_args: List[str]
 ) -> None:
    args_unpaper = ['unpaper', '-v', '--dpi', str(round(dpi, 6))] + mode_args

@@ -96,12 +96,12 @@ def run(
        try:
            with Image.open(output_pnm) as imout:
                imout.save(output_file, dpi=(dpi, dpi))
-        except OSError as e:
+        except (FileNotFoundError, OSError):
            raise SubprocessOutputError(
                "unpaper: failed to produce the expected output file. "
                + " Called with: "
                + str(args_unpaper)
-            ) from e
+            ) from None


 def validate_custom_args(args: str) -> List[str]:
@@ -114,7 +114,6 @@ def validate_custom_args(args: str) -> List[str]:
 def clean(
    input_file: Path,
    output_file: Path,
-    *,
    dpi: DecFloat,
    unpaper_args: Optional[List[str]] = None,
 ):
@@ -131,4 +130,4 @@ def clean(
    ]
    if not unpaper_args:
        unpaper_args = default_args
-    run(input_file, output_file, dpi=dpi, mode_args=unpaper_args)
+    run(input_file, output_file, dpi, unpaper_args)
@@ -6,50 +6,33 @@


 import logging
-import uuid
 from contextlib import suppress
 from pathlib import Path
 from typing import Optional

-from pikepdf import (
-    Dictionary,
-    Name,
-    Object,
-    Operator,
-    Page,
-    Pdf,
-    PdfError,
-    PdfMatrix,
-    Stream,
-    parse_content_stream,
-    unparse_content_stream,
-)
+import pikepdf

 log = logging.getLogger(__name__)
 MAX_REPLACE_PAGES = 100


-def _ensure_dictionary(obj, name):
-    if name not in obj:
-        obj[name] = Dictionary({})
-    return obj[name]
+def _update_page_resources(*, page, font, font_key, procset):
+    """Update this page's fonts with a reference to the Glyphless font"""

-
-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 '/Resources' not in page:
+        page['/Resources'] = pikepdf.Dictionary({})
+    resources = page['/Resources']
+    try:
+        fonts = resources['/Font']
+    except KeyError:
+        fonts = pikepdf.Dictionary({})
    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
-    if procset:
-        resources['/ProcSet'] = procset
+    resources['/ProcSet'] = procset


 def strip_invisible_text(pdf, page):
@@ -58,26 +41,43 @@ def strip_invisible_text(pdf, page):
    render_mode = 0
    text_objects = []

-    for operands, operator in parse_content_stream(page, ''):
+    page.page_contents_coalesce()
+    for operands, operator in pikepdf.parse_content_stream(page, ''):
        if not in_text_obj:
-            if operator == Operator('BT'):
+            if operator == pikepdf.Operator('BT'):
                in_text_obj = True
                render_mode = 0
                text_objects.append((operands, operator))
            else:
                stream.append((operands, operator))
        else:
-            if operator == Operator('Tr'):
+            if operator == pikepdf.Operator('Tr'):
                render_mode = operands[0]
            text_objects.append((operands, operator))
-            if operator == Operator('ET'):
+            if operator == pikepdf.Operator('ET'):
                in_text_obj = False
                if render_mode != 3:
                    stream.extend(text_objects)
                text_objects.clear()

-    content_stream = unparse_content_stream(stream)
-    page.Contents = Stream(pdf, content_stream)
+    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)


 class OcrGrafter:
@@ -85,14 +85,14 @@ class OcrGrafter:
        self.context = context
        self.path_base = context.origin

-        self.pdf_base = Pdf.open(self.path_base)
+        self.pdf_base = pikepdf.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(
-            Object.parse(b'[ /PDF /Text /ImageB /ImageC /ImageI ]')
+            pikepdf.Object.parse(b'[ /PDF /Text /ImageB /ImageC /ImageI ]')
        )

        self.emplacements = 1
@@ -116,7 +116,7 @@ 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 Pdf.open(path_image) as pdf_image:
+            with pikepdf.open(image) as pdf_image:
                self.emplacements += 1
                foreign_image_page = pdf_image.pages[0]
                self.pdf_base.pages.append(foreign_image_page)
@@ -169,13 +169,13 @@ class OcrGrafter:
        """

        page0 = self.pdf_base.pages[0]
-        _update_resources(
-            obj=page0, font=self.font, font_key=self.font_key, procset=self.procset
+        _update_page_resources(
+            page=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 updating
+        # {interim_count} is the opened file we were updateing
        # {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 +189,7 @@ class OcrGrafter:
        self.pdf_base.save(next_file)
        self.pdf_base.close()

-        self.pdf_base = Pdf.open(next_file)
+        self.pdf_base = pikepdf.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
@@ -205,12 +205,11 @@ class OcrGrafter:
        font, font_key = None, None
        possible_font_names = ('/f-0-0', '/F1')
        try:
-            with Pdf.open(text) as pdf_text:
+            with pikepdf.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:
@@ -219,7 +218,7 @@ class OcrGrafter:
                if pdf_text_font:
                    font = self.pdf_base.copy_foreign(pdf_text_font)
                return font, font_key
-        except (FileNotFoundError, PdfError):
+        except (FileNotFoundError, pikepdf.PdfError):
            # PdfError occurs if a 0-length file is written e.g. due to OCR timeout
            return None, None

@@ -228,9 +227,9 @@ class OcrGrafter:
        *,
        page_num: int,
        textpdf: Path,
-        font: Object,
-        font_key: Object,
-        procset: Object,
+        font: pikepdf.Object,
+        font_key: pikepdf.Object,
+        procset: pikepdf.Object,
        text_rotation: int,
        strip_old_text: bool,
    ):
@@ -241,7 +240,7 @@ class OcrGrafter:
            return

        # This is a pointer indicating a specific page in the base file
-        with Pdf.open(textpdf) as pdf_text:
+        with pikepdf.open(textpdf) as pdf_text:
            pdf_text_contents = pdf_text.pages[0].Contents.read_bytes()

            base_page = self.pdf_base.pages.p(page_num)
@@ -256,13 +255,13 @@ class OcrGrafter:
            mediabox = [float(base_page.MediaBox[v]) for v in range(4)]
            wp, hp = mediabox[2] - mediabox[0], mediabox[3] - mediabox[1]

-            translate = PdfMatrix().translated(-wt / 2, -ht / 2)
-            untranslate = PdfMatrix().translated(wp / 2, hp / 2)
-            corner = PdfMatrix().translated(mediabox[0], 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])
            # -rotation because the input is a clockwise angle and this formula
            # uses CCW
            text_rotation = -text_rotation % 360
-            rotate = PdfMatrix().rotated(text_rotation)
+            rotate = pikepdf.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
@@ -273,36 +272,24 @@ class OcrGrafter:
            scale_y = hp / ht

            # log.debug('%r', scale_x, scale_y)
-            scale = PdfMatrix().scaled(scale_x, scale_y)
+            scale = pikepdf.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

-            base_resources = _ensure_dictionary(base_page, Name.Resources)
-            base_xobjs = _ensure_dictionary(base_resources, Name.XObject)
-            text_xobj_name = Name('/' + str(uuid.uuid4()))
-            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]
+            pdf_text_contents = (
+                b'q %s cm\n' % ctm.encode() + pdf_text_contents + b'\nQ\n'
            )

-            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)
+            new_text_layer = pikepdf.Stream(self.pdf_base, pdf_text_contents)

            if strip_old_text:
                strip_invisible_text(self.pdf_base, base_page)

-            base_page.contents_add(new_text_layer, prepend=True)
+            base_page.page_contents_add(new_text_layer, prepend=True)

-            _update_resources(
-                obj=base_page, font=font, font_key=font_key, procset=procset
+            _update_page_resources(
+                page=base_page, font=font, font_key=font_key, procset=procset
            )
@@ -28,14 +28,22 @@ 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()
-        tqdm.write(msg.rstrip(), end='\n', file=self.file)
+        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)

    def flush(self):
        with suppress(AttributeError):
@@ -13,14 +13,15 @@ from contextlib import suppress
 from datetime import datetime, timezone
 from pathlib import Path
 from shutil import copyfileobj
-from typing import Dict, Iterable, Optional
+from typing import BinaryIO, Dict, Iterable, Optional, Union, cast

 import img2pdf
 import pikepdf
 from pikepdf.models.metadata import encode_pdf_date
 from PIL import Image, ImageColor, ImageDraw
+from tqdm import tqdm

-from ocrmypdf._concurrent import Executor
+from ocrmypdf import leptonica
 from ocrmypdf._exec import unpaper
 from ocrmypdf._jobcontext import PageContext, PdfContext
 from ocrmypdf._version import PROGRAM_NAME
@@ -47,7 +48,7 @@ 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 OSError as e:
+    except EnvironmentError as e:
        # Recover the original filename
        log.error(str(e).replace(str(input_file), str(options.input_file)))
        raise UnsupportedImageFormatError() from e
@@ -134,7 +135,7 @@ 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 OSError as e:
+    except EnvironmentError 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
@@ -145,8 +146,6 @@ def triage(original_filename, input_file, output_file, options):

 def get_pdfinfo(
    input_file,
-    *,
-    executor: Executor,
    detailed_analysis=False,
    progbar=False,
    max_workers=None,
@@ -159,12 +158,11 @@ def get_pdfinfo(
            progbar=progbar,
            max_workers=max_workers,
            check_pages=check_pages,
-            executor=executor,
        )
-    except pikepdf.PasswordError as e:
-        raise EncryptedPdfError() from e
-    except pikepdf.PdfError as e:
-        raise InputFileError() from e
+    except pikepdf.PasswordError:
+        raise EncryptedPdfError()
+    except pikepdf.PdfError:
+        raise InputFileError()


 def validate_pdfinfo_options(context: PdfContext):
@@ -465,8 +463,7 @@ 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)
-        raise NotImplementedError("--remove-background is temporarily not implemented")
+        leptonica.remove_background(input_file, output_file)
        return output_file
    else:
        log.info("background removal skipped on mono page")
@@ -476,30 +473,14 @@ 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)
-
-    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=Image.BICUBIC, fillcolor='white'
-        )
-        deskewed.save(output_file, dpi=dpi)
-
+    leptonica.deskew(input_file, output_file, dpi.x)
    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=dpi.x,
-        unpaper_args=page_context.options.unpaper_args,
-    )
+    unpaper.clean(input_file, output_file, dpi.x, page_context.options.unpaper_args)
    return output_file


@@ -511,6 +492,10 @@ 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:
@@ -520,7 +505,6 @@ 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
            ):
@@ -529,15 +513,25 @@ 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')
+                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

        filter_im = page_context.plugin_manager.hook.filter_ocr_image(
            page=page_context, image=im
@@ -617,10 +611,6 @@ def create_pdf_page_from_image(
        )
        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


@@ -630,9 +620,9 @@ 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_filename=hocr, dpi=dpi.x)  # square
+    hocrtransform = HocrTransform(hocr, dpi.x)  # square
    hocrtransform.to_pdf(
-        out_filename=output_file,
+        output_file,
        image_filename=None,
        show_bounding_boxes=False if not debug_mode else True,
        invisible_text=True if not debug_mode else False,
@@ -684,6 +674,11 @@ 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

@@ -731,11 +726,7 @@ 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=(
-            context.plugin_manager.hook.get_progressbar_class()
-            if options.progress_bar
-            else None
-        ),
+        progressbar_class=tqdm if options.progress_bar else None,
    )

    return output_file
@@ -748,24 +739,6 @@ 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
@@ -810,7 +783,9 @@ def metadata_fixup(working_file: Path, context: PdfContext):

        pdf.save(
            output_file,
-            **get_pdf_save_settings(options.output_type),
+            compress_streams=True,
+            preserve_pdfa=True,
+            object_stream_mode=pikepdf.ObjectStreamMode.generate,
            linearize=(  # Don't linearize if optimize() will be linearizing too
                should_linearize(working_file, context)
                if options.optimize == 0
@@ -821,40 +796,26 @@ def metadata_fixup(working_file: Path, context: PdfContext):
    return output_file


-def optimize_pdf(input_file: Path, context: PdfContext, executor: Executor):
+def optimize_pdf(input_file: Path, context: PdfContext):
    output_file = context.get_path('optimize.pdf')
    save_settings = dict(
+        compress_streams=True,
+        preserve_pdfa=True,
+        object_stream_mode=pikepdf.ObjectStreamMode.generate,
        linearize=should_linearize(input_file, context),
-        **get_pdf_save_settings(context.options.output_type),
    )
-    optimize(input_file, output_file, context, save_settings, executor)
+    optimize(input_file, output_file, context, save_settings)
    return output_file


-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 (frm, to), txt_file in enumerate_compress_ranges(txt_files):
-            if frm != 1:
+        for page_num, txt_file in enumerate(txt_files):
+            if page_num != 0:
                stream.write('\f')  # Form feed between pages
            if txt_file:
-                with open(txt_file, encoding="utf-8") as in_:
+                with open(txt_file, 'r', 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
@@ -864,11 +825,7 @@ def merge_sidecars(txt_files: Iterable[Optional[Path]], context: PdfContext):
                    else:
                        stream.write(txt)
            else:
-                if frm != to:
-                    pages = f'{frm}-{to}'
-                else:
-                    pages = f'{frm}'
-                stream.write(f'[OCR skipped on page(s) {pages}]')
+                stream.write(f'[OCR skipped on page {(page_num + 1)}]')
    return output_file


@@ -11,7 +11,7 @@ import importlib.util
 import pkgutil
 import sys
 from pathlib import Path
-from typing import List, Sequence, Tuple, Union
+from typing import List, Tuple, Union

 import pluggy

@@ -66,26 +66,15 @@ class OcrmypdfPluginManager(pluggy.PluginManager):

        # 1. Register builtins
        if self.__builtins:
-            for module in sorted(
-                pkgutil.iter_modules(ocrmypdf.builtin_plugins.__path__)
-            ):
+            for module in pkgutil.iter_modules(ocrmypdf.builtin_plugins.__path__):
                name = f'ocrmypdf.builtin_plugins.{module.name}'
                module = importlib.import_module(name)
                self.register(module)

-        # 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
+        # 2. Register setuptools plugins
        self.load_setuptools_entrypoints('ocrmypdf')

-        # 4. Register plugins specified on command line
+        # 3. Register plugins specified on command line
        for name in self.__plugins:
            if isinstance(name, Path) or name.endswith('.py'):
                # Import by filename
@@ -110,7 +99,7 @@ def get_plugin_manager(plugins: List[Union[str, Path]], builtins=True):


 def get_parser_options_plugins(
-    args: Sequence[str],
+    args,
 ) -> 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)
@@ -5,22 +5,20 @@
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.


-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, cast
+from typing import List, NamedTuple, Optional, Tuple

+import pikepdf
 import PIL

-from ocrmypdf._concurrent import Executor, setup_executor
+from ocrmypdf._concurrent import exec_progress_pool
 from ocrmypdf._graft import OcrGrafter
 from ocrmypdf._jobcontext import PageContext, PdfContext, cleanup_working_files
 from ocrmypdf._logging import PageNumberFilter
@@ -49,7 +47,7 @@ from ocrmypdf._pipeline import (
    triage,
    validate_pdfinfo_options,
 )
-from ocrmypdf._plugin_manager import OcrmypdfPluginManager, get_plugin_manager
+from ocrmypdf._plugin_manager import get_plugin_manager
 from ocrmypdf._validation import (
    check_requested_output_file,
    create_input_file,
@@ -68,7 +66,7 @@ from ocrmypdf.pdfa import file_claims_pdfa
 log = logging.getLogger(__name__)


-class PageResult(NamedTuple):  # pylint: disable=inherit-non-class
+class PageResult(NamedTuple):
    pageno: int
    pdf_page_from_image: Optional[Path]
    ocr: Optional[Path]
@@ -168,7 +166,7 @@ def make_intermediate_images(
    return ocr_image, preprocess_out


-def exec_page_sync(page_context: PageContext) -> PageResult:
+def exec_page_sync(page_context: PageContext):
    options = page_context.options
    tls.pageno = page_context.pageno + 1

@@ -226,17 +224,17 @@ def exec_page_sync(page_context: PageContext) -> PageResult:
    )


-def post_process(pdf_file: Path, context: PdfContext, executor: Executor) -> Path:
+def post_process(pdf_file, context: PdfContext):
    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, executor)
+    return optimize_pdf(pdf_out, context)


-def worker_init(max_pixels: int) -> None:
+def worker_init(max_pixels: int):
    # 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.
@@ -244,7 +242,7 @@ def worker_init(max_pixels: int) -> None:
    pikepdf_enable_mmap()


-def exec_concurrent(context: PdfContext, executor: Executor) -> None:
+def exec_concurrent(context: PdfContext):
    """Execute the pipeline concurrently"""

    # Run exec_page_sync on every page context
@@ -271,7 +269,7 @@ def exec_concurrent(context: PdfContext, executor: Executor) -> None:
        finally:
            tls.pageno = None

-    executor(
+    exec_progress_pool(
        use_threads=options.use_threads,
        max_workers=max_workers,
        tqdm_kwargs=dict(
@@ -281,7 +279,7 @@ def exec_concurrent(context: PdfContext, executor: Executor) -> None:
            unit_scale=0.5,
            disable=not options.progress_bar,
        ),
-        worker_initializer=partial(worker_init, PIL.Image.MAX_IMAGE_PIXELS),
+        task_initializer=partial(worker_init, PIL.Image.MAX_IMAGE_PIXELS),
        task=exec_page_sync,
        task_arguments=context.get_page_contexts(),
        task_finished=update_page,
@@ -296,18 +294,15 @@ def exec_concurrent(context: PdfContext, executor: Executor) -> None:
    # Merge layers to one single pdf
    pdf = ocrgraft.finalize()

-    if options.output_type != 'none':
-        # PDF/A and metadata
-        log.info("Postprocessing...")
-        pdf = post_process(pdf, context, executor)
+    # PDF/A and metadata
+    log.info("Postprocessing...")
+    pdf = post_process(pdf, context)

-        # Copy PDF file to destination
-        copy_final(pdf, options.output_file, context)
+    # Copy PDF file to destination
+    copy_final(pdf, options.output_file, context)


-def configure_debug_logging(
-    log_filename: Path, prefix: str = ''
-) -> logging.FileHandler:
+def configure_debug_logging(log_filename: Path, prefix: str = ''):
    """
    Create a debug log file at a specified location.

@@ -326,12 +321,7 @@ def configure_debug_logging(
    return log_file_handler


-def run_pipeline(
-    options: argparse.Namespace,
-    *,
-    plugin_manager: Optional[OcrmypdfPluginManager],
-    api: bool = False,
-) -> ExitCode:
+def run_pipeline(options, *, plugin_manager, api=False):
    # 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.)
@@ -356,7 +346,6 @@ def run_pipeline(

    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)
@@ -369,7 +358,6 @@ def run_pipeline(
        # 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
@@ -382,7 +370,7 @@ def run_pipeline(
        validate_pdfinfo_options(context)

        # Execute the pipeline
-        exec_concurrent(context, executor)
+        exec_concurrent(context)

        if options.output_file == '-':
            log.info("Output sent to stdout")
@@ -390,7 +378,7 @@ def run_pipeline(
            hasattr(options.output_file, 'writable') and options.output_file.writable()
        ):
            log.info("Output written to stream")
-        elif samefile(options.output_file, Path(os.devnull)):
+        elif samefile(options.output_file, os.devnull):
            pass  # Say nothing when sending to dev null
        else:
            if options.output_type.startswith('pdfa'):
@@ -410,40 +398,19 @@ def run_pipeline(
                return ExitCode.invalid_output_pdf
            report_output_file_size(options, start_input_file, options.output_file)

-    except (KeyboardInterrupt if not api else NeverRaise):
+    except (KeyboardInterrupt if not api else NeverRaise) as e:
        if options.verbose >= 1:
            log.exception("KeyboardInterrupt")
        else:
            log.error("KeyboardInterrupt")
        return ExitCode.ctrl_c
    except (ExitCodeException if not api else NeverRaise) as e:
-        e = cast(ExitCodeException, e)
-        if options.verbose >= 1:
-            log.exception("ExitCodeException")
-        elif str(e):
+        if str(e):
            log.error("%s: %s", type(e).__name__, str(e))
        else:
            log.error(type(e).__name__)
        return e.exit_code
-    except (PIL.Image.DecompressionBombError if not api else NeverRaise) as e:
-        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,
-    ) as e:
-        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
+    except (Exception if not api else NeverRaise) as e:  # pylint: disable=broad-except
        log.exception("An exception occurred while executing the pipeline")
        return ExitCode.other_error
    finally:
@@ -451,7 +418,7 @@ def run_pipeline(
            try:
                debug_log_handler.close()
                log.removeHandler(debug_log_handler)
-            except OSError as e:
+            except EnvironmentError as e:
                print(e, file=sys.stderr)
        cleanup_working_files(work_folder, options)

@@ -0,0 +1,118 @@
+# 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
+    )
@@ -13,31 +13,39 @@ import sys
 import unicodedata
 from pathlib import Path
 from shutil import copyfileobj
-from typing import List, Set, Tuple
+from typing import List, Set, Tuple, Union

 import pikepdf
 import PIL

 from ocrmypdf._exec import jbig2enc, pngquant, unpaper
+from ocrmypdf._unicodefun import verify_python3_env
 from ocrmypdf.exceptions import (
    BadArgsError,
    InputFileError,
    MissingDependencyError,
    OutputFileAccessError,
 )
-from ocrmypdf.helpers import is_file_writable, monotonic, safe_symlink, samefile
-from ocrmypdf.hocrtransform import HOCR_OK_LANGS
+from ocrmypdf.helpers import (
+    is_file_writable,
+    is_iterable_notstr,
+    monotonic,
+    safe_symlink,
+)
 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():
@@ -57,14 +65,13 @@ def check_options_languages(options, ocr_engine_languages):
            log.debug("No language specified; assuming --language %s", DEFAULT_LANGUAGE)
    if not ocr_engine_languages:
        return
-    missing_languages = options.languages - ocr_engine_languages
-    if missing_languages:
+    if not options.languages.issubset(ocr_engine_languages):
        msg = (
-            "OCR engine does not have language data for the following "
+            f"OCR engine does not have language data for the following "
            "requested languages: \n"
        )
-        msg += '\n'.join(lang for lang in missing_languages)
-        msg += '\nNote: most languages are identified by a 3-digit ISO 639-2 Code'
+        for lang in options.languages - ocr_engine_languages:
+            msg += lang + '\n'
        raise MissingDependencyError(msg)


@@ -72,18 +79,12 @@ def check_options_output(options):
    is_latin = options.languages.issubset(HOCR_OK_LANGS)

    if options.pdf_renderer.startswith('hocr') and not is_latin:
-        log.warning(
+        msg = (
            "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."
-        )
-
-    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."
+            "--pdf-renderer auto (the default) to avoid this issue."
        )
+        log.warning(msg)

    lossless_reconstruction = False
    if not any(
@@ -110,15 +111,7 @@ 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):
@@ -140,10 +133,12 @@ def check_options_preprocessing(options):
                    options.unpaper_args
                )
        except Exception as e:
-            raise BadArgsError("--unpaper-args: " + str(e)) from e
+            raise BadArgsError(str(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:
@@ -157,12 +152,10 @@ def _pages_from_ranges(ranges: str) -> Set[int]:
            try:
                new_pages = list(range(int(start) - 1, int(end)))
                if not new_pages:
-                    raise BadArgsError(
-                        f"invalid page subrange '{start}-{end}'"
-                    ) from None
+                    raise BadArgsError(f"invalid page subrange '{start}-{end}'")
                pages.extend(new_pages)
            except ValueError:
-                raise BadArgsError(f"invalid page subrange '{g}'") from None
+                raise BadArgsError("invalid page range") from None

    if not pages:
        raise BadArgsError(
@@ -184,11 +177,15 @@ 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)

@@ -272,6 +269,54 @@ 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
@@ -294,17 +339,7 @@ def create_input_file(options, work_folder: Path) -> Tuple[Path, str]:
            safe_symlink(options.input_file, target)
            return target, os.fspath(options.input_file)
        except FileNotFoundError:
-            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)
+            raise InputFileError(f"File not found - {options.input_file}")


 def check_requested_output_file(options):
@@ -5,12 +5,9 @@
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.


-try:
-    from importlib_metadata import version as _package_version
-except ImportError:
-    from importlib.metadata import version as _package_version
+import pkg_resources

 PROGRAM_NAME = 'ocrmypdf'

 # Official PEP 396
-__version__ = _package_version('ocrmypdf')
+__version__ = pkg_resources.get_distribution('ocrmypdf').version
@@ -28,7 +28,7 @@ except ModuleNotFoundError:
    coloredlogs = None


-StrPath = Union[Path, AnyStr]
+StrPath = Union[os.PathLike, AnyStr]
 PathOrIO = Union[BinaryIO, StrPath]

 _api_lock = threading.Lock()
@@ -45,23 +45,18 @@ 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. 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.
+    ``--verbose`` and applies those settings.

    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
@@ -79,11 +74,12 @@ def configure_logging(
    their own debug logging.

    Args:
-        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.
+        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

    Returns:
        The toplevel logger for ocrmypdf (or the root logger, if we are managing it).
@@ -94,11 +90,9 @@ def configure_logging(
    log = logging.getLogger(prefix)
    log.setLevel(logging.DEBUG)

-    console = None
-    if plugin_manager and progress_bar_friendly:
-        console = plugin_manager.hook.get_logging_console()
-
-    if not console:
+    if progress_bar_friendly:
+        console = logging.StreamHandler(stream=TqdmConsole(sys.stderr))
+    else:
        console = logging.StreamHandler(stream=sys.stderr)

    if verbosity < 0:
@@ -228,6 +222,7 @@ 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,
@@ -242,7 +237,6 @@ 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,
@@ -251,7 +245,6 @@ 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,
@@ -303,9 +296,6 @@ 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)):
@@ -325,8 +315,7 @@ def ocr(  # pylint: disable=unused-argument
        # they might install different plugins, and generally speaking we have areas
        # of code that use global state.

-        if not plugin_manager:
-            plugin_manager = get_plugin_manager(plugins)
+        plugin_manager = get_plugin_manager(plugins)
        plugin_manager.hook.add_options(parser=parser)  # pylint: disable=no-member

        if 'verbose' in kwargs:
@@ -335,17 +324,3 @@ 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',
-]
@@ -3,7 +3,3 @@
 # 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.
@@ -1,172 +0,0 @@
-# © 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/.
-
-# © 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/.
-
-
-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 multiprocessing.pool import Pool, ThreadPool
-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:
-    # 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):
-    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)
-                    pbar.update()
-            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))
@@ -1,14 +0,0 @@
-# © 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/.
-
-from ocrmypdf import hookimpl
-
-
-@hookimpl
-def filter_pdf_page(
-    page, image_filename, output_pdf
-):  # pylint: disable=unused-argument
-    return output_pdf
@@ -10,7 +10,8 @@ import os

 from ocrmypdf import hookimpl
 from ocrmypdf._exec import tesseract
-from ocrmypdf.cli import numeric, str_to_int
+from ocrmypdf.cli import numeric
+from ocrmypdf.exceptions import MissingDependencyError
 from ocrmypdf.helpers import clamp
 from ocrmypdf.pluginspec import OcrEngine
 from ocrmypdf.subprocess import check_external_program
@@ -43,27 +44,13 @@ 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,
@@ -93,24 +80,24 @@ def check_options(options):
        program='tesseract',
        package={'linux': 'tesseract-ocr'},
        version_checker=tesseract.version,
-        need_version='4.0.0-beta.1',  # using backport for Travis CI
-        version_parser=tesseract.TesseractVersion,
+        need_version='4.0.0',  # using backport for Travis CI
    )

    # 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 (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."
+            "Tesseract 4.0 ignores --user-words and --user-patterns, so these "
+            "arguments have no effect."
        )
    if options.tesseract_pagesegmode in (0, 2):
        log.warning(
@@ -162,15 +149,6 @@ 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(
@@ -182,7 +160,6 @@ 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,
        )
@@ -198,7 +175,6 @@ 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,
        )
@@ -6,47 +6,30 @@


 import argparse
-from typing import Any, Callable, Mapping, Optional, TypeVar
+from typing import Optional, Type, TypeVar

 from ocrmypdf._version import PROGRAM_NAME as _PROGRAM_NAME
 from ocrmypdf._version import __version__ as _VERSION

-T = TypeVar('T', int, float)
+T = TypeVar('T')


-def numeric(
-    basetype: Callable[[Any], T], min_: Optional[T] = None, max_: Optional[T] = None
-):
+def numeric(basetype: Type[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(s: str) -> T:
-        value = basetype(s)
+    def _numeric(string):
+        value = basetype(string)
        if (min_ is not None and value < min_) or (max_ is not None and value > max_):
-            raise argparse.ArgumentTypeError(
-                f"{s!r} not in valid range {(min_, max_)!r}"
-            )
+            msg = "%r not in valid range %r" % (string, (min_, max_))
+            raise argparse.ArgumentTypeError(msg)
        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())}"
-            )
-
-    return _str_to_int
-
-
 class ArgumentParser(argparse.ArgumentParser):
    """Override parser's default behavior of calling sys.exit()

@@ -162,7 +145,7 @@ Online documentation is located at:
    )
    parser.add_argument(
        '--output-type',
-        choices=['pdfa', 'pdf', 'pdfa-1', 'pdfa-2', 'pdfa-3', 'none'],
+        choices=['pdfa', 'pdf', 'pdfa-1', 'pdfa-2', 'pdfa-3'],
        default='pdfa',
        help="Choose output type. 'pdfa' creates a PDF/A-2b compliant file for "
        "long term archiving (default, recommended) but may not suitable "
@@ -170,8 +153,7 @@ 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. 'none' will produce no output, which may be helpful if "
-        "only the --sidecar is desired.",
+        "PDF/A3-b file.",
    )

    # Use null string '\0' as sentinel to indicate the user supplied no argument,
@@ -185,8 +167,7 @@ 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; the next "
-        "argument must NOT be the name of the input PDF. "
+        "If FILE is omitted, the sidecar file be named {output_file}.txt "
        "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.",
@@ -302,6 +283,14 @@ 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(
@@ -348,9 +337,8 @@ Online documentation is located at:
            "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."
+            "2 - do some lossy optimizations; "
+            "3 - do aggressive lossy optimizations (including lossy JBIG2)"
        ),
    )
    optimizing.add_argument(
@@ -388,8 +376,7 @@ Online documentation is located at:
        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."
+            "use cases - see documentation)."
        ),
    )
    optimizing.add_argument(
@@ -419,7 +406,7 @@ Online documentation is located at:
        metavar='MPixels',
        help="Set maximum number of pixels to unpack before treating an image as a "
        "decompression bomb",
-        default=250.0,
+        default=128.0,
    )
    advanced.add_argument(
        '--pdf-renderer',
@@ -1,8 +0,0 @@
-# © 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."""
@@ -1,195 +0,0 @@
-# © 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):
-    exception = auto()
-    result = auto()
-    complete = auto()
-
-
-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):
-    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:
-            conn.send((MessageType.exception, e))
-            break
-        else:
-            conn.send((MessageType.result, result))
-
-    conn.send((MessageType.complete, None))
-    conn.close()
-    return
-
-
-class LambdaExecutor(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 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 r in wait(connections):
-                    if not isinstance(r, Connection):
-                        raise NotImplementedError("We only support Connection()")
-                    try:
-                        msg_type, msg = r.recv()
-                    except EOFError:
-                        connections.remove(r)
-                        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(r)
-                    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
@@ -15,7 +15,7 @@ from collections.abc import Iterable
 from contextlib import suppress
 from functools import wraps
 from io import StringIO
-from math import isclose, isfinite
+from math import isclose
 from pathlib import Path
 from typing import Any, Sequence

@@ -25,35 +25,19 @@ log = logging.getLogger(__name__)


 class Resolution(namedtuple('Resolution', ('x', 'y'))):
-    """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.
-    """
+    """The number of pixels per inch in each 2D direction."""

    __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 self._isclose(self.x, self.y)
-
-    @property
-    def is_finite(self) -> bool:
-        return isfinite(self.x) and isfinite(self.y)
+        return isclose(self.x, self.y, rel_tol=1e-3)

    def take_max(self, vals, yvals=None):
        if yvals is not None:
@@ -70,16 +54,9 @@ class Resolution(namedtuple('Resolution', ('x', 'y'))):
    def __str__(self):
        return f"{self.x:f}x{self.y:f}"

-    def __repr__(self):  # pragma: no cover
+    def __repr__(self):
        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"""
@@ -110,7 +87,10 @@ 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")
-        os.unlink(soft_link_name)
+        try:
+            os.unlink(soft_link_name)
+        except OSError:
+            log.debug("Can't unlink %s", soft_link_name)

    if not os.path.exists(input_file):
        raise FileNotFoundError(f"trying to create a broken symlink to {input_file}")
@@ -189,7 +169,7 @@ def is_file_writable(test_file: os.PathLike) -> bool:
                with suppress(OSError):
                    p.unlink()
            return True
-    except (OSError, RuntimeError) as e:
+    except (EnvironmentError, RuntimeError) as e:
        log.debug(e)
        log.error(str(e))
        return False
@@ -201,36 +181,45 @@ 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:
-        with pdf:
-            messages = pdf.check()
-            for msg in messages:
-                if 'error' in msg.lower():
-                    log.error(msg)
-                else:
-                    log.warning(msg)
-
-            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
+        messages = pdf.check()
+        for msg in messages:
+            if 'error' in msg.lower():
+                log.error(msg)
            else:
-                linearize_msgs = sio.getvalue()
-                if linearize_msgs:
-                    log.warning(linearize_msgs)
+                log.warning(msg)

-            if not messages and not linearize_msgs:
-                return True
-            return 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:
+            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()


 def clamp(n, smallest, largest):  # mypy doesn't understand types for this
@@ -238,13 +227,6 @@ 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):
@@ -254,7 +236,7 @@ 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


@@ -265,7 +247,7 @@ def deprecated(func):
    def new_func(*args, **kwargs):
        warnings.simplefilter('always', DeprecationWarning)  # turn off filter
        warnings.warn(
-            f"Call to deprecated function {func.__name__}.",
+            "Call to deprecated function {}.".format(func.__name__),
            category=DeprecationWarning,
            stacklevel=2,
        )
@@ -31,6 +31,7 @@
 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
@@ -40,62 +41,6 @@ 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


@@ -132,7 +77,7 @@ 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))

@@ -237,7 +182,6 @@ class HocrTransform:

    def to_pdf(
        self,
-        *,
        out_filename: Path,
        image_filename: Optional[Path] = None,
        show_bounding_boxes: bool = False,
@@ -296,11 +240,13 @@ class HocrTransform:
                )

        found_lines = False
-        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'}
+        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,
        ):
            found_lines = True
            self._do_line(
@@ -349,7 +295,7 @@ class HocrTransform:
        interword_spaces: bool,
        show_bounding_boxes: bool,
    ):
-        if line is None:
+        if not line:
            return
        pxl_line_coords = self.element_coordinates(line)
        line_box = self.pt_from_pixel(pxl_line_coords)
@@ -487,10 +433,10 @@ if __name__ == "__main__":
    parser.add_argument('outputfile', help='Path to the PDF file to be generated')
    args = parser.parse_args()

-    hocr = HocrTransform(hocr_filename=args.hocrfile, dpi=args.resolution)
+    hocr = HocrTransform(args.hocrfile, args.resolution)
    hocr.to_pdf(
-        out_filename=args.outputfile,
-        image_filename=args.image,
-        show_bounding_boxes=args.boundingboxes,
+        args.outputfile,
+        args.image,
+        args.boundingboxes,
        interword_spaces=args.interword_spaces,
    )
@@ -1,7 +1,8 @@
-# © 2021 James R. Barlow: github.com/jbarlow83
+# © 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/.

-# Empty __init__.py file
+
+"""Bindings to external libraries"""
@@ -0,0 +1,516 @@
+#!/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()
@@ -9,6 +9,7 @@ import logging
 import sys
 import tempfile
 from collections import defaultdict
+from functools import partial
 from os import fspath
 from pathlib import Path
 from typing import (
@@ -25,23 +26,17 @@ from typing import (
 )

 import img2pdf
-from pikepdf import (
-    Dictionary,
-    Name,
-    Object,
-    ObjectStreamMode,
-    Pdf,
-    PdfImage,
-    Stream,
-    UnsupportedImageTypeError,
-)
+import pikepdf
+from pikepdf import Dictionary, Name, Object, Pdf, PdfImage
 from PIL import Image
+from tqdm import tqdm

-from ocrmypdf._concurrent import Executor, SerialExecutor
+from ocrmypdf import leptonica
+from ocrmypdf._concurrent import exec_progress_pool
 from ocrmypdf._exec import jbig2enc, pngquant
 from ocrmypdf._jobcontext import PdfContext
 from ocrmypdf.exceptions import OutputFileAccessError
-from ocrmypdf.helpers import safe_symlink
+from ocrmypdf.helpers import deprecated, safe_symlink

 log = logging.getLogger(__name__)

@@ -69,12 +64,13 @@ def jpg_name(root: Path, xref: Xref) -> Path:
    return img_name(root, xref, '.jpg')


-def extract_image_filter(
-    pike: Pdf, root: Path, image: Stream, xref: Xref
-) -> Optional[Tuple[PdfImage, Tuple[Name, Object]]]:
-    del pike  # unused args
-    del root
+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
+) -> Optional[Tuple[PdfImage, Tuple[Name, Object]]]:
    if image.Subtype != Name.Image:
        return None
    if image.Length < 100:
@@ -96,13 +92,9 @@ def extract_image_filter(
        return None  # Don't mess with wide gamut images

    if filtdp[0] == Name.JPXDecode:
-        log.debug(f"Skipping JPEG2000 image, xref {xref}")
+        log.debug(f"Skipping JPEG2000 iamge, xref {xref}")
        return None  # Don't do JPEG2000

-    if filtdp[0] == Name.CCITTFaxDecode and filtdp[1].get('/K', 0) >= 0:
-        log.debug(f"Skipping CCITT Group 3 image, xref {xref}")
-        return None  # pikepdf doesn't support Group 3 yet
-
    if Name.Decode in image:
        log.debug(f"Skipping image with Decode table, xref {xref}")
        return None  # Don't mess with custom Decode tables
@@ -111,10 +103,8 @@ def extract_image_filter(


 def extract_image_jbig2(
-    *, pike: Pdf, root: Path, image: Stream, xref: Xref, options
+    *, pike: pikepdf.Pdf, root: Path, image: Object, xref: Xref, options
 ) -> Optional[XrefExt]:
-    del options  # unused arg
-
    result = extract_image_filter(pike, root, image, xref)
    if result is None:
        return None
@@ -130,16 +120,16 @@ 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(Name.ColorSpace, None)
+        colorspace = pim.obj.get(pikepdf.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 = Name.DeviceGray
+                pim.obj.ColorSpace = pikepdf.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 UnsupportedImageTypeError:
+            except pikepdf.UnsupportedImageTypeError:
                return None
            finally:
                # Restore image colorspace after temporarily setting it to DeviceGray
@@ -152,7 +142,7 @@ def extract_image_jbig2(


 def extract_image_generic(
-    *, pike: Pdf, root: Path, image: Stream, xref: Xref, options
+    *, pike: Pdf, root: Path, image: PdfImage, xref: Xref, options
 ) -> Optional[XrefExt]:
    result = extract_image_filter(pike, root, image, xref)
    if result is None:
@@ -185,7 +175,7 @@ def extract_image_generic(
            with imgname.open('wb') as f:
                ext = pim.extract_to(stream=f)
            imgname.rename(imgname.with_suffix(ext))
-        except UnsupportedImageTypeError:
+        except pikepdf.UnsupportedImageTypeError:
            return None
        return XrefExt(xref, ext)
    elif (
@@ -268,8 +258,8 @@ def extract_images(
            result = extract_fn(
                pike=pike, root=root, image=image, xref=xref, options=options
            )
-        except Exception:  # pylint: disable=broad-except
-            log.exception(f"While extracting image xref {xref}, an error occurred")
+        except Exception as e:  # pylint: disable=broad-except
+            log.debug("Image xref %s, error %s", xref, repr(e))
            errors += 1
        else:
            if result:
@@ -307,17 +297,17 @@ def extract_images_jbig2(pike: Pdf, root: Path, options) -> Dict[int, List[XrefE


 def _produce_jbig2_images(
-    jbig2_groups: Dict[int, List[XrefExt]], root: Path, options, executor: Executor
+    jbig2_groups: Dict[int, List[XrefExt]], root: Path, options
 ) -> 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 (
-                fspath(root),  # =cwd
-                (img_name(root, xref, ext) for xref, ext in xref_exts),  # =infiles
-                prefix,  # =out_prefix
+            yield dict(
+                cwd=fspath(root),
+                infiles=(img_name(root, xref, ext) for xref, ext in xref_exts),
+                out_prefix=prefix,
            )

    def jbig2_single_args(root, groups: Dict[int, List[XrefExt]]):
@@ -326,20 +316,23 @@ 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 (
-                    fspath(root),
-                    img_name(root, xref, ext),
-                    root / f'{prefix}.{n:04d}',
+                yield dict(
+                    cwd=fspath(root),
+                    infile=img_name(root, xref, ext),
+                    outfile=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 = jbig2enc.convert_group_mp
+        jbig2_convert = partial(convert_generic, jbig2enc.convert_group)
    else:
        jbig2_args = jbig2_single_args
-        jbig2_convert = jbig2enc.convert_single_mp
+        jbig2_convert = partial(convert_generic, jbig2enc.convert_single)

-    executor(
+    exec_progress_pool(
        use_threads=True,
        max_workers=options.jobs,
        tqdm_kwargs=dict(
@@ -354,11 +347,7 @@ def _produce_jbig2_images(


 def convert_to_jbig2(
-    pike: Pdf,
-    jbig2_groups: Dict[int, List[XrefExt]],
-    root: Path,
-    options,
-    executor: Executor,
+    pike: Pdf, jbig2_groups: Dict[int, List[XrefExt]], root: Path, options
 ) -> None:
    """Convert images to JBIG2 and insert into PDF.

@@ -372,16 +361,15 @@ 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, executor)
+    _produce_jbig2_images(jbig2_groups, root, options)

    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 = Stream(pike, jbig2_globals_data)
+            jbig2_globals = pikepdf.Stream(pike, jbig2_globals_data)
            jbig2_globals_dict = Dictionary(JBIG2Globals=jbig2_globals)
        elif options.jbig2_page_group_size == 1:
            jbig2_globals_dict = None
@@ -398,53 +386,27 @@ def convert_to_jbig2(
            )


-def _optimize_jpeg(args: Tuple[Xref, Path, Path, int]) -> Tuple[Xref, Optional[Path]]:
-    xref, in_jpg, opt_jpg, jpeg_quality = args
+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')

-    # This may produce 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=jpeg_quality)
+        # 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)

-    if opt_jpg.stat().st_size > in_jpg.stat().st_size:
-        log.debug("xref %s, jpeg, made larger - skip", xref)
-        opt_jpg.unlink()
-        return xref, None
-    return xref, opt_jpg
+        if opt_jpg.stat().st_size > in_jpg.stat().st_size:
+            log.debug("xref %s, jpeg, made larger - skip", xref)
+            continue

-
-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="JPEGs",
-            total=len(jpegs),
-            unit='image',
-            disable=not options.progress_bar,
-        ),
-        task=_optimize_jpeg,
-        task_arguments=jpeg_args(),
-        task_finished=finish_jpeg,
-    )
+        compdata = leptonica.CompressedData.open(opt_jpg)
+        im_obj = pike.get_object(xref, 0)
+        im_obj.write(compdata.read(), filter=Name.DCTDecode)


 def _transcode_png(pike: Pdf, filename: Path, xref: Xref) -> bool:
@@ -452,8 +414,8 @@ def _transcode_png(pike: Pdf, filename: Path, xref: Xref) -> bool:
    with output.open('wb') as f:
        img2pdf.convert(fspath(filename), outputstream=f)

-    with Pdf.open(output) as pdf_image:
-        foreign_image = next(iter(pdf_image.pages[0].images.values()))
+    with pikepdf.open(output) as pdf_image:
+        foreign_image = next(pdf_image.pages[0].images.values())
        local_image = pike.copy_foreign(foreign_image)

        im_obj = pike.get_object(xref, 0)
@@ -494,7 +456,6 @@ def transcode_pngs(
    image_name_fn: Callable[[Path, Xref], Path],
    root: Path,
    options,
-    executor,
 ) -> None:
    modified: MutableSet[Xref] = set()
    if options.optimize >= 2:
@@ -514,7 +475,10 @@ def transcode_pngs(
                )
                modified.add(xref)

-        executor(
+        def pngquant_fn(args):
+            pngquant.quantize(*args)
+
+        exec_progress_pool(
            use_threads=True,
            max_workers=options.jobs,
            tqdm_kwargs=dict(
@@ -523,7 +487,7 @@ def transcode_pngs(
                unit='image',
                disable=not options.progress_bar,
            ),
-            task=pngquant.quantize_mp,
+            task=pngquant_fn,
            task_arguments=pngquant_args(),
        )

@@ -532,16 +496,82 @@ def transcode_pngs(
        _transcode_png(pike, filename, xref)


-DEFAULT_EXECUTOR = SerialExecutor()
+@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


-def optimize(
-    input_file: Path,
-    output_file: Path,
-    context,
-    save_settings,
-    executor: Executor = DEFAULT_EXECUTOR,
-) -> None:
+@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:
    options = context.options
    if options.optimize == 0:
        safe_symlink(input_file, output_file)
@@ -554,19 +584,19 @@ def optimize(
    if options.jbig2_page_group_size == 0:
        options.jbig2_page_group_size = 10 if options.jbig2_lossy else 1

-    with Pdf.open(input_file) as pike:
+    with pikepdf.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, executor)
+        transcode_jpegs(pike, jpegs, root, options)
        # if options.optimize >= 2:
        # Try pngifying the jpegs
        #    transcode_pngs(pike, jpegs, jpg_name, root, options)
-        transcode_pngs(pike, pngs, png_name, root, options, executor)
+        transcode_pngs(pike, pngs, png_name, root, options)

        jbig2_groups = extract_images_jbig2(pike, root, options)
-        convert_to_jbig2(pike, jbig2_groups, root, options, executor)
+        convert_to_jbig2(pike, jbig2_groups, root, options)

        target_file = output_file.with_suffix('.opt.pdf')
        pike.remove_unreferenced_resources()
@@ -584,12 +614,9 @@ def optimize(
    log.info(f"Optimize ratio: {ratio:.2f} savings: {(savings):.1%}")

    if savings < 0:
-        log.info(
-            "Image optimization did not improve the file - "
-            "optimizations will not be used"
-        )
+        log.info("Image optimization did not improve the file - discarded")
        # We still need to save the file
-        with Pdf.open(input_file) as pike:
+        with pikepdf.open(input_file) as pike:
            pike.remove_unreferenced_resources()
            pike.save(output_file, **save_settings)
    else:
@@ -636,7 +663,7 @@ def main(infile, outfile, level, jobs=1):
            dict(
                compress_streams=True,
                preserve_pdfa=True,
-                object_stream_mode=ObjectStreamMode.generate,
+                object_stream_mode=pikepdf.ObjectStreamMode.generate,
            ),
        )
        copy(fspath(tmpout), fspath(outfile))
@@ -13,14 +13,12 @@ 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

-SRGB_ICC_PROFILE_NAME = 'sRGB.icc'
+ICC_PROFILE_RELPATH = 'data/sRGB.icc'
+
+SRGB_ICC_PROFILE = pkg_resources.resource_filename('ocrmypdf', ICC_PROFILE_RELPATH)


 def _postscript_objdef(
@@ -99,12 +97,12 @@ def generate_pdfa_ps(target_filename: Path, icc: str = 'sRGB'):
    References:
        Adobe PDFMARK Reference: https://www.adobe.com/content/dam/acom/en/devnet/acrobat/pdfs/pdfmark_reference.pdf
    """
-    if icc != 'sRGB':
+    if icc == 'sRGB':
+        icc_profile = SRGB_ICC_PROFILE
+    else:
        raise NotImplementedError("Only supporting sRGB")

-    bytes_icc_profile = (
-        package_files('ocrmypdf.data') / SRGB_ICC_PROFILE_NAME
-    ).read_bytes()
+    bytes_icc_profile = Path(icc_profile).read_bytes()
    ps = '\n'.join(_make_postscript(icc, bytes_icc_profile, 3))

    # We should have encoded everything to pure ASCII by this point, and
@@ -6,40 +6,22 @@
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.


-import atexit
 import logging
 import re
-from collections import defaultdict
-from contextlib import ExitStack
+from collections import defaultdict, namedtuple
 from decimal import Decimal
 from enum import Enum
 from functools import partial
-from math import hypot, inf, isclose
+from math import hypot, isclose
 from os import PathLike
 from pathlib import Path
-from typing import (
-    Container,
-    Dict,
-    Iterator,
-    List,
-    Mapping,
-    NamedTuple,
-    Optional,
-    Tuple,
-    Union,
-)
+from typing import Container, Iterator, Optional, Tuple, Union
 from warnings import warn

-from pikepdf import (
-    Object,
-    Pdf,
-    PdfImage,
-    PdfInlineImage,
-    PdfMatrix,
-    parse_content_stream,
-)
+import pikepdf
+from pikepdf import Object, Pdf, PdfMatrix

-from ocrmypdf._concurrent import Executor, SerialExecutor
+from ocrmypdf._concurrent import exec_progress_pool
 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
@@ -52,7 +34,7 @@ Encoding = Enum(
    'Encoding', 'ccitt jpeg jpeg2000 jbig2 asciihex ascii85 lzw flate runlength'
 )

-FRIENDLY_COLORSPACE: Dict[str, Colorspace] = {
+FRIENDLY_COLORSPACE = {
    '/DeviceGray': Colorspace.gray,
    '/CalGray': Colorspace.gray,
    '/DeviceRGB': Colorspace.rgb,
@@ -70,7 +52,7 @@ FRIENDLY_COLORSPACE: Dict[str, Colorspace] = {
    '/I': Colorspace.index,
 }

-FRIENDLY_ENCODING: Dict[str, Encoding] = {
+FRIENDLY_ENCODING = {
    '/CCITTFaxDecode': Encoding.ccitt,
    '/DCTDecode': Encoding.jpeg,
    '/JPXDecode': Encoding.jpeg2000,
@@ -84,7 +66,7 @@ FRIENDLY_ENCODING: Dict[str, Encoding] = {
    '/RL': Encoding.runlength,
 }

-FRIENDLY_COMP: Dict[Colorspace, int] = {
+FRIENDLY_COMP = {
    Colorspace.gray: 1,
    Colorspace.rgb: 3,
    Colorspace.cmyk: 4,
@@ -102,30 +84,16 @@ def _is_unit_square(shorthand):
    return all(isclose(a, b, rel_tol=1e-3) for a, b in pairwise)


-class XobjectSettings(NamedTuple):
-    name: str
-    shorthand: Tuple[float, float, float, float, float, float]
-    stack_depth: int
+XobjectSettings = namedtuple('XobjectSettings', ['name', 'shorthand', 'stack_depth'])

+InlineSettings = namedtuple('InlineSettings', ['iimage', 'shorthand', 'stack_depth'])

-class InlineSettings(NamedTuple):
-    iimage: PdfInlineImage
-    shorthand: Tuple[float, float, float, float, float, float]
-    stack_depth: int
+ContentsInfo = namedtuple(
+    'ContentsInfo',
+    ['xobject_settings', 'inline_images', 'found_vector', 'found_text', 'name_index'],
+)

-
-class ContentsInfo(NamedTuple):
-    xobject_settings: List[XobjectSettings]
-    inline_images: List[InlineSettings]
-    found_vector: bool
-    found_text: bool
-    name_index: Mapping[str, List[XobjectSettings]]
-
-
-class TextboxInfo(NamedTuple):
-    bbox: Tuple[float, float, float, float]
-    is_visible: bool
-    is_corrupt: bool
+TextboxInfo = namedtuple('TextboxInfo', ['bbox', 'is_visible', 'is_corrupt'])


 class VectorMarker:
@@ -176,8 +144,8 @@ def _interpret_contents(contentstream: Object, initial_shorthand=UNIT_SQUARE):

    stack = []
    ctm = PdfMatrix(initial_shorthand)
-    xobject_settings: List[XobjectSettings] = []
-    inline_images: List[InlineSettings] = []
+    xobject_settings = []
+    inline_images = []
    name_index = defaultdict(lambda: [])
    found_vector = False
    found_text = False
@@ -187,7 +155,9 @@ 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(parse_content_stream(contentstream, operator_whitelist))
+        _normalize_stack(
+            pikepdf.parse_content_stream(contentstream, operator_whitelist)
+        )
    ):
        operands, operator = graphobj
        if operator == 'q':
@@ -213,7 +183,7 @@ 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[str(image_name)].append(settings)
+            name_index[image_name].append(settings)
        elif operator == 'INLINE IMAGE':  # BI/ID/EI are grouped into this
            iimage = operands[0]
            inline = InlineSettings(
@@ -284,43 +254,40 @@ def _get_dpi(ctm_shorthand, image_size) -> Resolution:
    a, b, c, d, _, _ = ctm_shorthand

    # Calculate the width and height of the image in PDF units
-    image_drawn = hypot(a, b), hypot(c, d)
+    image_drawn_width = hypot(a, b)
+    image_drawn_height = hypot(c, d)

-    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
+    # 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

-    dpi_w, dpi_h = (calc(image_drawn[n], image_size[n]) for n in range(2))
    return Resolution(dpi_w, dpi_h)


 class ImageInfo:
    DPI_PREC = Decimal('1.000')

-    _comp: Optional[int]
-    _name: str
-
    def __init__(
        self,
        *,
        name='',
        pdfimage: Optional[Object] = None,
-        inline: Optional[PdfInlineImage] = None,
+        inline: Optional[Object] = None,
        shorthand=None,
    ):
        self._name = str(name)
        self._shorthand = shorthand

-        pim: Union[PdfInlineImage, PdfImage]
-
        if inline is not None:
            self._origin = 'inline'
-            pim = inline
+            pim = inline.iimage
        elif pdfimage is not None:
            self._origin = 'xobject'
-            pim = PdfImage(pdfimage)
+            pim = pikepdf.PdfImage(pdfimage)
        else:
            raise ValueError("Either pdfimage or inline must be set")
        self._width = pim.width
@@ -336,14 +303,14 @@ class ImageInfo:

        self._bpc = int(pim.bits_per_component)
        try:
-            self._enc = FRIENDLY_ENCODING.get(pim.filters[0])
+            self._enc = FRIENDLY_ENCODING.get(pim.filters[0], 'image')
        except IndexError:
-            self._enc = None
+            self._enc = '?'

        try:
-            self._color = FRIENDLY_COLORSPACE.get(pim.colorspace or '')
+            self._color = FRIENDLY_COLORSPACE.get(pim.colorspace, '?')
        except NotImplementedError:
-            self._color = None
+            self._color = '?'
        if self._enc == Encoding.jpeg2000:
            self._color = Colorspace.jpeg2000

@@ -357,14 +324,11 @@ class ImageInfo:
            else:
                self._comp = 3
        else:
-            if isinstance(self._color, Colorspace):
-                self._comp = FRIENDLY_COMP.get(self._color)
-            else:
-                self._comp = None
+            self._comp = FRIENDLY_COMP.get(self._color, '?')

            # Bit of a hack... infer grayscale if component count is uncertain
            # but encoding only supports monochrome.
-            if self._comp is None and self._enc in (Encoding.ccitt, Encoding.jbig2):
+            if self._comp == '?' and self._enc in (Encoding.ccitt, Encoding.jbig2):
                self._comp = FRIENDLY_COMP[Colorspace.gray]

    @property
@@ -389,19 +353,15 @@ class ImageInfo:

    @property
    def color(self):
-        return self._color if self._color is not None else '?'
+        return self._color

    @property
    def comp(self):
-        return self._comp if self._comp is not None else '?'
+        return self._comp

    @property
    def enc(self):
-        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
+        return self._enc

    @property
    def dpi(self):
@@ -424,7 +384,7 @@ 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.iimage
+            name='inline-%02d' % n, shorthand=inline.shorthand, inline=inline
        )


@@ -449,7 +409,7 @@ def _image_xobjects(container) -> Iterator[Tuple[Object, str]]:
    xobjs = resources['/XObject'].as_dict()
    for xobj in xobjs:
        candidate: Object = xobjs[xobj]
-        if '/Subtype' not in candidate:
+        if not '/Subtype' in candidate:
            continue
        if candidate['/Subtype'] == '/Image':
            pdfimage = candidate
@@ -611,42 +571,29 @@ def simplify_textboxes(miner, textbox_getter) -> Iterator[TextboxInfo]:
 worker_pdf = None


-def _pdf_pageinfo_sync_init(pdf: Pdf, infile: Path, pdfminer_loglevel):
+def _pdf_pageinfo_sync_init(infile: Path, pdfminer_loglevel):
    global worker_pdf  # pylint: disable=global-statement
    pikepdf_enable_mmap()

    logging.getLogger('pdfminer').setLevel(pdfminer_loglevel)

-    # 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)
+    # 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)


 def _pdf_pageinfo_sync(args):
-    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
+    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


 def _pdf_pageinfo_concurrent(
-    pdf,
-    executor: Executor,
-    infile,
-    progbar,
-    max_workers,
-    check_pages,
-    detailed_analysis=False,
+    pdf, infile, progbar, max_workers, check_pages, detailed_analysis=False
 ):
+    global worker_pdf  # pylint: disable=global-statement
    pages = [None] * len(pdf.pages)

    def update_pageinfo(result, pbar):
@@ -660,6 +607,7 @@ 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)
@@ -668,39 +616,29 @@ def _pdf_pageinfo_concurrent(
        # a separate process.
        use_threads = True

-    # 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,
-    )
+    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()
    return pages


 class PageInfo:
-    _has_text: Optional[bool]
-    _has_vector: Optional[bool]
-    _images: List[ImageInfo]
-
    def __init__(
        self,
        pdf: Pdf,
@@ -772,13 +710,11 @@ class PageInfo:
        else:
            self._has_vector = None  # i.e. "no information"
            self._has_text = None
-            self._images = []
+            self._images = None

        self._dpi = None
        if self._images:
-            dpi = Resolution(0.0, 0.0).take_max(
-                image.dpi for image in self._images if image.renderable
-            )
+            dpi = Resolution(0.0, 0.0).take_max(image.dpi for image in self._images)
            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)))
@@ -789,7 +725,7 @@ class PageInfo:

    @property
    def has_text(self) -> bool:
-        return bool(self._has_text)
+        return self._has_text

    @property
    def has_corrupt_text(self) -> bool:
@@ -799,7 +735,7 @@ class PageInfo:

    @property
    def has_vector(self) -> bool:
-        return bool(self._has_vector)
+        return self._has_vector

    @property
    def width_inches(self) -> Decimal:
@@ -877,32 +813,26 @@ 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 Pdf.open(infile) as pdf:
+        with pikepdf.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,
@@ -135,7 +135,7 @@ class LTStateAwareChar(LTChar):
        return self._text

    def __repr__(self):
-        return '<{} {} matrix={} rendermode={!r} font={!r} adv={} text={!r}>'.format(
+        return '<%s %s matrix=%s rendermode=%r font=%r adv=%s text=%r>' % (
            self.__class__.__name__,
            bbox2str(self.bbox),
            matrix2str(self.matrix),
@@ -5,43 +5,27 @@
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.


-from abc import ABC, abstractmethod
+from abc import ABC, abstractmethod, abstractstaticmethod
 from argparse import ArgumentParser, Namespace
-from logging import Handler
+from collections import namedtuple
 from pathlib import Path
-from typing import TYPE_CHECKING, AbstractSet, List, NamedTuple, Optional
+from typing import TYPE_CHECKING, AbstractSet, List, Optional

 import pluggy

-from ocrmypdf._concurrent import Executor
 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 add_options(parser: ArgumentParser) -> None:
    """Allows the plugin to add its own command line and API arguments.
@@ -78,67 +62,6 @@ 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*.
@@ -196,7 +119,7 @@ def rasterize_pdf_page(


@hookspec(firstresult=True)
-def filter_ocr_image(page: 'PageContext', image: 'Image.Image') -> 'Image.Image':
+def filter_ocr_image(page: 'PageContext', 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
@@ -240,11 +163,11 @@ def filter_page_image(page: 'PageContext', image_filename: Path) -> Path:
    will be resized and the OCR layer misaligned. OCRmyPDF does not 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 (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.
+    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.

    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
@@ -262,62 +185,15 @@ def filter_page_image(page: 'PageContext', image_filename: Path) -> Path:
    """


-@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.
+OrientationConfidence = namedtuple('OrientationConfidence', ('angle', 'confidence'))
+"""Expresses an OCR engine's confidence in page rotation.

-    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
+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.
+"""


 class OcrEngine(ABC):
@@ -327,13 +203,11 @@ class OcrEngine(ABC):
    Tesseract OCR.
    """

-    @staticmethod
-    @abstractmethod
+    @abstractstaticmethod
    def version() -> str:
        """Returns the version of the OCR engine."""

-    @staticmethod
-    @abstractmethod
+    @abstractstaticmethod
    def creator_tag(options: Namespace) -> str:
        """Returns the creator tag to identify this software's role in creating the PDF.

@@ -353,33 +227,24 @@ class OcrEngine(ABC):
        to the user, usually in an error message.
        """

-    @staticmethod
-    @abstractmethod
+    @abstractstaticmethod
    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."""

-    @staticmethod
-    @abstractmethod
+    @abstractstaticmethod
    def get_orientation(input_file: Path, options: Namespace) -> OrientationConfidence:
        """Returns the orientation of the image."""

-    @staticmethod
-    def get_deskew(input_file: Path, options: Namespace) -> float:
-        """Returns the deskew angle of the image, in degrees."""
-        return 0.0
-
-    @staticmethod
-    @abstractmethod
+    @abstractstaticmethod
    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."""

-    @staticmethod
-    @abstractmethod
+    @abstractstaticmethod
    def generate_pdf(
        input_file: Path, output_pdf: Path, output_text: Path, options: Namespace
    ) -> None:
@@ -13,22 +13,18 @@ import re
 import sys
 from collections.abc import Mapping
 from contextlib import suppress
-from distutils.version import LooseVersion, Version
+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 ocrmypdf.exceptions import MissingDependencyError

-# pylint: disable=logging-format-interpolation
-
 log = logging.getLogger(__name__)


-def run(
-    args, *, env=None, logs_errors_to_stdout: bool = False, **kwargs
-) -> CompletedProcess:
+def run(args, *, env=None, logs_errors_to_stdout=False, **kwargs):
    """Wrapper around :py:func:`subprocess.run`

    The main purpose of this wrapper is to log subprocess output in an orderly
@@ -66,9 +62,7 @@ def run(
    return proc


-def run_polling_stderr(
-    args, *, callback: Callable[[str], None], check: bool = False, env=None, **kwargs
-) -> CompletedProcess:
+def run_polling_stderr(args, *, callback, check=False, env=None, **kwargs):
    """Run a process like ``ocrmypdf.subprocess.run``, and poll stderr.

    Every line of produced by stderr will be forwarded to the callback function.
@@ -86,8 +80,6 @@ def run_polling_stderr(
    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())
@@ -107,7 +99,7 @@ def _fix_process_args(args, env, kwargs):
        env = os.environ

    # Search in spoof path if necessary
-    program = str(args[0])
+    program = args[0]

    if os.name == 'nt':
        from ocrmypdf.subprocess._windows import fix_windows_args
@@ -117,14 +109,22 @@ 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 +145,7 @@ def get_version(
            check=True,
            env=env,
        )
-        output: str = proc.stdout
+        output = proc.stdout
    except FileNotFoundError as e:
        raise MissingDependencyError(
            f"Could not find program '{program}' on the PATH"
@@ -264,30 +264,13 @@ def _error_old_version(program, package, need_version, found_version, required_f

 def check_external_program(
    *,
-    program: str,
-    package: str,
-    version_checker: Union[str, Callable],
-    need_version: str,
-    required_for: Optional[str] = None,
+    program,
+    package,
+    version_checker,
+    need_version,
+    required_for=None,
    recommended=False,
-    version_parser: Type[Version] = LooseVersion,
 ):
-    """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()
@@ -296,7 +279,7 @@ def check_external_program(
    except (CalledProcessError, FileNotFoundError, MissingDependencyError):
        _error_missing_program(program, package, required_for, recommended)
        if not recommended:
-            raise MissingDependencyError(program)
+            raise MissingDependencyError()
        return

    def remove_leading_v(s):
@@ -307,9 +290,9 @@ def check_external_program(
    found_version = remove_leading_v(found_version)
    need_version = remove_leading_v(need_version)

-    if found_version and version_parser(found_version) < version_parser(need_version):
+    if found_version and LooseVersion(found_version) < LooseVersion(need_version):
        _error_old_version(program, package, need_version, found_version, required_for)
        if not recommended:
-            raise MissingDependencyError(program)
+            raise MissingDependencyError()

    log.debug('Found %s %s', program, found_version)
@@ -9,9 +9,9 @@ import os
 import shutil
 import sys
 from distutils.version import LooseVersion
-from itertools import chain
+from itertools import chain, filterfalse
 from pathlib import Path
-from typing import Any, Callable, Iterable, Iterator, Set, Tuple, TypeVar
+from typing import Any, Callable, Iterator, Optional, Tuple, TypeVar, cast

 try:
    import winreg
@@ -51,7 +51,7 @@ def registry_path_ghostscript(env=None) -> Iterator[Path]:
        with winreg.OpenKey(
            winreg.HKEY_LOCAL_MACHINE, r"SOFTWARE\Artifex\GPL Ghostscript"
        ) as k:
-            latest_gs = max(registry_subkeys(k), key=LooseVersion, default='0')
+            latest_gs = max(registry_subkeys(k), key=LooseVersion)
        with winreg.OpenKey(
            winreg.HKEY_LOCAL_MACHINE, fr"SOFTWARE\Artifex\GPL Ghostscript\{latest_gs}"
        ) as k:
@@ -113,7 +113,7 @@ SHIMS = [
 ]


-def fix_windows_args(program: str, args, env):
+def fix_windows_args(program, args, env):
    """Adjust our desired program and command line arguments for use on Windows"""

    if sys.version_info < (3, 8):
@@ -137,12 +137,14 @@ def fix_windows_args(program: str, args, env):
    return args


-def unique_everseen(iterable: Iterable[T], key: Callable[[T], T]) -> Iterator[T]:
-    "List unique elements, preserving order."
+def unique_everseen(iterable, key=None):
+    "List unique elements, preserving order. Remember all elements ever seen."
    # unique_everseen('AAAABBBCCDAABBB') --> A B C D
    # unique_everseen('ABBCcAD', str.lower) --> A B C D
-    seen: Set[T] = set()
+    seen = 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:
@@ -1 +1 @@
-Tesseract Open Source OCR Engine v5.0.0-beta-20210916-12-g19cc9 with Leptonica
+Tesseract Open Source OCR Engine v4.1.1 with Leptonica
@@ -1,6 +1,4 @@
-i a la Waterman
-
- 
+a la Waterman

 4 ons linzen

@@ -14,14 +12,14 @@ 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 Kruidnagel.
-Alle uien, kerrie en zgout 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 Kruicdnagel.
+Alle uien, kerrie en gout 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
-gfgaar Zijn is de soep klaar.
+gaar Zijn is de soep klaar.

@@ -1 +1 @@
-Tesseract Open Source OCR Engine v5.0.0-beta-20210916-12-g19cc9 with Leptonica
+Tesseract Open Source OCR Engine v4.1.1 with Leptonica
@@ -22,9 +22,9 @@ Mugerre

 Milafranga Komunikabideak

-BAIONA  zeiteninsiie —
+BAIONA i zeettnansise —

-7 Trenbideak -----
+1 Trenbideak -- ~~~

-t\ Basusarri — spmeans:20141004 ae: . _ ~
+t\ Basusarri — spmsans20141004 se: . a ~

@@ -1 +1 @@
-Tesseract Open Source OCR Engine v5.0.0-beta-20210916-12-g19cc9 with Leptonica
+Tesseract Open Source OCR Engine v4.1.1 with Leptonica
@@ -1 +1 @@
-Tesseract Open Source OCR Engine v5.0.0-beta-20210916-12-g19cc9 with Leptonica
+Tesseract Open Source OCR Engine v4.1.1 with Leptonica
@@ -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 Kruidnagel.
-Alle uien, kerrie en zgout 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 Kruicdnagel.
+Alle uien, kerrie en gout 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
-gfgaar Zijn is de soep klaar.
+gaar Zijn is de soep klaar.

@@ -5,11 +5,11 @@
 <head>
  <title></title>
  <meta http-equiv="Content-Type" content="text/html;charset=utf-8"/>
-  <meta name='ocr-system' content='tesseract 5.0.0-beta-20210916-12-g19cc9' />
+  <meta name='ocr-system' content='tesseract 4.1.1' />
  <meta name='ocr-capabilities' content='ocr_page ocr_carea ocr_par ocr_line ocrx_word ocrp_wconf'/>
 </head>
 <body>
-  <div class='ocr_page' id='page_1' title='image "/tmp/ocrmypdf.io.gzrr1v_b/000001_ocr.png"; bbox 0 0 1000 800; ppageno 0'>
+  <div class='ocr_page' id='page_1' title='image "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/com.github.ocrmypdf.mrqsewbu/000001_ocr.png"; bbox 0 0 1000 800; ppageno 0'>
   <div class='ocr_carea' id='block_1_1' title="bbox 296 96 704 504">
    <p class='ocr_par' id='par_1_1' lang='eng' title="bbox 296 96 704 504">
     <span class='ocr_line' id='line_1_1' title="bbox 296 96 704 504; baseline 0 296; x_size 169.33333; x_descenders 42.333332; x_ascenders 42.333336">
@@ -20,12 +20,12 @@
   <div class='ocr_carea' id='block_1_2' title="bbox 150 592 841 622">
    <p class='ocr_par' id='par_1_2' lang='eng' title="bbox 150 592 841 622">
     <span class='ocr_line' id='line_1_2' title="bbox 150 592 841 622; baseline 0 -6; x_size 30; x_descenders 6; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_2' title='bbox 150 592 230 616; x_wconf 95'>This</span>
+      <span class='ocrx_word' id='word_1_2' title='bbox 150 592 230 616; x_wconf 96'>This</span>
      <span class='ocrx_word' id='word_1_3' title='bbox 260 592 384 616; x_wconf 95'>should</span>
      <span class='ocrx_word' id='word_1_4' title='bbox 413 592 449 616; x_wconf 95'>be</span>
-      <span class='ocrx_word' id='word_1_5' title='bbox 479 600 493 616; x_wconf 94'>a</span>
-      <span class='ocrx_word' id='word_1_6' title='bbox 523 592 668 622; x_wconf 94'>perfect</span>
-      <span class='ocrx_word' id='word_1_7' title='bbox 698 592 841 616; x_wconf 61'>circle:</span>
+      <span class='ocrx_word' id='word_1_5' title='bbox 479 600 493 616; x_wconf 95'>a</span>
+      <span class='ocrx_word' id='word_1_6' title='bbox 523 592 668 622; x_wconf 95'>perfect</span>
+      <span class='ocrx_word' id='word_1_7' title='bbox 698 592 841 616; x_wconf 55'>circle:</span>
     </span>
    </p>
   </div>
@@ -1 +1 @@
-Tesseract Open Source OCR Engine v5.0.0-beta-20210916-12-g19cc9 with Leptonica
+Tesseract Open Source OCR Engine v4.1.1 with Leptonica
@@ -1 +1 @@
-Tesseract Open Source OCR Engine v5.0.0-beta-20210916-12-g19cc9 with Leptonica
+Tesseract Open Source OCR Engine v4.1.1 with Leptonica
@@ -5,11 +5,11 @@
 <head>
  <title></title>
  <meta http-equiv="Content-Type" content="text/html;charset=utf-8"/>
-  <meta name='ocr-system' content='tesseract 5.0.0-beta-20210916-12-g19cc9' />
+  <meta name='ocr-system' content='tesseract 4.1.1' />
  <meta name='ocr-capabilities' content='ocr_page ocr_carea ocr_par ocr_line ocrx_word ocrp_wconf'/>
 </head>
 <body>
-  <div class='ocr_page' id='page_1' title='image "/tmp/ocrmypdf.io.8je4vgpg/000001_ocr.png"; bbox 0 0 2550 3300; ppageno 0'>
+  <div class='ocr_page' id='page_1' title='image "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/com.github.ocrmypdf.ces85e5u/000001_ocr.png"; bbox 0 0 2550 3300; ppageno 0'>
   <div class='ocr_carea' id='block_1_1' title="bbox 582 131 1968 303">
    <p class='ocr_par' id='par_1_1' lang='eng' title="bbox 582 131 1968 303">
     <span class='ocr_header' id='line_1_1' title="bbox 882 131 1657 217; baseline 0.001 -17; x_size 85; x_descenders 16; x_ascenders 19">
@@ -18,8 +18,8 @@
     </span>
     <span class='ocr_header' id='line_1_2' title="bbox 582 215 1968 303; baseline 0 -17; x_size 87; x_descenders 16; x_ascenders 21">
      <span class='ocrx_word' id='word_1_3' title='bbox 582 215 674 286; x_wconf 96'>32</span>
-      <span class='ocrx_word' id='word_1_4' title='bbox 697 218 923 288; x_wconf 96'>Track</span>
-      <span class='ocrx_word' id='word_1_5' title='bbox 948 218 1181 287; x_wconf 95'>MIDI</span>
+      <span class='ocrx_word' id='word_1_4' title='bbox 697 218 923 288; x_wconf 95'>Track</span>
+      <span class='ocrx_word' id='word_1_5' title='bbox 948 218 1181 287; x_wconf 96'>MIDI</span>
      <span class='ocrx_word' id='word_1_6' title='bbox 1208 217 1575 303; x_wconf 96'>Sequence</span>
      <span class='ocrx_word' id='word_1_7' title='bbox 1600 218 1968 288; x_wconf 96'>Recorder</span>
     </span>
@@ -29,7 +29,7 @@
    <p class='ocr_par' id='par_1_2' lang='eng' title="bbox 347 380 2188 423">
     <span class='ocr_header' id='line_1_3' title="bbox 347 380 2188 423; baseline -0.001 -12; x_size 38; x_descenders 8; x_ascenders 10">
      <span class='ocrx_word' id='word_1_8' title='bbox 347 380 412 410; x_wconf 93'>The</span>
-      <span class='ocrx_word' id='word_1_9' title='bbox 424 380 676 417; x_wconf 90'>LinnSequencer</span>
+      <span class='ocrx_word' id='word_1_9' title='bbox 424 380 676 417; x_wconf 92'>LinnSequencer</span>
      <span class='ocrx_word' id='word_1_10' title='bbox 688 380 712 411; x_wconf 96'>is</span>
      <span class='ocrx_word' id='word_1_11' title='bbox 724 390 743 411; x_wconf 96'>a</span>
      <span class='ocrx_word' id='word_1_12' title='bbox 754 381 1005 423; x_wconf 96'>state-of-the-art</span>
@@ -37,7 +37,7 @@
      <span class='ocrx_word' id='word_1_14' title='bbox 1238 381 1299 411; x_wconf 96'>and</span>
      <span class='ocrx_word' id='word_1_15' title='bbox 1311 380 1525 418; x_wconf 96'>performance</span>
      <span class='ocrx_word' id='word_1_16' title='bbox 1536 380 1602 411; x_wconf 96'>tool</span>
-      <span class='ocrx_word' id='word_1_17' title='bbox 1615 380 1663 411; x_wconf 96'>for</span>
+      <span class='ocrx_word' id='word_1_17' title='bbox 1615 380 1663 411; x_wconf 97'>for</span>
      <span class='ocrx_word' id='word_1_18' title='bbox 1674 381 1725 410; x_wconf 96'>the</span>
      <span class='ocrx_word' id='word_1_19' title='bbox 1737 380 1940 417; x_wconf 95'>professional</span>
      <span class='ocrx_word' id='word_1_20' title='bbox 1952 380 2112 411; x_wconf 96'>musician.</span>
@@ -55,7 +55,7 @@
      <span class='ocrx_word' id='word_1_27' title='bbox 936 430 1044 468; x_wconf 96'>simple</span>
      <span class='ocrx_word' id='word_1_28' title='bbox 1055 435 1087 461; x_wconf 96'>to</span>
      <span class='ocrx_word' id='word_1_29' title='bbox 1099 431 1183 461; x_wconf 96'>learn</span>
-      <span class='ocrx_word' id='word_1_30' title='bbox 1195 431 1257 461; x_wconf 95'>and</span>
+      <span class='ocrx_word' id='word_1_30' title='bbox 1195 431 1257 461; x_wconf 96'>and</span>
      <span class='ocrx_word' id='word_1_31' title='bbox 1269 440 1329 461; x_wconf 95'>use.</span>
      <span class='ocrx_word' id='word_1_32' title='bbox 1344 431 1393 461; x_wconf 96'>It’s</span>
      <span class='ocrx_word' id='word_1_33' title='bbox 1406 440 1499 467; x_wconf 96'>many</span>
@@ -67,25 +67,25 @@

    <p class='ocr_par' id='par_1_4' lang='eng' title="bbox 350 482 2093 574">
     <span class='ocr_header' id='line_1_5' title="bbox 350 482 2093 527; baseline 0 -9; x_size 43; x_descenders 7; x_ascenders 11">
-      <span class='ocrx_word' id='word_1_37' title='bbox 350 490 368 508; x_wconf 72'>¢</span>
+      <span class='ocrx_word' id='word_1_37' title='bbox 350 490 368 508; x_wconf 73'>¢</span>
      <span class='ocrx_word' id='word_1_38' title='bbox 383 482 585 526; x_wconf 95'>Operation</span>
      <span class='ocrx_word' id='word_1_39' title='bbox 598 482 627 518; x_wconf 96'>is</span>
-      <span class='ocrx_word' id='word_1_40' title='bbox 641 482 776 518; x_wconf 95'>similar</span>
+      <span class='ocrx_word' id='word_1_40' title='bbox 641 482 776 518; x_wconf 96'>similar</span>
      <span class='ocrx_word' id='word_1_41' title='bbox 789 488 829 518; x_wconf 96'>to</span>
      <span class='ocrx_word' id='word_1_42' title='bbox 843 482 1062 519; x_wconf 96'>multi-track</span>
      <span class='ocrx_word' id='word_1_43' title='bbox 1076 488 1160 527; x_wconf 96'>tape</span>
      <span class='ocrx_word' id='word_1_44' title='bbox 1173 483 1336 519; x_wconf 96'>recorder</span>
-      <span class='ocrx_word' id='word_1_45' title='bbox 1350 482 1436 518; x_wconf 95'>with</span>
-      <span class='ocrx_word' id='word_1_46' title='bbox 1451 483 1580 525; x_wconf 96'>PLAY,</span>
+      <span class='ocrx_word' id='word_1_45' title='bbox 1350 482 1436 518; x_wconf 96'>with</span>
+      <span class='ocrx_word' id='word_1_46' title='bbox 1451 483 1580 525; x_wconf 95'>PLAY,</span>
      <span class='ocrx_word' id='word_1_47' title='bbox 1598 483 1724 525; x_wconf 96'>STOP,</span>
-      <span class='ocrx_word' id='word_1_48' title='bbox 1741 483 1957 525; x_wconf 95'>RECORD,</span>
+      <span class='ocrx_word' id='word_1_48' title='bbox 1741 483 1957 525; x_wconf 96'>RECORD,</span>
      <span class='ocrx_word' id='word_1_49' title='bbox 1974 483 2093 518; x_wconf 96'>FAST</span>
     </span>
     <span class='ocr_header' id='line_1_6' title="bbox 383 532 1345 574; baseline 0.001 -7; x_size 43; x_descenders 7; x_ascenders 11">
      <span class='ocrx_word' id='word_1_50' title='bbox 383 532 635 574; x_wconf 96'>FORWARD,</span>
-      <span class='ocrx_word' id='word_1_51' title='bbox 652 532 865 574; x_wconf 96'>REWIND,</span>
-      <span class='ocrx_word' id='word_1_52' title='bbox 882 532 956 568; x_wconf 96'>and</span>
-      <span class='ocrx_word' id='word_1_53' title='bbox 971 532 1163 568; x_wconf 96'>LOCATE</span>
+      <span class='ocrx_word' id='word_1_51' title='bbox 652 532 865 574; x_wconf 95'>REWIND,</span>
+      <span class='ocrx_word' id='word_1_52' title='bbox 882 532 956 568; x_wconf 95'>and</span>
+      <span class='ocrx_word' id='word_1_53' title='bbox 971 532 1163 568; x_wconf 95'>LOCATE</span>
      <span class='ocrx_word' id='word_1_54' title='bbox 1177 532 1345 568; x_wconf 95'>controls.</span>
     </span>
    </p>
@@ -93,7 +93,7 @@
   <div class='ocr_carea' id='block_1_3' title="bbox 349 589 2136 685">
    <p class='ocr_par' id='par_1_5' lang='eng' title="bbox 349 589 2136 685">
     <span class='ocr_header' id='line_1_7' title="bbox 349 589 2136 634; baseline 0.001 -9; x_size 44; x_descenders 8; x_ascenders 11">
-      <span class='ocrx_word' id='word_1_55' title='bbox 349 597 368 615; x_wconf 44'>¢</span>
+      <span class='ocrx_word' id='word_1_55' title='bbox 349 597 368 615; x_wconf 59'>e</span>
      <span class='ocrx_word' id='word_1_56' title='bbox 383 590 482 625; x_wconf 96'>Each</span>
      <span class='ocrx_word' id='word_1_57' title='bbox 496 589 539 625; x_wconf 96'>of</span>
      <span class='ocrx_word' id='word_1_58' title='bbox 549 590 611 626; x_wconf 96'>the</span>
@@ -109,7 +109,7 @@
      <span class='ocrx_word' id='word_1_68' title='bbox 2050 600 2136 634; x_wconf 96'>may</span>
     </span>
     <span class='ocr_header' id='line_1_8' title="bbox 383 639 2022 685; baseline 0.001 -10; x_size 44; x_descenders 8; x_ascenders 11">
-      <span class='ocrx_word' id='word_1_69' title='bbox 383 639 428 675; x_wconf 96'>be</span>
+      <span class='ocrx_word' id='word_1_69' title='bbox 383 639 428 675; x_wconf 95'>be</span>
      <span class='ocrx_word' id='word_1_70' title='bbox 442 639 607 684; x_wconf 95'>assigned</span>
      <span class='ocrx_word' id='word_1_71' title='bbox 621 645 659 676; x_wconf 96'>to</span>
      <span class='ocrx_word' id='word_1_72' title='bbox 674 650 745 676; x_wconf 96'>one</span>
@@ -117,7 +117,7 @@
      <span class='ocrx_word' id='word_1_74' title='bbox 813 641 851 676; x_wconf 96'>16</span>
      <span class='ocrx_word' id='word_1_75' title='bbox 864 641 980 675; x_wconf 96'>MIDI</span>
      <span class='ocrx_word' id='word_1_76' title='bbox 996 640 1176 676; x_wconf 95'>channels.</span>
-      <span class='ocrx_word' id='word_1_77' title='bbox 1194 640 1498 684; x_wconf 95'>Simultaneously</span>
+      <span class='ocrx_word' id='word_1_77' title='bbox 1194 640 1498 684; x_wconf 96'>Simultaneously</span>
      <span class='ocrx_word' id='word_1_78' title='bbox 1510 640 1609 685; x_wconf 96'>plays</span>
      <span class='ocrx_word' id='word_1_79' title='bbox 1624 651 1674 684; x_wconf 96'>up</span>
      <span class='ocrx_word' id='word_1_80' title='bbox 1688 645 1727 676; x_wconf 96'>to</span>
@@ -136,9 +136,9 @@
   <div class='ocr_carea' id='block_1_5' title="bbox 349 748 2117 793">
    <p class='ocr_par' id='par_1_7' lang='eng' title="bbox 349 748 2117 793">
     <span class='ocr_header' id='line_1_10' title="bbox 349 748 2117 793; baseline 0 -9; x_size 45; x_descenders 9; x_ascenders 11">
-      <span class='ocrx_word' id='word_1_84' title='bbox 349 755 367 774; x_wconf 42'>©</span>
-      <span class='ocrx_word' id='word_1_85' title='bbox 383 748 573 784; x_wconf 90'>Ultra-fast</span>
-      <span class='ocrx_word' id='word_1_86' title='bbox 616 749 677 784; x_wconf 9'>32”</span>
+      <span class='ocrx_word' id='word_1_84' title='bbox 349 755 367 774; x_wconf 58'>¢</span>
+      <span class='ocrx_word' id='word_1_85' title='bbox 383 748 573 784; x_wconf 91'>Ultra-fast</span>
+      <span class='ocrx_word' id='word_1_86' title='bbox 588 749 677 784; x_wconf 22'>3%”</span>
      <span class='ocrx_word' id='word_1_87' title='bbox 694 748 775 784; x_wconf 96'>disk</span>
      <span class='ocrx_word' id='word_1_88' title='bbox 790 748 887 784; x_wconf 96'>drive</span>
      <span class='ocrx_word' id='word_1_89' title='bbox 901 754 1012 785; x_wconf 96'>stores</span>
@@ -150,7 +150,7 @@
      <span class='ocrx_word' id='word_1_95' title='bbox 1638 748 1746 784; x_wconf 96'>holds</span>
      <span class='ocrx_word' id='word_1_96' title='bbox 1761 759 1844 784; x_wconf 96'>over</span>
      <span class='ocrx_word' id='word_1_97' title='bbox 1859 749 2000 791; x_wconf 96'>110,000</span>
-      <span class='ocrx_word' id='word_1_98' title='bbox 2013 753 2117 784; x_wconf 97'>notes</span>
+      <span class='ocrx_word' id='word_1_98' title='bbox 2013 753 2117 784; x_wconf 96'>notes</span>
     </span>
    </p>
   </div>
@@ -165,13 +165,13 @@
   <div class='ocr_carea' id='block_1_7' title="bbox 349 855 2030 1016">
    <p class='ocr_par' id='par_1_9' lang='eng' title="bbox 349 855 2030 1016">
     <span class='ocr_header' id='line_1_12' title="bbox 350 855 1638 900; baseline 0.001 -9; x_size 45; x_descenders 9; x_ascenders 11">
-      <span class='ocrx_word' id='word_1_101' title='bbox 350 863 367 881; x_wconf 51'>¢</span>
+      <span class='ocrx_word' id='word_1_101' title='bbox 350 863 367 881; x_wconf 45'>¢</span>
      <span class='ocrx_word' id='word_1_102' title='bbox 383 856 464 891; x_wconf 95'>One</span>
-      <span class='ocrx_word' id='word_1_103' title='bbox 478 866 520 891; x_wconf 95'>or</span>
+      <span class='ocrx_word' id='word_1_103' title='bbox 478 866 520 891; x_wconf 96'>or</span>
      <span class='ocrx_word' id='word_1_104' title='bbox 534 855 580 891; x_wconf 96'>all</span>
      <span class='ocrx_word' id='word_1_105' title='bbox 594 856 712 892; x_wconf 95'>tracks</span>
-      <span class='ocrx_word' id='word_1_106' title='bbox 726 867 811 900; x_wconf 96'>may</span>
-      <span class='ocrx_word' id='word_1_107' title='bbox 823 856 869 892; x_wconf 85'>be</span>
+      <span class='ocrx_word' id='word_1_106' title='bbox 726 867 811 900; x_wconf 95'>may</span>
+      <span class='ocrx_word' id='word_1_107' title='bbox 823 856 869 892; x_wconf 81'>be</span>
      <span class='ocrx_word' id='word_1_108' title='bbox 882 856 1212 892; x_wconf 96'>TRANSPOSED</span>
      <span class='ocrx_word' id='word_1_109' title='bbox 1227 861 1264 892; x_wconf 96'>at</span>
      <span class='ocrx_word' id='word_1_110' title='bbox 1277 856 1338 892; x_wconf 96'>the</span>
@@ -181,7 +181,7 @@
      <span class='ocrx_word' id='word_1_114' title='bbox 1568 856 1638 900; x_wconf 96'>key.</span>
     </span>
     <span class='ocr_header' id='line_1_13' title="bbox 350 913 1535 958; baseline 0.001 -9; x_size 44; x_descenders 8; x_ascenders 11">
-      <span class='ocrx_word' id='word_1_115' title='bbox 350 921 367 939; x_wconf 39'>¢</span>
+      <span class='ocrx_word' id='word_1_115' title='bbox 350 921 367 939; x_wconf 45'>e</span>
      <span class='ocrx_word' id='word_1_116' title='bbox 383 913 568 950; x_wconf 96'>Exclusive</span>
      <span class='ocrx_word' id='word_1_117' title='bbox 581 913 756 950; x_wconf 96'>real-time</span>
      <span class='ocrx_word' id='word_1_118' title='bbox 769 914 929 950; x_wconf 96'>ERASE</span>
@@ -191,11 +191,11 @@
      <span class='ocrx_word' id='word_1_122' title='bbox 1414 915 1535 950; x_wconf 95'>FAST.</span>
     </span>
     <span class='ocr_header' id='line_1_14' title="bbox 349 971 2030 1016; baseline 0.001 -10; x_size 44; x_descenders 8; x_ascenders 11">
-      <span class='ocrx_word' id='word_1_123' title='bbox 349 979 367 997; x_wconf 36'>©</span>
-      <span class='ocrx_word' id='word_1_124' title='bbox 382 971 568 1007; x_wconf 96'>Exclusive</span>
+      <span class='ocrx_word' id='word_1_123' title='bbox 349 979 367 997; x_wconf 0'>*</span>
+      <span class='ocrx_word' id='word_1_124' title='bbox 382 971 568 1007; x_wconf 95'>Exclusive</span>
      <span class='ocrx_word' id='word_1_125' title='bbox 582 972 773 1007; x_wconf 96'>REPEAT</span>
      <span class='ocrx_word' id='word_1_126' title='bbox 787 972 958 1008; x_wconf 96'>function</span>
-      <span class='ocrx_word' id='word_1_127' title='bbox 971 972 1245 1016; x_wconf 96'>automatically</span>
+      <span class='ocrx_word' id='word_1_127' title='bbox 971 972 1245 1016; x_wconf 95'>automatically</span>
      <span class='ocrx_word' id='word_1_128' title='bbox 1258 977 1396 1016; x_wconf 96'>repeats</span>
      <span class='ocrx_word' id='word_1_129' title='bbox 1410 983 1481 1016; x_wconf 96'>any</span>
      <span class='ocrx_word' id='word_1_130' title='bbox 1493 972 1578 1008; x_wconf 96'>held</span>
@@ -209,7 +209,7 @@
   <div class='ocr_carea' id='block_1_8' title="bbox 382 1021 689 1065">
    <p class='ocr_par' id='par_1_10' lang='eng' title="bbox 382 1021 689 1065">
     <span class='ocr_line' id='line_1_15' title="bbox 382 1021 689 1065; baseline 0.003 -8; x_size 45; x_descenders 8; x_ascenders 12">
-      <span class='ocrx_word' id='word_1_135' title='bbox 382 1021 564 1065; x_wconf 95'>rhythmic</span>
+      <span class='ocrx_word' id='word_1_135' title='bbox 382 1021 564 1065; x_wconf 96'>rhythmic</span>
      <span class='ocrx_word' id='word_1_136' title='bbox 577 1021 689 1058; x_wconf 96'>value.</span>
     </span>
    </p>
@@ -217,7 +217,7 @@
   <div class='ocr_carea' id='block_1_9' title="bbox 349 1080 2174 1125">
    <p class='ocr_par' id='par_1_11' lang='eng' title="bbox 349 1080 2174 1125">
     <span class='ocr_header' id='line_1_16' title="bbox 349 1080 2174 1125; baseline 0.001 -11; x_size 45; x_descenders 9; x_ascenders 11">
-      <span class='ocrx_word' id='word_1_137' title='bbox 349 1087 367 1105; x_wconf 82'>¢</span>
+      <span class='ocrx_word' id='word_1_137' title='bbox 349 1087 367 1105; x_wconf 80'>¢</span>
      <span class='ocrx_word' id='word_1_138' title='bbox 382 1080 567 1115; x_wconf 95'>TIMING</span>
      <span class='ocrx_word' id='word_1_139' title='bbox 582 1080 908 1116; x_wconf 95'>CORRECTION</span>
      <span class='ocrx_word' id='word_1_140' title='bbox 921 1080 1041 1116; x_wconf 96'>works</span>
@@ -226,7 +226,7 @@
      <span class='ocrx_word' id='word_1_143' title='bbox 1392 1080 1466 1116; x_wconf 96'>and</span>
      <span class='ocrx_word' id='word_1_144' title='bbox 1480 1085 1644 1124; x_wconf 96'>operates</span>
      <span class='ocrx_word' id='word_1_145' title='bbox 1658 1080 1814 1116; x_wconf 96'>without</span>
-      <span class='ocrx_word' id='word_1_146' title='bbox 1831 1080 2044 1125; x_wconf 93'>‘chopping’</span>
+      <span class='ocrx_word' id='word_1_146' title='bbox 1831 1080 2044 1125; x_wconf 95'>‘chopping’</span>
      <span class='ocrx_word' id='word_1_147' title='bbox 2061 1085 2174 1116; x_wconf 96'>notes.</span>
     </span>
    </p>
@@ -238,7 +238,7 @@
      <span class='ocrx_word' id='word_1_149' title='bbox 382 1137 560 1182; x_wconf 95'>Optional</span>
      <span class='ocrx_word' id='word_1_150' title='bbox 575 1138 739 1174; x_wconf 96'>SMPTE</span>
      <span class='ocrx_word' id='word_1_151' title='bbox 752 1138 839 1174; x_wconf 96'>time</span>
-      <span class='ocrx_word' id='word_1_152' title='bbox 853 1138 945 1174; x_wconf 96'>code</span>
+      <span class='ocrx_word' id='word_1_152' title='bbox 853 1138 945 1174; x_wconf 95'>code</span>
      <span class='ocrx_word' id='word_1_153' title='bbox 959 1138 1287 1182; x_wconf 96'>synchronization.</span>
     </span>
    </p>
@@ -246,7 +246,7 @@
   <div class='ocr_carea' id='block_1_11' title="bbox 349 1195 874 1240">
    <p class='ocr_par' id='par_1_13' lang='eng' title="bbox 349 1195 874 1240">
     <span class='ocr_line' id='line_1_18' title="bbox 349 1195 874 1240; baseline 0 -8; x_size 45; x_descenders 8; x_ascenders 12">
-      <span class='ocrx_word' id='word_1_154' title='bbox 349 1203 367 1222; x_wconf 73'>©</span>
+      <span class='ocrx_word' id='word_1_154' title='bbox 349 1203 367 1222; x_wconf 74'>©</span>
      <span class='ocrx_word' id='word_1_155' title='bbox 382 1195 560 1240; x_wconf 96'>Optional</span>
      <span class='ocrx_word' id='word_1_156' title='bbox 573 1201 709 1233; x_wconf 96'>remote</span>
      <span class='ocrx_word' id='word_1_157' title='bbox 723 1196 874 1233; x_wconf 95'>control.</span>
@@ -256,32 +256,32 @@
   <div class='ocr_carea' id='block_1_12' title="bbox 346 1288 1239 1491">
    <p class='ocr_par' id='par_1_14' lang='eng' title="bbox 346 1288 749 1329">
     <span class='ocr_line' id='line_1_19' title="bbox 346 1288 749 1329; baseline 0.002 -9; x_size 42; x_descenders 9; x_ascenders 10">
-      <span class='ocrx_word' id='word_1_158' title='bbox 346 1288 535 1329; x_wconf 96'>Recording</span>
-      <span class='ocrx_word' id='word_1_159' title='bbox 547 1298 567 1321; x_wconf 96'>a</span>
+      <span class='ocrx_word' id='word_1_158' title='bbox 346 1288 535 1329; x_wconf 95'>Recording</span>
+      <span class='ocrx_word' id='word_1_159' title='bbox 547 1298 567 1321; x_wconf 95'>a</span>
      <span class='ocrx_word' id='word_1_160' title='bbox 579 1288 749 1328; x_wconf 96'>Sequence</span>
     </span>
    </p>

    <p class='ocr_par' id='par_1_15' lang='eng' title="bbox 346 1339 1239 1491">
     <span class='ocr_line' id='line_1_20' title="bbox 383 1339 1239 1373; baseline 0 -7; x_size 33; x_descenders 7; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_161' title='bbox 383 1340 420 1366; x_wconf 96'>To</span>
+      <span class='ocrx_word' id='word_1_161' title='bbox 383 1340 420 1366; x_wconf 95'>To</span>
      <span class='ocrx_word' id='word_1_162' title='bbox 430 1339 524 1366; x_wconf 96'>record</span>
      <span class='ocrx_word' id='word_1_163' title='bbox 535 1347 551 1366; x_wconf 96'>a</span>
      <span class='ocrx_word' id='word_1_164' title='bbox 562 1347 704 1373; x_wconf 96'>sequence,</span>
      <span class='ocrx_word' id='word_1_165' title='bbox 716 1340 815 1373; x_wconf 96'>simply</span>
-      <span class='ocrx_word' id='word_1_166' title='bbox 825 1348 898 1373; x_wconf 96'>press</span>
+      <span class='ocrx_word' id='word_1_166' title='bbox 825 1348 898 1373; x_wconf 95'>press</span>
      <span class='ocrx_word' id='word_1_167' title='bbox 910 1340 1065 1367; x_wconf 96'>RECORD</span>
      <span class='ocrx_word' id='word_1_168' title='bbox 1075 1340 1131 1367; x_wconf 96'>and</span>
-      <span class='ocrx_word' id='word_1_169' title='bbox 1142 1341 1239 1372; x_wconf 95'>PLAY,</span>
+      <span class='ocrx_word' id='word_1_169' title='bbox 1142 1341 1239 1372; x_wconf 96'>PLAY,</span>
     </span>
     <span class='ocr_line' id='line_1_21' title="bbox 346 1378 1205 1412; baseline 0.001 -7; x_size 34; x_descenders 7; x_ascenders 8">
      <span class='ocrx_word' id='word_1_170' title='bbox 346 1379 411 1406; x_wconf 96'>then</span>
      <span class='ocrx_word' id='word_1_171' title='bbox 422 1378 483 1412; x_wconf 96'>play</span>
      <span class='ocrx_word' id='word_1_172' title='bbox 493 1387 562 1412; x_wconf 96'>your</span>
-      <span class='ocrx_word' id='word_1_173' title='bbox 572 1379 659 1405; x_wconf 95'>MIDI</span>
-      <span class='ocrx_word' id='word_1_174' title='bbox 671 1379 810 1412; x_wconf 95'>keyboard</span>
-      <span class='ocrx_word' id='word_1_175' title='bbox 821 1379 848 1406; x_wconf 96'>in</span>
-      <span class='ocrx_word' id='word_1_176' title='bbox 858 1379 923 1406; x_wconf 96'>time</span>
+      <span class='ocrx_word' id='word_1_173' title='bbox 572 1379 659 1405; x_wconf 96'>MIDI</span>
+      <span class='ocrx_word' id='word_1_174' title='bbox 671 1379 810 1412; x_wconf 96'>keyboard</span>
+      <span class='ocrx_word' id='word_1_175' title='bbox 821 1379 848 1406; x_wconf 95'>in</span>
+      <span class='ocrx_word' id='word_1_176' title='bbox 858 1379 923 1406; x_wconf 95'>time</span>
      <span class='ocrx_word' id='word_1_177' title='bbox 934 1384 963 1406; x_wconf 96'>to</span>
      <span class='ocrx_word' id='word_1_178' title='bbox 974 1379 1019 1406; x_wconf 93'>the</span>
      <span class='ocrx_word' id='word_1_179' title='bbox 1030 1379 1205 1412; x_wconf 92'>Sequencer’s</span>
@@ -297,11 +297,11 @@
      <span class='ocrx_word' id='word_1_187' title='bbox 995 1419 1101 1446; x_wconf 96'>around</span>
      <span class='ocrx_word' id='word_1_188' title='bbox 1112 1423 1141 1446; x_wconf 96'>to</span>
      <span class='ocrx_word' id='word_1_189' title='bbox 1152 1419 1201 1446; x_wconf 96'>bar</span>
-      <span class='ocrx_word' id='word_1_190' title='bbox 1213 1419 1232 1450; x_wconf 88'>1,</span>
+      <span class='ocrx_word' id='word_1_190' title='bbox 1213 1419 1232 1450; x_wconf 74'>1,</span>
     </span>
     <span class='ocr_line' id='line_1_23' title="bbox 346 1457 1223 1491; baseline 0.001 -7; x_size 34; x_descenders 7; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_191' title='bbox 346 1457 430 1490; x_wconf 16'>you’</span>
-      <span class='ocrx_word' id='word_1_192' title='bbox 410 1453 436 1496; x_wconf 16'>ll</span>
+      <span class='ocrx_word' id='word_1_191' title='bbox 346 1457 430 1490; x_wconf 14'>you’</span>
+      <span class='ocrx_word' id='word_1_192' title='bbox 406 1453 436 1496; x_wconf 14'>ll</span>
      <span class='ocrx_word' id='word_1_193' title='bbox 441 1457 506 1485; x_wconf 96'>hear</span>
      <span class='ocrx_word' id='word_1_194' title='bbox 517 1458 590 1485; x_wconf 96'>what</span>
      <span class='ocrx_word' id='word_1_195' title='bbox 600 1466 654 1491; x_wconf 93'>you</span>
@@ -323,7 +323,7 @@
      <span class='ocrx_word' id='word_1_205' title='bbox 802 1506 864 1531; x_wconf 96'>may</span>
      <span class='ocrx_word' id='word_1_206' title='bbox 875 1498 909 1525; x_wconf 96'>be</span>
      <span class='ocrx_word' id='word_1_207' title='bbox 920 1497 1047 1531; x_wconf 96'>adjusted</span>
-      <span class='ocrx_word' id='word_1_208' title='bbox 1058 1505 1089 1525; x_wconf 96'>or</span>
+      <span class='ocrx_word' id='word_1_208' title='bbox 1058 1505 1089 1525; x_wconf 97'>or</span>
      <span class='ocrx_word' id='word_1_209' title='bbox 1099 1497 1245 1531; x_wconf 96'>defeated).</span>
     </span>
    </p>
@@ -343,8 +343,8 @@
      <span class='ocrx_word' id='word_1_219' title='bbox 1111 1537 1186 1564; x_wconf 96'>track</span>
     </span>
     <span class='ocr_line' id='line_1_26' title="bbox 347 1575 1052 1610; baseline 0.001 -8; x_size 34; x_descenders 7; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_220' title='bbox 347 1591 369 1594; x_wconf 0'>—</span>
-      <span class='ocrx_word' id='word_1_221' title='bbox 375 1575 495 1609; x_wconf 0'>existing</span>
+      <span class='ocrx_word' id='word_1_220' title='bbox 347 1591 372 1594; x_wconf 0'>—</span>
+      <span class='ocrx_word' id='word_1_221' title='bbox 371 1575 495 1609; x_wconf 0'>existing</span>
      <span class='ocrx_word' id='word_1_222' title='bbox 505 1580 582 1603; x_wconf 96'>notes</span>
      <span class='ocrx_word' id='word_1_223' title='bbox 593 1584 637 1603; x_wconf 96'>are</span>
      <span class='ocrx_word' id='word_1_224' title='bbox 648 1580 696 1603; x_wconf 97'>not</span>
@@ -356,7 +356,7 @@

    <p class='ocr_par' id='par_1_18' lang='eng' title="bbox 346 1616 1205 1965">
     <span class='ocr_line' id='line_1_27' title="bbox 384 1616 1199 1648; baseline 0.001 -6; x_size 32; x_descenders 5; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_228' title='bbox 384 1616 471 1642; x_wconf 95'>FAST</span>
+      <span class='ocrx_word' id='word_1_228' title='bbox 384 1616 471 1642; x_wconf 96'>FAST</span>
      <span class='ocrx_word' id='word_1_229' title='bbox 481 1616 671 1648; x_wconf 96'>FORWARD,</span>
      <span class='ocrx_word' id='word_1_230' title='bbox 684 1617 844 1648; x_wconf 95'>REWIND,</span>
      <span class='ocrx_word' id='word_1_231' title='bbox 857 1616 912 1643; x_wconf 95'>and</span>
@@ -364,9 +364,9 @@
      <span class='ocrx_word' id='word_1_233' title='bbox 1079 1616 1199 1643; x_wconf 95'>controls</span>
     </span>
     <span class='ocr_line' id='line_1_28' title="bbox 346 1655 1202 1689; baseline 0 -7; x_size 34; x_descenders 6; x_ascenders 9">
-      <span class='ocrx_word' id='word_1_234' title='bbox 346 1663 409 1688; x_wconf 87'>may</span>
+      <span class='ocrx_word' id='word_1_234' title='bbox 346 1663 409 1688; x_wconf 92'>may</span>
      <span class='ocrx_word' id='word_1_235' title='bbox 419 1655 453 1682; x_wconf 96'>be</span>
-      <span class='ocrx_word' id='word_1_236' title='bbox 463 1655 530 1682; x_wconf 95'>used</span>
+      <span class='ocrx_word' id='word_1_236' title='bbox 463 1655 530 1682; x_wconf 96'>used</span>
      <span class='ocrx_word' id='word_1_237' title='bbox 541 1659 569 1682; x_wconf 96'>at</span>
      <span class='ocrx_word' id='word_1_238' title='bbox 580 1663 632 1688; x_wconf 96'>any</span>
      <span class='ocrx_word' id='word_1_239' title='bbox 642 1655 707 1683; x_wconf 96'>time</span>
@@ -380,10 +380,10 @@
     <span class='ocr_line' id='line_1_29' title="bbox 346 1694 1204 1728; baseline 0.001 -7; x_size 33; x_descenders 6; x_ascenders 8">
      <span class='ocrx_word' id='word_1_246' title='bbox 346 1702 414 1727; x_wconf 96'>your</span>
      <span class='ocrx_word' id='word_1_247' title='bbox 424 1702 558 1727; x_wconf 96'>sequence</span>
-      <span class='ocrx_word' id='word_1_248' title='bbox 570 1694 612 1721; x_wconf 92'>for</span>
-      <span class='ocrx_word' id='word_1_249' title='bbox 623 1695 847 1728; x_wconf 92'>spot-recording.</span>
+      <span class='ocrx_word' id='word_1_248' title='bbox 570 1694 612 1721; x_wconf 93'>for</span>
+      <span class='ocrx_word' id='word_1_249' title='bbox 623 1695 847 1728; x_wconf 91'>spot-recording.</span>
      <span class='ocrx_word' id='word_1_250' title='bbox 860 1696 897 1722; x_wconf 93'>To</span>
-      <span class='ocrx_word' id='word_1_251' title='bbox 908 1695 1028 1722; x_wconf 92'>overdub</span>
+      <span class='ocrx_word' id='word_1_251' title='bbox 908 1695 1028 1722; x_wconf 93'>overdub</span>
      <span class='ocrx_word' id='word_1_252' title='bbox 1039 1703 1056 1722; x_wconf 96'>a</span>
      <span class='ocrx_word' id='word_1_253' title='bbox 1066 1703 1125 1722; x_wconf 96'>new</span>
      <span class='ocrx_word' id='word_1_254' title='bbox 1135 1699 1204 1728; x_wconf 96'>part,</span>
@@ -400,13 +400,13 @@
     </span>
     <span class='ocr_line' id='line_1_31' title="bbox 346 1773 1203 1808; baseline 0.001 -8; x_size 33; x_descenders 6; x_ascenders 8">
      <span class='ocrx_word' id='word_1_263' title='bbox 346 1774 448 1806; x_wconf 96'>record,</span>
-      <span class='ocrx_word' id='word_1_264' title='bbox 460 1774 506 1801; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_265' title='bbox 510 1769 577 1812; x_wconf 95'>first</span>
-      <span class='ocrx_word' id='word_1_266' title='bbox 589 1774 658 1801; x_wconf 95'>track</span>
+      <span class='ocrx_word' id='word_1_264' title='bbox 460 1774 506 1801; x_wconf 97'>the</span>
+      <span class='ocrx_word' id='word_1_265' title='bbox 503 1769 577 1812; x_wconf 96'>first</span>
+      <span class='ocrx_word' id='word_1_266' title='bbox 581 1774 658 1801; x_wconf 96'>track</span>
      <span class='ocrx_word' id='word_1_267' title='bbox 673 1774 726 1801; x_wconf 96'>will</span>
      <span class='ocrx_word' id='word_1_268' title='bbox 736 1774 799 1807; x_wconf 96'>play</span>
      <span class='ocrx_word' id='word_1_269' title='bbox 809 1774 836 1801; x_wconf 96'>in</span>
-      <span class='ocrx_word' id='word_1_270' title='bbox 847 1774 949 1808; x_wconf 96'>perfect</span>
+      <span class='ocrx_word' id='word_1_270' title='bbox 847 1774 949 1808; x_wconf 97'>perfect</span>
      <span class='ocrx_word' id='word_1_271' title='bbox 961 1782 1026 1808; x_wconf 96'>sync</span>
      <span class='ocrx_word' id='word_1_272' title='bbox 1037 1774 1137 1808; x_wconf 96'>(unless</span>
      <span class='ocrx_word' id='word_1_273' title='bbox 1148 1782 1203 1807; x_wconf 96'>you</span>
@@ -442,7 +442,7 @@
      <span class='ocrx_word' id='word_1_297' title='bbox 580 1892 663 1924; x_wconf 96'>bend,</span>
      <span class='ocrx_word' id='word_1_298' title='bbox 675 1892 859 1924; x_wconf 96'>modulation,</span>
      <span class='ocrx_word' id='word_1_299' title='bbox 872 1892 991 1926; x_wconf 93'>velocity,</span>
-      <span class='ocrx_word' id='word_1_300' title='bbox 1004 1892 1168 1924; x_wconf 91'>aftertouch,</span>
+      <span class='ocrx_word' id='word_1_300' title='bbox 1004 1892 1168 1924; x_wconf 92'>aftertouch,</span>
     </span>
     <span class='ocr_line' id='line_1_35' title="bbox 346 1931 895 1965; baseline 0.002 -7; x_size 33; x_descenders 6; x_ascenders 8">
      <span class='ocrx_word' id='word_1_301' title='bbox 346 1931 448 1958; x_wconf 96'>sustain</span>
@@ -463,7 +463,7 @@
    <p class='ocr_par' id='par_1_20' lang='eng' title="bbox 346 2050 1212 2163">
     <span class='ocr_line' id='line_1_37' title="bbox 383 2050 1186 2084; baseline 0.001 -8; x_size 34; x_descenders 7; x_ascenders 8">
      <span class='ocrx_word' id='word_1_307' title='bbox 383 2050 419 2076; x_wconf 96'>To</span>
-      <span class='ocrx_word' id='word_1_308' title='bbox 430 2057 503 2076; x_wconf 96'>erase</span>
+      <span class='ocrx_word' id='word_1_308' title='bbox 430 2057 503 2076; x_wconf 95'>erase</span>
      <span class='ocrx_word' id='word_1_309' title='bbox 514 2058 530 2077; x_wconf 96'>a</span>
      <span class='ocrx_word' id='word_1_310' title='bbox 540 2058 634 2083; x_wconf 96'>wrong</span>
      <span class='ocrx_word' id='word_1_311' title='bbox 644 2054 717 2082; x_wconf 96'>note,</span>
@@ -477,7 +477,7 @@
      <span class='ocrx_word' id='word_1_317' title='bbox 346 2089 391 2116; x_wconf 96'>the</span>
      <span class='ocrx_word' id='word_1_318' title='bbox 402 2094 465 2117; x_wconf 96'>note</span>
      <span class='ocrx_word' id='word_1_319' title='bbox 475 2094 504 2117; x_wconf 96'>to</span>
-      <span class='ocrx_word' id='word_1_320' title='bbox 515 2090 549 2117; x_wconf 96'>be</span>
+      <span class='ocrx_word' id='word_1_320' title='bbox 515 2090 549 2117; x_wconf 97'>be</span>
      <span class='ocrx_word' id='word_1_321' title='bbox 559 2090 652 2117; x_wconf 96'>erased</span>
      <span class='ocrx_word' id='word_1_322' title='bbox 661 2090 718 2123; x_wconf 96'>just</span>
      <span class='ocrx_word' id='word_1_323' title='bbox 729 2090 822 2117; x_wconf 96'>before</span>
@@ -485,7 +485,7 @@
      <span class='ocrx_word' id='word_1_325' title='bbox 862 2090 937 2124; x_wconf 96'>plays</span>
      <span class='ocrx_word' id='word_1_326' title='bbox 947 2090 975 2117; x_wconf 96'>in</span>
      <span class='ocrx_word' id='word_1_327' title='bbox 986 2090 1032 2118; x_wconf 93'>the</span>
-      <span class='ocrx_word' id='word_1_328' title='bbox 1043 2098 1212 2124; x_wconf 91'>sequence—</span>
+      <span class='ocrx_word' id='word_1_328' title='bbox 1043 2098 1212 2124; x_wconf 88'>sequence—</span>
     </span>
     <span class='ocr_line' id='line_1_39' title="bbox 346 2129 1134 2163; baseline 0.003 -8; x_size 34; x_descenders 7; x_ascenders 8">
      <span class='ocrx_word' id='word_1_329' title='bbox 346 2129 425 2156; x_wconf 96'>when</span>
@@ -510,21 +510,21 @@
      <span class='ocrx_word' id='word_1_342' title='bbox 572 2177 604 2196; x_wconf 96'>or</span>
      <span class='ocrx_word' id='word_1_343' title='bbox 614 2169 739 2202; x_wconf 96'>changed</span>
      <span class='ocrx_word' id='word_1_344' title='bbox 749 2169 829 2203; x_wconf 96'>using</span>
-      <span class='ocrx_word' id='word_1_345' title='bbox 839 2169 885 2196; x_wconf 95'>the</span>
-      <span class='ocrx_word' id='word_1_346' title='bbox 896 2170 1031 2196; x_wconf 95'>SINGLE</span>
-      <span class='ocrx_word' id='word_1_347' title='bbox 1042 2170 1131 2196; x_wconf 93'>STEP</span>
-      <span class='ocrx_word' id='word_1_348' title='bbox 1143 2169 1220 2196; x_wconf 92'>func-</span>
+      <span class='ocrx_word' id='word_1_345' title='bbox 839 2169 885 2196; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_346' title='bbox 896 2170 1031 2196; x_wconf 96'>SINGLE</span>
+      <span class='ocrx_word' id='word_1_347' title='bbox 1042 2170 1131 2196; x_wconf 91'>STEP</span>
+      <span class='ocrx_word' id='word_1_348' title='bbox 1143 2169 1220 2196; x_wconf 91'>func-</span>
     </span>
     <span class='ocr_line' id='line_1_41' title="bbox 345 2207 1228 2242; baseline 0.002 -8; x_size 35; x_descenders 7; x_ascenders 9">
      <span class='ocrx_word' id='word_1_349' title='bbox 345 2207 412 2234; x_wconf 96'>tion.</span>
      <span class='ocrx_word' id='word_1_350' title='bbox 424 2208 461 2235; x_wconf 93'>To</span>
-      <span class='ocrx_word' id='word_1_351' title='bbox 472 2208 592 2235; x_wconf 92'>overdub</span>
+      <span class='ocrx_word' id='word_1_351' title='bbox 472 2208 592 2235; x_wconf 91'>overdub</span>
      <span class='ocrx_word' id='word_1_352' title='bbox 603 2212 680 2235; x_wconf 96'>notes</span>
      <span class='ocrx_word' id='word_1_353' title='bbox 691 2212 718 2235; x_wconf 96'>at</span>
      <span class='ocrx_word' id='word_1_354' title='bbox 729 2208 841 2242; x_wconf 96'>specific</span>
      <span class='ocrx_word' id='word_1_355' title='bbox 851 2209 943 2242; x_wconf 97'>points</span>
-      <span class='ocrx_word' id='word_1_356' title='bbox 955 2208 1049 2236; x_wconf 97'>within</span>
-      <span class='ocrx_word' id='word_1_357' title='bbox 1060 2217 1076 2236; x_wconf 97'>a</span>
+      <span class='ocrx_word' id='word_1_356' title='bbox 955 2208 1049 2236; x_wconf 96'>within</span>
+      <span class='ocrx_word' id='word_1_357' title='bbox 1060 2217 1076 2236; x_wconf 96'>a</span>
      <span class='ocrx_word' id='word_1_358' title='bbox 1086 2216 1228 2242; x_wconf 96'>sequence,</span>
     </span>
    </p>
@@ -551,7 +551,7 @@
     </span>
     <span class='ocr_line' id='line_1_44' title="bbox 1297 1328 2033 1362; baseline 0.001 -7; x_size 32; x_descenders 5; x_ascenders 8">
      <span class='ocrx_word' id='word_1_369' title='bbox 1297 1328 1356 1355; x_wconf 96'>find</span>
-      <span class='ocrx_word' id='word_1_370' title='bbox 1366 1329 1412 1355; x_wconf 97'>the</span>
+      <span class='ocrx_word' id='word_1_370' title='bbox 1366 1329 1412 1355; x_wconf 96'>the</span>
      <span class='ocrx_word' id='word_1_371' title='bbox 1423 1328 1527 1356; x_wconf 96'>desired</span>
      <span class='ocrx_word' id='word_1_372' title='bbox 1537 1329 1587 1356; x_wconf 96'>bar</span>
      <span class='ocrx_word' id='word_1_373' title='bbox 1598 1329 1720 1361; x_wconf 96'>number,</span>
@@ -570,13 +570,13 @@
      <span class='ocrx_word' id='word_1_381' title='bbox 1904 1376 1958 1402; x_wconf 96'>you</span>
      <span class='ocrx_word' id='word_1_382' title='bbox 1968 1373 1997 1395; x_wconf 96'>to</span>
      <span class='ocrx_word' id='word_1_383' title='bbox 2008 1377 2087 1396; x_wconf 96'>move</span>
-      <span class='ocrx_word' id='word_1_384' title='bbox 2097 1369 2160 1396; x_wconf 96'>bars</span>
+      <span class='ocrx_word' id='word_1_384' title='bbox 2097 1369 2160 1396; x_wconf 97'>bars</span>
     </span>
     <span class='ocr_line' id='line_1_46' title="bbox 1297 1407 2151 1441; baseline 0.002 -8; x_size 33; x_descenders 6; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_385' title='bbox 1297 1407 1369 1434; x_wconf 95'>from</span>
+      <span class='ocrx_word' id='word_1_385' title='bbox 1297 1407 1369 1434; x_wconf 96'>from</span>
      <span class='ocrx_word' id='word_1_386' title='bbox 1380 1415 1433 1434; x_wconf 95'>one</span>
      <span class='ocrx_word' id='word_1_387' title='bbox 1443 1407 1565 1435; x_wconf 95'>location</span>
-      <span class='ocrx_word' id='word_1_388' title='bbox 1576 1411 1605 1434; x_wconf 91'>to</span>
+      <span class='ocrx_word' id='word_1_388' title='bbox 1576 1411 1605 1434; x_wconf 92'>to</span>
      <span class='ocrx_word' id='word_1_389' title='bbox 1616 1407 1796 1435; x_wconf 91'>another—in</span>
      <span class='ocrx_word' id='word_1_390' title='bbox 1806 1408 1852 1435; x_wconf 96'>the</span>
      <span class='ocrx_word' id='word_1_391' title='bbox 1863 1416 1937 1435; x_wconf 96'>same</span>
@@ -615,12 +615,12 @@
      <span class='ocrx_word' id='word_1_418' title='bbox 1691 1527 1737 1553; x_wconf 96'>the</span>
      <span class='ocrx_word' id='word_1_419' title='bbox 1748 1535 1823 1554; x_wconf 96'>same</span>
      <span class='ocrx_word' id='word_1_420' title='bbox 1833 1535 1891 1560; x_wconf 96'>way</span>
-      <span class='ocrx_word' id='word_1_421' title='bbox 1901 1531 1930 1554; x_wconf 95'>to</span>
+      <span class='ocrx_word' id='word_1_421' title='bbox 1901 1531 1930 1554; x_wconf 96'>to</span>
      <span class='ocrx_word' id='word_1_422' title='bbox 1940 1535 2047 1554; x_wconf 96'>remove</span>
     </span>
     <span class='ocr_line' id='line_1_50' title="bbox 1295 1565 1577 1593; baseline 0.004 -1; x_size 34.748871; x_descenders 6.7488689; x_ascenders 9">
      <span class='ocrx_word' id='word_1_423' title='bbox 1295 1565 1441 1592; x_wconf 96'>unwanted</span>
-      <span class='ocrx_word' id='word_1_424' title='bbox 1452 1565 1577 1593; x_wconf 94'>sections,</span>
+      <span class='ocrx_word' id='word_1_424' title='bbox 1452 1565 1577 1593; x_wconf 95'>sections,</span>
     </span>
    </p>
   </div>
@@ -640,8 +640,8 @@
      <span class='ocrx_word' id='word_1_430' title='bbox 1472 1694 1500 1717; x_wconf 96'>to</span>
      <span class='ocrx_word' id='word_1_431' title='bbox 1511 1694 1598 1717; x_wconf 96'>create</span>
      <span class='ocrx_word' id='word_1_432' title='bbox 1608 1698 1625 1717; x_wconf 96'>a</span>
-      <span class='ocrx_word' id='word_1_433' title='bbox 1635 1698 1704 1723; x_wconf 96'>song</span>
-      <span class='ocrx_word' id='word_1_434' title='bbox 1715 1690 1736 1717; x_wconf 96'>is</span>
+      <span class='ocrx_word' id='word_1_433' title='bbox 1635 1698 1704 1723; x_wconf 95'>song</span>
+      <span class='ocrx_word' id='word_1_434' title='bbox 1715 1690 1736 1717; x_wconf 95'>is</span>
      <span class='ocrx_word' id='word_1_435' title='bbox 1747 1694 1776 1717; x_wconf 96'>to</span>
      <span class='ocrx_word' id='word_1_436' title='bbox 1787 1690 1880 1717; x_wconf 96'>record</span>
      <span class='ocrx_word' id='word_1_437' title='bbox 1891 1690 1958 1717; x_wconf 96'>each</span>
@@ -652,14 +652,14 @@
     <span class='ocr_line' id='line_1_53' title="bbox 1295 1729 2121 1762; baseline 0.001 -6; x_size 33; x_descenders 6; x_ascenders 8">
      <span class='ocrx_word' id='word_1_441' title='bbox 1295 1737 1353 1762; x_wconf 96'>way</span>
      <span class='ocrx_word' id='word_1_442' title='bbox 1362 1729 1481 1762; x_wconf 96'>through</span>
-      <span class='ocrx_word' id='word_1_443' title='bbox 1493 1729 1541 1762; x_wconf 96'>(up</span>
-      <span class='ocrx_word' id='word_1_444' title='bbox 1552 1733 1581 1756; x_wconf 96'>to</span>
+      <span class='ocrx_word' id='word_1_443' title='bbox 1493 1729 1541 1762; x_wconf 95'>(up</span>
+      <span class='ocrx_word' id='word_1_444' title='bbox 1552 1733 1581 1756; x_wconf 95'>to</span>
      <span class='ocrx_word' id='word_1_445' title='bbox 1592 1730 1644 1756; x_wconf 96'>999</span>
      <span class='ocrx_word' id='word_1_446' title='bbox 1654 1729 1738 1762; x_wconf 96'>bars).</span>
      <span class='ocrx_word' id='word_1_447' title='bbox 1751 1729 1878 1757; x_wconf 96'>Another</span>
      <span class='ocrx_word' id='word_1_448' title='bbox 1888 1737 1945 1762; x_wconf 96'>way</span>
      <span class='ocrx_word' id='word_1_449' title='bbox 1956 1729 1977 1757; x_wconf 96'>is</span>
-      <span class='ocrx_word' id='word_1_450' title='bbox 1987 1733 2016 1757; x_wconf 97'>to</span>
+      <span class='ocrx_word' id='word_1_450' title='bbox 1987 1733 2016 1757; x_wconf 96'>to</span>
      <span class='ocrx_word' id='word_1_451' title='bbox 2027 1729 2121 1757; x_wconf 96'>record</span>
     </span>
     <span class='ocr_line' id='line_1_54' title="bbox 1296 1768 2066 1802; baseline 0 -6; x_size 33; x_descenders 5; x_ascenders 9">
@@ -668,8 +668,8 @@
      <span class='ocrx_word' id='word_1_454' title='bbox 1458 1768 1562 1796; x_wconf 96'>section</span>
      <span class='ocrx_word' id='word_1_455' title='bbox 1574 1769 1666 1802; x_wconf 96'>(verse,</span>
      <span class='ocrx_word' id='word_1_456' title='bbox 1679 1769 1788 1801; x_wconf 96'>chorus,</span>
-      <span class='ocrx_word' id='word_1_457' title='bbox 1800 1769 1865 1802; x_wconf 95'>etc.)</span>
-      <span class='ocrx_word' id='word_1_458' title='bbox 1876 1768 1904 1795; x_wconf 95'>in</span>
+      <span class='ocrx_word' id='word_1_457' title='bbox 1800 1769 1865 1802; x_wconf 96'>etc.)</span>
+      <span class='ocrx_word' id='word_1_458' title='bbox 1876 1768 1904 1795; x_wconf 96'>in</span>
      <span class='ocrx_word' id='word_1_459' title='bbox 1914 1768 2066 1796; x_wconf 96'>individual</span>
     </span>
     <span class='ocr_line' id='line_1_55' title="bbox 1296 1808 2215 1841; baseline 0 -6; x_size 32; x_descenders 5; x_ascenders 8">
@@ -677,7 +677,7 @@
      <span class='ocrx_word' id='word_1_461' title='bbox 1463 1808 1528 1835; x_wconf 96'>then</span>
      <span class='ocrx_word' id='word_1_462' title='bbox 1538 1816 1587 1835; x_wconf 96'>use</span>
      <span class='ocrx_word' id='word_1_463' title='bbox 1597 1808 1643 1835; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_464' title='bbox 1653 1809 1799 1835; x_wconf 95'>CREATE</span>
+      <span class='ocrx_word' id='word_1_464' title='bbox 1653 1809 1799 1835; x_wconf 96'>CREATE</span>
      <span class='ocrx_word' id='word_1_465' title='bbox 1810 1808 1911 1835; x_wconf 96'>SONG</span>
      <span class='ocrx_word' id='word_1_466' title='bbox 1923 1808 2050 1836; x_wconf 96'>function</span>
      <span class='ocrx_word' id='word_1_467' title='bbox 2060 1812 2089 1835; x_wconf 96'>to</span>
@@ -686,7 +686,7 @@
     <span class='ocr_line' id='line_1_56' title="bbox 1295 1847 2135 1881; baseline 0.001 -7; x_size 34; x_descenders 7; x_ascenders 8">
      <span class='ocrx_word' id='word_1_469' title='bbox 1295 1848 1370 1874; x_wconf 96'>them</span>
      <span class='ocrx_word' id='word_1_470' title='bbox 1381 1848 1508 1881; x_wconf 95'>together.</span>
-      <span class='ocrx_word' id='word_1_471' title='bbox 1521 1848 1667 1875; x_wconf 95'>CREATE</span>
+      <span class='ocrx_word' id='word_1_471' title='bbox 1521 1848 1667 1875; x_wconf 96'>CREATE</span>
      <span class='ocrx_word' id='word_1_472' title='bbox 1678 1848 1779 1875; x_wconf 96'>SONG</span>
      <span class='ocrx_word' id='word_1_473' title='bbox 1789 1847 1842 1874; x_wconf 96'>will</span>
      <span class='ocrx_word' id='word_1_474' title='bbox 1853 1848 1918 1875; x_wconf 96'>then</span>
@@ -697,7 +697,7 @@
      <span class='ocrx_word' id='word_1_477' title='bbox 1377 1887 1412 1914; x_wconf 96'>all</span>
      <span class='ocrx_word' id='word_1_478' title='bbox 1422 1887 1468 1914; x_wconf 96'>the</span>
      <span class='ocrx_word' id='word_1_479' title='bbox 1478 1891 1552 1920; x_wconf 96'>parts</span>
-      <span class='ocrx_word' id='word_1_480' title='bbox 1563 1887 1621 1914; x_wconf 96'>into</span>
+      <span class='ocrx_word' id='word_1_480' title='bbox 1563 1887 1621 1914; x_wconf 95'>into</span>
      <span class='ocrx_word' id='word_1_481' title='bbox 1632 1895 1649 1914; x_wconf 96'>a</span>
      <span class='ocrx_word' id='word_1_482' title='bbox 1659 1895 1718 1914; x_wconf 96'>new</span>
      <span class='ocrx_word' id='word_1_483' title='bbox 1729 1895 1870 1920; x_wconf 96'>sequence.</span>
@@ -716,8 +716,8 @@
      <span class='ocrx_word' id='word_1_494' title='bbox 1675 1931 1704 1954; x_wconf 96'>to</span>
      <span class='ocrx_word' id='word_1_495' title='bbox 1715 1931 1806 1960; x_wconf 96'>repeat</span>
      <span class='ocrx_word' id='word_1_496' title='bbox 1816 1926 1955 1960; x_wconf 96'>infinitely,</span>
-      <span class='ocrx_word' id='word_1_497' title='bbox 1968 1926 2011 1954; x_wconf 96'>for</span>
-      <span class='ocrx_word' id='word_1_498' title='bbox 2022 1935 2038 1954; x_wconf 92'>a</span>
+      <span class='ocrx_word' id='word_1_497' title='bbox 1968 1926 2011 1954; x_wconf 95'>for</span>
+      <span class='ocrx_word' id='word_1_498' title='bbox 2022 1935 2038 1954; x_wconf 93'>a</span>
      <span class='ocrx_word' id='word_1_499' title='bbox 2049 1927 2169 1954; x_wconf 92'>fadeout.</span>
     </span>
    </p>
@@ -725,7 +725,7 @@
   <div class='ocr_carea' id='block_1_20' title="bbox 1293 2000 2179 2242">
    <p class='ocr_par' id='par_1_27' lang='eng' title="bbox 1294 2000 1948 2040">
     <span class='ocr_line' id='line_1_59' title="bbox 1294 2000 1948 2040; baseline 0.002 -8; x_size 40; x_descenders 8; x_ascenders 10">
-      <span class='ocrx_word' id='word_1_500' title='bbox 1294 2000 1532 2040; x_wconf 96'>Composition</span>
+      <span class='ocrx_word' id='word_1_500' title='bbox 1294 2000 1532 2040; x_wconf 95'>Composition</span>
      <span class='ocrx_word' id='word_1_501' title='bbox 1544 2000 1699 2033; x_wconf 96'>Without</span>
      <span class='ocrx_word' id='word_1_502' title='bbox 1711 2000 1948 2040; x_wconf 96'>Compromise</span>
     </span>
@@ -741,24 +741,24 @@
      <span class='ocrx_word' id='word_1_508' title='bbox 1808 2059 1887 2078; x_wconf 96'>never</span>
      <span class='ocrx_word' id='word_1_509' title='bbox 1897 2052 1932 2079; x_wconf 96'>be</span>
      <span class='ocrx_word' id='word_1_510' title='bbox 1942 2059 1973 2079; x_wconf 96'>so</span>
-      <span class='ocrx_word' id='word_1_511' title='bbox 1984 2051 2110 2085; x_wconf 89'>complex</span>
+      <span class='ocrx_word' id='word_1_511' title='bbox 1984 2051 2110 2085; x_wconf 67'>complex</span>
      <span class='ocrx_word' id='word_1_512' title='bbox 2120 2052 2179 2078; x_wconf 96'>that</span>
     </span>
     <span class='ocr_line' id='line_1_61' title="bbox 1294 2090 2157 2124; baseline 0 -6; x_size 34; x_descenders 7; x_ascenders 8">
      <span class='ocrx_word' id='word_1_513' title='bbox 1294 2090 1313 2118; x_wconf 96'>it</span>
      <span class='ocrx_word' id='word_1_514' title='bbox 1323 2091 1459 2118; x_wconf 96'>interferes</span>
      <span class='ocrx_word' id='word_1_515' title='bbox 1470 2091 1535 2118; x_wconf 96'>with</span>
-      <span class='ocrx_word' id='word_1_516' title='bbox 1545 2091 1591 2118; x_wconf 97'>the</span>
+      <span class='ocrx_word' id='word_1_516' title='bbox 1545 2091 1591 2118; x_wconf 96'>the</span>
      <span class='ocrx_word' id='word_1_517' title='bbox 1602 2091 1715 2118; x_wconf 96'>creative</span>
      <span class='ocrx_word' id='word_1_518' title='bbox 1725 2099 1841 2124; x_wconf 93'>process.</span>
-      <span class='ocrx_word' id='word_1_519' title='bbox 1854 2091 1947 2118; x_wconf 91'>That’s</span>
+      <span class='ocrx_word' id='word_1_519' title='bbox 1854 2091 1947 2118; x_wconf 92'>That’s</span>
      <span class='ocrx_word' id='word_1_520' title='bbox 1957 2091 2086 2124; x_wconf 96'>precisely</span>
      <span class='ocrx_word' id='word_1_521' title='bbox 2096 2091 2157 2124; x_wconf 96'>why</span>
     </span>
     <span class='ocr_line' id='line_1_62' title="bbox 1293 2130 2156 2164; baseline 0 -7; x_size 34; x_descenders 7; x_ascenders 8">
      <span class='ocrx_word' id='word_1_522' title='bbox 1293 2130 1339 2157; x_wconf 93'>the</span>
      <span class='ocrx_word' id='word_1_523' title='bbox 1350 2130 1576 2164; x_wconf 90'>LinnSequencer</span>
-      <span class='ocrx_word' id='word_1_524' title='bbox 1586 2130 1607 2157; x_wconf 96'>is</span>
+      <span class='ocrx_word' id='word_1_524' title='bbox 1586 2130 1607 2157; x_wconf 97'>is</span>
      <span class='ocrx_word' id='word_1_525' title='bbox 1619 2130 1747 2164; x_wconf 96'>designed</span>
      <span class='ocrx_word' id='word_1_526' title='bbox 1758 2134 1787 2157; x_wconf 96'>to</span>
      <span class='ocrx_word' id='word_1_527' title='bbox 1798 2130 1834 2157; x_wconf 96'>let</span>
@@ -782,18 +782,18 @@
      <span class='ocrx_word' id='word_1_541' title='bbox 1402 2209 1451 2236; x_wconf 96'>See</span>
      <span class='ocrx_word' id='word_1_542' title='bbox 1460 2217 1529 2242; x_wconf 96'>your</span>
      <span class='ocrx_word' id='word_1_543' title='bbox 1540 2209 1611 2236; x_wconf 96'>Linn</span>
-      <span class='ocrx_word' id='word_1_544' title='bbox 1621 2209 1712 2236; x_wconf 96'>dealer</span>
-      <span class='ocrx_word' id='word_1_545' title='bbox 1722 2209 1806 2242; x_wconf 96'>today</span>
+      <span class='ocrx_word' id='word_1_544' title='bbox 1621 2209 1712 2236; x_wconf 95'>dealer</span>
+      <span class='ocrx_word' id='word_1_545' title='bbox 1722 2209 1806 2242; x_wconf 95'>today</span>
      <span class='ocrx_word' id='word_1_546' title='bbox 1817 2209 1859 2236; x_wconf 95'>for</span>
      <span class='ocrx_word' id='word_1_547' title='bbox 1870 2217 1887 2236; x_wconf 96'>a</span>
-      <span class='ocrx_word' id='word_1_548' title='bbox 1897 2209 2126 2236; x_wconf 95'>demonstration!</span>
+      <span class='ocrx_word' id='word_1_548' title='bbox 1897 2209 2126 2236; x_wconf 96'>demonstration!</span>
     </span>
    </p>
   </div>
   <div class='ocr_carea' id='block_1_21' title="bbox 347 2343 2171 2378">
    <p class='ocr_par' id='par_1_29' lang='eng' title="bbox 347 2343 2171 2378">
     <span class='ocr_header' id='line_1_65' title="bbox 347 2343 2171 2378; baseline 0.001 -8; x_size 34; x_descenders 7; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_549' title='bbox 347 2350 361 2363; x_wconf 43'>*</span>
+      <span class='ocrx_word' id='word_1_549' title='bbox 347 2350 361 2363; x_wconf 58'>*</span>
      <span class='ocrx_word' id='word_1_550' title='bbox 373 2343 483 2377; x_wconf 96'>Simple,</span>
      <span class='ocrx_word' id='word_1_551' title='bbox 495 2352 559 2377; x_wconf 96'>easy</span>
      <span class='ocrx_word' id='word_1_552' title='bbox 569 2348 598 2371; x_wconf 96'>to</span>
@@ -805,11 +805,11 @@
      <span class='ocrx_word' id='word_1_558' title='bbox 1211 2345 1316 2378; x_wconf 96'>display</span>
      <span class='ocrx_word' id='word_1_559' title='bbox 1326 2345 1424 2378; x_wconf 97'>clearly</span>
      <span class='ocrx_word' id='word_1_560' title='bbox 1434 2345 1528 2378; x_wconf 96'>guides</span>
-      <span class='ocrx_word' id='word_1_561' title='bbox 1539 2353 1594 2378; x_wconf 96'>you</span>
+      <span class='ocrx_word' id='word_1_561' title='bbox 1539 2353 1594 2378; x_wconf 97'>you</span>
      <span class='ocrx_word' id='word_1_562' title='bbox 1604 2345 1724 2378; x_wconf 96'>through</span>
      <span class='ocrx_word' id='word_1_563' title='bbox 1735 2344 1770 2371; x_wconf 96'>all</span>
      <span class='ocrx_word' id='word_1_564' title='bbox 1781 2344 1947 2377; x_wconf 96'>operations.</span>
-      <span class='ocrx_word' id='word_1_565' title='bbox 1961 2344 1989 2371; x_wconf 97'>If</span>
+      <span class='ocrx_word' id='word_1_565' title='bbox 1961 2344 1989 2371; x_wconf 96'>If</span>
      <span class='ocrx_word' id='word_1_566' title='bbox 1997 2344 2112 2376; x_wconf 96'>needed,</span>
      <span class='ocrx_word' id='word_1_567' title='bbox 2125 2345 2171 2371; x_wconf 96'>the</span>
     </span>
@@ -821,7 +821,7 @@
      <span class='ocrx_word' id='word_1_568' title='bbox 373 2381 472 2407; x_wconf 96'>HELP</span>
      <span class='ocrx_word' id='word_1_569' title='bbox 483 2381 583 2408; x_wconf 96'>button</span>
      <span class='ocrx_word' id='word_1_570' title='bbox 594 2381 711 2415; x_wconf 96'>displays</span>
-      <span class='ocrx_word' id='word_1_571' title='bbox 722 2382 875 2409; x_wconf 95'>additional</span>
+      <span class='ocrx_word' id='word_1_571' title='bbox 722 2382 875 2409; x_wconf 96'>additional</span>
      <span class='ocrx_word' id='word_1_572' title='bbox 886 2382 1083 2415; x_wconf 96'>explanations.</span>
     </span>
    </p>
@@ -829,20 +829,20 @@
   <div class='ocr_carea' id='block_1_23' title="bbox 347 2427 2145 2507">
    <p class='ocr_par' id='par_1_31' lang='eng' title="bbox 347 2427 2145 2507">
     <span class='ocr_header' id='line_1_67' title="bbox 347 2427 1468 2461; baseline 0.002 -8; x_size 34; x_descenders 7; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_573' title='bbox 347 2432 361 2446; x_wconf 77'>*</span>
+      <span class='ocrx_word' id='word_1_573' title='bbox 347 2432 361 2446; x_wconf 70'>*</span>
      <span class='ocrx_word' id='word_1_574' title='bbox 373 2427 612 2454; x_wconf 91'>Non-destructive</span>
-      <span class='ocrx_word' id='word_1_575' title='bbox 622 2427 914 2461; x_wconf 84'>recording—existing</span>
+      <span class='ocrx_word' id='word_1_575' title='bbox 622 2427 914 2461; x_wconf 89'>recording—existing</span>
      <span class='ocrx_word' id='word_1_576' title='bbox 924 2432 1002 2455; x_wconf 96'>notes</span>
      <span class='ocrx_word' id='word_1_577' title='bbox 1013 2436 1057 2455; x_wconf 96'>are</span>
      <span class='ocrx_word' id='word_1_578' title='bbox 1068 2432 1116 2455; x_wconf 96'>not</span>
      <span class='ocrx_word' id='word_1_579' title='bbox 1127 2428 1220 2455; x_wconf 96'>erased</span>
      <span class='ocrx_word' id='word_1_580' title='bbox 1231 2428 1309 2455; x_wconf 96'>while</span>
-      <span class='ocrx_word' id='word_1_581' title='bbox 1319 2428 1468 2461; x_wconf 94'>recording.</span>
+      <span class='ocrx_word' id='word_1_581' title='bbox 1319 2428 1468 2461; x_wconf 92'>recording.</span>
     </span>
     <span class='ocr_header' id='line_1_68' title="bbox 347 2473 2145 2507; baseline 0.001 -8; x_size 35; x_descenders 7; x_ascenders 9">
-      <span class='ocrx_word' id='word_1_582' title='bbox 347 2478 361 2492; x_wconf 40'>¢</span>
+      <span class='ocrx_word' id='word_1_582' title='bbox 347 2478 361 2492; x_wconf 70'>¢</span>
      <span class='ocrx_word' id='word_1_583' title='bbox 372 2473 433 2500; x_wconf 93'>Two</span>
-      <span class='ocrx_word' id='word_1_584' title='bbox 444 2473 689 2500; x_wconf 91'>FOOTSWITCH</span>
+      <span class='ocrx_word' id='word_1_584' title='bbox 444 2473 689 2500; x_wconf 90'>FOOTSWITCH</span>
      <span class='ocrx_word' id='word_1_585' title='bbox 701 2474 837 2501; x_wconf 95'>INPUTS</span>
      <span class='ocrx_word' id='word_1_586' title='bbox 848 2481 910 2507; x_wconf 96'>may</span>
      <span class='ocrx_word' id='word_1_587' title='bbox 921 2473 955 2500; x_wconf 96'>be</span>
@@ -864,8 +864,8 @@
    <p class='ocr_par' id='par_1_32' lang='eng' title="bbox 372 2510 1090 2543">
     <span class='ocr_line' id='line_1_69' title="bbox 372 2510 1090 2543; baseline 0.001 -6; x_size 35.625; x_descenders 8.90625; x_ascenders 8.90625">
      <span class='ocrx_word' id='word_1_599' title='bbox 372 2510 500 2542; x_wconf 96'>ERASE,</span>
-      <span class='ocrx_word' id='word_1_600' title='bbox 513 2511 660 2542; x_wconf 93'>REPEAT,</span>
-      <span class='ocrx_word' id='word_1_601' title='bbox 673 2511 883 2543; x_wconf 91'>PLAY/STOP,</span>
+      <span class='ocrx_word' id='word_1_600' title='bbox 513 2511 660 2542; x_wconf 92'>REPEAT,</span>
+      <span class='ocrx_word' id='word_1_601' title='bbox 673 2511 883 2543; x_wconf 89'>PLAY/STOP,</span>
      <span class='ocrx_word' id='word_1_602' title='bbox 896 2519 927 2538; x_wconf 96'>or</span>
      <span class='ocrx_word' id='word_1_603' title='bbox 939 2511 1090 2538; x_wconf 96'>LOCATE.</span>
     </span>
@@ -874,9 +874,9 @@
   <div class='ocr_carea' id='block_1_25' title="bbox 347 2556 1768 2590">
    <p class='ocr_par' id='par_1_33' lang='eng' title="bbox 347 2556 1768 2590">
     <span class='ocr_header' id='line_1_70' title="bbox 347 2556 1768 2590; baseline 0.001 -8; x_size 34; x_descenders 7; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_604' title='bbox 347 2561 361 2575; x_wconf 87'>¢</span>
-      <span class='ocrx_word' id='word_1_605' title='bbox 372 2556 433 2583; x_wconf 79'>Two</span>
-      <span class='ocrx_word' id='word_1_606' title='bbox 443 2556 612 2583; x_wconf 95'>TRIGGER</span>
+      <span class='ocrx_word' id='word_1_604' title='bbox 347 2561 361 2575; x_wconf 86'>¢</span>
+      <span class='ocrx_word' id='word_1_605' title='bbox 372 2556 433 2583; x_wconf 85'>Iwo</span>
+      <span class='ocrx_word' id='word_1_606' title='bbox 443 2556 612 2583; x_wconf 96'>TRIGGER</span>
      <span class='ocrx_word' id='word_1_607' title='bbox 623 2556 797 2584; x_wconf 96'>OUTPUTS</span>
      <span class='ocrx_word' id='word_1_608' title='bbox 808 2565 871 2590; x_wconf 96'>may</span>
      <span class='ocrx_word' id='word_1_609' title='bbox 881 2557 915 2584; x_wconf 96'>be</span>
@@ -887,7 +887,7 @@
      <span class='ocrx_word' id='word_1_614' title='bbox 1381 2561 1409 2584; x_wconf 96'>at</span>
      <span class='ocrx_word' id='word_1_615' title='bbox 1419 2565 1472 2590; x_wconf 96'>any</span>
      <span class='ocrx_word' id='word_1_616' title='bbox 1483 2557 1598 2584; x_wconf 96'>selected</span>
-      <span class='ocrx_word' id='word_1_617' title='bbox 1608 2561 1673 2584; x_wconf 96'>note</span>
+      <span class='ocrx_word' id='word_1_617' title='bbox 1608 2561 1673 2584; x_wconf 97'>note</span>
      <span class='ocrx_word' id='word_1_618' title='bbox 1683 2556 1768 2583; x_wconf 96'>value.</span>
     </span>
    </p>
@@ -895,12 +895,12 @@
   <div class='ocr_carea' id='block_1_26' title="bbox 347 2601 1226 2635">
    <p class='ocr_par' id='par_1_34' lang='eng' title="bbox 347 2601 1226 2635">
     <span class='ocr_line' id='line_1_71' title="bbox 347 2601 1226 2635; baseline 0.001 -7; x_size 33; x_descenders 6; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_619' title='bbox 347 2607 361 2620; x_wconf 50'>©</span>
+      <span class='ocrx_word' id='word_1_619' title='bbox 347 2607 361 2620; x_wconf 58'>©</span>
      <span class='ocrx_word' id='word_1_620' title='bbox 372 2601 434 2628; x_wconf 96'>Will</span>
-      <span class='ocrx_word' id='word_1_621' title='bbox 445 2609 510 2634; x_wconf 95'>sync</span>
+      <span class='ocrx_word' id='word_1_621' title='bbox 445 2609 510 2634; x_wconf 96'>sync</span>
      <span class='ocrx_word' id='word_1_622' title='bbox 521 2605 549 2628; x_wconf 96'>to</span>
      <span class='ocrx_word' id='word_1_623' title='bbox 561 2602 690 2629; x_wconf 93'>standard</span>
-      <span class='ocrx_word' id='word_1_624' title='bbox 701 2602 864 2629; x_wconf 92'>LinnDrum</span>
+      <span class='ocrx_word' id='word_1_624' title='bbox 701 2602 864 2629; x_wconf 91'>LinnDrum</span>
      <span class='ocrx_word' id='word_1_625' title='bbox 875 2610 907 2629; x_wconf 96'>or</span>
      <span class='ocrx_word' id='word_1_626' title='bbox 918 2602 989 2629; x_wconf 96'>Linn</span>
      <span class='ocrx_word' id='word_1_627' title='bbox 1000 2603 1069 2629; x_wconf 95'>9000</span>
@@ -912,13 +912,13 @@
   <div class='ocr_carea' id='block_1_27' title="bbox 347 2648 2100 2727">
    <p class='ocr_par' id='par_1_35' lang='eng' title="bbox 347 2648 2100 2727">
     <span class='ocr_header' id='line_1_72' title="bbox 347 2648 1664 2682; baseline 0.001 -7; x_size 33; x_descenders 6; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_630' title='bbox 347 2654 360 2667; x_wconf 47'>®</span>
-      <span class='ocrx_word' id='word_1_631' title='bbox 372 2648 483 2675; x_wconf 96'>Utilizes</span>
+      <span class='ocrx_word' id='word_1_630' title='bbox 347 2654 360 2667; x_wconf 45'>©</span>
+      <span class='ocrx_word' id='word_1_631' title='bbox 372 2648 483 2675; x_wconf 95'>Utilizes</span>
      <span class='ocrx_word' id='word_1_632' title='bbox 493 2648 564 2680; x_wconf 96'>ultra</span>
      <span class='ocrx_word' id='word_1_633' title='bbox 573 2648 744 2682; x_wconf 96'>high-speed,</span>
-      <span class='ocrx_word' id='word_1_634' title='bbox 766 2649 772 2676; x_wconf 95'>8</span>
+      <span class='ocrx_word' id='word_1_634' title='bbox 757 2649 772 2676; x_wconf 95'>8</span>
      <span class='ocrx_word' id='word_1_635' title='bbox 783 2649 862 2675; x_wconf 94'>MHz</span>
-      <span class='ocrx_word' id='word_1_636' title='bbox 873 2649 954 2676; x_wconf 95'>80186</span>
+      <span class='ocrx_word' id='word_1_636' title='bbox 873 2649 954 2676; x_wconf 96'>80186</span>
      <span class='ocrx_word' id='word_1_637' title='bbox 965 2649 994 2676; x_wconf 96'>16</span>
      <span class='ocrx_word' id='word_1_638' title='bbox 1004 2648 1043 2676; x_wconf 96'>bit</span>
      <span class='ocrx_word' id='word_1_639' title='bbox 1054 2653 1197 2682; x_wconf 96'>computer</span>
@@ -928,15 +928,15 @@
      <span class='ocrx_word' id='word_1_643' title='bbox 1512 2648 1664 2682; x_wconf 96'>operation.</span>
     </span>
     <span class='ocr_header' id='line_1_73' title="bbox 347 2694 2100 2727; baseline 0.001 -7; x_size 33; x_descenders 6; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_644' title='bbox 347 2699 361 2713; x_wconf 52'>*</span>
+      <span class='ocrx_word' id='word_1_644' title='bbox 347 2699 361 2713; x_wconf 31'>*</span>
      <span class='ocrx_word' id='word_1_645' title='bbox 372 2694 504 2721; x_wconf 96'>TEMPO</span>
      <span class='ocrx_word' id='word_1_646' title='bbox 515 2702 578 2727; x_wconf 96'>may</span>
-      <span class='ocrx_word' id='word_1_647' title='bbox 589 2694 623 2721; x_wconf 96'>be</span>
+      <span class='ocrx_word' id='word_1_647' title='bbox 589 2694 623 2721; x_wconf 95'>be</span>
      <span class='ocrx_word' id='word_1_648' title='bbox 633 2694 764 2727; x_wconf 95'>specified</span>
-      <span class='ocrx_word' id='word_1_649' title='bbox 774 2694 802 2721; x_wconf 92'>in</span>
+      <span class='ocrx_word' id='word_1_649' title='bbox 774 2694 802 2721; x_wconf 93'>in</span>
      <span class='ocrx_word' id='word_1_650' title='bbox 814 2695 1172 2722; x_wconf 91'>BEATS-PER-MINUTE</span>
      <span class='ocrx_word' id='word_1_651' title='bbox 1183 2703 1215 2722; x_wconf 93'>or</span>
-      <span class='ocrx_word' id='word_1_652' title='bbox 1225 2695 1567 2722; x_wconf 92'>FRAMES-PER-BEAT</span>
+      <span class='ocrx_word' id='word_1_652' title='bbox 1225 2695 1567 2722; x_wconf 91'>FRAMES-PER-BEAT</span>
      <span class='ocrx_word' id='word_1_653' title='bbox 1577 2698 1605 2721; x_wconf 96'>at</span>
      <span class='ocrx_word' id='word_1_654' title='bbox 1616 2695 1659 2726; x_wconf 96'>24,</span>
      <span class='ocrx_word' id='word_1_655' title='bbox 1672 2695 1716 2726; x_wconf 96'>25,</span>
@@ -960,18 +960,18 @@
   <div class='ocr_carea' id='block_1_29' title="bbox 347 2777 2174 2811">
    <p class='ocr_par' id='par_1_37' lang='eng' title="bbox 347 2777 2174 2811">
     <span class='ocr_header' id='line_1_75' title="bbox 347 2777 2174 2811; baseline 0.001 -8; x_size 33; x_descenders 5; x_ascenders 9">
-      <span class='ocrx_word' id='word_1_664' title='bbox 347 2782 360 2796; x_wconf 78'>¢</span>
-      <span class='ocrx_word' id='word_1_665' title='bbox 372 2777 504 2804; x_wconf 95'>TEMPO</span>
+      <span class='ocrx_word' id='word_1_664' title='bbox 347 2782 360 2796; x_wconf 81'>¢</span>
+      <span class='ocrx_word' id='word_1_665' title='bbox 372 2777 504 2804; x_wconf 94'>TEMPO</span>
      <span class='ocrx_word' id='word_1_666' title='bbox 515 2785 578 2810; x_wconf 96'>may</span>
-      <span class='ocrx_word' id='word_1_667' title='bbox 588 2777 622 2804; x_wconf 95'>be</span>
+      <span class='ocrx_word' id='word_1_667' title='bbox 588 2777 622 2804; x_wconf 96'>be</span>
      <span class='ocrx_word' id='word_1_668' title='bbox 633 2778 741 2804; x_wconf 95'>entered</span>
      <span class='ocrx_word' id='word_1_669' title='bbox 751 2777 934 2811; x_wconf 96'>numerically,</span>
-      <span class='ocrx_word' id='word_1_670' title='bbox 946 2777 1101 2811; x_wconf 96'>adjustable</span>
+      <span class='ocrx_word' id='word_1_670' title='bbox 946 2777 1101 2811; x_wconf 95'>adjustable</span>
      <span class='ocrx_word' id='word_1_671' title='bbox 1111 2777 1139 2804; x_wconf 95'>in</span>
-      <span class='ocrx_word' id='word_1_672' title='bbox 1149 2778 1239 2805; x_wconf 95'>tenths</span>
+      <span class='ocrx_word' id='word_1_672' title='bbox 1149 2778 1239 2805; x_wconf 96'>tenths</span>
      <span class='ocrx_word' id='word_1_673' title='bbox 1250 2778 1282 2805; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_674' title='bbox 1290 2786 1307 2805; x_wconf 93'>a</span>
-      <span class='ocrx_word' id='word_1_675' title='bbox 1317 2777 1567 2805; x_wconf 92'>Beat-Per-Minute</span>
+      <span class='ocrx_word' id='word_1_674' title='bbox 1290 2786 1307 2805; x_wconf 91'>a</span>
+      <span class='ocrx_word' id='word_1_675' title='bbox 1317 2777 1567 2805; x_wconf 91'>Beat-Per-Minute</span>
      <span class='ocrx_word' id='word_1_676' title='bbox 1577 2777 1748 2809; x_wconf 96'>increments,</span>
      <span class='ocrx_word' id='word_1_677' title='bbox 1760 2785 1792 2804; x_wconf 96'>or</span>
      <span class='ocrx_word' id='word_1_678' title='bbox 1803 2777 1839 2810; x_wconf 96'>by</span>
@@ -995,23 +995,23 @@
   <div class='ocr_carea' id='block_1_31' title="bbox 347 2861 1792 2940">
    <p class='ocr_par' id='par_1_39' lang='eng' title="bbox 347 2861 1792 2940">
     <span class='ocr_header' id='line_1_77' title="bbox 347 2861 1792 2895; baseline 0.001 -8; x_size 33; x_descenders 6; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_687' title='bbox 347 2866 360 2880; x_wconf 62'>¢</span>
+      <span class='ocrx_word' id='word_1_687' title='bbox 347 2866 360 2880; x_wconf 43'>¢</span>
      <span class='ocrx_word' id='word_1_688' title='bbox 372 2861 504 2887; x_wconf 96'>TEMPO</span>
      <span class='ocrx_word' id='word_1_689' title='bbox 515 2861 696 2888; x_wconf 96'>CHANGES</span>
      <span class='ocrx_word' id='word_1_690' title='bbox 707 2869 771 2894; x_wconf 96'>may</span>
      <span class='ocrx_word' id='word_1_691' title='bbox 781 2861 815 2888; x_wconf 96'>be</span>
-      <span class='ocrx_word' id='word_1_692' title='bbox 825 2861 1019 2894; x_wconf 95'>programmed</span>
+      <span class='ocrx_word' id='word_1_692' title='bbox 825 2861 1019 2894; x_wconf 96'>programmed</span>
      <span class='ocrx_word' id='word_1_693' title='bbox 1030 2861 1087 2888; x_wconf 96'>into</span>
      <span class='ocrx_word' id='word_1_694' title='bbox 1099 2869 1115 2888; x_wconf 95'>a</span>
      <span class='ocrx_word' id='word_1_695' title='bbox 1126 2870 1268 2895; x_wconf 96'>sequence,</span>
      <span class='ocrx_word' id='word_1_696' title='bbox 1280 2861 1344 2888; x_wconf 96'>with</span>
-      <span class='ocrx_word' id='word_1_697' title='bbox 1356 2862 1467 2888; x_wconf 96'>smooth</span>
+      <span class='ocrx_word' id='word_1_697' title='bbox 1356 2862 1467 2888; x_wconf 95'>smooth</span>
      <span class='ocrx_word' id='word_1_698' title='bbox 1478 2861 1635 2888; x_wconf 96'>transitions</span>
      <span class='ocrx_word' id='word_1_699' title='bbox 1646 2861 1670 2887; x_wconf 96'>if</span>
-      <span class='ocrx_word' id='word_1_700' title='bbox 1679 2861 1792 2888; x_wconf 84'>desired.</span>
+      <span class='ocrx_word' id='word_1_700' title='bbox 1679 2861 1792 2888; x_wconf 87'>desired.</span>
     </span>
     <span class='ocr_header' id='line_1_78' title="bbox 347 2906 1507 2940; baseline 0.002 -8; x_size 33; x_descenders 7; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_701' title='bbox 347 2911 360 2925; x_wconf 76'>¢</span>
+      <span class='ocrx_word' id='word_1_701' title='bbox 347 2911 360 2925; x_wconf 69'>¢</span>
      <span class='ocrx_word' id='word_1_702' title='bbox 371 2906 434 2938; x_wconf 96'>Any</span>
      <span class='ocrx_word' id='word_1_703' title='bbox 444 2906 539 2932; x_wconf 96'>TIME</span>
      <span class='ocrx_word' id='word_1_704' title='bbox 550 2906 763 2933; x_wconf 96'>SIGNATURE</span>
@@ -1022,8 +1022,8 @@
      <span class='ocrx_word' id='word_1_709' title='bbox 1046 2915 1109 2940; x_wconf 96'>may</span>
      <span class='ocrx_word' id='word_1_710' title='bbox 1120 2907 1154 2934; x_wconf 96'>be</span>
      <span class='ocrx_word' id='word_1_711' title='bbox 1164 2907 1288 2940; x_wconf 96'>changed</span>
-      <span class='ocrx_word' id='word_1_712' title='bbox 1299 2907 1393 2934; x_wconf 96'>within</span>
-      <span class='ocrx_word' id='word_1_713' title='bbox 1404 2915 1420 2934; x_wconf 96'>a</span>
+      <span class='ocrx_word' id='word_1_712' title='bbox 1299 2907 1393 2934; x_wconf 95'>within</span>
+      <span class='ocrx_word' id='word_1_713' title='bbox 1404 2915 1420 2934; x_wconf 95'>a</span>
      <span class='ocrx_word' id='word_1_714' title='bbox 1431 2915 1507 2940; x_wconf 96'>song.</span>
     </span>
    </p>
@@ -1047,15 +1047,15 @@
      <span class='ocrx_word' id='word_1_720' title='bbox 1648 3150 1761 3177; x_wconf 96'>Oxnard</span>
      <span class='ocrx_word' id='word_1_721' title='bbox 1772 3150 1866 3182; x_wconf 96'>Street,</span>
      <span class='ocrx_word' id='word_1_722' title='bbox 1878 3150 2006 3182; x_wconf 96'>Tarzana,</span>
-      <span class='ocrx_word' id='word_1_723' title='bbox 2019 3150 2071 3177; x_wconf 96'>CA</span>
+      <span class='ocrx_word' id='word_1_723' title='bbox 2019 3150 2071 3177; x_wconf 95'>CA</span>
      <span class='ocrx_word' id='word_1_724' title='bbox 2082 3150 2163 3177; x_wconf 96'>91356</span>
     </span>
     <span class='ocr_line' id='line_1_82' title="bbox 1554 3192 2188 3226; baseline 0 -7; x_size 33.5; x_descenders 5.5; x_ascenders 8.5">
      <span class='ocrx_word' id='word_1_725' title='bbox 1554 3193 1622 3226; x_wconf 96'>(818)</span>
-      <span class='ocrx_word' id='word_1_726' title='bbox 1633 3193 1755 3219; x_wconf 96'>708-8131</span>
-      <span class='ocrx_word' id='word_1_727' title='bbox 1789 3193 1888 3219; x_wconf 95'>TELEX</span>
-      <span class='ocrx_word' id='word_1_728' title='bbox 1899 3192 2022 3219; x_wconf 96'>#298949</span>
-      <span class='ocrx_word' id='word_1_729' title='bbox 2033 3193 2125 3219; x_wconf 96'>LINN</span>
+      <span class='ocrx_word' id='word_1_726' title='bbox 1633 3193 1755 3219; x_wconf 95'>708-8131</span>
+      <span class='ocrx_word' id='word_1_727' title='bbox 1765 3193 1888 3219; x_wconf 95'>TELEX</span>
+      <span class='ocrx_word' id='word_1_728' title='bbox 1899 3192 2022 3219; x_wconf 95'>#298949</span>
+      <span class='ocrx_word' id='word_1_729' title='bbox 2033 3193 2125 3219; x_wconf 95'>LINN</span>
      <span class='ocrx_word' id='word_1_730' title='bbox 2135 3193 2188 3219; x_wconf 96'>UR</span>
     </span>
    </p>
@@ -1 +1 @@
-Tesseract Open Source OCR Engine v5.0.0-beta-20210916-12-g19cc9 with Leptonica
+Tesseract Open Source OCR Engine v4.1.1 with Leptonica
@@ -8,18 +8,18 @@ extremely powerful, yet amazingly simple to learn and use. It’s many remarkabl
 ¢ Operation is similar to multi-track tape recorder with PLAY, STOP, RECORD, FAST
 FORWARD, REWIND, and LOCATE controls.

-¢ Each of the 100 sequences contains 32 simultaneous, polyphonic tracks. Each track may
+e Each of the 100 sequences contains 32 simultaneous, polyphonic tracks. Each track may
 be assigned to one of 16 MIDI channels. Simultaneously plays up to 16 polyphonic

 synthesizers!

-© Ultra-fast 32” disk drive stores complex songs in seconds and holds over 110,000 notes
+¢ Ultra-fast 3%” disk drive stores complex songs in seconds and holds over 110,000 notes

 per disk!

 ¢ One or all tracks may be TRANSPOSED at the touch of a key.
-¢ Exclusive real-time ERASE function makes editing FAST.
-© Exclusive REPEAT function automatically repeats any held notes at a pre-selected
+e Exclusive real-time ERASE function makes editing FAST.
+* Exclusive REPEAT function automatically repeats any held notes at a pre-selected

 rhythmic value.

@@ -99,11 +99,11 @@ HELP button displays additional explanations.

 ERASE, REPEAT, PLAY/STOP, or LOCATE.

-¢ Two TRIGGER OUTPUTS may be programmed to output pulses at any selected note value.
+¢ Iwo TRIGGER OUTPUTS may be programmed to output pulses at any selected note value.

 © Will sync to standard LinnDrum or Linn 9000 sync tone.

-® Utilizes ultra high-speed, 8 MHz 80186 16 bit computer internally for FAST operation.
+© Utilizes ultra high-speed, 8 MHz 80186 16 bit computer internally for FAST operation.
 * TEMPO may be specified in BEATS-PER-MINUTE or FRAMES-PER-BEAT at 24, 25, or 30 frames per second,

 (even drop frame!)
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
James R. Barlow	d0d2948f9a	Merge branch 'master' into scriptsender	2021-02-18 01:48:00 -08:00
James R. Barlow	cac64b5ed2	Update pyproj	2021-02-18 01:08:49 -08:00