Redo dpi calc to avoid 'math.ulp'

This reverts commit d89a633ba7.

.coveragerc

@@ -1,24 +0,0 @@
-[paths]
-source =
-	src
-	*/site-packages
-
-[run]
-branch = true
-parallel = true
-concurrency =
-	thread
-	multiprocessing
-source =
-	src/ocrmypdf
-
-[report]
-exclude_lines =
-	pragma: no cover
-	def __repr__
-	raise AssertionError
-	raise NotImplementedError
-	if 0:
-	if False:
-	if __name__ == .__main__.:
-	if TYPE_CHECKING:

.docker/Dockerfile

@@ -10,8 +10,10 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
   build-essential autoconf automake libtool \
   libleptonica-dev \
   zlib1g-dev \
-  python3 \
+  python3-dev \
   python3-distutils \
+  libffi-dev \
+  libqpdf-dev \
   ca-certificates \
   curl \
   git
@@ -62,7 +64,8 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
   tesseract-ocr-fra \
   tesseract-ocr-por \
   tesseract-ocr-spa \
-  unpaper
+  unpaper \
+  && rm -rf /var/lib/apt/lists/*
 
 WORKDIR /app
 

.github/ISSUE_TEMPLATE/1-general-issues.md

@@ -21,8 +21,12 @@ If applicable, add screenshots to help explain your problem.
 
 **System (please complete the following information):**
  - OS:
- - Python version: 
+ - Python version:
  - OCRmyPDF version:
 
+**Installation**
+How did you install OCRmyPDF? Did you install it from your operating system's
+package manager, or using pip?
+
 **Additional context**
 Add any other context about the problem here.

.github/workflows/build.yml

@@ -0,0 +1,275 @@
+name: Test and deploy
+
+on:
+  push:
+    branches:
+      - master
+      - ci
+      - release/*
+    tags:
+      - v*
+    paths-ignore:
+      - README*
+  pull_request:
+
+jobs:
+  test_linux:
+    name: Test ${{ matrix.os }} with Python ${{ matrix.python }}
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os: [ubuntu-18.04] #, ubuntu-20.04]
+        python: ["3.6"] #, "3.7", "3.8", "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 common packages
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y --no-install-recommends \
+            curl \
+            ghostscript \
+            img2pdf \
+            libffi-dev \
+            liblept5 \
+            libsm6 libxext6 libxrender-dev \
+            pngquant \
+            poppler-utils \
+            tesseract-ocr \
+            tesseract-ocr-deu \
+            tesseract-ocr-eng \
+            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'
+        run: |
+          sudo apt-get install -y --no-install-recommends \
+            libexempi8
+
+      - name: Install Python packages
+        run: |
+          python -m pip install -r requirements/main.txt -r requirements/test.txt .
+
+      - 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 \
+            leptonica \
+            openjpeg \
+            pngquant \
+            tesseract
+
+      - name: Install Python packages
+        run: |
+          python -m pip install --upgrade pip
+          python -m pip install -r requirements/main.txt -r requirements/test.txt .
+
+      - 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 ghostscript
+          choco install --yes --no-progress pngquant
+
+      - name: Install Python packages
+        run: |
+          python -m pip install --upgrade pip
+          python -m pip install -r requirements/main.txt -r requirements/test.txt .
+
+      - 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: "3.6"
+
+      - 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: ${{ secrets.TOKEN_PYPI }}
+          # repository_url: https://test.pypi.org/legacy/
+
+  docker:
+    name: Build Docker images
+    needs: [wheel_sdist_linux, test_linux, test_macos, test_windows]
+    runs-on: ubuntu-latest
+    steps:
+      - name: Set image tag to release or branch
+        run: echo "DOCKER_IMAGE_TAG=${GITHUB_REF##*/}" >> $GITHUB_ENV
+
+      - name: If master, set to latest
+        run: echo 'DOCKER_IMAGE_TAG=latest' >> $GITHUB_ENV
+        if: env.DOCKER_IMAGE_TAG == 'master'
+
+      - name: Set Docker Hub repository to username
+        run: echo "DOCKER_REPOSITORY=jbarlow83" >> $GITHUB_ENV
+
+      - name: Set image name
+        run: echo "DOCKER_IMAGE_NAME=ocrmypdf" >> $GITHUB_ENV
+
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: "0" # 0=all, needed for setuptools-scm to resolve version tags
+
+      - name: Login to Docker Hub
+        uses: docker/login-action@v1
+        with:
+          username: jbarlow83
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
+
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v1
+
+      - name: Set up Docker Buildx
+        id: buildx
+        uses: docker/setup-buildx-action@v1
+
+      - name: Print image tag
+        run: echo "Building image ${DOCKER_REPOSITORY}/${DOCKER_IMAGE_NAME}:${DOCKER_IMAGE_TAG}"
+
+      - name: Build
+        run: |
+          docker buildx build \
+            --push \
+            --platform linux/arm64/v8,linux/amd64  \
+            --tag "${DOCKER_REPOSITORY}/${DOCKER_IMAGE_NAME}:${DOCKER_IMAGE_TAG}" \
+            --file .docker/Dockerfile .

.gitignore

@@ -24,6 +24,7 @@ venv*/
 /debug_tests.py
 *.traineddata
 /private
+/coverage.xml
 
 # Package building
 *.egg-info/

.pre-commit-config.yaml

@@ -1,6 +1,6 @@
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v3.1.0
+    rev: v3.4.0
     hooks:
       - id: check-case-conflict
       - id: check-merge-conflict
@@ -12,12 +12,12 @@ repos:
     hooks:
       - id: seed-isort-config
   - repo: https://github.com/pre-commit/mirrors-isort
-    rev: v5.0.5 # pick the isort version you'd like to use from https://github.com/pre-commit/mirrors-isort/releases
+    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
   - repo: https://github.com/psf/black
-    rev: 19.10b0
+    rev: 20.8b1
     hooks:
       - id: black
-        language_version: python3.8
+        language_version: python
         exclude: ^src/ocrmypdf/lib/_leptonica.py

README.md

@@ -1,6 +1,6 @@
 <img src="docs/images/logo.svg" width="240" alt="OCRmyPDF">
 
-[![Build Status][azure]](https://dev.azure.com/jim0585/ocrmypdf/_build/latest?definitionId=2&branchName=master) [![PyPI version][pypi]](https://pypi.org/project/ocrmypdf/) ![Homebrew version][homebrew] ![ReadTheDocs][docs] ![Python versions][pyversions]
+[![Build Status](https://github.com/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]
 
 [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"
@@ -59,23 +59,15 @@ I searched the web for a free command line tool to OCR PDF files: I found many,
 
 Linux, Windows, macOS and FreeBSD are supported. Docker images are also available.
 
-Users of Debian 9 or later or Ubuntu 16.10 or later may simply
-
-```bash
-apt-get install ocrmypdf
-```
-
-and users of Fedora 29 or later may simply
-
-```bash
-dnf install ocrmypdf
-```
-
-and Homebrew users (macOS, Linux, Windows Subsystem for Linux) may simply
-
-```bash
-brew install ocrmypdf
-```
+| Operating system              | Install command               |
+| ----------------------------- | ------------------------------|
+| Debian, Ubuntu                | ``apt install ocrmypdf``      |
+| Windows Subsystem for Linux   | ``apt install ocrmypdf``      |
+| Fedora                        | ``dnf install ocrmypdf``      |
+| macOS                         | ``brew install ocrmypdf``     |
+| LinuxBrew                     | ``brew install ocrmypdf``     |
+| FreeBSD                       | ``pkg install py37-ocrmypdf`` |
+| Conda                         | ``conda install ocrmypdf``    |
 
 For everyone else, [see our documentation](https://ocrmypdf.readthedocs.io/en/latest/installation.html) for installation steps.
 
@@ -92,6 +84,9 @@ apt-get install tesseract-ocr-chi-sim  # Example: Install Chinese Simplified lan
 
 # Arch Linux users
 pacman -S tesseract-data-eng tesseract-data-deu # Example: Install the English and German language packs
+
+# brew macOS users
+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.

azure-pipelines.yml

@@ -1,249 +0,0 @@
-trigger:
-  tags:
-    include:
-      - v*
-  branches:
-    include:
-      - "*"
-    exclude:
-      - "travis"
-
-stages:
-  - stage: "Test"
-    jobs:
-      - job: Windows
-        pool:
-          vmImage: "vs2017-win2016"
-        strategy:
-          matrix:
-            Python36:
-              python.version: "3.6"
-            Python37:
-              python.version: "3.7"
-            Python38:
-              python.version: "3.8"
-        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 $(python.version)"
-            condition: succeededOrFailed()
-      - job: "Ubuntu_1804"
-        pool:
-          vmImage: "ubuntu-18.04"
-        strategy:
-          matrix:
-            Python36:
-              python.version: "3.6"
-            Python37:
-              python.version: "3.7"
-            Python38:
-              python.version: "3.8"
-        steps:
-          - task: UsePythonVersion@0
-            inputs:
-              versionSpec: "$(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 unlink python@2
-              brew upgrade python
-              echo "Using Python `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')))

docs/advanced.rst

@@ -125,8 +125,7 @@ include:
 .. envvar:: OMP_THREAD_LIMIT
 
    Controls the number of threads Tesseract will use. OCRmyPDF will
-   manage this environment if it is not already set. (Currently, it will
-   set it to 1 because this gives the best results in testing.)
+   manage this environment variable if it is not already set.
 
 For example, if you have a development build of Tesseract don't wish to
 use the system installation, you can launch OCRmyPDF as follows:
@@ -315,7 +314,7 @@ message is:
 .. code-block:: none
 
     Temporary working files retained at:
-    /tmp/com.github.ocrmypdf.u20wpz07
+    /tmp/ocrmypdf.io.u20wpz07
 
 The organization of this folder is an implementation detail and subject
 to change between releases. However the general organization is that

docs/api.rst

@@ -20,7 +20,8 @@ and largely have the same functions.
 
     import ocrmypdf
 
-    ocrmypdf.ocr('input.pdf', 'output.pdf', deskew=True)
+    if __name__ == '__main__':  # To ensure correct behavior on Windows and macOS
+        ocrmypdf.ocr('input.pdf', 'output.pdf', deskew=True)
 
 With a few exceptions, all of the command line arguments are available
 and may be passed as equivalent keywords.
@@ -35,33 +36,37 @@ The :func:`ocrmypdf.ocr` function runs OCRmyPDF similar to command line
 execution. To do this, it will:
 
 - create a monitoring thread
-- create worker processes (forking itself)
-- manage the signal flags of worker processes
+- create worker processes (on Linux, forking itself; on Windows and macOS, by
+  spawning)
+- manage the signal flags of its worker processes
 - execute other subprocesses (forking and executing other programs)
 
 The Python process that calls ``ocrmypdf.ocr()`` must be sufficiently
-privileged to perform these actions. If it is not, ``ocrmypdf()`` will
-fail.
+privileged to perform these actions.
 
 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.
 
-Forking a child process to call ``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 does not.
+OCRmyPDF fails for any reason.
 
 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.
 
+``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, the script that calls ``ocrmypdf.ocr()`` must be protected
-    by an "ifmain" guard (``if __name__ == '__main__'``) or you must use
-    ``ocrmypdf.ocr(...use_threads=True)``. If you do not take at least one
-    of these steps, Windows process semantics will prevent OCRmyPDF from working
-    correctly.
+    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.
 
 Logging
 -------

docs/apiref.rst

@@ -5,6 +5,15 @@ API Reference
 This page summarizes the rest of the public API. Generally speaking this
 should mainly of interest to plugin developers.
 
+ocrmypdf
+========
+
+.. autoclass:: ocrmypdf.PageContext
+    :members:
+
+.. autoclass:: ocrmypdf.PdfContext
+    :members:
+
 ocrmypdf.exceptions
 ===================
 
@@ -17,6 +26,9 @@ ocrmypdf.helpers
 
 .. automodule:: ocrmypdf.helpers
     :members:
+    :noindex: deprecated
+
+    .. autodecorator:: deprecated
 
 ocrmypdf.hocrtransform
 ======================

docs/batch.rst

@@ -127,7 +127,7 @@ Users may need to customize the script to meet their requirements.
     "OCR_ON_SUCCESS_DELETE", "This will delete the input file if the exit code is 0 (OK)"
     "OCR_OUTPUT_DIRECTORY_YEAR_MONTH", "This will place files in the output in ``{output}/{year}/{month}/{filename}``"
     "OCR_DESKEW", "Apply deskew to crooked input PDFs"
-    "OCR_JSON_SETTINGS", "A JSON string specifying any other arguments for ``ocrmypdf.ocr``, e.g. ``'OCR_JSON_SETTINGS={"rotate_pages": true}'``.
+    "OCR_JSON_SETTINGS", "A JSON string specifying any other arguments for ``ocrmypdf.ocr``, e.g. ``'OCR_JSON_SETTINGS={""rotate_pages"": true}'``."
     "OCR_POLL_NEW_FILE_SECONDS", "Polling interval"
     "OCR_LOGLEVEL", "Level of log messages to report"
 
@@ -202,13 +202,6 @@ Alternatives
 -  `Watchman <https://facebook.github.io/watchman/>`__ is a more
    powerful alternative to ``watchmedo``.
 
-AWS Lambda is not viable
-------------------------
-
-AWS Lambda and its equivalents have low limits on execution time and payload
-size, relative to OCRmyPDF's needs. As of this writing, the request/response
-payload for AWS Lambda was 6 MB, which means many PDFs will not fit.
-
 macOS Automator
 ===============
 

docs/docker.rst

@@ -1,3 +1,5 @@
+.. _docker:
+
 =====================
 OCRmyPDF Docker image
 =====================

docs/index.rst

@@ -1,7 +1,7 @@
 OCRmyPDF documentation
 ======================
 
-OCRmyPDF adds an optical charcter recognition (OCR) text layer to scanned PDF
+OCRmyPDF adds an optical character recognition (OCR) text layer to scanned PDF
 files, allowing them to be searched.
 
 PDF is the best format for storing and exchanging scanned documents.

docs/installation.rst

@@ -12,19 +12,21 @@ system/platform. This version may be out of date, however.
 
 These platforms have one-liner installs:
 
-+-----------------------------+-------------------------------+
-| Debian, Ubuntu              | ``apt install ocrmypdf``      |
-+-----------------------------+-------------------------------+
-| Windows Subsystem for Linux | ``apt install ocrmypdf``      |
-+-----------------------------+-------------------------------+
-| Fedora                      | ``dnf install ocrmypdf``      |
-+-----------------------------+-------------------------------+
-| macOS                       | ``brew install ocrmypdf``     |
-+-----------------------------+-------------------------------+
-| LinuxBrew                   | ``brew install ocrmypdf``     |
-+-----------------------------+-------------------------------+
-| FreeBSD                     | ``pkg install py37-ocrmypdf`` |
-+-----------------------------+-------------------------------+
++-------------------------------+-------------------------------+
+| Debian, Ubuntu                | ``apt install ocrmypdf``      |
++-------------------------------+-------------------------------+
+| Windows Subsystem for Linux   | ``apt install ocrmypdf``      |
++-------------------------------+-------------------------------+
+| Fedora                        | ``dnf install ocrmypdf``      |
++-------------------------------+-------------------------------+
+| macOS                         | ``brew install ocrmypdf``     |
++-------------------------------+-------------------------------+
+| LinuxBrew                     | ``brew install ocrmypdf``     |
++-------------------------------+-------------------------------+
+| FreeBSD                       | ``pkg install py37-ocrmypdf`` |
++-------------------------------+-------------------------------+
+| Conda (WSL, macOS, Linux)     | ``conda install ocrmypdf``    |
++-------------------------------+-------------------------------+
 
 More detailed procedures are outlined below. If you want to do a manual
 install, or install a more recent version than your platform provides, read on.
@@ -54,6 +56,9 @@ Debian and Ubuntu 18.04 or newer
 .. |ubu-2004| image:: https://repology.org/badge/version-for-repo/ubuntu_20_04/ocrmypdf.svg
     :alt: Ubuntu 20.04 LTS
 
+.. |ubu-2010| image:: https://repology.org/badge/version-for-repo/ubuntu_20_10/ocrmypdf.svg
+    :alt: Ubuntu 20.10
+
 +-----------------------------------------------+
 | **OCRmyPDF versions in Debian & Ubuntu**      |
 +-----------------------------------------------+
@@ -61,7 +66,7 @@ Debian and Ubuntu 18.04 or newer
 +-----------------------------------------------+
 | |deb-stable| |deb-testing| |deb-unstable|     |
 +-----------------------------------------------+
-| |ubu-1804| |ubu-2004|                         |
+| |ubu-1804| |ubu-2004| |ubu-2010|              |
 +-----------------------------------------------+
 
 Users of Debian 9 ("stretch") or later, or Ubuntu 18.04 or later, including users
@@ -90,15 +95,15 @@ 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 29 or newer
-------------------
-
-.. |fedora-31| image:: https://repology.org/badge/version-for-repo/fedora_31/ocrmypdf.svg
-    :alt: Fedora 31
+Fedora
+------
 
 .. |fedora-32| image:: https://repology.org/badge/version-for-repo/fedora_32/ocrmypdf.svg
     :alt: Fedora 32
 
+.. |fedora-33| image:: https://repology.org/badge/version-for-repo/fedora_33/ocrmypdf.svg
+    :alt: Fedora 33
+
 .. |fedora-rawhide| image:: https://repology.org/badge/version-for-repo/fedora_rawhide/ocrmypdf.svg
     :alt: Fedore Rawhide
 
@@ -107,7 +112,7 @@ Fedora 29 or newer
 +-----------------------------------------------+
 | |latest|                                      |
 +-----------------------------------------------+
-| |fedora-31| |fedora-32| |fedora-rawhide|      |
+| |fedora-32| |fedora-33| |fedora-rawhide|      |
 +-----------------------------------------------+
 
 Users of Fedora 29 or later may simply
@@ -158,7 +163,7 @@ To install ocrmypdf for the system:
 
 .. code-block:: bash
 
-    sudo pip3 install ocrmypdf
+    pip3 install ocrmypdf
 
 To install for the current user only:
 
@@ -190,7 +195,8 @@ of ocrmypdf, and install the following dependencies:
         python3-reportlab \
         qpdf \
         tesseract-ocr \
-        zlib1g
+        zlib1g \
+        unpaper
 
 We will need a newer version of ``pip`` then was available for Ubuntu 18.04:
 
@@ -354,7 +360,8 @@ To install OCRmyPDF for Alpine Linux:
 Mageia 7
 --------
 
-Install the following dependencies:
+There is no OS-level packaging available for Mageia, so you must install the
+dependencies:
 
 .. code-block:: bash
 
@@ -379,12 +386,16 @@ Install the following 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
 
@@ -453,7 +464,7 @@ Update Homebrew:
 
 Install or upgrade the required Homebrew packages, if any are missing.
 To do this, use ``brew edit ocrmypdf`` to obtain a recent list of Homebrew
-dependencies. You could also check the ``azure-pipelines.yml``.
+dependencies. You could also check the ``.workflows/build.yml``.
 
 This will include the English, French, German and Spanish language
 packs. If you need other languages you can optionally install them all:
@@ -494,10 +505,6 @@ Installing on Windows
 Native Windows
 --------------
 
-.. note::
-
-    It is easier to install OCRmyPDF on Windows Subsystem for Linux.
-
 .. note::
 
     Administrator privileges will be required for some of these steps.
@@ -508,31 +515,41 @@ You must install the following for Windows:
 * Tesseract 4.0 or later
 * Ghostscript 9.50 or later
 
-You can install these with the Chocolatey package manager:
+Using the `Chocolatey <https://chocolatey.org/>`_ package manager, install the
+following when running in an Administrator command prompt:
 
 * ``choco install python3``
 * ``choco install --pre tesseract``
 * ``choco install ghostscript``
+* ``choco install pngquant`` (optional)
 
-Also consider adding:
+The commands above will install Python 3.x (latest version), Tesseract, Ghostscript
+and pngquant. Chocolatey may also need to install the Windows Visual C++ Runtime
+DLLs or other Windows patches, and may require a reboot.
 
-* ``choco install pngquant``
-
-Windows 10 64-bit and 64-bit versions of applications are recommended. Earlier
-versions of Windows and 32-bit versions of these programs are not tested, and not
-supported at this time.
-
-OCRmyPDF will check for Tesseract-OCR and Ghostscript in your Program Files folder.
-If they are in some other location, you may need to modify the ``PATH``
-environment variable so Tesseract, Ghostscript, and other any optional executables can
-be found. You can enter it in the command line or
-`follow these directions <https://www.computerhope.com/issues/ch000549.htm#dospath>`_
-to make the change persistent and system-wide.
-
-You may then use pip to install ocrmypdf:
+You may then use ``pip`` to install ocrmypdf. (This can performed by a user or
+Administrator.):
 
 * ``pip install ocrmypdf``
 
+Chocolatey automatically selects appropriate versions of these applications. If you
+are installing them manually, please install 64-bit versions of all applications for
+64-bit Windows, or 32-bit versions of all applications for 32-bit Windows. Mixing
+the "bitness" of these programs will lead to errors.
+
+OCRmyPDF will check the Windows Registry and standard locations in your Program Files
+for third party software it needs (specifically, Tesseract and Ghostscript). To
+override the versions OCRmyPDF selects, you can modify the ``PATH`` environment
+variable. `Follow these directions <https://www.computerhope.com/issues/ch000549.htm#dospath>`_
+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.
+
 Windows Subsystem for Linux
 ---------------------------
 
@@ -603,7 +620,7 @@ However, the OCR-to-text-layer functionality is available.
 Docker
 ------
 
-You can also :ref:`Install the Docker <docker-install>` container on Windows. Ensure that
+You can also :ref:`Install the Docker <docker>` container on Windows. Ensure that
 your command prompt can run the docker "hello world" container.
 
 Installing on FreeBSD
@@ -629,7 +646,7 @@ Installing the Docker image
 For some users, installing the Docker image will be easier than
 installing all of OCRmyPDF's dependencies.
 
-See `OCRmyPDF Docker Image <docker>`__ for more information.
+See :ref:`docker` for more information.
 
 Installing with Python pip
 ==========================
@@ -637,7 +654,22 @@ Installing with Python pip
 OCRmyPDF is delivered by PyPI because it is a convenient way to install
 the latest version. However, PyPI and ``pip`` cannot address the fact
 that ``ocrmypdf`` depends on certain non-Python system libraries and
-programs being instsalled.
+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

docs/languages.rst

@@ -70,4 +70,7 @@ derived Docker image as
 Windows users
 =============
 
-The Tesseract installer provided by Chocolatey already includes 100 languages.
+The Tesseract installer provided by Chocolatey currently includes only English language. 
+To install other languages, download the respective language pack (``.traineddata`` file) 
+from https://github.com/tesseract-ocr/tessdata/ and place it in 
+``C:\\Program Files\\Tesseract-OCR\\tessdata`` (or wherever Tesseract OCR is installed).

docs/plugins.rst

@@ -65,6 +65,27 @@ similar to ``pytest`` packages such as ``pytest-cov`` (the package) and
     ``ocrmypdf-`` (for the package name on PyPI) and ``ocrmypdf_`` (for the
     module), just like pytest plugins.
 
+Setuptools plugins
+==================
+
+You can also create a plugin that OCRmyPDF will always automatically load if both are
+installed in the same virtual environment, using a setuptools entrypoint.
+
+Your package's ``setup.py`` would need to contain the following, for a plugin
+named ``ocrmypdf-exampleplugin``:
+
+.. code-block:: python
+
+    # sample ./setup.py file
+    from setuptools import setup
+
+    setup(
+        name="ocrmypdf-exampleplugin",
+        packages=["exampleplugin"],
+        # the following makes a plugin available to pytest
+        entry_points={"ocrmypdf": ["exampleplugin = exampleplugin.pluginmodule"]},
+    )
+
 Plugin requirements
 ===================
 
@@ -130,6 +151,18 @@ Custom command line arguments
 
 .. autofunction:: ocrmypdf.pluginspec.check_options
 
+Execution and progress reporting
+--------------------------------
+
+.. autoclass: ocrmypdf.pluginspec.Executor
+    :members:
+
+.. autofunction:: ocrmypdf.pluginspec.get_logging_console
+
+.. autofunction:: ocrmypdf.pluginspec.get_executor
+
+.. autofunction:: ocrmypdf.pluginspec.get_progressbar_class
+
 Applying special behavior before processing
 -------------------------------------------
 
@@ -147,6 +180,8 @@ Modifying intermediate images
 
 .. autofunction:: ocrmypdf.pluginspec.filter_page_image
 
+.. autofunction:: ocrmypdf.pluginspec.filter_pdf_page
+
 OCR engine
 ----------
 

docs/release_notes.rst

@@ -12,6 +12,269 @@ 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.
 
+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 poor implemented PDF text
+   extraction algorithms may fail to find 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.
+-  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.
+-  Using the new plugin hooks, it is now possible to run OCRmyPDF on alternative
+   execution environments that do not have interprocess semaphores, such as
+   AWS Lambda and Android Termux.
+-  Continuous integration moved to GitHub Actions.
+-  We now generate an ARM64-compatible Docker image alongside the x64 image.
+   Thanks to @andkrause for contributing the change and @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. (#761)
+
+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. (#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`` (#730).
+
+v11.6.1
+=======
+
+-  Fixed an issue with attempting optimize unusually narrow-width images by excluding
+   these images from optimization (#732).
+-  Remove an obsolete compatibility shim for a version of pikepdf that is no longer
+   supported.
+
+v11.6.0
+=======
+
+-  OCRmyPDF will now automatically register plugins from the same virtual environment
+   with an appropriate setuptools entrypoint.
+-  Refactor the plugin manager to remove unnecessary complications and make plugin
+   registration more automatic.
+-  ``PageContext`` and ``PdfContext`` are now formally part of the API, as they
+   should have been, since they were part of ``ocrmypdf.pluginspec``.
+
+v11.5.0
+=======
+
+-  Fixed an issue where the output page size might differ by a fractional amount
+   due to rounding, when ``--force-ocr`` was used and the page contained objects
+   with multiple resolutions.
+-  When determining the resolution at which to rasterize a page, we now consider
+   printed text on the page as requiring a higher resolution. This fixes issues
+   with certain pages being rendered with unacceptably low resolution text, but
+   may increase output file sizes in some workflows where low resolution text
+   is acceptable.
+-  Added a workaround to fix an exception that occurs when trying to
+   ``import ocrmypdf.leptonica`` on Apple ARM silicon (or potentially, other
+   platforms that do not permit write+executable memory).
+
+v11.4.5
+=======
+
+-  Fixed an issue where files may not be closed when the API is used.
+-  Improved ``setup.cfg`` with better settings for test coverage.
+
+v11.4.4
+=======
+
+-  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
+   documentation.
+
+v11.4.3
+=======
+
+-  Removed a redundant debug message.
+-  Test suite now asserts that most patched functions are called when they should be.
+-  Test suite now skips a test that fails on two particular versions of piekpdf.
+
+v11.4.2
+=======
+
+-  Fixed support for Cygwin, hopefully.
+-  watcher.py: Fixed an issue with the OCR_LOGLEVEL not being interpreted.
+
+v11.4.1
+=======
+
+-  Fixed an issue where invalid pages ranges passed using the ``pages`` argument,
+   such as "1-0" would cause unhandled exceptions.
+-  Accepted a user-contributed to the Synology demo script in misc/synology.py.
+-  Clarified documentation about change of temporary file location ``ocrmypdf.io``.
+-  Fixed Python wheel tag which was incorrectly set to py35 even though we long
+   since dropped support for Python 3.5.
+
+v11.4.0
+=======
+
+-  When looking for Tesseract and Ghostscript, we now check the Windows Registry to
+   see if their installers registered the location of their executables. This should
+   help Windows users who have installed these programs to non-standard
+   locations.
+-  We now report on the progress of PDF/A conversion, since this operation is
+   sometimes slow.
+-  Improved command line completions.
+-  The prefix of the temporary folder OCRmyPDF creates has been changed from
+   ``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
+   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.
+
+v11.3.4
+=======
+
+-  Fixed an error message 'called readLinearizationData for file that is not
+   linearized' that may occur when pikepdf 2.1.0 is used. (Upgrading to pikepdf
+   2.1.1 also fixes the issue.)
+-  File watcher now automatically includes ``.PDF`` in addition to ``.pdf`` to
+   better support case sensitive file systems.
+-  Some documentation and comment improvements.
+
+v11.3.3
+=======
+
+-  If unpaper outputs non-UTF-8 data, quietly fix this rather than choke on the
+   conversion. (Possibly addresses #671.)
+
+v11.3.2
+=======
+
+-  Explicitly require pikepdf 2.0.0 or newer when running on Python 3.9. (There are
+   concerns about the stability of pybind11 2.5.x with Python 3.9, which is used in
+   pikepdf 1.x.)
+-  Fixed another issue related to page rotation.
+-  Fixed an issue where image marked as image masks were not properly considered
+   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 #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.
+
+v11.3.1
+=======
+
+-  Declare support for new versions: pdfminer.six 20201018 and pikepdf 2.x
+-  Fix warning related to ``--pdfa-image-compression`` that appears at the wrong
+   time.
+
+v11.3.0
+=======
+
+-  The "OCR" step is describing as "Image processing" in the output messages when
+   OCR is disabled, to better explain the application's behavior.
+-  Debug logs are now only created when run as a command line, and not when OCR
+   is performed for an API call. It is the calling application's responsibility
+   to set up logging.
+-  For PDFs with a low number of pages, we gathered information about the input PDF
+   in a thread rather than process (when there are more pages). When run as a
+   thread, we did not close the file handle to the working PDF, leaking one file
+   handle per call of ``ocrmypdf.ocr``.
+-  Fixed an issue where debug messages send by child worker processes did not match
+   the log settings of parent process, causing messages to be dropped. This affected
+   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 (#658) due to the issue above.
+
+v11.2.1
+=======
+
+-  Fixed an issue where optimization of a 1-bit image with a color palette or
+   associated ICC that was optimized to JBIG2 could have its colors inverted.
+
+v11.2.0
+=======
+
+-  Fixed an issue with optimizing PNG-type images that had soft masks or image masks.
+   This is a regression introduced in (or about) v11.1.0.
+-  Improved type checking of the ``plugins`` parameter for the ``ocrmypdf.ocr``
+   API call.
+
+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. (#642)
+
 v11.1.1
 =======
 

misc/completion/ocrmypdf.fish

@@ -59,7 +59,8 @@ complete -c ocrmypdf -x -l output-type -a '(__fish_ocrmypdf_output_type)' -d "se
 
 function __fish_ocrmypdf_pdf_renderer
     echo -e "auto\t"(_ "auto select PDF renderer")
-    echo -e "hocr\t"(_ "use hocr renderer")
+    echo -e "hocr\t"(_ "use hOCR renderer")
+    echo -e "hocrdebug\t"(_ "uses hOCR renderer in debug mode, showing recognized text")
     echo -e "sandwich\t"(_ "use sandwich renderer")
 end
 complete -c ocrmypdf -x -l pdf-renderer -a '(__fish_ocrmypdf_pdf_renderer)' -d "select PDF renderer options"
@@ -135,4 +136,4 @@ complete -c ocrmypdf -r -l user-words -d "specify location of user words file"
 complete -c ocrmypdf -r -l user-patterns -d "specify location of user patterns file"
 complete -c ocrmypdf -x -l fast-web-view -d "if file size if above this amount in MB, linearize PDF"
 
-complete -c ocrmypdf -x -a "(__fish_complete_suffix .pdf)"
+complete -c ocrmypdf -x -a "(__fish_complete_suffix .pdf; __fish_complete_suffix .PDF; __fish_complete_suffix .jpg; __fish_complete_suffix .png)"

misc/docker-compose.example.yml

@@ -9,7 +9,7 @@ services:
       - "/media/scan:/input"
       - "/mnt/scan:/output"
     environment:
-      - OCR_OUTPUT_DIRECTORY_YEAR_MONT=0
+      - OCR_OUTPUT_DIRECTORY_YEAR_MONTH=0
     user: "<SET TO YOUR USER ID>:<SET TO YOUR GROUP ID>"
     entrypoint: python3
     command: watcher.py

misc/example_plugin.py

@@ -18,6 +18,25 @@
 # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 # SOFTWARE.
 
+"""
+An example of an OCRmyPDF plugin.
+
+This plugin adds two new command line arguments
+    --grayscale-ocr: converts the image to grayscale before performing OCR on it
+        (This is occasionally useful for images whose color confounds OCR. It only
+        affects the image shown to OCR. The image is not saved.)
+    --mono-page: converts pages all pages in the output file to black and white
+
+To use this from the command line:
+    ocrmypdf --plugin path/to/example_plugin.py --mono-page input.pdf output.pdf
+
+To use this as an API:
+    import ocrmypdf
+    ocrmypdf.ocr('input.pdf', 'output.pdf',
+        plugins=['path/to/example_plugin.py'], mono_page=True
+    )
+"""
+
 import logging
 
 from PIL import Image

misc/synology.py

@@ -79,8 +79,10 @@ for dir_name, subdirs, file_list in os.walk(start_dir):
                 stdout=output_file,
                 stderr=subprocess.PIPE,
                 check=False,
+                text=True,
+                errors='ignore',
             )
-        logging.info(proc.stderr.read())
+        logging.info(proc.stderr)
         os.chmod(full_path_ocr, 0o664)
         os.chmod(full_path, 0o664)
         full_path_ocr_archive = sys.argv[2]

misc/watcher.py

@@ -44,8 +44,8 @@ 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 = bool(os.getenv('OCR_USE_POLLING', ''))
-LOGLEVEL = os.getenv('OCR_LOGLEVEL', 'INFO').upper()
-PATTERNS = ['*.pdf']
+LOGLEVEL = os.getenv('OCR_LOGLEVEL', 'INFO')
+PATTERNS = ['*.pdf', '*.PDF']
 
 log = logging.getLogger('ocrmypdf-watcher')
 
@@ -117,7 +117,12 @@ class HandleObserverEvent(PatternMatchingEventHandler):
 
 def main():
     ocrmypdf.configure_logging(
-        verbosity=ocrmypdf.Verbosity.default, manage_root_logger=True
+        verbosity=(
+            ocrmypdf.Verbosity.default
+            if LOGLEVEL != 'DEBUG'
+            else ocrmypdf.Verbosity.debug
+        ),
+        manage_root_logger=True,
     )
     log.setLevel(LOGLEVEL)
     log.info(
@@ -135,7 +140,7 @@ def main():
         f"ARGS: {OCR_JSON_SETTINGS}\n"
         f"POLL_NEW_FILE_SECONDS: {POLL_NEW_FILE_SECONDS}\n"
         f"USE_POLLING: {USE_POLLING}\n"
-        f"LOGLEVEL: {LOGLEVEL}\n"
+        f"LOGLEVEL: {LOGLEVEL}"
     )
 
     if 'input_file' in OCR_JSON_SETTINGS or 'output_file' in OCR_JSON_SETTINGS:

requirements/main.txt

@@ -1,12 +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 == 1.14.0
-coloredlogs == 14.0  # technically optional
-img2pdf == 0.3.6
-pdfminer.six == 20200517
-pikepdf == 1.16.1
+cffi == 1.14.5
+coloredlogs == 15.0  # technically optional
+img2pdf == 0.4.0
+pdfminer.six == 20201018
+pikepdf == 2.10.0
 pluggy == 0.13.1
-Pillow == 7.1.2
-reportlab == 3.5.42
-tqdm == 4.46.1
+Pillow == 8.1.2
+reportlab == 3.5.66
+tqdm == 4.59.0

requirements/test.txt

@@ -1,7 +1,6 @@
-pytest >= 5.0.0
-pytest-helpers-namespace >= 2019.1.8
-pytest-xdist >= 1.31.0
-pytest-cov >= 2.10.0
+pytest >= 6.0.0
+pytest-xdist >= 2.2.0
+pytest-cov >= 2.11.1
 python-xmp-toolkit == 2.0.1  # requires apt-get install libexempi3
                      # or brew install exempi
 #PyMuPDF == 1.13.4    # optional

requirements/watcher.txt

@@ -1 +1 @@
-watchdog == 0.10.2
+watchdog == 1.0.2

setup.cfg

@@ -1,5 +1,5 @@
 [bdist_wheel]
-python-tag = py35
+python-tag = py36
 
 [aliases]
 test=pytest
@@ -15,6 +15,8 @@ filterwarnings =
 	ignore:.*XMLParser.*:DeprecationWarning
 markers =
 	slow
+addopts =
+	-n auto
 
 [isort]
 multi_line_output=3
@@ -27,3 +29,28 @@ known_third_party = PIL,_cffi_backend,cffi,flask,img2pdf,pdfminer,pikepdf,pkg_re
 
 [metadata]
 license_file = LICENSE
+
+[coverage:paths]
+source =
+	src/ocrmypdf
+
+[coverage:run]
+branch = true
+parallel = true
+concurrency = multiprocessing
+
+[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:

setup.py

@@ -17,10 +17,6 @@ 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()
 
 
@@ -63,7 +59,6 @@ setup(
     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
-        'pytest-runner',  # to enable python setup.py test
         'setuptools_scm',  # so that version will work
         'setuptools_scm_git_archive',  # enable version from github tarballs
     ],
@@ -73,16 +68,17 @@ setup(
         '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, <= 20200726',
-        'pikepdf >= 1.14.0, < 2',
-        'Pillow >= 7.0.0',
-        'pluggy >= 0.13.0',
-        'reportlab >= 3.3.0',  # oldest released version with sane image handling
+        'pdfminer.six >= 20191110, != 20200720, <= 20201018',
+        "pikepdf >= 2.10.0",
+        'Pillow >= 8.1.2',
+        'pluggy >= 0.13.0, < 1.0',
+        'reportlab >= 3.5.66',
+        'setuptools',
         'tqdm >= 4',
     ],
     tests_require=tests_require,
     entry_points={'console_scripts': ['ocrmypdf = ocrmypdf.__main__:run']},
-    package_data={'ocrmypdf': ['data/sRGB.icc']},
+    package_data={'ocrmypdf': ['data/sRGB.icc', 'py.typed']},
     include_package_data=True,
     zip_safe=False,
     project_urls={

src/ocrmypdf/RELEASE.md

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

src/ocrmypdf/__init__.py

@@ -8,6 +8,8 @@
 from pluggy import HookimplMarker as _HookimplMarker
 
 from ocrmypdf import helpers, hocrtransform, leptonica, pdfa, pdfinfo
+from ocrmypdf._concurrent import Executor
+from ocrmypdf._jobcontext import PageContext, PdfContext
 from ocrmypdf._version import PROGRAM_NAME, __version__
 from ocrmypdf.api import Verbosity, configure_logging, ocr
 from ocrmypdf.exceptions import (

src/ocrmypdf/__main__.py

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

src/ocrmypdf/_concurrent.py

@@ -4,135 +4,130 @@
 # 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 multiprocessing import Pool as ProcessPool
-from multiprocessing.dummy import Pool as ThreadPool
+from abc import ABC, abstractmethod
 from typing import Callable, Iterable, Optional
 
-from tqdm import tqdm
 
-from ocrmypdf.exceptions import InputFileError
+def _task_noop(*_args, **_kwargs):
+    return
 
 
-def log_listener(queue):
-    """Listen to the worker processes and forward the messages to logging
+class NullProgressBar:
+    def __init__(self, **kwargs):
+        pass
 
-    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 __enter__(self):
+        return self
 
-    See https://docs.python.org/3/howto/logging-cookbook.html#logging-to-a-single-file-from-multiple-processes
+    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.
     """
 
-    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):
-    """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)
-    if hasattr(signal, '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.handlers = []
-    root.addHandler(h)
-
-    if user_init:
-        user_init()
-
-
-def thread_init(_queue, user_init):
-    # As a thread, block SIGBUS so the main thread deals with it...
-    if hasattr(signal, '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),
-        )
-        try:
-            results = pool.imap_unordered(task, task_arguments)
-            while True:
-                try:
-                    result = results.next()
-                    if task_finished:
-                        task_finished(result, pbar)
-                    else:
-                        pbar.update()
-                except StopIteration:
-                    break
-        except KeyboardInterrupt:
-            # Terminate pool so we exit instantly
-            pool.terminate()
-            # Don't try listener.join() here, will deadlock
-            raise
-        except Exception:
-            if not os.environ.get("PYTEST_CURRENT_TEST", ""):
-                # Unless inside pytest, exit immediately because no one wants
-                # to wait for child processes to finalize results that will be
-                # thrown away. Inside pytest, we want child processes to exit
-                # cleanly so that they output an error messages or coverage data
-                # we need from them.
-                pool.terminate()
-            raise
-        finally:
-            # Terminate log listener
-            log_queue.put_nowait(None)
-            pool.close()
-            pool.join()
-
-    listener.join()
+    def _execute(
+        self,
+        *,
+        use_threads: bool,
+        max_workers: int,
+        tqdm_kwargs: dict,
+        worker_initializer: Callable,
+        task: Callable,
+        task_arguments: Iterable,
+        task_finished: Callable,
+    ):  # pylint: disable=unused-argument
+        with self.pbar_class(**tqdm_kwargs) as pbar:
+            for args in task_arguments:
+                result = task(args)
+                task_finished(result, pbar)

src/ocrmypdf/_exec/ghostscript.py

@@ -15,34 +15,34 @@ from os import fspath
 from pathlib import Path
 from shutil import which
 from subprocess import PIPE, CalledProcessError
-from typing import Optional, cast
+from typing import Optional
 
 from PIL import Image
 
 from ocrmypdf.exceptions import MissingDependencyError, SubprocessOutputError
 from ocrmypdf.helpers import Resolution
-from ocrmypdf.subprocess import get_version, run
+from ocrmypdf.subprocess import get_version, run, run_polling_stderr
 
 log = logging.getLogger(__name__)
 
+missing_gs_error = """
+---------------------------------------------------------------------
+This error normally occurs when ocrmypdf find can't Ghostscript.
+Please ensure Ghostscript is installed and its location is added to
+the system PATH environment variable.
+
+For details see:
+    https://ocrmypdf.readthedocs.io/en/latest/installation.html
+---------------------------------------------------------------------
+"""
+
 _gswin = None
 if os.name == 'nt':
     _gswin = which('gswin64c')
     if not _gswin:
         _gswin = which('gswin32c')
         if not _gswin:
-            raise MissingDependencyError(
-                """
-                ---------------------------------------------------------------------
-                This error normally occurs when ocrmypdf can't Ghostscript.  Please
-                ensure Ghostscript is installed and its location is added to the
-                system PATH environment variable.
-
-                For details see:
-                    https://ocrmypdf.readthedocs.io/en/latest/installation.html
-                ---------------------------------------------------------------------
-                """
-            )
+            raise MissingDependencyError(missing_gs_error)
     _gswin = Path(_gswin).stem
 
 GS = _gswin if _gswin else 'gs'
@@ -56,16 +56,14 @@ def version():
 def jpeg_passthrough_available() -> bool:
     """Returns True if the installed version of Ghostscript supports JPEG passthru
 
-    Prior to 9.23, Ghostscript decode and re-encoded JPEGs internally. In 9.23
+    Prior to 9.23, Ghostscript decoded and re-encoded JPEGs internally. In 9.23
     it gained the ability to keep JPEGs unmodified. However, the 9.23
     implementation was buggy and would deletes the last two bytes of images in
     some cases, as reported here.
     https://bugs.ghostscript.com/show_bug.cgi?id=699216
 
     The issue was fixed for 9.24, hence that is the first version we consider
-    the feature available. (However, we don't use 9.24 at all, so the first
-    version that allows JPEG passthrough is 9.25.
-
+    the feature available. (Ghostscript 9.24 has its own problems is blacklisted.)
     """
     return version() >= '9.24'
 
@@ -81,8 +79,8 @@ def rasterize_pdf(
     raster_device: str,
     raster_dpi: Resolution,
     pageno: int = 1,
-    page_dpi: Resolution = None,
-    rotation: int = None,
+    page_dpi: Optional[Resolution] = None,
+    rotation: Optional[int] = None,
     filter_vector: bool = False,
 ):
     """Rasterize one page of a PDF at resolution raster_dpi in canvas units."""
@@ -122,8 +120,6 @@ def rasterize_pdf(
         stderr = p.stderr.decode(errors='replace')
         if _gs_error_reported(stderr):
             log.error(stderr)
-        elif stderr:
-            log.debug(stderr)
 
     with Image.open(BytesIO(p.stdout)) as im:
         if rotation is not None:
@@ -141,13 +137,44 @@ def rasterize_pdf(
         im.save(fspath(output_file), dpi=page_dpi)
 
 
+class GhostscriptFollower:
+    re_process = re.compile(r"Processing pages \d+ through (\d+).")
+    re_page = re.compile(r"Page (\d+)")
+
+    def __init__(self, progressbar_class):
+        self.count = 0
+        self.progressbar_class = progressbar_class
+        self.progressbar = None
+
+    def __call__(self, line):
+        if not self.progressbar_class:
+            return
+        if not self.progressbar:
+            m = self.re_process.match(line.strip())
+            if m:
+                self.count = int(m.group(1))
+                self.progressbar = self.progressbar_class(
+                    total=self.count, desc="PDF/A conversion", unit='page'
+                )
+                return
+        else:
+            m = self.re_page.match(line.strip())
+            if m:
+                self.progressbar.update()
+
+
 def generate_pdfa(
     pdf_pages,
     output_file: os.PathLike,
+    *,
     compression: str,
     pdf_version: str = '1.5',
     pdfa_part: str = '2',
+    progressbar_class=None,
 ):
+    # Ghostscript's compression is all or nothing. We can either force all images
+    # to JPEG, force all to Flate/PNG, or let it decide how to encode the images.
+    # In most case it's best to let it decide.
     compression_args = []
     if compression == 'jpeg':
         compression_args = [
@@ -169,14 +196,16 @@ def generate_pdfa(
             "-dAutoFilterGrayImages=true",
         ]
 
+    strategy = 'LeaveColorUnchanged'
     # Older versions of Ghostscript expect a leading slash in
     # sColorConversionStrategy, newer ones should not have it. See Ghostscript
     # git commit fe1c025d.
-    strategy = 'RGB' if version() >= '9.19' else '/RGB'
+    strategy = ('/' + strategy) if version() < '9.19' else strategy
 
     if version() == '9.23':
-        # 9.23: new feature JPEG passthrough is broken in some cases, best to
-        # disable it always
+        # 9.23: added JPEG passthrough as a new feature, but with a bug that
+        # incorrectly formats some images. Fixed as of 9.24. So we disable this
+        # feature for 9.23.
         # https://bugs.ghostscript.com/show_bug.cgi?id=699216
         compression_args.append('-dPassThroughJPEGImages=false')
 
@@ -186,7 +215,6 @@ def generate_pdfa(
     args_gs = (
         [
             GS,
-            "-dQUIET",
             "-dBATCH",
             "-dNOPAUSE",
             "-dSAFER",
@@ -206,16 +234,28 @@ def generate_pdfa(
         ]
     )
     args_gs.extend(fspath(s) for s in pdf_pages)  # Stringify Path objs
+
     try:
         with Path(output_file).open('wb') as output:
-            p = run(args_gs, stdout=output, stderr=PIPE, check=True)
+            p = run_polling_stderr(
+                args_gs,
+                stdout=output,
+                stderr=PIPE,
+                check=True,
+                text=True,
+                encoding='utf-8',
+                errors='replace',
+                callback=GhostscriptFollower(progressbar_class),
+            )
     except CalledProcessError as e:
         # Ghostscript does not change return code when it fails to create
         # PDF/A - check PDF/A status elsewhere
-        log.error(e.stderr.decode(errors='replace'))
-        raise SubprocessOutputError('Ghostscript PDF/A rendering failed')
+        log.error(e.stderr)
+        raise SubprocessOutputError('Ghostscript PDF/A rendering failed') from e
     else:
-        stderr = p.stderr.decode('utf-8', errors='replace')
+        stderr = p.stderr
+        # If there is an error we log the whole stderr, except for filtering
+        # duplicates.
         if _gs_error_reported(stderr):
             last_part = None
             repcount = 0
@@ -228,11 +268,3 @@ def generate_pdfa(
                 else:
                     repcount += 1
                 last_part = part
-        elif 'overprint mode not set' in stderr:
-            # Unless someone is going to print PDF/A documents on a
-            # magical sRGB printer I can't see the removal of overprinting
-            # being a problem....
-            log.debug(
-                "Ghostscript had to remove PDF 'overprinting' from the "
-                "input file to complete PDF/A conversion. "
-            )

src/ocrmypdf/_exec/jbig2enc.py

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

src/ocrmypdf/_exec/pngquant.py

@@ -9,10 +9,8 @@
 
 from contextlib import contextmanager
 from io import BytesIO
-from os import fspath
 from pathlib import Path
 from subprocess import PIPE
-from tempfile import NamedTemporaryFile
 
 from PIL import Image
 
@@ -61,3 +59,7 @@ def quantize(input_file: Path, output_file: Path, quality_min: int, quality_max:
     if result.returncode == 0:
         # input_file could be the same as output_file, so we defer the write
         output_file.write_bytes(result.stdout)
+
+
+def quantize_mp(args):
+    return quantize(*args)

src/ocrmypdf/_exec/tesseract.py

@@ -9,12 +9,14 @@
 
 import logging
 import os
+import re
 import shutil
 from collections import namedtuple
+from distutils.version import StrictVersion
 from os import fspath
 from pathlib import Path
 from subprocess import PIPE, STDOUT, CalledProcessError, TimeoutExpired
-from typing import List
+from typing import List, Optional
 
 from PIL import Image
 
@@ -53,31 +55,29 @@ 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+)?)?  # 5/prerelease, 6/prerelease_num
+        $
+        ''',
+        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():
     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?
 
@@ -100,7 +100,12 @@ def get_languages():
     args_tess = ['tesseract', '--list-langs']
     try:
         proc = run(
-            args_tess, universal_newlines=True, stdout=PIPE, stderr=STDOUT, check=True
+            args_tess,
+            text=True,
+            stdout=PIPE,
+            stderr=STDOUT,
+            logs_errors_to_stdout=True,
+            check=True,
         )
         output = proc.stdout
     except CalledProcessError as e:
@@ -113,7 +118,7 @@ def get_languages():
     return set(lang.strip() for lang in rest)
 
 
-def tess_base_args(langs: List[str], engine_mode: int) -> List[str]:
+def tess_base_args(langs: List[str], engine_mode: Optional[int]) -> List[str]:
     args = ['tesseract']
     if langs:
         args.extend(['-l', '+'.join(langs)])
@@ -122,7 +127,7 @@ def tess_base_args(langs: List[str], engine_mode: int) -> List[str]:
     return args
 
 
-def get_orientation(input_file: Path, engine_mode: int, timeout: float):
+def get_orientation(input_file: Path, engine_mode: Optional[int], timeout: float):
     args_tesseract = tess_base_args(['osd'], engine_mode) + [
         '--psm',
         '0',
@@ -216,6 +221,7 @@ def _generate_null_hocr(output_hocr, output_text, image):
 
 
 def generate_hocr(
+    *,
     input_file: Path,
     output_hocr: Path,
     output_text: Path,

src/ocrmypdf/_exec/unpaper.py

@@ -13,10 +13,11 @@
 import logging
 import os
 import shlex
+from decimal import Decimal
 from pathlib import Path
-from subprocess import PIPE, STDOUT, CalledProcessError
+from subprocess import PIPE, STDOUT
 from tempfile import TemporaryDirectory
-from typing import Tuple
+from typing import List, Optional, Tuple, Union
 
 from PIL import Image
 
@@ -24,10 +25,12 @@ from ocrmypdf.exceptions import MissingDependencyError, SubprocessOutputError
 from ocrmypdf.subprocess import get_version
 from ocrmypdf.subprocess import run as external_run
 
+DecFloat = Union[Decimal, float]
+
 log = logging.getLogger(__name__)
 
 
-def version():
+def version() -> str:
     return get_version('unpaper')
 
 
@@ -53,23 +56,25 @@ def _setup_unpaper_io(tmpdir: Path, input_file: Path) -> Tuple[Path, Path]:
         except KeyError:
             raise MissingDependencyError(
                 "Failed to convert image to a supported format."
-            ) from e
+            ) from None
 
-        if im_modified or input_file.suffix != '.png':
-            input_png = tmpdir / 'input.png'
-            im.save(input_png, format='PNG', compress_level=1)
+        if im_modified or input_file.suffix != '.pnm':
+            input_pnm = tmpdir / 'input.pnm'
+            im.save(input_pnm, format='PPM')
         else:
             # No changes, PNG input, just use the file we already have
-            input_png = input_file
+            input_pnm = input_file
         output_pnm = tmpdir / f'output{suffix}'
-    return input_png, output_pnm
+    return input_pnm, output_pnm
 
 
-def run(input_file, output_file, dpi, mode_args):
-    args_unpaper = ['unpaper', '-v', '--dpi', str(dpi)] + mode_args
+def run(
+    input_file: Path, output_file: Path, *, dpi: DecFloat, mode_args: List[str]
+) -> None:
+    args_unpaper = ['unpaper', '-v', '--dpi', str(round(dpi, 6))] + mode_args
 
     with TemporaryDirectory() as tmpdir:
-        input_png, output_pnm = _setup_unpaper_io(Path(tmpdir), input_file)
+        input_pnm, output_pnm = _setup_unpaper_io(Path(tmpdir), input_file)
 
         # To prevent any shenanigans from accepting arbitrary parameters in
         # --unpaper-args, we:
@@ -78,41 +83,41 @@ def run(input_file, output_file, dpi, mode_args):
         # 3) append absolute paths for the input and output file
         # This should ensure that a user cannot clobber some other file with
         # their unpaper arguments (whether intentionally or otherwise)
-        args_unpaper.extend([os.fspath(input_png), os.fspath(output_pnm)])
+        args_unpaper.extend([os.fspath(input_pnm), os.fspath(output_pnm)])
+        external_run(
+            args_unpaper,
+            close_fds=True,
+            check=True,
+            stderr=STDOUT,  # unpaper writes logging output to stdout and stderr
+            stdout=PIPE,  # and cannot send file output to stdout
+            cwd=tmpdir,
+            logs_errors_to_stdout=True,
+        )
         try:
-            proc = external_run(
-                args_unpaper,
-                check=True,
-                close_fds=True,
-                universal_newlines=True,
-                stderr=STDOUT,  # unpaper writes logging output to stdout and stderr
-                cwd=tmpdir,  # and cannot send file output to stdout
-                stdout=PIPE,
-            )
-        except CalledProcessError as e:
-            log.debug(e.stderr)
-            raise e from e
-        else:
-            log.debug(proc.stderr)
-            try:
-                with Image.open(output_pnm) as imout:
-                    imout.save(output_file, dpi=(dpi, dpi))
-            except (FileNotFoundError, OSError):
-                raise SubprocessOutputError(
-                    "unpaper: failed to produce the expected output file. "
-                    + " Called with: "
-                    + str(args_unpaper)
-                ) from None
+            with Image.open(output_pnm) as imout:
+                imout.save(output_file, dpi=(dpi, dpi))
+        except (FileNotFoundError, OSError):
+            raise SubprocessOutputError(
+                "unpaper: failed to produce the expected output file. "
+                + " Called with: "
+                + str(args_unpaper)
+            ) from None
 
 
-def validate_custom_args(args: str):
+def validate_custom_args(args: str) -> List[str]:
     unpaper_args = shlex.split(args)
     if any(('/' in arg or arg == '.' or arg == '..') for arg in unpaper_args):
         raise ValueError('No filenames allowed in --unpaper-args')
     return unpaper_args
 
 
-def clean(input_file, output_file, dpi, unpaper_args=None):
+def clean(
+    input_file: Path,
+    output_file: Path,
+    *,
+    dpi: DecFloat,
+    unpaper_args: Optional[List[str]] = None,
+):
     default_args = [
         '--layout',
         'none',
@@ -126,4 +131,4 @@ def clean(input_file, output_file, dpi, unpaper_args=None):
     ]
     if not unpaper_args:
         unpaper_args = default_args
-    run(input_file, output_file, dpi, unpaper_args)
+    run(input_file, output_file, dpi=dpi, mode_args=unpaper_args)

src/ocrmypdf/_graft.py

@@ -6,33 +6,39 @@
 
 
 import logging
+import uuid
 from contextlib import suppress
 from pathlib import Path
 from typing import Optional
 
 import pikepdf
+from pikepdf.objects import Dictionary, Name
 
 log = logging.getLogger(__name__)
 MAX_REPLACE_PAGES = 100
 
 
-def _update_page_resources(*, page, font, font_key, procset):
-    """Update this page's fonts with a reference to the Glyphless font"""
+def _ensure_dictionary(obj, name):
+    if name not in obj:
+        obj[name] = Dictionary({})
+    return obj[name]
 
-    if '/Resources' not in page:
-        page['/Resources'] = pikepdf.Dictionary({})
-    resources = page['/Resources']
-    try:
-        fonts = resources['/Font']
-    except KeyError:
-        fonts = pikepdf.Dictionary({})
+
+def _update_resources(*, obj, font, font_key, procset):
+    """Update this obj's fonts with a reference to the Glyphless font.
+
+    obj can be a page or Form XObject.
+    """
+
+    resources = _ensure_dictionary(obj, Name.Resources)
+    fonts = _ensure_dictionary(resources, Name.Font)
     if font_key is not None and font_key not in fonts:
         fonts[font_key] = font
-    resources['/Font'] = fonts
 
     # Reassign /ProcSet to one that just lists everything - ProcSet is
     # obsolete and doesn't matter but recommended for old viewer support
-    resources['/ProcSet'] = procset
+    if procset:
+        resources['/ProcSet'] = procset
 
 
 def strip_invisible_text(pdf, page):
@@ -109,6 +115,7 @@ class OcrGrafter:
         if textpdf and not self.font:
             self.font, self.font_key = self._find_font(textpdf)
 
+        emplaced_page = False
         content_rotation = self.pdfinfo[pageno].rotation
         path_image = Path(image).resolve() if image else None
         if path_image is not None and path_image != self.path_base:
@@ -122,24 +129,21 @@ class OcrGrafter:
                 local_image_page = self.pdf_base.pages[-1]
                 self.pdf_base.pages[pageno].emplace(local_image_page)
                 del self.pdf_base.pages[-1]
-            # The pdf_image_page will always be created with any /Rotate applied
-            # applied already
-            content_rotation = 0
+            emplaced_page = True
 
-        if content_rotation != 0:
-            # Text can be misaligned on a /Rotate'd page.
-            # That is because we rasterize pages with /Rotate applied,
-            # so that the OCR image text is upright and comes back upright.
-            text_misaligned = (autorotate_correction - content_rotation) % 360
-            log.debug(
-                f"Text rotation: (autorotate, content) -> text misalignment = "
-                f"({autorotate_correction}, {content_rotation}) -> {text_misaligned}"
-            )
-        else:
-            text_misaligned = 0
+        # Calculate if the text is misaligned compared to the content
+        if emplaced_page:
+            content_rotation = autorotate_correction
+        text_rotation = autorotate_correction
+        text_misaligned = (text_rotation - content_rotation) % 360
+        log.debug(
+            f"Text rotation: (text, autorotate, content) -> text misalignment = "
+            f"({text_rotation}, {autorotate_correction}, {content_rotation}) -> {text_misaligned}"
+        )
 
         if textpdf and self.font:
-            # Graft the text layer onto this page, whether new or old
+            # Graft the text layer onto this page, whether new or old, possibly
+            # rotating the text layer by the amount is misaligned.
             strip_old = self.context.options.redo_ocr
             self._graft_text_layer(
                 page_num=pageno + 1,
@@ -151,14 +155,14 @@ class OcrGrafter:
                 strip_old_text=strip_old,
             )
 
-        # Correct the page rotation
+        # Correct the overall page rotation if needed, now that the text and content
+        # are aligned
         page_rotation = (content_rotation - autorotate_correction) % 360
         self.pdf_base.pages[pageno].Rotate = page_rotation
         log.debug(
             f"Page rotation: (content, auto) -> page = "
             f"({content_rotation}, {autorotate_correction}) -> {page_rotation}"
         )
-
         if self.emplacements % MAX_REPLACE_PAGES == 0:
             self.save_and_reload()
 
@@ -171,13 +175,13 @@ class OcrGrafter:
         """
 
         page0 = self.pdf_base.pages[0]
-        _update_page_resources(
-            page=page0, font=self.font, font_key=self.font_key, procset=self.procset
+        _update_resources(
+            obj=page0, font=self.font, font_key=self.font_key, procset=self.procset
         )
 
         # We cannot read and write the same file, that will corrupt it
         # but we don't to keep more copies than we need to. Delete intermediates.
-        # {interim_count} is the opened file we were updateing
+        # {interim_count} is the opened file we were updating
         # {interim_count - 1} can be deleted
         # {interim_count + 1} is the new file will produce and open
         old_file = self.output_file.with_suffix(f'.working{self.interim_count - 1}.pdf')
@@ -212,6 +216,7 @@ class OcrGrafter:
                     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:
@@ -281,17 +286,29 @@ class OcrGrafter:
             # finally move the lower left corner to match the mediabox
             ctm = translate @ rotate @ scale @ untranslate @ corner
 
-            pdf_text_contents = (
-                b'q %s cm\n' % ctm.encode() + pdf_text_contents + b'\nQ\n'
+            base_resources = _ensure_dictionary(base_page, Name.Resources)
+            base_xobjs = _ensure_dictionary(base_resources, Name.XObject)
+            text_xobj_name = Name('/' + 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]
             )
 
-            new_text_layer = pikepdf.Stream(self.pdf_base, pdf_text_contents)
+            pdf_draw_xobj = (
+                (b'q %s cm\n' % ctm.encode()) + (b'%s Do\n' % text_xobj_name) + b'\nQ\n'
+            )
+            new_text_layer = pikepdf.Stream(self.pdf_base, pdf_draw_xobj)
 
             if strip_old_text:
                 strip_invisible_text(self.pdf_base, base_page)
 
             base_page.page_contents_add(new_text_layer, prepend=True)
 
-            _update_page_resources(
-                page=base_page, font=font, font_key=font_key, procset=procset
+            _update_resources(
+                obj=base_page, font=font, font_key=font_key, procset=procset
             )

src/ocrmypdf/_jobcontext.py

@@ -10,15 +10,22 @@ import shutil
 import sys
 from argparse import Namespace
 from copy import copy
-from io import IOBase
 from pathlib import Path
 from typing import Iterator
 
+from pluggy import PluginManager
+
 from ocrmypdf.pdfinfo import PdfInfo
+from ocrmypdf.pdfinfo.info import PageInfo
 
 
 class PdfContext:
-    """Holds our context for a particular run of the pipeline"""
+    """Holds the context for a particular run of the pipeline."""
+
+    options: Namespace  #: The specified options for processing this PDF.
+    origin: Path  #: The filename of the original input file.
+    pdfinfo: PdfInfo  #: Detailed data for this PDF.
+    plugin_manager: PluginManager  #: PluginManager for processing the current PDF.
 
     def __init__(
         self,
@@ -35,21 +42,33 @@ class PdfContext:
         self.plugin_manager = plugin_manager
 
     def get_path(self, name: str) -> Path:
+        """Generate a ``Path`` for an intermediate file involved in processing.
+
+        The path will be in a temporary folder that is common for all processing
+        of this particular PDF.
+        """
         return self.work_folder / name
 
     def get_page_contexts(self) -> Iterator['PageContext']:
+        """Get all ``PageContext`` for this PDF."""
         npages = len(self.pdfinfo)
         for n in range(npages):
             yield PageContext(self, n)
 
 
 class PageContext:
-    """Holds our context for a page
+    """Holds our context for a page.
 
     Must be pickable, so stores only intrinsic/simple data elements or those
-    capable of their serializing themselves via __getstate__.
+    capable of their serializing themselves via ``__getstate__``.
     """
 
+    options: Namespace  #: The specified options for processing this PDF.
+    origin: Path  #: The filename of the original input file.
+    pageno: int  #: This page number (zero-based).
+    pageinfo: PageInfo  #: Information on this page.
+    plugin_manager: PluginManager  #: PluginManager for processing the current PDF.
+
     def __init__(self, pdf_context: PdfContext, pageno):
         self.work_folder = pdf_context.work_folder
         self.origin = pdf_context.origin
@@ -59,6 +78,11 @@ class PageContext:
         self.plugin_manager = pdf_context.plugin_manager
 
     def get_path(self, name: str) -> Path:
+        """Generate a ``Path`` for a file that is part of processing this page.
+
+        The path will be based in a common temporary folder and have a prefix based
+        on the page number.
+        """
         return self.work_folder / ("%06d_%s" % (self.pageno + 1, name))
 
     def __getstate__(self):

src/ocrmypdf/_pipeline.py

@@ -13,7 +13,7 @@ from contextlib import suppress
 from datetime import datetime, timezone
 from pathlib import Path
 from shutil import copyfileobj
-from typing import BinaryIO, Dict, Iterable, Optional, Union, cast
+from typing import Dict, Iterable, Optional
 
 import img2pdf
 import pikepdf
@@ -21,6 +21,7 @@ from pikepdf.models.metadata import encode_pdf_date
 from PIL import Image, ImageColor, ImageDraw
 
 from ocrmypdf import leptonica
+from ocrmypdf._concurrent import Executor
 from ocrmypdf._exec import unpaper
 from ocrmypdf._jobcontext import PageContext, PdfContext
 from ocrmypdf._version import PROGRAM_NAME
@@ -145,11 +146,13 @@ def triage(original_filename, input_file, output_file, options):
 
 def get_pdfinfo(
     input_file,
+    *,
+    executor: Executor,
     detailed_analysis=False,
     progbar=False,
     max_workers=None,
     check_pages=None,
-):
+) -> PdfInfo:
     try:
         return PdfInfo(
             input_file,
@@ -157,11 +160,12 @@ def get_pdfinfo(
             progbar=progbar,
             max_workers=max_workers,
             check_pages=check_pages,
+            executor=executor,
         )
-    except pikepdf.PasswordError:
-        raise EncryptedPdfError()
-    except pikepdf.PdfError:
-        raise InputFileError()
+    except pikepdf.PasswordError as e:
+        raise EncryptedPdfError() from e
+    except pikepdf.PdfError as e:
+        raise InputFileError() from e
 
 
 def validate_pdfinfo_options(context: PdfContext):
@@ -205,17 +209,21 @@ def validate_pdfinfo_options(context: PdfContext):
     context.plugin_manager.hook.validate(pdfinfo=pdfinfo, options=options)
 
 
+def _vector_page_dpi(pageinfo):
+    return VECTOR_PAGE_DPI if pageinfo.has_vector or pageinfo.has_text else 0.0
+
+
 def get_page_dpi(pageinfo, options):
     "Get the DPI when nonsquare DPI is tolerable"
     xres = max(
         pageinfo.dpi.x or VECTOR_PAGE_DPI,
         options.oversample or 0.0,
-        VECTOR_PAGE_DPI if pageinfo.has_vector else 0.0,
+        _vector_page_dpi(pageinfo),
     )
     yres = max(
         pageinfo.dpi.y or VECTOR_PAGE_DPI,
         options.oversample or 0,
-        VECTOR_PAGE_DPI if pageinfo.has_vector else 0.0,
+        _vector_page_dpi(pageinfo),
     )
     return Resolution(float(xres), float(yres))
 
@@ -229,7 +237,7 @@ def get_page_square_dpi(pageinfo, options) -> Resolution:
         max(
             (xres * userunit) or VECTOR_PAGE_DPI,
             (yres * userunit) or VECTOR_PAGE_DPI,
-            VECTOR_PAGE_DPI if pageinfo.has_vector else 0.0,
+            _vector_page_dpi(pageinfo),
             options.oversample or 0.0,
         )
     )
@@ -242,7 +250,7 @@ def get_canvas_square_dpi(pageinfo, options) -> Resolution:
         max(
             (pageinfo.dpi.x) or VECTOR_PAGE_DPI,
             (pageinfo.dpi.y) or VECTOR_PAGE_DPI,
-            VECTOR_PAGE_DPI if pageinfo.has_vector else 0.0,
+            _vector_page_dpi(pageinfo),
             options.oversample or 0.0,
         )
     )
@@ -332,8 +340,10 @@ def rasterize_preview(input_file: Path, page_context: PageContext):
         output_file=output_file,
         raster_device='jpeggray',
         raster_dpi=canvas_dpi,
-        page_dpi=page_dpi,
         pageno=page_context.pageinfo.pageno + 1,
+        page_dpi=page_dpi,
+        rotation=0,
+        filter_vector=False,
     )
     return output_file
 
@@ -433,7 +443,7 @@ def rasterize(
 
     device = colorspaces[device_idx]
 
-    log.debug(f"Rasterize with {device}")
+    log.debug(f"Rasterize with {device}, rotation {correction}")
 
     # Produce the page image with square resolution or else deskew and OCR
     # will not work properly.
@@ -473,7 +483,12 @@ def preprocess_deskew(input_file: Path, page_context: PageContext):
 def preprocess_clean(input_file: Path, page_context: PageContext):
     output_file = page_context.get_path('pp_clean.png')
     dpi = get_page_square_dpi(page_context.pageinfo, page_context.options)
-    unpaper.clean(input_file, output_file, dpi.x, page_context.options.unpaper_args)
+    unpaper.clean(
+        input_file,
+        output_file,
+        dpi=dpi.x,
+        unpaper_args=page_context.options.unpaper_args,
+    )
     return output_file
 
 
@@ -577,7 +592,9 @@ def create_visible_page_jpg(image: Path, page_context: PageContext) -> Path:
     return output_file
 
 
-def create_pdf_page_from_image(image: Path, page_context: PageContext):
+def create_pdf_page_from_image(
+    image: Path, page_context: PageContext, orientation_correction
+):
     # We rasterize a square DPI version of each page because most image
     # processing tools don't support rectangular DPI. Use the square DPI as it
     # accurately describes the image. It would be possible to resample the image
@@ -585,28 +602,42 @@ def create_pdf_page_from_image(image: Path, page_context: PageContext):
     # except that the hocr renderer does not understand non-square DPI. The
     # sandwich renderer would be fine.
     output_file = page_context.get_path('visible.pdf')
-    dpi = get_page_square_dpi(page_context.pageinfo, page_context.options)
-    layout_fun = img2pdf.get_fixed_dpi_layout_fun(dpi)
+
+    pageinfo = page_context.pageinfo
+    pagesize = 72.0 * float(pageinfo.width_inches), 72.0 * float(pageinfo.height_inches)
+    effective_rotation = (pageinfo.rotation - orientation_correction) % 360
+    if effective_rotation % 180 == 90:
+        pagesize = pagesize[1], pagesize[0]
 
     # This create a single page PDF
     with open(image, 'rb') as imfile, open(output_file, 'wb') as pdf:
         log.debug('convert')
+
+        layout_fun = img2pdf.get_layout_fun(pagesize)
         img2pdf.convert(
             imfile, with_pdfrw=False, layout_fun=layout_fun, outputstream=pdf
         )
         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
 
 
 def render_hocr_page(hocr: Path, page_context: PageContext):
+    options = page_context.options
     output_file = page_context.get_path('ocr_hocr.pdf')
-    dpi = get_page_square_dpi(page_context.pageinfo, page_context.options)
-    hocrtransform = HocrTransform(hocr, dpi.x)  # square
+    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.to_pdf(
-        output_file,
+        out_filename=output_file,
         image_filename=None,
-        show_bounding_boxes=False,
-        invisible_text=True,
+        show_bounding_boxes=False if not debug_mode else True,
+        invisible_text=True if not debug_mode else False,
         interword_spaces=True,
     )
     return output_file
@@ -655,11 +686,6 @@ def get_docinfo(base_pdf: pikepdf.Pdf, context: PdfContext) -> Dict[str, str]:
 
     pdfmark['/Creator'] = f'{PROGRAM_NAME} {VERSION} / {creator_tag}'
     pdfmark['/Producer'] = f'pikepdf {pikepdf.__version__}'
-    if 'OCRMYPDF_CREATOR' in os.environ:
-        pdfmark['/Creator'] = os.environ['OCRMYPDF_CREATOR']
-    if 'OCRMYPDF_PRODUCER' in os.environ:
-        pdfmark['/Producer'] = os.environ['OCRMYPDF_PRODUCER']
-
     pdfmark['/ModDate'] = encode_pdf_date(datetime.now(timezone.utc))
     return pdfmark
 
@@ -707,6 +733,11 @@ def convert_to_pdfa(input_pdf: Path, input_ps_stub: Path, context: PdfContext):
         output_file=output_file,
         compression=options.pdfa_image_compression,
         pdfa_part=options.output_type[-1],  # is pdfa-1, pdfa-2, or pdfa-3
+        progressbar_class=(
+            context.plugin_manager.hook.get_progressbar_class()
+            if options.progress_bar
+            else None
+        ),
     )
 
     return output_file
@@ -719,6 +750,24 @@ def should_linearize(working_file: Path, context: PdfContext):
     return False
 
 
+def get_pdf_save_settings(output_type: str):
+    if output_type == 'pdfa-1':
+        # Trigger recompression to ensure object streams are removed, because
+        # Acrobat complains about them in PDF/A-1b validation.
+        return dict(
+            preserve_pdfa=True,
+            compress_streams=True,
+            stream_decode_level=pikepdf.StreamDecodeLevel.generalized,
+            object_stream_mode=pikepdf.ObjectStreamMode.disable,
+        )
+    else:
+        return dict(
+            preserve_pdfa=True,
+            compress_streams=True,
+            object_stream_mode=(pikepdf.ObjectStreamMode.generate),
+        )
+
+
 def metadata_fixup(working_file: Path, context: PdfContext):
     output_file = context.get_path('metafix.pdf')
     options = context.options
@@ -749,25 +798,21 @@ def metadata_fixup(working_file: Path, context: PdfContext):
             if 'xmp:CreateDate' not in meta:
                 meta['xmp:CreateDate'] = meta.get('xmp:ModifyDate', '')
 
-            # Ghostscript likes to set title to Untitled if omitted from input.
-            # Reverse this, because PDF/A TechNote 0003:Metadata in PDF/A-1
-            # and the XMP Spec do not make this recommendation.
-            if meta.get('dc:title') == 'Untitled':
-                with original.open_metadata(
-                    set_pikepdf_as_editor=False, update_docinfo=False
-                ) as original_meta:
-                    if 'dc:title' not in original_meta:
+            with original.open_metadata(
+                set_pikepdf_as_editor=False, update_docinfo=False, strict=False
+            ) as meta_original:
+                if meta.get('dc:title') == 'Untitled':
+                    # Ghostscript likes to set title to Untitled if omitted from input.
+                    # Reverse this, because PDF/A TechNote 0003:Metadata in PDF/A-1
+                    # and the XMP Spec do not make this recommendation.
+                    if 'dc:title' not in meta_original:
                         del meta['dc:title']
-
-            meta_original = original.open_metadata()
-            missing = set(meta_original.keys()) - set(meta.keys())
-            report_on_metadata(missing)
+                missing = set(meta_original.keys()) - set(meta.keys())
+                report_on_metadata(missing)
 
         pdf.save(
             output_file,
-            compress_streams=True,
-            preserve_pdfa=True,
-            object_stream_mode=pikepdf.ObjectStreamMode.generate,
+            **get_pdf_save_settings(options.output_type),
             linearize=(  # Don't linearize if optimize() will be linearizing too
                 should_linearize(working_file, context)
                 if options.optimize == 0
@@ -778,23 +823,37 @@ def metadata_fixup(working_file: Path, context: PdfContext):
     return output_file
 
 
-def optimize_pdf(input_file: Path, context: PdfContext):
+def optimize_pdf(input_file: Path, context: PdfContext, executor: Executor):
     output_file = context.get_path('optimize.pdf')
     save_settings = dict(
-        compress_streams=True,
-        preserve_pdfa=True,
-        object_stream_mode=pikepdf.ObjectStreamMode.generate,
         linearize=should_linearize(input_file, context),
+        **get_pdf_save_settings(context.options.output_type),
     )
-    optimize(input_file, output_file, context, save_settings)
+    optimize(input_file, output_file, context, save_settings, executor)
     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 page_num, txt_file in enumerate(txt_files):
-            if page_num != 0:
+        for (frm, to), txt_file in enumerate_compress_ranges(txt_files):
+            if frm != 1:
                 stream.write('\f')  # Form feed between pages
             if txt_file:
                 with open(txt_file, 'r', encoding="utf-8") as in_:
@@ -807,7 +866,11 @@ def merge_sidecars(txt_files: Iterable[Optional[Path]], context: PdfContext):
                     else:
                         stream.write(txt)
             else:
-                stream.write(f'[OCR skipped on page {(page_num + 1)}]')
+                if frm != to:
+                    pages = f'{frm}-{to}'
+                else:
+                    pages = f'{frm}'
+                stream.write(f'[OCR skipped on page(s) {pages}]')
     return output_file
 
 

src/ocrmypdf/_plugin_manager.py

@@ -8,13 +8,14 @@
 import argparse
 import importlib
 import importlib.util
+import pkgutil
 import sys
-from functools import partial
 from pathlib import Path
-from typing import Callable, List, Tuple, Union
+from typing import List, Tuple, Union
 
 import pluggy
 
+import ocrmypdf.builtin_plugins
 from ocrmypdf import pluginspec
 from ocrmypdf.cli import get_parser, plugins_only_parser
 
@@ -30,62 +31,71 @@ class OcrmypdfPluginManager(pluggy.PluginManager):
     """
 
     def __init__(
-        self, *args, setup_func: Callable[[pluggy.PluginManager], None], **kwargs
+        self,
+        *args,
+        plugins: List[Union[str, Path]],
+        builtins: bool = True,
+        **kwargs,
     ):
-        self._init_args = args
-        self._setup_func = setup_func
-        self._init_kwargs = kwargs
+        self.__init_args = args
+        self.__init_kwargs = kwargs
+        self.__plugins = plugins
+        self.__builtins = builtins
         super().__init__(*args, **kwargs)
-        setup_func(self)
+        self.setup_plugins()
 
     def __getstate__(self):
         state = dict(
-            _init_args=self._init_args,
-            _setup_func=self._setup_func,
-            _init_kwargs=self._init_kwargs,
+            init_args=self.__init_args,
+            plugins=self.__plugins,
+            builtins=self.__builtins,
+            init_kwargs=self.__init_kwargs,
         )
         return state
 
     def __setstate__(self, state):
         self.__init__(
-            *state['_init_args'],
-            setup_func=state['_setup_func'],
-            **state['_init_kwargs'],
+            *state['init_args'],
+            plugins=state['plugins'],
+            builtins=state['builtins'],
+            **state['init_kwargs'],
         )
 
+    def setup_plugins(self):
+        self.add_hookspecs(pluginspec)
 
-def _setup_plugins(
-    pm: pluggy.PluginManager, plugins: List[Union[str, Path]], builtins: bool = True
-):
-    pm.add_hookspecs(pluginspec)
+        # 1. Register builtins
+        if self.__builtins:
+            for module in sorted(
+                pkgutil.iter_modules(ocrmypdf.builtin_plugins.__path__)
+            ):
+                name = f'ocrmypdf.builtin_plugins.{module.name}'
+                module = importlib.import_module(name)
+                self.register(module)
 
-    all_plugins: List[Union[str, Path]] = []
-    if builtins:
-        all_plugins.extend(
-            [
-                'ocrmypdf.builtin_plugins.ghostscript',
-                'ocrmypdf.builtin_plugins.tesseract_ocr',
-            ]
-        )
-    all_plugins.extend(plugins)
-    for name in all_plugins:
-        if isinstance(name, Path) or name.endswith('.py'):
-            # Import by filename
-            module_name = Path(name).stem
-            spec = importlib.util.spec_from_file_location(module_name, name)
-            module = importlib.util.module_from_spec(spec)
-            sys.modules[module_name] = module
-            spec.loader.exec_module(module)
-        else:
-            # Import by dotted module name
-            module = importlib.import_module(name)
-        pm.register(module)
+        # 2. Register setuptools plugins
+        self.load_setuptools_entrypoints('ocrmypdf')
+
+        # 3. Register plugins specified on command line
+        for name in self.__plugins:
+            if isinstance(name, Path) or name.endswith('.py'):
+                # Import by filename
+                module_name = Path(name).stem
+                spec = importlib.util.spec_from_file_location(module_name, name)
+                module = importlib.util.module_from_spec(spec)
+                sys.modules[module_name] = module
+                spec.loader.exec_module(module)
+            else:
+                # Import by dotted module name
+                module = importlib.import_module(name)
+            self.register(module)
 
 
-def get_plugin_manager(plugins: List[str], builtins=True):
+def get_plugin_manager(plugins: List[Union[str, Path]], builtins=True):
     pm = OcrmypdfPluginManager(
         project_name='ocrmypdf',
-        setup_func=partial(_setup_plugins, plugins=plugins, builtins=builtins),
+        plugins=plugins,
+        builtins=builtins,
     )
     return pm
 

src/ocrmypdf/_sync.py

@@ -15,10 +15,9 @@ from pathlib import Path
 from tempfile import mkdtemp
 from typing import List, NamedTuple, Optional, Tuple
 
-import pikepdf
 import PIL
 
-from ocrmypdf._concurrent import exec_progress_pool
+from ocrmypdf._concurrent import Executor, setup_executor
 from ocrmypdf._graft import OcrGrafter
 from ocrmypdf._jobcontext import PageContext, PdfContext, cleanup_working_files
 from ocrmypdf._logging import PageNumberFilter
@@ -55,6 +54,7 @@ from ocrmypdf._validation import (
 )
 from ocrmypdf.exceptions import ExitCode, ExitCodeException
 from ocrmypdf.helpers import (
+    NeverRaise,
     available_cpu_count,
     check_pdf,
     pikepdf_enable_mmap,
@@ -65,7 +65,7 @@ from ocrmypdf.pdfa import file_claims_pdfa
 log = logging.getLogger(__name__)
 
 
-class PageResult(NamedTuple):
+class PageResult(NamedTuple):  # pylint: disable=inherit-non-class
     pageno: int
     pdf_page_from_image: Optional[Path]
     ocr: Optional[Path]
@@ -203,15 +203,16 @@ def exec_page_sync(page_context: PageContext):
         if filtered_image:
             visible_image_out = filtered_image
         pdf_page_from_image_out = create_pdf_page_from_image(
-            visible_image_out, page_context
+            visible_image_out, page_context, orientation_correction
         )
 
-    if options.pdf_renderer == 'hocr':
+    if options.pdf_renderer.startswith('hocr'):
         (hocr_out, text_out) = ocr_engine_hocr(ocr_image_out, page_context)
         ocr_out = render_hocr_page(hocr_out, page_context)
-
-    if options.pdf_renderer == 'sandwich':
+    elif options.pdf_renderer == 'sandwich':
         (ocr_out, text_out) = ocr_engine_textonly_pdf(ocr_image_out, page_context)
+    else:
+        raise NotImplementedError(f"pdf_renderer {options.pdf_renderer}")
 
     return PageResult(
         pageno=page_context.pageno,
@@ -222,14 +223,14 @@ def exec_page_sync(page_context: PageContext):
     )
 
 
-def post_process(pdf_file, context: PdfContext):
+def post_process(pdf_file, context: PdfContext, executor: Executor):
     pdf_out = pdf_file
     if context.options.output_type.startswith('pdfa'):
         ps_stub_out = generate_postscript_stub(context)
         pdf_out = convert_to_pdfa(pdf_out, ps_stub_out, context)
 
     pdf_out = metadata_fixup(pdf_out, context)
-    return optimize_pdf(pdf_out, context)
+    return optimize_pdf(pdf_out, context, executor)
 
 
 def worker_init(max_pixels: int):
@@ -240,11 +241,12 @@ def worker_init(max_pixels: int):
     pikepdf_enable_mmap()
 
 
-def exec_concurrent(context: PdfContext):
+def exec_concurrent(context: PdfContext, executor: Executor):
     """Execute the pipeline concurrently"""
 
     # Run exec_page_sync on every page context
-    max_workers = min(len(context.pdfinfo), context.options.jobs)
+    options = context.options
+    max_workers = min(len(context.pdfinfo), options.jobs)
     if max_workers > 1:
         log.info("Start processing %d pages concurrently", max_workers)
 
@@ -266,46 +268,47 @@ def exec_concurrent(context: PdfContext):
         finally:
             tls.pageno = None
 
-    exec_progress_pool(
-        use_threads=context.options.use_threads,
+    executor(
+        use_threads=options.use_threads,
         max_workers=max_workers,
         tqdm_kwargs=dict(
             total=(2 * len(context.pdfinfo)),
-            desc='OCR',
+            desc='OCR' if options.tesseract_timeout > 0 else 'Image processing',
             unit='page',
             unit_scale=0.5,
-            disable=not context.options.progress_bar,
+            disable=not options.progress_bar,
         ),
-        task_initializer=partial(worker_init, PIL.Image.MAX_IMAGE_PIXELS),
+        worker_initializer=partial(worker_init, PIL.Image.MAX_IMAGE_PIXELS),
         task=exec_page_sync,
         task_arguments=context.get_page_contexts(),
         task_finished=update_page,
     )
 
     # Output sidecar text
-    if context.options.sidecar:
+    if options.sidecar:
         text = merge_sidecars(sidecars, context)
         # Copy text file to destination
-        copy_final(text, context.options.sidecar, context)
+        copy_final(text, options.sidecar, context)
 
     # Merge layers to one single pdf
     pdf = ocrgraft.finalize()
 
     # PDF/A and metadata
     log.info("Postprocessing...")
-    pdf = post_process(pdf, context)
+    pdf = post_process(pdf, context, executor)
 
     # Copy PDF file to destination
-    copy_final(pdf, context.options.output_file, context)
+    copy_final(pdf, options.output_file, context)
 
 
-class NeverRaise(Exception):
-    """An exception that is never raised"""
+def configure_debug_logging(log_filename: Path, prefix: str = ''):
+    """
+    Create a debug log file at a specified location.
 
-    pass  # pylint: disable=unnecessary-pass
-
-
-def configure_debug_logging(log_filename, prefix=''):
+    Arguments:
+        log_filename: Where to the put the log file.
+        prefix: The logging domain prefix that should be sent to the log.
+    """
     log_file_handler = logging.FileHandler(log_filename, delay=True)
     log_file_handler.setLevel(logging.DEBUG)
     formatter = logging.Formatter(
@@ -326,15 +329,23 @@ def run_pipeline(options, *, plugin_manager, api=False):
     if not plugin_manager:
         plugin_manager = get_plugin_manager(options.plugins)
 
-    work_folder = Path(mkdtemp(prefix="com.github.ocrmypdf."))
+    work_folder = Path(mkdtemp(prefix="ocrmypdf.io."))
     debug_log_handler = None
-    if (options.keep_temporary_files or options.verbose >= 1) and not os.environ.get(
-        'PYTEST_CURRENT_TEST', ''
+    if (
+        (options.keep_temporary_files or options.verbose >= 1)
+        and not os.environ.get('PYTEST_CURRENT_TEST', '')
+        and not api
     ):
-        debug_log_handler = configure_debug_logging(Path(work_folder) / "debug.log")
+        # Debug log for command line interface only with verbose output
+        # See https://github.com/pytest-dev/pytest/issues/5502 for why we skip this
+        # when pytest is running
+        debug_log_handler = configure_debug_logging(
+            Path(work_folder) / "debug.log"
+        )  # pragma: no cover
 
     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)
@@ -347,6 +358,7 @@ def run_pipeline(options, *, plugin_manager, api=False):
         # Gather pdfinfo and create context
         pdfinfo = get_pdfinfo(
             origin_pdf,
+            executor=executor,
             detailed_analysis=options.redo_ocr,
             progbar=options.progress_bar,
             max_workers=options.jobs if not options.use_threads else 1,  # To help debug
@@ -359,7 +371,7 @@ def run_pipeline(options, *, plugin_manager, api=False):
         validate_pdfinfo_options(context)
 
         # Execute the pipeline
-        exec_concurrent(context)
+        exec_concurrent(context, executor)
 
         if options.output_file == '-':
             log.info("Output sent to stdout")

src/ocrmypdf/_validation.py

@@ -13,7 +13,7 @@ import sys
 import unicodedata
 from pathlib import Path
 from shutil import copyfileobj
-from typing import Tuple
+from typing import List, Set, Tuple, Union
 
 import pikepdf
 import PIL
@@ -78,7 +78,7 @@ def check_options_languages(options, ocr_engine_languages):
 def check_options_output(options):
     is_latin = options.languages.issubset(HOCR_OK_LANGS)
 
-    if options.pdf_renderer == 'hocr' and not is_latin:
+    if options.pdf_renderer.startswith('hocr') and not is_latin:
         msg = (
             "The 'hocr' PDF renderer is known to cause problems with one "
             "or more of the languages in your document.  Use "
@@ -112,6 +112,10 @@ def check_options_sidecar(options):
                 "--sidecar filename must be specified when output file is stdout."
             )
         options.sidecar = options.output_file + '.txt'
+    if options.sidecar == options.input_file or options.sidecar == options.output_file:
+        raise BadArgsError(
+            "--sidecar file must be different from the input and output files"
+        )
 
 
 def check_options_preprocessing(options):
@@ -133,13 +137,13 @@ def check_options_preprocessing(options):
                     options.unpaper_args
                 )
         except Exception as e:
-            raise BadArgsError(str(e))
+            raise BadArgsError("--unpaper-args: " + str(e)) from e
 
 
-def _pages_from_ranges(ranges):
+def _pages_from_ranges(ranges: str) -> Set[int]:
     if is_iterable_notstr(ranges):
         return set(ranges)
-    pages = []
+    pages: List[int] = []
     page_groups = ranges.replace(' ', '').split(',')
     for g in page_groups:
         if not g:
@@ -150,9 +154,18 @@ def _pages_from_ranges(ranges):
             pages.append(int(g) - 1)
         else:
             try:
-                pages.extend(range(int(start) - 1, int(end)))
+                new_pages = list(range(int(start) - 1, int(end)))
+                if not new_pages:
+                    raise BadArgsError(f"invalid page subrange '{start}-{end}'")
+                pages.extend(new_pages)
             except ValueError:
-                raise BadArgsError("invalid page range")
+                raise BadArgsError("invalid page range") from None
+
+    if not pages:
+        raise BadArgsError(
+            f"The string of page ranges '{ranges}' did not contain any recognizable "
+            f"page ranges."
+        )
 
     if not monotonic(pages):
         log.warning(
@@ -175,8 +188,6 @@ def check_options_ocr_behavior(options):
     )
     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)
 
@@ -213,12 +224,12 @@ def check_options_optimizing(options):
 
 
 def check_options_advanced(options):
-    if options.pdfa_image_compression != 'auto' and options.output_type.startswith(
+    if options.pdfa_image_compression != 'auto' and not options.output_type.startswith(
         'pdfa'
     ):
         log.warning(
-            "--pdfa-image-compression argument has no effect when "
-            "--output-type is not 'pdfa', 'pdfa-1', or 'pdfa-2'"
+            "--pdfa-image-compression argument only applies when "
+            "--output-type is one of 'pdfa', 'pdfa-1', or 'pdfa-2'"
         )
 
 

src/ocrmypdf/api.py

@@ -8,11 +8,17 @@
 import logging
 import os
 import sys
+import threading
 from enum import IntEnum
+from io import IOBase
 from pathlib import Path
-from typing import BinaryIO, Iterable, Union
+from typing import AnyStr, BinaryIO, Iterable, Optional, Union
+from warnings import warn
 
-from ocrmypdf._logging import PageNumberFilter, TqdmConsole
+from ocrmypdf._logging import (  # pylint: disable=unused-import
+    PageNumberFilter,
+    TqdmConsole,
+)
 from ocrmypdf._plugin_manager import get_plugin_manager
 from ocrmypdf._sync import run_pipeline
 from ocrmypdf._validation import check_options
@@ -25,7 +31,10 @@ except ModuleNotFoundError:
     coloredlogs = None
 
 
-PathOrIO = Union[BinaryIO, os.PathLike, str, bytes]
+StrPath = Union[os.PathLike, AnyStr]
+PathOrIO = Union[BinaryIO, StrPath]
+
+_api_lock = threading.Lock()
 
 
 class Verbosity(IntEnum):
@@ -39,29 +48,45 @@ class Verbosity(IntEnum):
 
 def configure_logging(
     verbosity: Verbosity,
+    *,
     progress_bar_friendly: bool = True,
     manage_root_logger: bool = False,
+    plugin_manager=None,
 ):
     """Set up logging.
 
-    Library users may wish to use this function if they want their log output to be
-    similar to ocrmypdf command line interface. If not used, the external application
-    should configure logging on its own.
+    Before calling :func:`ocrmypdf.ocr()`, you can use this function to
+    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.
 
-    ocrmypdf will perform all of its logging under the ``"ocrmypdf"`` logging namespace.
-    In addition, ocrmypdf imports pdfminer, which logs under ``"pdfminer"``. A library
-    user may wish to configure both; note that pdfminer is extremely chatty at the log
-    level ``logging.INFO``.
+    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
+    the Python standard library's logging module. If this function is called,
+    the caller may of course make further adjustments to logging.
 
-    Library users may perform additional configuration afterwards.
+    Regardless of whether this function is called, ocrmypdf will perform all of
+    its logging under the ``"ocrmypdf"`` logging namespace. In addition,
+    ocrmypdf imports pdfminer, which logs under ``"pdfminer"``. A library user
+    may wish to configure both; note that pdfminer is extremely chatty at the
+    log level ``logging.INFO``.
+
+    This function does not set up the ``debug.log`` log file that the command
+    line interface does at certain verbosity levels. Applications should configure
+    their own debug logging.
 
     Args:
-        verbosity (Verbosity): Verbosity level.
-        progress_bar_friendly (bool): Install the TqdmConsole log handler, which is
-            compatible with the tqdm progress bar; without this log messages will
-            overwrite the progress bar
-        manage_root_logger (bool): Configure the process's root logger, to ensure
-            all log output is sent through
+        verbosity: Verbosity level.
+        progress_bar_friendly: If True (the default), install a custom log handler
+            that is compatible with progress bars and colored output.
+        manage_root_logger: Configure the process's root logger.
+        plugin_manager: The plugin manager, used for obtaining the custom log handler.
 
     Returns:
         The toplevel logger for ocrmypdf (or the root logger, if we are managing it).
@@ -72,9 +97,11 @@ def configure_logging(
     log = logging.getLogger(prefix)
     log.setLevel(logging.DEBUG)
 
-    if progress_bar_friendly:
-        console = logging.StreamHandler(stream=TqdmConsole(sys.stderr))
-    else:
+    console = None
+    if plugin_manager and progress_bar_friendly:
+        console = plugin_manager.hook.get_logging_console()
+
+    if not console:
         console = logging.StreamHandler(stream=sys.stderr)
 
     if verbosity < 0:
@@ -160,14 +187,14 @@ def create_options(
         else:
             raise TypeError(f"{arg}: {val} ({type(val)})")
 
-    try:
-        cmdline.append(os.fspath(input_file))
-    except TypeError:
+    if isinstance(input_file, (BinaryIO, IOBase)):
         cmdline.append('stream://input_file')
-    try:
-        cmdline.append(os.fspath(output_file))
-    except TypeError:
+    else:
+        cmdline.append(os.fspath(input_file))
+    if isinstance(output_file, (BinaryIO, IOBase)):
         cmdline.append('stream://output_file')
+    else:
+        cmdline.append(os.fspath(output_file))
 
     parser._api_mode = True
     options = parser.parse_args(cmdline)
@@ -189,7 +216,7 @@ def ocr(  # pylint: disable=unused-argument
     language: Iterable[str] = None,
     image_dpi: int = None,
     output_type=None,
-    sidecar: os.PathLike = None,
+    sidecar: Optional[StrPath] = None,
     jobs: int = None,
     use_threads: bool = None,
     title: str = None,
@@ -226,7 +253,8 @@ def ocr(  # pylint: disable=unused-argument
     user_words: os.PathLike = None,
     user_patterns: os.PathLike = None,
     fast_web_view: float = None,
-    plugins: Iterable[str] = None,
+    plugins: Iterable[StrPath] = None,
+    plugin_manager=None,
     keep_temporary_files: bool = None,
     progress_bar: bool = None,
     **kwargs,
@@ -248,7 +276,7 @@ def ocr(  # pylint: disable=unused-argument
             read.
         output_file: If a :class:`pathlib.Path`, ``str`` or ``bytes``, this is
             interpreted as file system path to the output file. If the object
-            appears to be a writable stream (with methods such as ``.read()`` and
+            appears to be a writable stream (with methods such as ``.write()`` and
             ``.seek()``), the output will be written to this stream. If
             ``output_file`` is ``"-"``, the output will be written to ``sys.stdout``
             (provided that standard output does not seem to be a terminal device).
@@ -278,20 +306,35 @@ 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)):
+        plugins = [plugins]
     else:
         plugins = list(plugins)
 
-    parser = get_parser()
-    _plugin_manager = get_plugin_manager(plugins)
-    _plugin_manager.hook.add_options(parser=parser)  # pylint: disable=no-member
-
-    create_options_kwargs = {
-        k: v for k, v in locals().items() if not k.startswith('_') and k != 'kwargs'
-    }
+    # No new variable names should be assigned until these two steps are run
+    create_options_kwargs = {k: v for k, v in locals().items() if k != 'kwargs'}
     create_options_kwargs.update(kwargs)
 
-    options = create_options(**create_options_kwargs)
-    check_options(options, _plugin_manager)
-    return run_pipeline(options=options, plugin_manager=_plugin_manager, api=True)
+    parser = get_parser()
+    create_options_kwargs['parser'] = parser
+
+    with _api_lock:
+        # We can't allow multiple ocrmypdf.ocr() threads to run in parallel, because
+        # 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.hook.add_options(parser=parser)  # pylint: disable=no-member
+
+        if 'verbose' in kwargs:
+            warn("ocrmypdf.ocr(verbose=) is ignored. Use ocrmypdf.configure_logging().")
+
+        options = create_options(**create_options_kwargs)
+        check_options(options, plugin_manager)
+        return run_pipeline(options=options, plugin_manager=plugin_manager, api=True)

src/ocrmypdf/builtin_plugins/__init__.py

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

src/ocrmypdf/builtin_plugins/concurrency.py

@@ -0,0 +1,172 @@
+# © 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 contextlib import suppress
+from multiprocessing import Pool as ProcessPool
+from multiprocessing.pool import ThreadPool
+from typing import Callable, Iterable, 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
+
+Queue = Union[multiprocessing.Queue, queue.Queue]
+
+
+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: Callable[[], None], 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
+        # 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(_queue: Queue, user_init: Callable[[], None], _loglevel):
+    # 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(-1)
+            pool_class = ThreadPool
+            initializer = thread_init
+        else:
+            log_queue = multiprocessing.Queue(-1)
+            pool_class = ProcessPool
+            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:
+            pool = pool_class(
+                processes=max_workers,
+                initializer=initializer,
+                initargs=(log_queue, worker_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()
+
+
+@hookimpl
+def get_executor(progressbar_class):
+    return StandardExecutor(pbar_class=progressbar_class)
+
+
+@hookimpl
+def get_progressbar_class():
+    return tqdm
+
+
+@hookimpl
+def get_logging_console():
+    return logging.StreamHandler(stream=TqdmConsole(sys.stderr))

src/ocrmypdf/builtin_plugins/default_filters.py

@@ -0,0 +1,14 @@
+# © 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

src/ocrmypdf/builtin_plugins/ghostscript.py

@@ -61,9 +61,9 @@ def rasterize_pdf_page(
     raster_device,
     raster_dpi,
     pageno,
-    page_dpi=None,
-    rotation=None,
-    filter_vector=False,
+    page_dpi,
+    rotation,
+    filter_vector,
 ):
     ghostscript.rasterize_pdf(
         input_file,
@@ -79,12 +79,21 @@ def rasterize_pdf_page(
 
 
 @hookimpl
-def generate_pdfa(pdf_pages, pdfmark, output_file, compression, pdf_version, pdfa_part):
+def generate_pdfa(
+    pdf_pages,
+    pdfmark,
+    output_file,
+    compression,
+    pdf_version,
+    pdfa_part,
+    progressbar_class,
+):
     ghostscript.generate_pdfa(
         pdf_pages=[*pdf_pages, pdfmark],
         output_file=output_file,
         compression=compression,
         pdf_version=pdf_version,
         pdfa_part=pdfa_part,
+        progressbar_class=progressbar_class,
     )
     return output_file

src/ocrmypdf/builtin_plugins/tesseract_ocr.py

@@ -80,20 +80,14 @@ def check_options(options):
         program='tesseract',
         package={'linux': 'tesseract-ocr'},
         version_checker=tesseract.version,
-        need_version='4.0.0',  # using backport for Travis CI
+        need_version='4.0.0-beta.1',  # using backport for Travis CI
+        version_parser=tesseract.TesseractVersion,
     )
 
     # Decide on what renderer to use
     if options.pdf_renderer == 'auto':
         options.pdf_renderer = 'sandwich'
 
-    if options.pdf_renderer == 'sandwich' and not tesseract.has_textonly_pdf(
-        set(options.languages)
-    ):
-        raise MissingDependencyError(
-            "You are using an alpha version of Tesseract 4.0 that does not support "
-            "the textonly_pdf parameter. We don't support versions this old."
-        )
     if not tesseract.has_user_words() and (options.user_words or options.user_patterns):
         log.warning(
             "Tesseract 4.0 ignores --user-words and --user-patterns, so these "
@@ -121,9 +115,7 @@ def validate(pdfinfo, options):
         os.environ['OMP_THREAD_LIMIT'] = str(tess_threads)
     else:
         tess_threads = int(os.environ['OMP_THREAD_LIMIT'])
-
-    if tess_threads > 1:
-        log.info("Using Tesseract OpenMP thread limit %d", tess_threads)
+    log.debug("Using Tesseract OpenMP thread limit %d", tess_threads)
 
 
 class TesseractOcrEngine(OcrEngine):

src/ocrmypdf/cli.py

@@ -6,12 +6,15 @@
 
 
 import argparse
+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')
 
-def numeric(basetype, min_=None, max_=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
@@ -164,7 +167,8 @@ Online documentation is located at:
         metavar='FILE',
         help="Generate sidecar text files that contain the same text recognized "
         "by Tesseract. This may be useful for building a OCR text database. "
-        "If FILE is omitted, the sidecar file be named {output_file}.txt "
+        "If FILE is omitted, the sidecar file be named {output_file}.txt; the next "
+        "argument must NOT be the name of the input PDF. "
         "If FILE is set to '-', the sidecar is written to stdout (a "
         "convenient way to preview OCR quality). The output file and sidecar "
         "may not both use stdout at the same time.",
@@ -407,7 +411,7 @@ Online documentation is located at:
     )
     advanced.add_argument(
         '--pdf-renderer',
-        choices=['auto', 'hocr', 'sandwich'],
+        choices=['auto', 'hocr', 'sandwich', 'hocrdebug'],
         default='auto',
         help="Choose OCR PDF renderer - the default option is to let OCRmyPDF "
         "choose.  See documentation for discussion.",

src/ocrmypdf/extra_plugins/awslambda.py

@@ -0,0 +1,193 @@
+# © 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/.
+
+
+"""Alternate executor to support OCRmyPDF in AWS Lambda.
+
+AWS Lambda does not support the standard multiprocessing module because it
+environment.
+
+This alternate executor avoids that. However, it has drawbacks. Most notably,
+it divvies up work among worker processes at the beginning, to avoid coordinating
+effort in ways that require shared queues and semaphores. In this implementation,
+there is no shared queue, so no possible lock contention between workers. The
+main process has a shared pipe with each worker.
+
+If some tasks are larger than others, some workers will may fall far behind
+while others have deep queues. The last worker may end up with fewer tasks.
+"""
+
+
+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
+from unittest.mock import Mock
+
+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:
+        super().__init__(None)
+        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 = []
+        connections = []
+        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):
+                    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

src/ocrmypdf/helpers.py

@@ -15,9 +15,9 @@ from collections.abc import Iterable
 from contextlib import suppress
 from functools import wraps
 from io import StringIO
-from math import isclose
+from math import isclose, isfinite
 from pathlib import Path
-from typing import Any, Sequence, TypeVar
+from typing import Any, Sequence
 
 import pikepdf
 
@@ -39,6 +39,10 @@ class Resolution(namedtuple('Resolution', ('x', 'y'))):
     def is_square(self) -> bool:
         return isclose(self.x, self.y, rel_tol=1e-3)
 
+    @property
+    def is_finite(self) -> bool:
+        return isfinite(self.x) and isfinite(self.y)
+
     def take_max(self, vals, yvals=None):
         if yvals is not None:
             return Resolution(max(self.x, *vals), max(self.y, *yvals))
@@ -54,13 +58,22 @@ class Resolution(namedtuple('Resolution', ('x', 'y'))):
     def __str__(self):
         return f"{self.x:f}x{self.y:f}"
 
-    def __repr__(self):
+    def __repr__(self):  # pragma: no cover
         return f"Resolution({self.x}x{self.y} dpi)"
 
 
+class NeverRaise(Exception):
+    """An exception that is never raised"""
+
+
 def safe_symlink(input_file: os.PathLike, soft_link_name: os.PathLike):
-    """
-    Helper function: relinks soft symbolic link if necessary
+    """Create a symbolic link at ``soft_link_name``, which references ``input_file``.
+
+    Think of this as copying ``input_file`` to ``soft_link_name`` with less overhead.
+
+    Use symlinks safely. Self-linking loops are prevented. On Windows, file copy is
+    used since symlinks may require administrator privileges. An existing link at the
+    destination is removed.
     """
     input_file = os.fspath(input_file)
     soft_link_name = os.fspath(soft_link_name)
@@ -68,8 +81,8 @@ def safe_symlink(input_file: os.PathLike, soft_link_name: os.PathLike):
     # Guard against soft linking to oneself
     if input_file == soft_link_name:
         log.warning(
-            "No symbolic link made. You are using "
-            "the original data directory as the working directory."
+            "No symbolic link created. You are using  the original data directory "
+            "as the working directory."
         )
         return
 
@@ -78,10 +91,7 @@ def safe_symlink(input_file: os.PathLike, soft_link_name: os.PathLike):
         # do not delete or overwrite real (non-soft link) file
         if not os.path.islink(soft_link_name):
             raise FileExistsError(f"{soft_link_name} exists and is not a link")
-        try:
-            os.unlink(soft_link_name)
-        except OSError:
-            log.debug("Can't unlink %s", soft_link_name)
+        os.unlink(soft_link_name)
 
     if not os.path.exists(input_file):
         raise FileNotFoundError(f"trying to create a broken symlink to {input_file}")
@@ -110,7 +120,7 @@ def is_iterable_notstr(thing: Any) -> bool:
 
 
 def monotonic(L: Sequence) -> bool:
-    """Does list increase monotonically?"""
+    """Does this sequence increase monotonically?"""
     return all(b > a for a, b in zip(L, L[1:]))
 
 
@@ -169,49 +179,61 @@ def is_file_writable(test_file: os.PathLike) -> bool:
 def check_pdf(input_file: Path) -> bool:
     """Check if a PDF complies with the PDF specification.
 
-    Checks for proper formatting and proper linearization.
+    Checks for proper formatting and proper linearization. Uses pikepdf (which in
+    turn, uses QPDF) to perform the checks.
     """
-    pdf = None
     try:
         pdf = pikepdf.open(input_file)
     except pikepdf.PdfError as e:
         log.error(e)
         return False
     else:
-        messages = pdf.check()
-        for msg in messages:
-            if 'error' in msg.lower():
-                log.error(msg)
+        with pdf:
+            messages = pdf.check()
+            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:
+                pass
+            except (
+                # Workaround for a problematic pikepdf version
+                # pragma: no cover
+                getattr(pikepdf, 'ForeignObjectError')
+                if pikepdf.__version__ == '2.1.0'
+                else NeverRaise
+            ):
+                pass
             else:
-                log.warning(msg)
+                linearize_msgs = sio.getvalue()
+                if linearize_msgs:
+                    log.warning(linearize_msgs)
 
-        sio = StringIO()
-        linearize = None
-        try:
-            pdf.check_linearization(sio)
-        except RuntimeError:
-            pass
-        else:
-            linearize = sio.getvalue()
-            if linearize:
-                log.warning(linearize)
-
-        if not messages and not linearize:
-            return True
-        return False
-    finally:
-        if pdf:
-            pdf.close()
+            if not messages and not linearize_msgs:
+                return True
+            return False
 
 
-T = TypeVar('T')
-
-
-def clamp(n: T, smallest: T, largest: T) -> T:
-    """Clamps the value of n to between smallest and largest."""
+def clamp(n, smallest, largest):  # mypy doesn't understand types for this
+    """Clamps the value of ``n`` to between ``smallest`` and ``largest``."""
     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):
@@ -220,12 +242,13 @@ def pikepdf_enable_mmap():
     #     log.debug("pikepdf mmap not available")
     # 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.
-    log.debug("pikepdf mmap disabled")
+    # Fix is not in pybind11 2.6.0
+    # log.debug("pikepdf mmap disabled")
     return
 
 
 def deprecated(func):
-    """Warn that function is deprecated"""
+    """Warn that function is deprecated."""
 
     @wraps(func)
     def new_func(*args, **kwargs):

src/ocrmypdf/hocrtransform.py

@@ -31,18 +31,26 @@
 import argparse
 import os
 import re
-from collections import namedtuple
 from itertools import chain
 from math import atan, cos, sin
 from pathlib import Path
-from typing import Optional, Tuple, Union
+from typing import Any, NamedTuple, Optional, Tuple, Union
 from xml.etree import ElementTree
 
 from reportlab.lib.colors import black, cyan, magenta, red
 from reportlab.lib.units import inch
 from reportlab.pdfgen.canvas import Canvas
 
-Rect = namedtuple('Rect', ['x1', 'y1', 'x2', 'y2'])
+Element = ElementTree.Element
+
+
+class Rect(NamedTuple):  # pylint: disable=inherit-non-class
+    """A rectangle for managing PDF coordinates."""
+
+    x1: Any
+    y1: Any
+    x2: Any
+    y2: Any
 
 
 class HocrTransformError(Exception):
@@ -69,7 +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))
 
@@ -105,7 +113,7 @@ class HocrTransform:
         else:
             return ''
 
-    def _get_element_text(self, element):
+    def _get_element_text(self, element: Element):
         """
         Return the textual content of the element and its children
         """
@@ -119,7 +127,7 @@ class HocrTransform:
         return text
 
     @classmethod
-    def element_coordinates(cls, element) -> Rect:
+    def element_coordinates(cls, element: Element) -> Rect:
         """
         Returns a tuple containing the coordinates of the bounding box around
         an element
@@ -133,7 +141,7 @@ class HocrTransform:
         return out
 
     @classmethod
-    def baseline(cls, element) -> Tuple[float, float]:
+    def baseline(cls, element: Element) -> Tuple[float, float]:
         """
         Returns a tuple containing the baseline slope and intercept.
         """
@@ -149,7 +157,7 @@ class HocrTransform:
         """
         return Rect._make((c / self.dpi * inch) for c in pxl)
 
-    def _child_xpath(self, html_tag, html_class=None):
+    def _child_xpath(self, html_tag: str, html_class: Optional[str] = None) -> str:
         xpath = f".//{self.xmlns}{html_tag}"
         if html_class:
             xpath += f"[@class='{html_class}']"
@@ -167,10 +175,14 @@ class HocrTransform:
     def topdown_position(self, element):
         pxl_line_coords = self.element_coordinates(element)
         line_box = self.pt_from_pixel(pxl_line_coords)
-        return -line_box.y2
+        # Coordinates here are still in the hocr coordinate system, so 0 on the y axis
+        # is the top of the page and increasing values of y will move towards the
+        # bottom of the page.
+        return line_box.y2
 
     def to_pdf(
         self,
+        *,
         out_filename: Path,
         image_filename: Optional[Path] = None,
         show_bounding_boxes: bool = False,
@@ -277,13 +289,15 @@ class HocrTransform:
     def _do_line(
         self,
         pdf: Canvas,
-        line,
+        line: Optional[Element],
         elemclass: str,
         fontname: str,
         invisible_text: bool,
         interword_spaces: bool,
         show_bounding_boxes: bool,
     ):
+        if not line:
+            return
         pxl_line_coords = self.element_coordinates(line)
         line_box = self.pt_from_pixel(pxl_line_coords)
         line_height = line_box.y2 - line_box.y1
@@ -420,10 +434,10 @@ if __name__ == "__main__":
     parser.add_argument('outputfile', help='Path to the PDF file to be generated')
     args = parser.parse_args()
 
-    hocr = HocrTransform(args.hocrfile, args.resolution)
+    hocr = HocrTransform(hocr_filename=args.hocrfile, dpi=args.resolution)
     hocr.to_pdf(
-        args.outputfile,
-        args.image,
-        args.boundingboxes,
+        out_filename=args.outputfile,
+        image_filename=args.image,
+        show_bounding_boxes=args.boundingboxes,
         interword_spaces=args.interword_spaces,
     )

src/ocrmypdf/leptonica.py

@@ -15,7 +15,6 @@ import logging
 import os
 import sys
 import threading
-import warnings
 from collections import deque
 from collections.abc import Sequence
 from contextlib import suppress
@@ -24,18 +23,20 @@ from functools import lru_cache
 from io import BytesIO, UnsupportedOperation
 from os import fspath
 from tempfile import TemporaryFile
+from warnings import warn
 
 from ocrmypdf.exceptions import MissingDependencyError
 from ocrmypdf.lib._leptonica import ffi
-from ocrmypdf.subprocess import shim_paths_with_program_files
 
 # pylint: disable=protected-access
 
 logger = logging.getLogger(__name__)
 
 if os.name == 'nt':
+    from ocrmypdf.subprocess._windows import shim_env_path
+
     libname = 'liblept-5'
-    os.environ['PATH'] = shim_paths_with_program_files()
+    os.environ['PATH'] = shim_env_path()
 else:
     libname = 'lept'
 _libpath = find_library(libname)
@@ -58,9 +59,9 @@ if not _libpath:
         ---------------------------------------------------------------------
         """
     )
-if os.name == 'nt':  
-    # On Windows, recent versions of libpng require zlib. We have to make sure 
-    # the zlib version being loaded is the same one that libpng was built with. 
+if os.name == 'nt':
+    # On Windows, recent versions of libpng require zlib. We have to make sure
+    # the zlib version being loaded is the same one that libpng was built with.
     # This tries to import zlib from Tesseract's installation folder, falling back
     # to find_library() if liblept is being loaded from somewhere else.
     # Loading zlib from other places could cause a version mismatch
@@ -87,9 +88,9 @@ except ffi.error as e:
 
 class _LeptonicaErrorTrap_Redirect:
     """
-    Context manager to trap errors reported by Leptonica.
+    Context manager to trap errors reported by Leptonica < 1.79 or on Apple Silicon.
 
-    Leptonica's error return codes don't provide much informatino about what
+    Leptonica's error return codes don't provide much information about what
     went wrong. Leptonica does, however, write more detailed errors to stderr
     (provided this is not disabled at compile time). The Leptonica source
     code is very consistent in its use of macros to generate errors.
@@ -113,8 +114,10 @@ class _LeptonicaErrorTrap_Redirect:
         # Save the old stderr, and redirect stderr to temporary file
         self.leptonica_lock.acquire()
         try:
-            with suppress(AttributeError):
-                sys.stderr.flush()
+            # It would make sense to do sys.stderr.flush() here, but that can deadlock
+            # due to https://bugs.python.org/issue6721. So don't flush. Pretend
+            # there's nothing important in sys.stderr. If the user cared they would
+            # be using Leptonica 1.79 or later anyway to avoid this mess.
             self.copy_of_stderr = os.dup(sys.stderr.fileno())
             os.dup2(self.tmpfile.fileno(), sys.stderr.fileno(), inheritable=False)
         except AttributeError:
@@ -169,20 +172,6 @@ tls = threading.local()
 tls.trap = None
 
 
-@ffi.callback("void(char *)")
-def _stderr_handler(cstr):
-    msg = ffi.string(cstr).decode(errors='replace')
-    if msg.startswith("Error"):
-        logger.error(msg)
-    elif msg.startswith("Warning"):
-        logger.warning(msg)
-    else:
-        logger.debug(msg)
-    if tls.trap is not None:
-        tls.trap.append(msg)
-    return
-
-
 class _LeptonicaErrorTrap_Queue:
     def __init__(self):
         self.queue = deque()
@@ -212,9 +201,25 @@ class _LeptonicaErrorTrap_Queue:
 
 
 try:
+
+    @ffi.callback("void(char *)")
+    def _stderr_handler(cstr):
+        msg = ffi.string(cstr).decode(errors='replace')
+        if msg.startswith("Error"):
+            logger.error(msg)
+        elif msg.startswith("Warning"):
+            logger.warning(msg)
+        else:
+            logger.debug(msg)
+        if tls.trap is not None:
+            tls.trap.append(msg)
+        return
+
     lept.leptSetStderrHandler(_stderr_handler)
-except ffi.error:
+except (ffi.error, MemoryError):
     # Pre-1.79 Leptonica does not have leptSetStderrHandler
+    # And some platforms, notably Apple ARM 64, do not allow the write+execute
+    # memory needed to set up the callback function.
     _LeptonicaErrorTrap = _LeptonicaErrorTrap_Redirect
 else:
     # 1.79 have this new symbol
@@ -386,7 +391,7 @@ class Pix(LeptonicaObject):
 
     @classmethod
     def read(cls, path):
-        warnings.warn('Use Pix.open() instead', DeprecationWarning)
+        warn('Use Pix.open() instead', DeprecationWarning)
         return cls.open(path)
 
     @classmethod

src/ocrmypdf/optimize.py

@@ -9,12 +9,9 @@ import logging
 import sys
 import tempfile
 from collections import defaultdict
-from functools import partial
-from io import BytesIO
 from os import fspath
 from pathlib import Path
 from typing import (
-    Any,
     Callable,
     Dict,
     Iterator,
@@ -25,21 +22,19 @@ from typing import (
     Optional,
     Sequence,
     Tuple,
-    Union,
 )
 
 import img2pdf
 import pikepdf
 from pikepdf import Dictionary, Name, Object, Pdf, PdfImage
 from PIL import Image
-from tqdm import tqdm
 
 from ocrmypdf import leptonica
-from ocrmypdf._concurrent import exec_progress_pool
+from ocrmypdf._concurrent import Executor, SerialExecutor
 from ocrmypdf._exec import jbig2enc, pngquant
 from ocrmypdf._jobcontext import PdfContext
 from ocrmypdf.exceptions import OutputFileAccessError
-from ocrmypdf.helpers import deprecated, safe_symlink
+from ocrmypdf.helpers import safe_symlink
 
 log = logging.getLogger(__name__)
 
@@ -50,7 +45,7 @@ DEFAULT_PNG_QUALITY = 70
 Xref = NewType('Xref', int)
 
 
-class XrefExt(NamedTuple):
+class XrefExt(NamedTuple):  # pylint: disable=inherit-non-class
     xref: Xref
     ext: str
 
@@ -67,33 +62,42 @@ def jpg_name(root: Path, xref: Xref) -> Path:
     return img_name(root, xref, '.jpg')
 
 
-def tif_name(root: Path, xref: Xref) -> Path:
-    return img_name(root, xref, '.tif')
-
-
 def extract_image_filter(
     pike: Pdf, root: Path, image: Object, xref: Xref
 ) -> Optional[Tuple[PdfImage, Tuple[Name, Object]]]:
+    del pike  # unused args
+    del root
+
     if image.Subtype != Name.Image:
         return None
     if image.Length < 100:
-        log.debug("Skipping small image, xref %s", xref)
+        log.debug(f"Skipping small image, xref {xref}")
+        return None
+    if image.Width < 8 or image.Height < 8:  # Issue 732
+        log.debug(f"Skipping oddly sized image, xref {xref}")
         return None
 
     pim = PdfImage(image)
 
     if len(pim.filter_decodeparms) > 1:
-        log.debug("Skipping multiply filtered, xref %s", xref)
+        log.debug(f"Skipping multiply filtered image, xref {xref}")
         return None
     filtdp = pim.filter_decodeparms[0]
 
     if pim.bits_per_component > 8:
+        log.debug(f"Skipping wide gamut image, xref {xref}")
         return None  # Don't mess with wide gamut images
 
     if filtdp[0] == Name.JPXDecode:
+        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
 
     return pim, filtdp
@@ -102,6 +106,8 @@ def extract_image_filter(
 def extract_image_jbig2(
     *, 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
@@ -112,14 +118,29 @@ def extract_image_jbig2(
         and filtdp[0] != Name.JBIG2Decode
         and jbig2enc.available()
     ):
-        try:
-            imgname = root / f'{xref:08d}'
-            with imgname.open('wb') as f:
-                ext = pim.extract_to(stream=f)
-            imgname.rename(imgname.with_suffix(ext))
-        except pikepdf.UnsupportedImageTypeError:
-            return None
-        return XrefExt(xref, ext)
+        # Save any colorspace associated with the image, so that we
+        # will export a pure 1-bit PNG with no palette or ICC profile.
+        # Showing the palette or ICC to jbig2enc will cause it to perform
+        # colorspace transform to 1bpp, which will conflict the palette or
+        # ICC if it exists.
+        colorspace = pim.obj.get(pikepdf.Name.ColorSpace, None)
+        if colorspace is not None or pim.image_mask:
+            try:
+                # Set to DeviceGray temporarily; we already in 1 bpc.
+                pim.obj.ColorSpace = pikepdf.Name.DeviceGray
+                imgname = root / f'{xref:08d}'
+                with imgname.open('wb') as f:
+                    ext = pim.extract_to(stream=f)
+                imgname.rename(imgname.with_suffix(ext))
+            except pikepdf.UnsupportedImageTypeError:
+                return None
+            finally:
+                # Restore image colorspace after temporarily setting it to DeviceGray
+                if colorspace is not None:
+                    pim.obj.ColorSpace = colorspace
+                else:
+                    del pim.obj.ColorSpace
+            return XrefExt(xref, ext)
     return None
 
 
@@ -135,11 +156,6 @@ def extract_image_generic(
     if pim.bits_per_component == 1:
         return None
 
-    try:
-        pim.indexed  # pikepdf 1.6.3 can't handle [/Indexed [/Array...]]
-    except NotImplementedError:
-        return None
-
     if filtdp[0] == Name.DCTDecode and options.optimize >= 2:
         # This is a simple heuristic derived from some training data, that has
         # about a 70% chance of guessing whether the JPEG is high quality,
@@ -195,7 +211,10 @@ def extract_image_generic(
 
 
 def extract_images(
-    pike: Pdf, root: Path, options, extract_fn: Callable[..., Optional[XrefExt]],
+    pike: Pdf,
+    root: Path,
+    options,
+    extract_fn: Callable[..., Optional[XrefExt]],
 ) -> Iterator[Tuple[int, XrefExt]]:
     """Extract image using extract_fn
 
@@ -229,7 +248,9 @@ def extract_images(
                 # Ignore soft masks
                 smask_xref = Xref(image.SMask.objgen[0])
                 exclude_xrefs.add(smask_xref)
+                log.debug(f"Skipping image {smask_xref} because it is an SMask")
             include_xrefs.add(xref)
+            log.debug(f"Treating {xref} as an optimization candidate")
             if xref not in pageno_for_xref:
                 pageno_for_xref[xref] = pageno
 
@@ -240,8 +261,8 @@ def extract_images(
             result = extract_fn(
                 pike=pike, root=root, image=image, xref=xref, options=options
             )
-        except Exception as e:  # pylint: disable=broad-except
-            log.debug("Image xref %s, error %s", xref, repr(e))
+        except Exception:  # pylint: disable=broad-except
+            log.exception(f"While extracting image xref {xref}, an error occurred")
             errors += 1
         else:
             if result:
@@ -274,26 +295,22 @@ def extract_images_jbig2(pike: Pdf, root: Path, options) -> Dict[int, List[XrefE
         group = pageno // options.jbig2_page_group_size
         jbig2_groups[group].append(xref_ext)
 
-    # Elide empty groups
-    jbig2_groups = {
-        group: xrefs for group, xrefs in jbig2_groups.items() if len(xrefs) > 0
-    }
     log.debug("Optimizable images: JBIG2 groups: %s", (len(jbig2_groups),))
     return jbig2_groups
 
 
 def _produce_jbig2_images(
-    jbig2_groups: Dict[int, List[XrefExt]], root: Path, options
+    jbig2_groups: Dict[int, List[XrefExt]], root: Path, options, executor: Executor
 ) -> None:
     """Produce JBIG2 images from their groups"""
 
     def jbig2_group_args(root: Path, groups: Dict[int, List[XrefExt]]):
         for group, xref_exts in groups.items():
             prefix = f'group{group:08d}'
-            yield dict(
-                cwd=fspath(root),
-                infiles=(img_name(root, xref, ext) for xref, ext in xref_exts),
-                out_prefix=prefix,
+            yield (
+                fspath(root),  # =cwd
+                (img_name(root, xref, ext) for xref, ext in xref_exts),  # =infiles
+                prefix,  # =out_prefix
             )
 
     def jbig2_single_args(root, groups: Dict[int, List[XrefExt]]):
@@ -302,23 +319,20 @@ def _produce_jbig2_images(
             # Second loop is to ensure multiple images per page are unpacked
             for n, xref_ext in enumerate(xref_exts):
                 xref, ext = xref_ext
-                yield dict(
-                    cwd=fspath(root),
-                    infile=img_name(root, xref, ext),
-                    outfile=root / f'{prefix}.{n:04d}',
+                yield (
+                    fspath(root),
+                    img_name(root, xref, ext),
+                    root / f'{prefix}.{n:04d}',
                 )
 
-    def convert_generic(fn, kwargs_dict):
-        return fn(**kwargs_dict)
-
     if options.jbig2_page_group_size > 1:
         jbig2_args = jbig2_group_args
-        jbig2_convert = partial(convert_generic, jbig2enc.convert_group)
+        jbig2_convert = jbig2enc.convert_group_mp
     else:
         jbig2_args = jbig2_single_args
-        jbig2_convert = partial(convert_generic, jbig2enc.convert_single)
+        jbig2_convert = jbig2enc.convert_single_mp
 
-    exec_progress_pool(
+    executor(
         use_threads=True,
         max_workers=options.jobs,
         tqdm_kwargs=dict(
@@ -333,7 +347,11 @@ def _produce_jbig2_images(
 
 
 def convert_to_jbig2(
-    pike: Pdf, jbig2_groups: Dict[int, List[XrefExt]], root: Path, options
+    pike: Pdf,
+    jbig2_groups: Dict[int, List[XrefExt]],
+    root: Path,
+    options,
+    executor: Executor,
 ) -> None:
     """Convert images to JBIG2 and insert into PDF.
 
@@ -348,7 +366,7 @@ def convert_to_jbig2(
     and needs no dictionary. Currently this must be lossless JBIG2.
     """
 
-    _produce_jbig2_images(jbig2_groups, root, options)
+    _produce_jbig2_images(jbig2_groups, root, options, executor)
 
     for group, xref_exts in jbig2_groups.items():
         prefix = f'group{group:08d}'
@@ -372,27 +390,53 @@ def convert_to_jbig2(
             )
 
 
-def transcode_jpegs(pike: Pdf, jpegs: Sequence[Xref], root: Path, options) -> None:
-    for xref in tqdm(
-        jpegs, desc="JPEGs", unit='image', disable=not options.progress_bar
-    ):
-        in_jpg = jpg_name(root, xref)
-        opt_jpg = in_jpg.with_suffix('.opt.jpg')
+def _optimize_jpeg(args):
+    xref, in_jpg, opt_jpg, jpeg_quality = args
 
-        # This produces a debug warning from PIL
-        # DEBUG:PIL.Image:Error closing: 'NoneType' object has no attribute
-        # 'close'.  Seems to be mostly harmless
-        # https://github.com/python-pillow/Pillow/issues/1144
-        with Image.open(in_jpg) as im:
-            im.save(opt_jpg, optimize=True, quality=options.jpeg_quality)
+    # 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)
 
-        if opt_jpg.stat().st_size > in_jpg.stat().st_size:
-            log.debug("xref %s, jpeg, made larger - skip", xref)
-            continue
+    if opt_jpg.stat().st_size > in_jpg.stat().st_size:
+        log.debug("xref %s, jpeg, made larger - skip", xref)
+        opt_jpg.unlink()
+        opt_jpg = None
+    return xref, opt_jpg
 
-        compdata = leptonica.CompressedData.open(opt_jpg)
-        im_obj = pike.get_object(xref, 0)
-        im_obj.write(compdata.read(), filter=Name.DCTDecode)
+
+def transcode_jpegs(
+    pike: Pdf, jpegs: Sequence[Xref], root: Path, options, executor
+) -> None:
+    def jpeg_args():
+        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, pbar):
+        xref, opt_jpg = result
+        if opt_jpg:
+            compdata = leptonica.CompressedData.open(opt_jpg)
+            im_obj = pike.get_object(xref, 0)
+            im_obj.write(compdata.read(), 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,
+    )
 
 
 def _transcode_png(pike: Pdf, filename: Path, xref: Xref) -> bool:
@@ -411,9 +455,25 @@ def _transcode_png(pike: Pdf, filename: Path, xref: Xref) -> bool:
             decode_parms=local_image.DecodeParms,
         )
 
+        # Don't copy keys from the new image...
         del_keys = set(im_obj.keys()) - set(local_image.keys())
+        # ...except for the keep_fields, which are essential to displaying
+        # the image correctly and preserving its metadata. (/Decode arrays
+        # and /SMaskInData are implicitly discarded prior to this point.)
+        keep_fields = {
+            '/ID',
+            '/Intent',
+            '/Interpolate',
+            '/Mask',
+            '/Metadata',
+            '/OC',
+            '/OPI',
+            '/SMask',
+            '/StructParent',
+        }
+        del_keys -= keep_fields
         for key in local_image.keys():
-            if key != Name.Length:
+            if key != Name.Length and str(key) not in keep_fields:
                 im_obj[key] = local_image[key]
         for key in del_keys:
             del im_obj[key]
@@ -426,6 +486,7 @@ def transcode_pngs(
     image_name_fn: Callable[[Path, Xref], Path],
     root: Path,
     options,
+    executor,
 ) -> None:
     modified: MutableSet[Xref] = set()
     if options.optimize >= 2:
@@ -445,10 +506,7 @@ def transcode_pngs(
                 )
                 modified.add(xref)
 
-        def pngquant_fn(args):
-            pngquant.quantize(*args)
-
-        exec_progress_pool(
+        executor(
             use_threads=True,
             max_workers=options.jobs,
             tqdm_kwargs=dict(
@@ -457,7 +515,7 @@ def transcode_pngs(
                 unit='image',
                 disable=not options.progress_bar,
             ),
-            task=pngquant_fn,
+            task=pngquant.quantize_mp,
             task_arguments=pngquant_args(),
         )
 
@@ -466,82 +524,13 @@ def transcode_pngs(
         _transcode_png(pike, filename, xref)
 
 
-@deprecated
-def rewrite_png_as_g4(pike: Pdf, im_obj: Object, compdata) -> None:
-    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
-
-
-@deprecated
-def rewrite_png(pike: Pdf, im_obj: Object, compdata) -> None:
-    # 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 == 3:
-            cs = Name.DeviceRGB
-        elif compdata.spp == 4:
-            cs = Name.DeviceCMYK
-    im_obj.ColorSpace = cs
-    im_obj.write(compdata.read(), filter=Name.FlateDecode, decode_parms=dparms)
-
-
-def optimize(input_file: Path, output_file: Path, context, save_settings) -> None:
+def optimize(
+    input_file: Path,
+    output_file: Path,
+    context,
+    save_settings,
+    executor: Executor = SerialExecutor(),
+) -> None:
     options = context.options
     if options.optimize == 0:
         safe_symlink(input_file, output_file)
@@ -559,14 +548,14 @@ def optimize(input_file: Path, output_file: Path, context, save_settings) -> Non
         root.mkdir(exist_ok=True)
 
         jpegs, pngs = extract_images_generic(pike, root, options)
-        transcode_jpegs(pike, jpegs, root, options)
+        transcode_jpegs(pike, jpegs, root, options, executor)
         # if options.optimize >= 2:
         # Try pngifying the jpegs
         #    transcode_pngs(pike, jpegs, jpg_name, root, options)
-        transcode_pngs(pike, pngs, png_name, root, options)
+        transcode_pngs(pike, pngs, png_name, root, options, executor)
 
         jbig2_groups = extract_images_jbig2(pike, root, options)
-        convert_to_jbig2(pike, jbig2_groups, root, options)
+        convert_to_jbig2(pike, jbig2_groups, root, options, executor)
 
         target_file = output_file.with_suffix('.opt.pdf')
         pike.remove_unreferenced_resources()
@@ -581,7 +570,7 @@ def optimize(input_file: Path, output_file: Path, context, save_settings) -> Non
         )
     ratio = input_size / output_size
     savings = 1 - output_size / input_size
-    log.info(f"Optimize ratio: {ratio:.2f} savings: {(100 * savings):.1f}%")
+    log.info(f"Optimize ratio: {ratio:.2f} savings: {(savings):.1%}")
 
     if savings < 0:
         log.info("Image optimization did not improve the file - discarded")

src/ocrmypdf/pdfa.py

@@ -33,6 +33,7 @@ def _postscript_objdef(
     objtype = '/stream' if stream_name else '/dict'
 
     if stream_name:
+        assert stream_data is not None
         a85_data = base64.a85encode(stream_data, adobe=True).decode('ascii')
         yield f'{stream_name} ' + a85_data
         yield 'def'
@@ -111,7 +112,7 @@ def generate_pdfa_ps(target_filename: Path, icc: str = 'sRGB'):
 
 
 def file_claims_pdfa(filename: Path):
-    """Determines if the file claims to be PDF/A compliant
+    """Determines if the file claims to be PDF/A compliant.
 
     This only checks if the XMP metadata contains a PDF/A marker. It does not
     do full PDF/A validation.

src/ocrmypdf/pdfinfo/info.py

@@ -6,23 +6,25 @@
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 
 
+import atexit
 import logging
 import re
 from collections import defaultdict, namedtuple
+from contextlib import ExitStack
 from decimal import Decimal
 from enum import Enum
 from functools import partial
-from math import hypot, isclose
+from math import hypot, inf, isclose
 from os import PathLike
 from pathlib import Path
-from typing import Any, Dict, List, Optional, Union
+from typing import Container, Iterator, Optional, Tuple, Union
 from warnings import warn
 
 import pikepdf
-from pikepdf import PdfMatrix
+from pikepdf import Object, Pdf, PdfMatrix
 
-from ocrmypdf._concurrent import exec_progress_pool
-from ocrmypdf.exceptions import EncryptedPdfError
+from ocrmypdf._concurrent import Executor, SerialExecutor
+from ocrmypdf.exceptions import EncryptedPdfError, InputFileError
 from ocrmypdf.helpers import Resolution, available_cpu_count, pikepdf_enable_mmap
 from ocrmypdf.pdfinfo.layout import get_page_analysis, get_text_boxes
 
@@ -115,7 +117,7 @@ def _normalize_stack(graphobjs):
             yield (operands, operator)
 
 
-def _interpret_contents(contentstream, initial_shorthand=UNIT_SQUARE):
+def _interpret_contents(contentstream: Object, initial_shorthand=UNIT_SQUARE):
     """Interpret the PDF content stream.
 
     The stack represents the state of the PDF graphics stack.  We are only
@@ -204,7 +206,7 @@ def _interpret_contents(contentstream, initial_shorthand=UNIT_SQUARE):
     )
 
 
-def _get_dpi(ctm_shorthand, image_size):
+def _get_dpi(ctm_shorthand, image_size) -> Resolution:
     """Given the transformation matrix and image size, find the image DPI.
 
     PDFs do not include image resolution information within image data.
@@ -254,25 +256,29 @@ def _get_dpi(ctm_shorthand, image_size):
     a, b, c, d, _, _ = ctm_shorthand
 
     # Calculate the width and height of the image in PDF units
-    image_drawn_width = hypot(a, b)
-    image_drawn_height = hypot(c, d)
+    image_drawn = hypot(a, b), hypot(c, d)
 
-    # The scale of the image is pixels per unit of default user space (1/72")
-    scale_w = image_size[0] / image_drawn_width
-    scale_h = image_size[1] / image_drawn_height
-
-    # DPI = scale * 72
-    dpi_w = scale_w * 72.0
-    dpi_h = scale_h * 72.0
+    def calc(drawn, pixels, inches_per_pt=72.0):
+        # The scale of the image is pixels per unit of default user space (1/72")
+        scale = pixels / drawn if drawn != 0 else inf
+        dpi = scale * inches_per_pt
+        return dpi
 
+    dpi_w, dpi_h = (calc(image_drawn[n], image_size[n]) for n in range(2))
     return Resolution(dpi_w, dpi_h)
 
 
 class ImageInfo:
     DPI_PREC = Decimal('1.000')
 
-    def __init__(self, *, name='', pdfimage=None, inline=None, shorthand=None):
-
+    def __init__(
+        self,
+        *,
+        name='',
+        pdfimage: Optional[Object] = None,
+        inline: Optional[Object] = None,
+        shorthand=None,
+    ):
         self._name = str(name)
         self._shorthand = shorthand
 
@@ -282,6 +288,8 @@ class ImageInfo:
         elif pdfimage is not None:
             self._origin = 'xobject'
             pim = pikepdf.PdfImage(pdfimage)
+        else:
+            raise ValueError("Either pdfimage or inline must be set")
         self._width = pim.width
         self._height = pim.height
 
@@ -355,6 +363,10 @@ class ImageInfo:
     def enc(self):
         return self._enc
 
+    @property
+    def renderable(self):
+        return self.dpi.is_finite and self.width >= 0 and self.height >= 0
+
     @property
     def dpi(self):
         return _get_dpi(self._shorthand, (self._width, self._height))
@@ -371,7 +383,7 @@ class ImageInfo:
         ).format(**class_locals)
 
 
-def _find_inline_images(contentsinfo):
+def _find_inline_images(contentsinfo: ContentsInfo) -> Iterator[ImageInfo]:
     "Find inline images in the contentstream"
 
     for n, inline in enumerate(contentsinfo.inline_images):
@@ -380,7 +392,7 @@ def _find_inline_images(contentsinfo):
         )
 
 
-def _image_xobjects(container):
+def _image_xobjects(container) -> Iterator[Tuple[Object, str]]:
     """Search for all XObject-based images in the container
 
     Usually the container is a page, but it could also be a Form XObject
@@ -400,7 +412,7 @@ def _image_xobjects(container):
         return
     xobjs = resources['/XObject'].as_dict()
     for xobj in xobjs:
-        candidate = xobjs[xobj]
+        candidate: Object = xobjs[xobj]
         if not '/Subtype' in candidate:
             continue
         if candidate['/Subtype'] == '/Image':
@@ -408,7 +420,9 @@ def _image_xobjects(container):
             yield (pdfimage, xobj)
 
 
-def _find_regular_images(container, contentsinfo):
+def _find_regular_images(
+    container: Object, contentsinfo: ContentsInfo
+) -> Iterator[ImageInfo]:
     """Find images stored in the container's /Resources /XObject
 
     Usually the container is a page, but it could also be a Form XObject
@@ -432,7 +446,7 @@ def _find_regular_images(container, contentsinfo):
             yield ImageInfo(name=draw.name, pdfimage=pdfimage, shorthand=draw.shorthand)
 
 
-def _find_form_xobject_images(pdf, container, contentsinfo):
+def _find_form_xobject_images(pdf: Pdf, container: Object, contentsinfo: ContentsInfo):
     """Find any images that are in Form XObjects in the container
 
     The container may be a page, or a parent Form XObject.
@@ -464,7 +478,9 @@ def _find_form_xobject_images(pdf, container, contentsinfo):
             )
 
 
-def _process_content_streams(*, pdf, container, shorthand=None):
+def _process_content_streams(
+    *, pdf: Pdf, container: Object, shorthand=None
+) -> Iterator[Union[VectorMarker, TextMarker, ImageInfo]]:
     """Find all individual instances of images drawn in the container
 
     Usually the container is a page, but it may also be a Form XObject.
@@ -526,7 +542,7 @@ def _page_has_text(text_blocks, page_width, page_height) -> bool:
         margin_ratio * ph,  # bottom  (first quadrant: bottom < top)
     )
 
-    def rects_intersect(a, b):
+    def rects_intersect(a, b) -> bool:
         """
         Where (a,b) are 4-tuple rects (left-0, top-1, right-2, bottom-3)
         https://stackoverflow.com/questions/306316/determine-if-two-rectangles-overlap-each-other
@@ -542,7 +558,7 @@ def _page_has_text(text_blocks, page_width, page_height) -> bool:
     return has_text
 
 
-def simplify_textboxes(miner, textbox_getter):
+def simplify_textboxes(miner, textbox_getter) -> Iterator[TextboxInfo]:
     """Extract only limited content from text boxes
 
     We do this to save memory and ensure that our objects are pickleable.
@@ -556,97 +572,51 @@ def simplify_textboxes(miner, textbox_getter):
         yield TextboxInfo(box.bbox, visible, corrupt)
 
 
-def _pdf_get_pageinfo(
-    pdf, pageno: int, infile: PathLike, check_pages, detailed_analysis: bool
-):
-    pageinfo: Dict[str, Any] = {}
-    pageinfo['pageno'] = pageno
-    pageinfo['images'] = []
-
-    page = pdf.pages[pageno]
-    mediabox = [Decimal(d) for d in page.MediaBox.as_list()]
-    width_pt = mediabox[2] - mediabox[0]
-    height_pt = mediabox[3] - mediabox[1]
-
-    check_this_page = pageno in check_pages
-
-    if check_this_page and detailed_analysis:
-        pscript5_mode = str(pdf.docinfo.get('/Creator')).startswith('PScript5')
-        miner = get_page_analysis(infile, pageno, pscript5_mode)
-        pageinfo['textboxes'] = list(simplify_textboxes(miner, get_text_boxes))
-        bboxes = (box.bbox for box in pageinfo['textboxes'])
-
-        pageinfo['has_text'] = _page_has_text(bboxes, width_pt, height_pt)
-    else:
-        pageinfo['textboxes'] = []
-        pageinfo['has_text'] = None  # i.e. "no information"
-
-    userunit = page.get('/UserUnit', Decimal(1.0))
-    if not isinstance(userunit, Decimal):
-        userunit = Decimal(userunit)
-    pageinfo['userunit'] = userunit
-    pageinfo['width_inches'] = width_pt * userunit / Decimal(72.0)
-    pageinfo['height_inches'] = height_pt * userunit / Decimal(72.0)
-
-    try:
-        pageinfo['rotate'] = int(page['/Rotate'])
-    except KeyError:
-        pageinfo['rotate'] = 0
-
-    userunit_shorthand = (userunit, 0, 0, userunit, 0, 0)
-
-    if check_this_page:
-        pageinfo['has_vector'] = False
-        pageinfo['has_text'] = False
-        pageinfo['images'] = []
-        for ci in _process_content_streams(
-            pdf=pdf, container=page, shorthand=userunit_shorthand
-        ):
-            if isinstance(ci, VectorMarker):
-                pageinfo['has_vector'] = True
-            elif isinstance(ci, TextMarker):
-                pageinfo['has_text'] = True
-            elif isinstance(ci, ImageInfo):
-                pageinfo['images'].append(ci)
-            else:
-                raise NotImplementedError()
-    else:
-        pageinfo['has_vector'] = None  # i.e. "no information"
-        pageinfo['has_text'] = None
-        pageinfo['images'] = None
-
-    if pageinfo['images']:
-        dpi = Resolution(0.0, 0.0).take_max(image.dpi for image in pageinfo['images'])
-        pageinfo['dpi'] = dpi
-        pageinfo['width_pixels'] = int(round(dpi.x * float(pageinfo['width_inches'])))
-        pageinfo['height_pixels'] = int(round(dpi.y * float(pageinfo['height_inches'])))
-
-    return pageinfo
-
-
 worker_pdf = None
 
 
-def _pdf_pageinfo_sync_init(infile):
+def _pdf_pageinfo_sync_init(pdf: Pdf, infile: Path, pdfminer_loglevel):
     global worker_pdf  # pylint: disable=global-statement
     pikepdf_enable_mmap()
-    worker_pdf = pikepdf.open(infile)
+
+    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 = pikepdf.open(infile)
+
+        def on_process_close():
+            worker_pdf.close()
+
+        # Close when this process exits
+        atexit.register(on_process_close)
 
 
 def _pdf_pageinfo_sync(args):
-    global worker_pdf  # pylint: disable=global-statement
-    pageno, infile, check_pages, detailed_analysis = args
-    page = PageInfo(worker_pdf, pageno, infile, check_pages, detailed_analysis)
-    return page
+    pageno, thread_pdf, infile, check_pages, detailed_analysis = args
+    pdf = thread_pdf if thread_pdf is not None else worker_pdf
+    with ExitStack() as stack:
+        if not pdf:  # When called with SerialExecutor
+            pdf = stack.enter_context(pikepdf.open(infile))
+        page = PageInfo(pdf, pageno, infile, check_pages, detailed_analysis)
+        return page
 
 
 def _pdf_pageinfo_concurrent(
-    pdf, infile, progbar, max_workers, check_pages, detailed_analysis=False
+    pdf,
+    executor: Executor,
+    infile,
+    progbar,
+    max_workers,
+    check_pages,
+    detailed_analysis=False,
 ):
     pages = [None] * len(pdf.pages)
 
     def update_pageinfo(result, pbar):
         page = result
+        if not page:
+            raise InputFileError("Could read a page in the PDF")
         pages[page.pageno] = page
         pbar.update()
 
@@ -654,7 +624,6 @@ def _pdf_pageinfo_concurrent(
         max_workers = available_cpu_count()
 
     total = len(pdf.pages)
-    contexts = ((n, infile, check_pages, detailed_analysis) for n in range(total))
 
     use_threads = False  # No performance gain if threaded due to GIL
     n_workers = min(1 + len(pages) // 4, max_workers)
@@ -663,13 +632,27 @@ def _pdf_pageinfo_concurrent(
         # a separate process.
         use_threads = True
 
-    exec_progress_pool(
+    # 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
         ),
-        task_initializer=partial(_pdf_pageinfo_sync_init, infile),
+        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,
@@ -678,13 +661,87 @@ def _pdf_pageinfo_concurrent(
 
 
 class PageInfo:
-    def __init__(self, pdf, pageno, infile, check_pages, detailed_analysis=False):
+    def __init__(
+        self,
+        pdf: Pdf,
+        pageno: int,
+        infile: PathLike,
+        check_pages: Container[int],
+        detailed_analysis: bool = False,
+    ):
         self._pageno = pageno
         self._infile = infile
         self._detailed_analysis = detailed_analysis
-        self._pageinfo = _pdf_get_pageinfo(
-            pdf, pageno, infile, check_pages, detailed_analysis
-        )
+        self._gather_pageinfo(pdf, pageno, infile, check_pages, detailed_analysis)
+
+    def _gather_pageinfo(
+        self,
+        pdf: Pdf,
+        pageno: int,
+        infile: PathLike,
+        check_pages: Container[int],
+        detailed_analysis: bool,
+    ):
+        page = pdf.pages[pageno]
+        mediabox = [Decimal(d) for d in page.MediaBox.as_list()]
+        width_pt = mediabox[2] - mediabox[0]
+        height_pt = mediabox[3] - mediabox[1]
+
+        check_this_page = pageno in check_pages
+
+        if check_this_page and detailed_analysis:
+            pscript5_mode = str(pdf.docinfo.get('/Creator')).startswith('PScript5')
+            miner = get_page_analysis(infile, pageno, pscript5_mode)
+            self._textboxes = list(simplify_textboxes(miner, get_text_boxes))
+            bboxes = (box.bbox for box in self._textboxes)
+
+            self._has_text = _page_has_text(bboxes, width_pt, height_pt)
+        else:
+            self._textboxes = []
+            self._has_text = None  # i.e. "no information"
+
+        userunit = page.get('/UserUnit', Decimal(1.0))
+        if not isinstance(userunit, Decimal):
+            userunit = Decimal(userunit)
+        self._userunit = userunit
+        self._width_inches = width_pt * userunit / Decimal(72.0)
+        self._height_inches = height_pt * userunit / Decimal(72.0)
+
+        try:
+            self._rotate = int(page['/Rotate'])
+        except KeyError:
+            self._rotate = 0
+
+        userunit_shorthand = (userunit, 0, 0, userunit, 0, 0)
+
+        if check_this_page:
+            self._has_vector = False
+            self._has_text = False
+            self._images = []
+            for ci in _process_content_streams(
+                pdf=pdf, container=page, shorthand=userunit_shorthand
+            ):
+                if isinstance(ci, VectorMarker):
+                    self._has_vector = True
+                elif isinstance(ci, TextMarker):
+                    self._has_text = True
+                elif isinstance(ci, ImageInfo):
+                    self._images.append(ci)
+                else:
+                    raise NotImplementedError()
+        else:
+            self._has_vector = None  # i.e. "no information"
+            self._has_text = None
+            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
+            )
+            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)))
 
     @property
     def pageno(self) -> int:
@@ -692,25 +749,25 @@ class PageInfo:
 
     @property
     def has_text(self) -> bool:
-        return self._pageinfo['has_text']
+        return self._has_text
 
     @property
     def has_corrupt_text(self) -> bool:
         if not self._detailed_analysis:
             raise NotImplementedError('Did not do detailed analysis')
-        return any(tbox.is_corrupt for tbox in self._pageinfo['textboxes'])
+        return any(tbox.is_corrupt for tbox in self._textboxes)
 
     @property
     def has_vector(self) -> bool:
-        return self._pageinfo['has_vector']
+        return self._has_vector
 
     @property
     def width_inches(self) -> Decimal:
-        return self._pageinfo['width_inches']
+        return self._width_inches
 
     @property
     def height_inches(self) -> Decimal:
-        return self._pageinfo['height_inches']
+        return self._height_inches
 
     @property
     def width_pixels(self) -> int:
@@ -722,18 +779,18 @@ class PageInfo:
 
     @property
     def rotation(self) -> int:
-        return self._pageinfo.get('rotate', None)
+        return self._rotate
 
     @rotation.setter
     def rotation(self, value):
         if value in (0, 90, 180, 270, 360, -90, -180, -270):
-            self._pageinfo['rotate'] = value
+            self._rotate = value
         else:
             raise ValueError("rotation must be a cardinal angle")
 
     @property
     def images(self):
-        return self._pageinfo['images']
+        return self._images
 
     def get_textareas(
         self, visible: Optional[bool] = None, corrupt: Optional[bool] = None
@@ -748,24 +805,22 @@ class PageInfo:
                     result = False
             return result
 
-        if 'textboxes' not in self._pageinfo:
+        if not self._textboxes:
             if visible is not None and corrupt is not None:
                 raise NotImplementedError('Incomplete information on textboxes')
-            return self._pageinfo['bboxes']
+            return self._textboxes
 
-        return (
-            obj.bbox
-            for obj in self._pageinfo['textboxes']
-            if predicate(obj, visible, corrupt)
-        )
+        return (obj.bbox for obj in self._textboxes if predicate(obj, visible, corrupt))
 
     @property
     def dpi(self) -> Resolution:
-        return self._pageinfo.get('dpi', Resolution(0.0, 0.0))
+        if self._dpi is None:
+            return Resolution(0.0, 0.0)
+        return self._dpi
 
     @property
     def userunit(self) -> Decimal:
-        return self._pageinfo.get('userunit', None)
+        return self._userunit
 
     @property
     def min_version(self) -> str:
@@ -788,10 +843,12 @@ class PdfInfo:
     def __init__(
         self,
         infile,
+        *,
         detailed_analysis: bool = False,
         progbar: bool = False,
         max_workers: int = None,
         check_pages=None,
+        executor: Executor = SerialExecutor(),
     ):
         self._infile = infile
         if check_pages is None:
@@ -802,18 +859,19 @@ class PdfInfo:
                 raise EncryptedPdfError()  # Triggered by encryption with empty passwd
             self._pages = _pdf_pageinfo_concurrent(
                 pdf,
+                executor,
                 infile,
                 progbar,
                 max_workers,
                 check_pages=check_pages,
                 detailed_analysis=detailed_analysis,
             )
-            self._needs_rendering = pdf.root.get('/NeedsRendering', False)
+            self._needs_rendering = pdf.Root.get('/NeedsRendering', False)
             self._has_acroform = False
-            if '/AcroForm' in pdf.root:
-                if len(pdf.root.AcroForm.get('/Fields', [])) > 0:
+            if '/AcroForm' in pdf.Root:
+                if len(pdf.Root.AcroForm.get('/Fields', [])) > 0:
                     self._has_acroform = True
-                elif '/XFA' in pdf.root.AcroForm:
+                elif '/XFA' in pdf.Root.AcroForm:
                     self._has_acroform = True
 
     @property

src/ocrmypdf/pdfinfo/layout.py

@@ -21,7 +21,7 @@ from pdfminer.pdffont import PDFSimpleFont, PDFUnicodeNotDefined
 from pdfminer.pdfpage import PDFPage
 from pdfminer.utils import bbox2str, matrix2str
 
-from ocrmypdf.exceptions import EncryptedPdfError
+from ocrmypdf.exceptions import EncryptedPdfError, InputFileError
 
 STRIP_NAME = re.compile(r'[0-9]+')
 
@@ -223,6 +223,7 @@ def get_page_analysis(infile, pageno, pscript5_mode):
     )
     interp = pdfminer.pdfinterp.PDFPageInterpreter(rman, dev)
 
+    patcher = None
     if pscript5_mode:
         patcher = patch.multiple(
             'pdfminer.pdffont.PDFType3Font',
@@ -235,12 +236,17 @@ def get_page_analysis(infile, pageno, pscript5_mode):
 
     try:
         with Path(infile).open('rb') as f:
-            page = PDFPage.get_pages(f, pagenos=[pageno], maxpages=0)
-            interp.process_page(next(page))
-    except PDFTextExtractionNotAllowed:
-        raise EncryptedPdfError()
+            page_iter = PDFPage.get_pages(f, pagenos=[pageno], maxpages=0)
+            page = next(page_iter, None)
+            if page is None:
+                raise InputFileError(
+                    f"pdfminer could not process page {pageno} (counting from 0)."
+                )
+            interp.process_page(page)
+    except PDFTextExtractionNotAllowed as e:
+        raise EncryptedPdfError() from e
     finally:
-        if pscript5_mode:
+        if patcher is not None:
             patcher.stop()
 
     return dev.get_result()

src/ocrmypdf/pluginspec.py

@@ -8,24 +8,41 @@
 from abc import ABC, abstractmethod, abstractstaticmethod
 from argparse import ArgumentParser, Namespace
 from collections import namedtuple
+from logging import Handler
 from pathlib import Path
 from typing import TYPE_CHECKING, AbstractSet, List, Optional
 
 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.
@@ -59,7 +76,68 @@ def check_options(options: Namespace) -> None:
     Note:
         This hook will be called from the main process, and may modify global state
         before child worker processes are forked.
-	"""
+    """
+
+
+@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
@@ -89,9 +167,9 @@ def rasterize_pdf_page(
     raster_device: str,
     raster_dpi: Resolution,
     pageno: int,
-    page_dpi: Optional[Resolution] = None,
-    rotation: Optional[int] = None,
-    filter_vector: bool = False,
+    page_dpi: Optional[Resolution],
+    rotation: Optional[int],
+    filter_vector: bool,
 ) -> Path:
     """Rasterize one page of a PDF at resolution raster_dpi in canvas units.
 
@@ -163,11 +241,11 @@ 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. If you
-    convert the image to a JPEG, the output page will be created as a JPEG, etc.
-    If you change the colorspace, that change will be kept. Note that the
-    OCRmyPDF image optimization stage, if enabled, may ultimately chose a
-    different format.
+    OCRmyPDF will create the PDF page based on the image format used (unless the
+    hook is overriden). If you convert the image to a JPEG, the output page will
+    be created as a JPEG, etc. If you change the colorspace, that change will be
+    kept. Note that the OCRmyPDF image optimization stage, if enabled, may
+    ultimately chose a different format.
 
     If the return value is a file that does not exist, ``FileNotFoundError``
     will occur. The return value should be a path to a file in the same folder
@@ -185,6 +263,50 @@ 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.
+
+    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>`.
+    """
+
+
 OrientationConfidence = namedtuple('OrientationConfidence', ('angle', 'confidence'))
 """Expresses an OCR engine's confidence in page rotation.
 
@@ -280,6 +402,7 @@ def generate_pdfa(
     compression: str,
     pdf_version: str,
     pdfa_part: str,
+    progressbar_class,
 ) -> Path:
     """Generate a PDF/A.
 
@@ -302,10 +425,21 @@ def generate_pdfa(
             At its own discretion, the PDF/A generator may raise the version,
             but should not lower it.
         pdfa_part: The desired PDF/A compliance level, such as ``'2B'``.
+        progressbar_class: The class of a progress bar with a tqdm-like API. An
+            instance of this class will be initialized when PDF/A conversion
+            begins, using
+            ``instance = progressbar_class(total: int, desc: str, unit:str)``,
+            defining the number of work units, a user-visible description,
+            and the name of the work units ("page"). Then ``instance.update()``
+            will be called when a work unit is completed. If ``None``, no
+            progress information is reported.
 
     Returns:
         Path: If successful, the hook should return ``output_file``.
 
     Note:
         This is a :ref:`firstresult hook<firstresult>`.
+
+    See also:
+        https://github.com/tqdm/tqdm
     """

src/ocrmypdf/py.typed

@@ -0,0 +1 @@
+# ocrmypdf is typed

src/ocrmypdf/subprocess/__init__.py

@@ -10,29 +10,94 @@
 import logging
 import os
 import re
-import shutil
 import sys
 from collections.abc import Mapping
 from contextlib import suppress
-from distutils.version import LooseVersion
+from distutils.version import LooseVersion, Version
 from functools import lru_cache
 from pathlib import Path
-from subprocess import PIPE, STDOUT, CalledProcessError
+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, **kwargs):
+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
     fashion that indentifies the responsible subprocess. An additional
     task is that this function goes to greater lengths to find possible Windows
     locations of our dependencies when they are not on the system PATH.
+
+    Arguments should be identical to ``subprocess.run``, except for following:
+
+    Arguments:
+        logs_errors_to_stdout: If True, indicates that the process writes its error
+            messages to stdout rather than stderr, so stdout should be logged
+            if there is an error. If False, stderr is logged. Could be used with
+            stderr=STDOUT, stdout=PIPE for example.
     """
+    args, env, process_log, _text = _fix_process_args(args, env, kwargs)
+
+    stderr = None
+    stderr_name = 'stderr' if not logs_errors_to_stdout else 'stdout'
+    try:
+        proc = subprocess_run(args, env=env, **kwargs)
+    except CalledProcessError as e:
+        stderr = getattr(e, stderr_name, None)
+        raise
+    else:
+        stderr = getattr(proc, stderr_name, None)
+    finally:
+        if process_log.isEnabledFor(logging.DEBUG) and stderr:
+            with suppress(AttributeError, UnicodeDecodeError):
+                stderr = stderr.decode('utf-8', 'replace')
+            if logs_errors_to_stdout:
+                process_log.debug("stdout/stderr = %s", stderr)
+            else:
+                process_log.debug("stderr = %s", stderr)
+    return proc
+
+
+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.
+    The intended use is monitoring progress of subprocesses that output their
+    own progress indicators. In addition, each line will be logged if debug
+    logging is enabled.
+
+    Requires stderr to be opened in text mode for ease of handling errors. In
+    addition the expected encoding= and errors= arguments should be set. Note
+    that if stdout is already set up, it need not be binary.
+    """
+    args, env, process_log, text = _fix_process_args(args, env, kwargs)
+    assert text, "Must use text=True"
+
+    with Popen(args, env=env, **kwargs) as proc:
+        lines = []
+        while proc.poll() is None:
+            for msg in iter(proc.stderr.readline, ''):
+                if process_log.isEnabledFor(logging.DEBUG):
+                    process_log.debug(msg.strip())
+                callback(msg)
+                lines.append(msg)
+        stderr = ''.join(lines)
+
+        if check and proc.returncode != 0:
+            raise CalledProcessError(proc.returncode, args, output=None, stderr=stderr)
+        return CompletedProcess(args, proc.returncode, None, stderr=stderr)
+
+
+def _fix_process_args(args, env, kwargs):
+    assert 'universal_newlines' not in kwargs, "Use text= instead of universal_newlines"
+
     if not env:
         env = os.environ
 
@@ -40,53 +105,23 @@ def run(args, *, env=None, **kwargs):
     program = args[0]
 
     if os.name == 'nt':
-        args = _fix_windows_args(program, args, env)
+        from ocrmypdf.subprocess._windows import fix_windows_args
+
+        args = fix_windows_args(program, args, env)
 
     log.debug("Running: %s", args)
-    process_log = log.getChild('subprocess.' + os.path.basename(program))
-    if sys.version_info < (3, 7) and 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
-
-    stderr = None
-    try:
-        proc = subprocess_run(args, env=env, **kwargs)
-    except CalledProcessError as e:
-        stderr = getattr(e, 'stderr', None)
-        raise
-    else:
-        stderr = getattr(proc, 'stderr', None)
-    finally:
-        if process_log.isEnabledFor(logging.DEBUG) and stderr:
-            with suppress(AttributeError, UnicodeDecodeError):
-                stderr = stderr.decode('utf-8', 'replace')
-            process_log.debug("stderr = %s", stderr)
-    return proc
-
-
-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):
-        # bpo-33617 - Windows needs manual Path -> str conversion
-        args = [os.fspath(arg) for arg in args]
-        program = os.fspath(program)
-
-    # If we are running a .py on Windows, ensure we call it with this Python
-    # (to support test suite shims)
-    if program.lower().endswith('.py'):
-        args = [sys.executable] + args
-
-    paths = os.pathsep.join(os.get_exec_path(env))
-    if not shutil.which(args[0], path=paths):
-        # If the program we want is not on the PATH, add some interesting
-        # locations in %PROGRAMFILES% to the PATH and try again
-        shimmed_path = shim_paths_with_program_files(env)
-        new_args0 = shutil.which(args[0], path=shimmed_path)
-        if new_args0:
-            args[0] = new_args0
-    return 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)
@@ -107,7 +142,7 @@ def get_version(
         proc = run(
             args_prog,
             close_fds=True,
-            universal_newlines=True,
+            text=True,
             stdout=PIPE,
             stderr=STDOUT,
             check=True,
@@ -126,44 +161,18 @@ def get_version(
         raise MissingDependencyError(
             f"Could not find program '{program}' on the PATH"
         ) from e
-    try:
-        version = re.match(regex, output.strip()).group(1)
-    except AttributeError as e:
+
+    match = re.match(regex, output.strip())
+    if not match:
         raise MissingDependencyError(
             f"The program '{program}' did not report its version. "
             f"Message was:\n{output}"
         )
+    version = match.group(1)
 
     return version
 
 
-def shim_paths_with_program_files(env=None):
-    if not env:
-        env = os.environ
-    program_files = env.get('PROGRAMFILES', '')
-    if not program_files:
-        return env.get('PATH', '')
-
-    def path_walker():
-        for path in Path(program_files).iterdir():
-            if not path.is_dir():
-                continue
-            if path.name.lower() == 'tesseract-ocr':
-                yield path
-            elif path.name.lower() == 'gs':
-                yield from (p for p in path.glob('**/bin') if p.is_dir())
-
-    paths = sorted(
-        (p for p in path_walker()), key=lambda p: (p.name, p.parent.name), reverse=True
-    )
-    paths.extend(
-        Path(str_path)
-        for str_path in os.get_exec_path(env)
-        if Path(str_path) not in set(paths)
-    )
-    return os.pathsep.join(str(p) for p in paths)
-
-
 missing_program = '''
 The program '{program}' could not be executed or was not found on your
 system PATH.
@@ -258,13 +267,30 @@ def _error_old_version(program, package, need_version, found_version, required_f
 
 def check_external_program(
     *,
-    program,
-    package,
-    version_checker,
-    need_version,
-    required_for=None,
+    program: str,
+    package: str,
+    version_checker: Union[str, Callable],
+    need_version: str,
+    required_for: Optional[str] = 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()
@@ -273,7 +299,7 @@ def check_external_program(
     except (CalledProcessError, FileNotFoundError, MissingDependencyError):
         _error_missing_program(program, package, required_for, recommended)
         if not recommended:
-            raise MissingDependencyError()
+            raise MissingDependencyError(program)
         return
 
     def remove_leading_v(s):
@@ -284,9 +310,9 @@ def check_external_program(
     found_version = remove_leading_v(found_version)
     need_version = remove_leading_v(need_version)
 
-    if found_version and LooseVersion(found_version) < LooseVersion(need_version):
+    if found_version and version_parser(found_version) < version_parser(need_version):
         _error_old_version(program, package, need_version, found_version, required_for)
         if not recommended:
-            raise MissingDependencyError()
+            raise MissingDependencyError(program)
 
     log.debug('Found %s %s', program, found_version)

src/ocrmypdf/subprocess/_windows.py

@@ -0,0 +1,162 @@
+# © 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 os
+import shutil
+import sys
+from distutils.version import LooseVersion
+from itertools import chain, filterfalse
+from pathlib import Path
+from typing import Any, Callable, Iterator, Optional, Tuple, TypeVar, cast
+
+try:
+    import winreg
+except ModuleNotFoundError as e:
+    raise ModuleNotFoundError("This module is for Windows only") from e
+
+log = logging.getLogger(__name__)
+
+T = TypeVar('T')
+
+
+def registry_enum(
+    key: winreg.HKEYType, enum_fn: Callable[[winreg.HKEYType, int], T]
+) -> Iterator[T]:
+    LIMIT = 999
+    n = 0
+    while n < LIMIT:
+        try:
+            yield enum_fn(key, n)
+            n += 1
+        except OSError:
+            break
+    if n == LIMIT:
+        raise ValueError(f"Too many registry keys under {key}")
+
+
+def registry_subkeys(key: winreg.HKEYType) -> Iterator[str]:
+    return registry_enum(key, winreg.EnumKey)
+
+
+def registry_values(key: winreg.HKEYType) -> Iterator[Tuple[str, Any, int]]:
+    return registry_enum(key, winreg.EnumValue)
+
+
+def registry_path_ghostscript(env=None) -> Iterator[Path]:
+    try:
+        with winreg.OpenKey(
+            winreg.HKEY_LOCAL_MACHINE, r"SOFTWARE\Artifex\GPL Ghostscript"
+        ) as k:
+            latest_gs = max(registry_subkeys(k), key=LooseVersion)
+        with winreg.OpenKey(
+            winreg.HKEY_LOCAL_MACHINE, fr"SOFTWARE\Artifex\GPL Ghostscript\{latest_gs}"
+        ) as k:
+            _, gs_path, _ = next(registry_values(k))
+            yield Path(gs_path) / 'bin'
+    except OSError as e:
+        log.warning(e)
+
+
+def registry_path_tesseract(env=None) -> Iterator[Path]:
+    try:
+        with winreg.OpenKey(winreg.HKEY_LOCAL_MACHINE, r"SOFTWARE\Tesseract-OCR") as k:
+            for subkey, val, _valtype in registry_values(k):
+                if subkey == 'InstallDir':
+                    tesseract_path = Path(val)
+                    yield tesseract_path
+    except OSError as e:
+        log.warning(e)
+
+
+def program_files_paths(env=None) -> Iterator[Path]:
+    if not env:
+        env = os.environ
+    program_files = env.get('PROGRAMFILES', '')
+
+    def path_walker() -> Iterator[Path]:
+        for path in Path(program_files).iterdir():
+            if not path.is_dir():
+                continue
+            if path.name.lower() == 'tesseract-ocr':
+                yield path
+            elif path.name.lower() == 'gs':
+                yield from (p for p in path.glob('**/bin') if p.is_dir())
+
+    return iter(
+        sorted(
+            (p for p in path_walker()),
+            key=lambda p: (p.name, p.parent.name),
+            reverse=True,
+        )
+    )
+
+
+def paths_from_env(env=None) -> Iterator[Path]:
+    return (Path(p) for p in os.get_exec_path(env) if p)
+
+
+def shim_path(new_paths: Callable[[Any], Iterator[Path]], env=None) -> str:
+    if not env:
+        env = os.environ
+    return os.pathsep.join(str(p) for p in new_paths(env) if p)
+
+
+SHIMS = [
+    paths_from_env,
+    registry_path_ghostscript,
+    registry_path_tesseract,
+    program_files_paths,
+]
+
+
+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):
+        # bpo-33617 - Windows needs manual Path -> str conversion
+        args = [os.fspath(arg) for arg in args]
+        program = os.fspath(program)
+
+    # If we are running a .py on Windows, ensure we call it with this Python
+    # (to support test suite shims)
+    if program.lower().endswith('.py'):
+        args = [sys.executable] + args
+
+    # If the program we want is not on the PATH, check elsewhere
+    for shim in SHIMS:
+        shimmed_path = shim_path(shim, env)
+        new_args0 = shutil.which(args[0], path=shimmed_path)
+        if new_args0:
+            args[0] = new_args0
+            break
+
+    return args
+
+
+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()
+    seen_add = seen.add
+    if key is None:
+        key = lambda x: x
+    for element in iterable:
+        k = key(element)
+        if k not in seen:
+            seen_add(k)
+            yield element
+
+
+def shim_env_path(env=None):
+    if env is None:
+        env = os.environ
+
+    shim_paths = chain.from_iterable(shim(env) for shim in SHIMS)
+    return os.pathsep.join(
+        str(p) for p in unique_everseen(shim_paths, key=lambda p: str.casefold(str(p)))
+    )

tests/__init__.py

@@ -0,0 +1,7 @@
+# © 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/.
+
+# Empty __init__.py file

tests/cache/aspect/__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt/hocr.bin

@@ -9,7 +9,7 @@
   <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 "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/com.github.ocrmypdf.mrqsewbu/000001_ocr.png"; bbox 0 0 1000 800; ppageno 0'>
+  <div class='ocr_page' id='page_1' title='image "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/ocrmypdf.io.h7hsz1g7/000001_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">

tests/cache/cardinal/__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt/hocr.bin

@@ -9,7 +9,7 @@
   <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 "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/com.github.ocrmypdf.ces85e5u/000001_ocr.png"; bbox 0 0 2550 3300; ppageno 0'>
+  <div class='ocr_page' id='page_1' title='image "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/ocrmypdf.io.tgp04npj/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">

tests/cache/cardinal/__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt/txt.bin

@@ -103,7 +103,7 @@ ERASE, REPEAT, PLAY/STOP, or LOCATE.
 
 © 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!)
@@ -115,9 +115,9 @@ on the TAP TEMPO button.
 ¢ TEMPO CHANGES may be programmed into a sequence, with smooth transitions if desired.
 ¢ Any TIME SIGNATURE may be used, and may be changed within a song.
 
-nn
-
+linn
 Linn Electronics, Inc.
+
 18720 Oxnard Street, Tarzana, CA 91356
 (818) 708-8131 TELEX #298949 LINN UR
 

tests/cache/cardinal/__-l__eng__000003_ocr.png__000003_ocr_hocr__hocr__txt/txt.bin

@@ -1,128 +1,123 @@
-2A NNI‘I 6F6867# XATALL IE18-80L (818)
+The LinnSequencer
+32 Track MIDI Sequence Recorder
 
-9SEI6 VO “BUBZIRY, “J0aNS PIPUXO OZLEI
-“Uy ‘soTUOMOI,q UUrT
+The LinnSequencer is a state-of-the-art composition and performance tool for the professional musician. It is
 
-uut]
+extremely powerful, yet amazingly simple to learn and use. It’s many remarkable features include:
 
-“‘SUOS B UIJIM pasueyo oq ABU pue ‘posn oq AWW AYN IVNOIS AWLL AUV
-“parlsop Jr SUOTIISUBI} YIOOUIS YIM “BoueNbas eB OJUI pourtueIZOId 9q ABU SFONWHO OdINAL e
+¢ Operation is similar to multi-track tape recorder with PLAY, STOP, RECORD, FAST
+FORWARD, REWIND, and LOCATE controls.
 
-‘uonng OdNAL dV L 9) uO
+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
 
-sojou Jayienb Suiddy} Aq 10 ‘syUSTIOIOUI oINUTIAI-J8g-Jesg & JO sys} UL ofquisn(pe ‘ATTeouIAUINU paiajus oq ABU OdINALL e
+synthesizers!
 
-(jouer doup u3a9)
+¢ Ultra-fast 3%” disk drive stores complex songs in seconds and holds over 110,000 notes
 
-“puooes Jed souely O€ 10 “SZ “pz 18 [LVAG-MAd-SHN VU 10 ALOANIWAAd-SLVAd U! patyoeds aq kewl OAL «
-‘uoTe1odo [SVx JO} Aj[eusoyUT JoyndUIOd 11g 9] 98108 ZHI 8g ‘poeds-ysry Bann soz] e
+per disk!
 
-"9U0} DUAS 0006 UUL] Jo wNIqUUr] prepue}s 0} OUAS [ITAA ©
+¢ One or all tracks may be TRANSPOSED at the touch of a key.
+e Exclusive real-time ERASE function makes editing FAST.
+* Exclusive REPEAT function automatically repeats any held notes at a pre-selected
 
-“ONYBA 9}OU poloapes Aue Je sas—nd jndyno 07 pewureigold 3q ACW SL Ad LNO YADONAL OML
+rhythmic value.
 
-"ALVOOT 10 GOLS/AV 1d ‘LWddad “ASV
+¢ TIMING CORRECTION works during playback and operates without ‘chopping’ notes.
 
-SUIpNpoUr ‘suOTIOUN] posn A[UOUILUOS 94] JO AUBUT [O1]UOD AJ9]OWIAI 0} PousIsse oq ACUI ST AdNI HOLIMSLOO OME «
-“SUIPIONAI I[IYM P2sesd JOU Iv $3}OU BUTISIXO—ZUIPIOIA SATON.ASOP-UON
+¢ Optional SMPTE time code synchronization.
 
-‘suoneurldxa peuoyippe sdeydsip uowng g1TqH
+© Optional remote control.
 
-oy] ‘pepsau JI ‘suoneiodo [ye yYsnosy] NOA sapins ApIespo Avfdsip QO] Joey Z7¢ 9y3—uoeIodo Urea] 0} Ased ‘aus «
+Recording a Sequence
 
-jUorel]suowtap & IO} Aepol Jayeap uur’] INOA dag ‘dISHUL
-INOA 0} UONUS}]¥ PaplAIPUN INOA SUTJOASp ITY ps pue
-p1osai ‘asoduod no Jay 0} pausisap st 1s0uenbesuur’] oy)
-Aum Aposiooid $,Jeu], ‘SS9d0Id SATTBS1D OY} YIM SOIOJIOIUT
-yey) xo]dwWI0d Os dq JOA9U P[NoUsS osn NOA AZopOuYdE} oy
+To record a sequence, simply press RECORD and PLAY,
+then play your MIDI keyboard in time to the Sequencer’s
+click track. When the sequence loops back around to bar 1,
+you’ ll hear what you played—only all timing errors will be
 
-ISTUMOIAUIO?) NOAA UOHISOdWIO)
+corrected! (Timing correction may be adjusted or defeated).
 
-"NOSpr] B Oy ‘AONUTJUT yada 0} seq Maz Se] BY] Jas UdAd
-uvd NOA ‘palisap JJ ‘souanbes Mou ¥B OVUT sjied ou] [Te Adoo
-ATesrewO Ne WI) [IM ONOS ALVAAO JeyIe80} wey}
-,deyd,, 0} UOTOUNJ ONOS ALVA ou] asn usy] ‘saouanbes
-JENPIAIpUt UI (“949 ‘snJOYD ‘aS1OA) UOTIDIS JIseq Yes
-Pl0da1 OF ST ABM JOuIOUY “(812g 666 01 dn) ysnory) ABM
+Any additional notes played will be added into the track
+— existing notes are not erased while recording!
 
-dU} [fe YORI] YORs p10991 0} ST SUOS B 9789I9 0} ABM SUG,
+FAST FORWARD, REWIND, and LOCATE controls
+may be used at any time to quickly access any location in
+your sequence for spot-recording. To overdub a new part,
+select a different track and start recording—while you
+record, the first track will play in perfect sync (unless you
+MUTE it, or SOLO another track). In this way, up to 32
+tracks may be overdubbed! All MIDI effects are recorded
+including pitch bend, modulation, velocity, aftertouch,
+sustain pedal, and program changes!
 
-SUOS & SUTVAID
+Editing
 
-*suoT}oes poJUBMUN
+To erase a wrong note, simply hold ERASE and press
+the note to be erased just before it plays in the sequence—
+when played back, it will be gone. Notes may also be
 
-SAOUIOI 0} ABM SWS dU} SoyeIodo SUV ALATAaG
+added, erased, or changed using the SINGLE STEP func-
+tion. To overdub notes at specific points within a sequence,
 
-“OBPLIq dy} PUB SNIOY PUOdAS dT]] Ud9MIAQ SIDA ISI
+Additional Features
 
-ay) Jo Adoo B JJasuT WYSE NOAA ‘afdwexs 10.f ‘UO JUSIN]JIP
+simply use LOCATE, FAST FORWARD, or REWIND to
+find the desired bar number, then start recording.
 
-B IO aouaNbas sues OY} UI—JOY OUP 0} UOTIEIO] 9UO WOT]
-$1Bq JAOUI OF NOA sMOTIe WOTIOUNS AdOO/IMASNI OULL
+The INSERT/COPY function allows you to move bars
+from one location to another—in the same sequence or a
+different one. For example, you might insert a copy of the
+first verse between the second chorus and the bridge.
+DELETE BARS operates the same way to remove
+unwanted sections,
 
-‘SUIPIONAI JIVIS Udy) “OQuINU eq porisop ay} puy
+Creating a Song
 
-0} CNIMAY 10 ‘CYVM Od LSWA “AEVOOT esn Apduns
+One way to create a song is to record each track all the
+way through (up to 999 bars). Another way is to record
+each basic section (verse, chorus, etc.) in individual
+sequences, then use the CREATE SONG function to “chain”
+them together. CREATE SONG will then automatically
+copy all the parts into a new sequence. If desired, you can
+even set the last few bars to repeat infinitely, for a fadeout.
 
-sainjeay [PUOHIPPY
+Composition Without Compromise
 
-‘gouanbas & UTYIIM s]UTOd a1y1dads 3¥ $9100 QnPIOAO OL "UOT}
--ouns dALLS ATONIS 24) Suisn pasueyo Jo ‘pasesa ‘pappe
-aq osye ABUT S9]ON ‘U0 9q ]IIM 1 “yoeq podeyd uayM
-—aouanbas oy] ul skeyd 71 a10J9q Isnf posers oq 0} d]0U ayy
-ssaid pue ASvwug ploy Aydunis ‘jou Suomm & aseso OL
+The technology you use should never be so complex that
+it interferes with the creative process. That’s precisely why
+the LinnSequencer is designed to let you compose, record
+and edit while devoting your undivided attention to your
+music. See your Linn dealer today for a demonstration!
 
-sunipa
+* Simple, easy to learn operation—the 32 character LCD display clearly guides you through all operations. If needed, the
 
-jsesdueyo ureisoid pue ‘fepod ureysns
-‘yonoplalje ‘AWOOTOA ‘UOTyeTNpow ‘pusg youd Surpnyour
-pep10del are $199JJ2 TCTIN [WV iPeqqnpseao aq Aeur syoen
-Ze 07 dn ‘Kem sie Uy *(foeI} JOyOUR OJOS 10 ALLAN
-NOA ssofum) duAS yOaysod ul Avy [[IM Yow] ISI 93 “prooar
-NOA 3[IYM—SUIPIOIA LIBIS PU YORI) TUdIOTJIP B JOaTas
-*y1ed MOU B QNPIsA0 OL, “SuIps0daJ-jods 10} aouanbes mno0k
-UI UOHBIO] Aue ssad0e ATYOIND 0} owt} Aue ye pasn aq AvUE
-SJONUOD FLIVOOT pur ‘ANIMA ‘CYVMaYOd LSVd
-{SUIPIOSAI {IY posesa JOU se So]OU SuTsTXO—
-yous} 3U} OUT poppe aq JIM poteyd sajou yeuonippe Auy
+HELP button displays additional explanations.
 
-*(povesjap 10 poysn{pe oq ABW UOTIIII0D BUTUTT]) j{paqoeLI09
+* Non-destructive recording—existing notes are not erased while recording.
+¢ Two FOOTSWITCH INPUTS may be assigned to remotely control many of the commonly used functions, including
 
-2q ][IM S1OLIe Sur [fe ATUO—patey]d nod Jey Jedy ]],NOA
+ERASE, REPEAT, PLAY/STOP, or LOCATE.
 
-‘] req 0] punose yoeq sdoo] sduanbas ay] Udy AA “YOu Yor
+¢ Iwo TRIGGER OUTPUTS may be programmed to output pulses at any selected note value.
 
-§,sa0uaNbas at} O] SUIT) UI preogday [IW] INO Avy usy3
-AV'1d pue (YOON ssoid Ayduus ‘aousnbes & p1o09es OF,
+© Will sync to standard LinnDrum or Linn 9000 sync tone.
 
-g0uaNbas & SUIP10I0y]
+© 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,
 
-‘JONWOD s}JouNaI TeuONdGO e
+(even drop frame!)
 
-"UOTJEZIUOIYUAS OPOS UIT} FLAWS [euondo e
+¢ TEMPO may be entered numerically, adjustable in tenths of a Beat-Per-Minute increments, or by tapping quarter notes
 
-‘sou .sulddoys, noyyM sayelodo pue yoegdvyd ZuLINp S¥IOM NOLLOANNYOO ONIWILL e
+on the TAP TEMPO button.
 
-‘onqea ory AY
+¢ TEMPO CHANGES may be programmed into a sequence, with smooth transitions if desired.
+¢ Any TIME SIGNATURE may be used, and may be changed within a song.
 
-pojoojes-oid & ye sajou pyoy Aue syeadas ATTeONewWO Ne UOTOUNS [WAdAY OAISNOX e
-‘LSVJ SUnIpS soyeu UOTOUN ASV UA OUlN-[eal SAISNIOXY e
-‘Koy B JO YONO} 941 12 CASOdSNVALL 0g ABU Syde] [Te 10 9UC e
+linn
+Linn Electronics, Inc.
 
-i ASIP Jed
-
-S9}0U OOO‘OTT JOA SpfOy puv SpUOdeS UT SBUOS Xa[AUIOD So10}S DALIP YSIP , 74 € ISCJ-CNIN
-
-jSIOZISOUJUAS
-
-stuoydAjod of 0} dn skeyd A[snoourynuls ‘spouueYd [IW 9T JO duo 0} pousisse oq
-ABUL YORI] YOR ‘syous) oruoydAjod ‘snoouelnurs 7¢ SuTeJUOS ssouUaNbas QO] OY} JO YORA e
-
-‘SJONUOS ATWOOT pur ‘GNIMAY ‘GaVM OA
-LSVd ‘GYOOde AOLS ‘AV Td YIM Jopsocas ade} Yowsj-N[NU O} eps st UOTLISdO @
-LOPNOUT SaINjeoy s[quyIeUlss AUB S.JJ ‘OSN pue UIes] 0} o[duns A[suIzeUe JOA ‘PnJsomod APOUIOITXO
-St 1] “UeIOIsNUL feUOIssajoid oY} 10 JOO} soUBULIOJIJAd pue UOTIsOduIOS 11e-dY1-JO-9}e)s B SI IONUANbDaguUT] ay
-
-JOps1odady soUINbIS [GTI YVAL ZE
-Jgouanbaguury oy
+18720 Oxnard Street, Tarzana, CA 91356
+(818) 708-8131 TELEX #298949 LINN UR
 

tests/cache/cardinal/__-l__eng__000004_ocr.png__000004_ocr_hocr__hocr__txt/txt.bin

@@ -1,124 +1,123 @@
-2A NNI‘I 6F6867# XATALL IE18-80L (818)
+The LinnSequencer
+32 Track MIDI Sequence Recorder
 
-9SEI6 VO “BUBZIRY, “J0aNS PIPUXO OZLEI
-“Uy ‘soTUOMOI,q UUrT
+The LinnSequencer is a state-of-the-art composition and performance tool for the professional musician. It is
 
-uu
+extremely powerful, yet amazingly simple to learn and use. It’s many remarkable features include:
 
-“‘SUOS B UIJIM pasueyo oq ABU pue ‘posn oq AWW AYN IVNOIS AWLL AUV
-“parlsop Jr SUOTIISUBI} YIOOUIS YIM “BoueNbas eB OJUI pourtueIZOId 9q ABU SFONWHO OdINAL e
+¢ Operation is similar to multi-track tape recorder with PLAY, STOP, RECORD, FAST
+FORWARD, REWIND, and LOCATE controls.
 
-‘uonng OdNAL dV L 9) uO
+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
 
-sojou Jayienb Suiddy} Aq 10 ‘syUSTIOIOUI oINUTIAI-J8g-Jesg & JO sys} UL ofquisn(pe ‘ATTeouIAUINU paiajus oq ABU OdINALL e
+synthesizers!
 
-(jouer doup u3a9)
+¢ Ultra-fast 3%” disk drive stores complex songs in seconds and holds over 110,000 notes
 
-“puooes Jed souely O€ 10 “SZ “pz 18 [LVAG-MAd-SHN VU 10 ALOANIWAAd-SLVAd U! patyoeds aq kewl OAL «
-‘uoTe1odo [SVx JO} Aj[eusoyUT JoyndUIOd 11g 9] 98108 ZHI 8g ‘poeds-ysry Bann soz] e
+per disk!
 
-"9U0} DUAS 0006 UUL] Jo wNIqUUr] prepue}s 0} OUAS [ITAA ©
+¢ One or all tracks may be TRANSPOSED at the touch of a key.
+e Exclusive real-time ERASE function makes editing FAST.
+* Exclusive REPEAT function automatically repeats any held notes at a pre-selected
 
-“ONYBA 9}OU poloapes Aue Je sas—nd jndyno 07 pewureigold 3q ACW SL Ad LNO YADONAL OML
+rhythmic value.
 
-"ALVOOT 10 GOLS/AV 1d ‘LWddad “ASV
+¢ TIMING CORRECTION works during playback and operates without ‘chopping’ notes.
 
-SUIpNpoUr ‘suOTIOUN] posn A[UOUILUOS 94] JO AUBUT [O1]UOD AJ9]OWIAI 0} PousIsse oq ACUI ST AdNI HOLIMSLOO OME «
-“SUIPIONAI I[IYM P2sesd JOU Iv $3}OU BUTISIXO—ZUIPIOIA SATON.ASOP-UON
+¢ Optional SMPTE time code synchronization.
 
-‘suoneurldxa peuoyippe sdeydsip uowng g1TqH
+© Optional remote control.
 
-oy] ‘pepsau JI ‘suoneiodo [ye yYsnosy] NOA sapins ApIespo Avfdsip QO] Joey Z7¢ 9y3—uoeIodo Urea] 0} Ased ‘aus «
+Recording a Sequence
 
-jUorel]suowtap & IO} Aepol Jayeap uur’] INOA dag ‘dISHUL
-INOA 0} UONUS}]¥ PaplAIPUN INOA SUTJOASp ITY ps pue
-p1osai ‘asoduod no Jay 0} pausisap st 1s0uenbesuur’] oy)
-Aum Aposiooid $,Jeu], ‘SS9d0Id SATTBS1D OY} YIM SOIOJIOIUT
-yey) xo]dwWI0d Os dq JOA9U P[NoUsS osn NOA AZopOuYdE} oy
+To record a sequence, simply press RECORD and PLAY,
+then play your MIDI keyboard in time to the Sequencer’s
+click track. When the sequence loops back around to bar 1,
+you’ ll hear what you played—only all timing errors will be
 
-ISTUMOIAUIO?) NOAA UOHISOdWIO)
+corrected! (Timing correction may be adjusted or defeated).
 
-"NOSpr] B Oy ‘AONUTJUT yada 0} seq Maz Se] BY] Jas UdAd
-uvd NOA ‘palisap JJ ‘souanbes Mou ¥B OVUT sjied ou] [Te Adoo
-ATesrewO Ne WI) [IM ONOS ALVAAO JeyIe80} wey}
-,deyd,, 0} UOTOUNJ ONOS ALVA ou] asn usy] ‘saouanbes
-JENPIAIpUt UI (“949 ‘snJOYD ‘aS1OA) UOTIDIS JIseq Yes
-Pl0da1 OF ST ABM JOuIOUY “(812g 666 01 dn) ysnory) ABM
+Any additional notes played will be added into the track
+— existing notes are not erased while recording!
 
-dU} [fe YORI] YORs p10991 0} ST SUOS B 9789I9 0} ABM SUG,
+FAST FORWARD, REWIND, and LOCATE controls
+may be used at any time to quickly access any location in
+your sequence for spot-recording. To overdub a new part,
+select a different track and start recording—while you
+record, the first track will play in perfect sync (unless you
+MUTE it, or SOLO another track). In this way, up to 32
+tracks may be overdubbed! All MIDI effects are recorded
+including pitch bend, modulation, velocity, aftertouch,
+sustain pedal, and program changes!
 
-SUOS & SUTVAID
+Editing
 
-*suoT}oes poJUBMUN
+To erase a wrong note, simply hold ERASE and press
+the note to be erased just before it plays in the sequence—
+when played back, it will be gone. Notes may also be
 
-SAOUIOI 0} ABM SWS dU} SoyeIodo SUV ALATAaG
+added, erased, or changed using the SINGLE STEP func-
+tion. To overdub notes at specific points within a sequence,
 
-“OBPLIq dy} PUB SNIOY PUOdAS dT]] Ud9MIAQ SIDA ISI
+Additional Features
 
-ay) Jo Adoo B JJasuT WYSE NOAA ‘afdwexs 10.f ‘UO JUSIN]JIP
+simply use LOCATE, FAST FORWARD, or REWIND to
+find the desired bar number, then start recording.
 
-B IO aouaNbas sues OY} UI—JOY OUP 0} UOTIEIO] 9UO WOT]
-$1Bq JAOUI OF NOA sMOTIe WOTIOUNS AdOO/IMASNI OULL
+The INSERT/COPY function allows you to move bars
+from one location to another—in the same sequence or a
+different one. For example, you might insert a copy of the
+first verse between the second chorus and the bridge.
+DELETE BARS operates the same way to remove
+unwanted sections,
 
-‘SUIPIONAI JIVIS Udy) “OQuINU eq porisop ay} puy
+Creating a Song
 
-0} CNIMAY 10 ‘CYVM Od LSWA “AEVOOT esn Apduns
+One way to create a song is to record each track all the
+way through (up to 999 bars). Another way is to record
+each basic section (verse, chorus, etc.) in individual
+sequences, then use the CREATE SONG function to “chain”
+them together. CREATE SONG will then automatically
+copy all the parts into a new sequence. If desired, you can
+even set the last few bars to repeat infinitely, for a fadeout.
 
-sainjeay [PUOHIPPY
+Composition Without Compromise
 
-‘gouanbas & UTYIIM s]UTOd a1y1dads 3¥ $9100 QnPIOAO OL "UOT}
--ouns dALLS ATONIS 24) Suisn pasueyo Jo ‘pasesa ‘pappe
-aq osye ABUT S9]ON ‘U0 9q ]IIM 1 “yoeq podeyd uayM
-—aouanbas oy] ul skeyd 71 a10J9q Isnf posers oq 0} d]0U ayy
-ssaid pue ASvwug ploy Aydunis ‘jou Suomm & aseso OL
+The technology you use should never be so complex that
+it interferes with the creative process. That’s precisely why
+the LinnSequencer is designed to let you compose, record
+and edit while devoting your undivided attention to your
+music. See your Linn dealer today for a demonstration!
 
-sunipa
+* Simple, easy to learn operation—the 32 character LCD display clearly guides you through all operations. If needed, the
 
-jsesdueyo ureisoid pue ‘fepod ureysns
-‘yonoplalje ‘AWOOTOA ‘UOTyeTNpow ‘pusg youd Surpnyour
-pep10del are $199JJ2 TCTIN [WV iPeqqnpseao aq Aeur syoen
-Ze 07 dn ‘Kem sie Uy *(foeI} JOyOUR OJOS 10 ALLAN
-NOA ssofum) duAS yOaysod ul Avy [[IM Yow] ISI 93 “prooar
-NOA 3[IYM—SUIPIOIA LIBIS PU YORI) TUdIOTJIP B JOaTas
-*y1ed MOU B QNPIsA0 OL, “SuIps0daJ-jods 10} aouanbes mno0k
-UI UOHBIO] Aue ssad0e ATYOIND 0} owt} Aue ye pasn aq AvUE
-SJONUOD FLIVOOT pur ‘ANIMA ‘CYVMaYOd LSVd
-{SUIPIOSAI {IY posesa JOU se So]OU SuTsTXO—
-yous} 3U} OUT poppe aq JIM poteyd sajou yeuonippe Auy
-*(povesjap 10 poysn{pe oq ABW UOTIIII0D BUTUTT]) j{paqoeLI09
-2q ][IM S1OLIe Sur [fe ATUO—patey]d nod Jey Jedy ]],NOA
-‘] req 0] punose yoeq sdoo] sduanbas ay] Udy AA “YOu Yor
-§,sa0uaNbas at} O] SUIT) UI preogday [IW] INO Avy usy3
-AV'1d pue (YOON ssoid Ayduus ‘aousnbes & p1o09es OF,
+HELP button displays additional explanations.
 
-g0uaNbas & SUIP10I0y]
+* Non-destructive recording—existing notes are not erased while recording.
+¢ Two FOOTSWITCH INPUTS may be assigned to remotely control many of the commonly used functions, including
 
-‘JONWOD s}JouNaI TeuONdGO e
+ERASE, REPEAT, PLAY/STOP, or LOCATE.
 
-"UOTJEZIUOIYUAS OPOS UIT} FLAWS [euondo e
+¢ Iwo TRIGGER OUTPUTS may be programmed to output pulses at any selected note value.
 
-‘sou .sulddoys, noyyM sayelodo pue yoegdvyd ZuLINp S¥IOM NOLLOANNYOO ONIWILL e
+© Will sync to standard LinnDrum or Linn 9000 sync tone.
 
-‘onqea ory AY
+© 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,
 
-pojoojes-oid & ye sajou pyoy Aue syeadas ATTeONewWO Ne UOTOUNS [WAdAY OAISNOX e
-‘LSVJ SUnIpS soyeu UOTOUN ASV UA OUlN-[eal SAISNIOXY e
-‘Koy B JO YONO} 941 12 CASOdSNVALL 0g ABU Syde] [Te 10 9UC e
+(even drop frame!)
 
-i ASIP Jed
+¢ TEMPO may be entered numerically, adjustable in tenths of a Beat-Per-Minute increments, or by tapping quarter notes
 
-S9}0U OOO‘OTT JOA SpfOy puv SpUOdeS UT SBUOS Xa[AUIOD So10}S DALIP YSIP , 74 € ISCJ-CNIN
+on the TAP TEMPO button.
 
-jSIOZISOUJUAS
+¢ TEMPO CHANGES may be programmed into a sequence, with smooth transitions if desired.
+¢ Any TIME SIGNATURE may be used, and may be changed within a song.
 
-stuoydAjod of 0} dn skeyd A[snoourynuls ‘spouueYd [IW 9T JO duo 0} pousisse oq
-ABUL YORI] YOR ‘syous) oruoydAjod ‘snoouelnurs 7¢ SuTeJUOS ssouUaNbas QO] OY} JO YORA e
+linn
+Linn Electronics, Inc.
 
-‘SJONUOS ATWOOT pur ‘GNIMAY ‘GaVM OA
-LSVd ‘GYOOde AOLS ‘AV Td YIM Jopsocas ade} Yowsj-N[NU O} eps st UOTLISdO @
-LOPNOUT SaINjeoy s[quyIeUlss AUB S.JJ ‘OSN pue UIes] 0} o[duns A[suIzeUe JOA ‘PnJsomod APOUIOITXO
-St 1] “UeIOIsNUL feUOIssajoid oY} 10 JOO} soUBULIOJIJAd pue UOTIsOduIOS 11e-dY1-JO-9}e)s B SI IONUANbDaguUT] ay
-
-JOps1odady soUINbIS [GTI YVAL ZE
-Jgouanbaguury oy
+18720 Oxnard Street, Tarzana, CA 91356
+(818) 708-8131 TELEX #298949 LINN UR
 

tests/cache/ccitt/__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt/hocr.bin

@@ -9,7 +9,7 @@
   <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 "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/com.github.ocrmypdf.ucmqv4tt/000001_ocr.png"; bbox 0 0 2550 3300; ppageno 0'>
+  <div class='ocr_page' id='page_1' title='image "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/ocrmypdf.io.sly7uqzv/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">
@@ -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 92'>LinnSequencer</span>
+      <span class='ocrx_word' id='word_1_9' title='bbox 424 380 676 417; x_wconf 91'>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>
@@ -39,7 +39,7 @@
       <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 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_19' title='bbox 1737 380 1940 417; x_wconf 96'>professional</span>
       <span class='ocrx_word' id='word_1_20' title='bbox 1952 380 2112 411; x_wconf 96'>musician.</span>
       <span class='ocrx_word' id='word_1_21' title='bbox 2127 381 2152 410; x_wconf 96'>It</span>
       <span class='ocrx_word' id='word_1_22' title='bbox 2164 380 2188 410; x_wconf 96'>is</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 96'>rhythmic</span>
+      <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_136' title='bbox 577 1021 689 1058; x_wconf 96'>value.</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 74'>©</span>
+      <span class='ocrx_word' id='word_1_154' title='bbox 349 1203 367 1222; x_wconf 70'>©</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>
@@ -300,8 +300,8 @@
       <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 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_191' title='bbox 346 1457 430 1490; x_wconf 15'>you’</span>
+      <span class='ocrx_word' id='word_1_192' title='bbox 406 1453 439 1496; x_wconf 15'>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>
@@ -361,7 +361,7 @@
       <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>
       <span class='ocrx_word' id='word_1_232' title='bbox 924 1616 1068 1643; x_wconf 96'>LOCATE</span>
-      <span class='ocrx_word' id='word_1_233' title='bbox 1079 1616 1199 1643; x_wconf 95'>controls</span>
+      <span class='ocrx_word' id='word_1_233' title='bbox 1079 1616 1199 1643; x_wconf 96'>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 92'>may</span>
@@ -400,9 +400,9 @@
      </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 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_264' title='bbox 460 1774 506 1801; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_265' title='bbox 517 1773 577 1801; x_wconf 96'>first</span>
+      <span class='ocrx_word' id='word_1_266' title='bbox 587 1774 662 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>
@@ -412,11 +412,11 @@
       <span class='ocrx_word' id='word_1_273' title='bbox 1148 1782 1203 1807; x_wconf 96'>you</span>
      </span>
      <span class='ocr_line' id='line_1_32' title="bbox 346 1813 1191 1847; baseline 0.002 -8; x_size 33; x_descenders 5; x_ascenders 9">
-      <span class='ocrx_word' id='word_1_274' title='bbox 346 1813 454 1840; x_wconf 95'>MUTE</span>
-      <span class='ocrx_word' id='word_1_275' title='bbox 464 1813 492 1845; x_wconf 95'>it,</span>
+      <span class='ocrx_word' id='word_1_274' title='bbox 346 1813 454 1840; x_wconf 94'>MUTE</span>
+      <span class='ocrx_word' id='word_1_275' title='bbox 464 1813 492 1845; x_wconf 94'>it,</span>
       <span class='ocrx_word' id='word_1_276' title='bbox 505 1821 537 1840; x_wconf 95'>or</span>
       <span class='ocrx_word' id='word_1_277' title='bbox 547 1813 642 1840; x_wconf 95'>SOLO</span>
-      <span class='ocrx_word' id='word_1_278' title='bbox 653 1814 769 1841; x_wconf 95'>another</span>
+      <span class='ocrx_word' id='word_1_278' title='bbox 653 1814 769 1841; x_wconf 96'>another</span>
       <span class='ocrx_word' id='word_1_279' title='bbox 779 1814 875 1847; x_wconf 94'>track).</span>
       <span class='ocrx_word' id='word_1_280' title='bbox 889 1814 920 1840; x_wconf 96'>In</span>
       <span class='ocrx_word' id='word_1_281' title='bbox 930 1813 984 1841; x_wconf 96'>this</span>
@@ -518,13 +518,13 @@
      <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 91'>overdub</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_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 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_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_358' title='bbox 1086 2216 1228 2242; x_wconf 96'>sequence,</span>
      </span>
     </p>
@@ -681,7 +681,7 @@
       <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>
-      <span class='ocrx_word' id='word_1_468' title='bbox 2103 1808 2215 1836; x_wconf 93'>“chain”</span>
+      <span class='ocrx_word' id='word_1_468' title='bbox 2103 1808 2215 1836; x_wconf 91'>“chain”</span>
      </span>
      <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>
@@ -874,10 +874,10 @@
    <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 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_604' title='bbox 347 2561 361 2575; x_wconf 85'>¢</span>
+      <span class='ocrx_word' id='word_1_605' title='bbox 372 2556 433 2583; x_wconf 84'>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_607' title='bbox 623 2556 797 2584; x_wconf 95'>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>
       <span class='ocrx_word' id='word_1_610' title='bbox 925 2557 1119 2590; x_wconf 96'>programmed</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 97'>note</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_618' title='bbox 1683 2556 1768 2583; x_wconf 96'>value.</span>
      </span>
     </p>
@@ -911,15 +911,15 @@
    </div>
    <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 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='ocr_header' id='line_1_72' title="bbox 347 2648 1664 2682; baseline 0 -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 51'>©</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_632' title='bbox 493 2648 564 2677; 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 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 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_636' title='bbox 873 2649 954 2676; x_wconf 95'>80186</span>
+      <span class='ocrx_word' id='word_1_637' title='bbox 965 2649 994 2676; x_wconf 95'>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>
       <span class='ocrx_word' id='word_1_640' title='bbox 1208 2649 1350 2682; x_wconf 96'>internally</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 81'>¢</span>
+      <span class='ocrx_word' id='word_1_664' title='bbox 347 2782 360 2796; x_wconf 79'>¢</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 96'>be</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_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 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_670' title='bbox 946 2777 1101 2811; x_wconf 96'>adjustable</span>
+      <span class='ocrx_word' id='word_1_671' title='bbox 1111 2777 1139 2804; x_wconf 96'>in</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 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_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_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>

tests/cache/jbig2/__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt/hocr.bin

@@ -9,459 +9,467 @@
   <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 "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/com.github.ocrmypdf.1ca2h7iv/000001_ocr.png"; bbox 0 0 1000 1520; ppageno 0'>
-   <div class='ocr_carea' id='block_1_1' title="bbox 126 101 547 145">
-    <p class='ocr_par' id='par_1_1' lang='eng' title="bbox 126 101 547 145">
-     <span class='ocr_line' id='line_1_1' title="bbox 126 101 547 125; baseline 0.005 -7; x_size 23; x_descenders 6; x_ascenders 4">
-      <span class='ocrx_word' id='word_1_1' title='bbox 126 101 173 118; x_wconf 37'>41st</span>
-      <span class='ocrx_word' id='word_1_2' title='bbox 181 101 303 124; x_wconf 16'>CONGRESS;</span>
-      <span class='ocrx_word' id='word_1_3' title='bbox 329 123 331 125; x_wconf 42'>|</span>
-      <span class='ocrx_word' id='word_1_4' title='bbox 443 103 547 120; x_wconf 94'>SENATE.</span>
+  <div class='ocr_page' id='page_1' title='image "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/ocrmypdf.io.6nqy9raf/000001_ocr.png"; bbox 0 0 1000 1520; ppageno 0'>
+   <div class='ocr_carea' id='block_1_1' title="bbox 122 99 848 146">
+    <p class='ocr_par' id='par_1_1' lang='eng' title="bbox 122 99 848 146">
+     <span class='ocr_line' id='line_1_1' title="bbox 122 99 848 146; baseline 0 -26; x_size 22; x_descenders 5; x_ascenders 4">
+      <span class='ocrx_word' id='word_1_1' title='bbox 122 104 169 121; x_wconf 29'>4ist</span>
+      <span class='ocrx_word' id='word_1_2' title='bbox 177 103 291 125; x_wconf 54'>ConGREss,</span>
+      <span class='ocrx_word' id='word_1_3' title='bbox 300 105 308 146; x_wconf 46'>}</span>
+      <span class='ocrx_word' id='word_1_4' title='bbox 439 99 543 130; x_wconf 94'>SENATE.</span>
+      <span class='ocrx_word' id='word_1_5' title='bbox 736 105 745 146; x_wconf 90'>{</span>
+      <span class='ocrx_word' id='word_1_6' title='bbox 753 103 788 120; x_wconf 94'>Ex.</span>
+      <span class='ocrx_word' id='word_1_7' title='bbox 798 103 848 121; x_wconf 95'>Doc,</span>
      </span>
-     <span class='ocr_line' id='line_1_2' title="bbox 156 104 312 145; baseline 0.006 -5; x_size 21; x_descenders 4; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_5' title='bbox 156 124 180 140; x_wconf 89'>3d</span>
-      <span class='ocrx_word' id='word_1_6' title='bbox 189 124 264 141; x_wconf 91'>Session.</span>
-      <span class='ocrx_word' id='word_1_7' title='bbox 304 104 312 145; x_wconf 52'>}</span>
+     <span class='ocr_line' id='line_1_2' title="bbox 152 125 838 143; baseline -0.001 0; x_size 21.943821; x_descenders 4.94382; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_8' title='bbox 152 126 176 142; x_wconf 93'>3d</span>
+      <span class='ocrx_word' id='word_1_9' title='bbox 185 126 260 143; x_wconf 88'>Session.</span>
+      <span class='ocrx_word' id='word_1_10' title='bbox 767 125 802 142; x_wconf 86'>No.</span>
+      <span class='ocrx_word' id='word_1_11' title='bbox 810 126 838 142; x_wconf 76'>25.</span>
      </span>
     </p>
    </div>
-   <div class='ocr_carea' id='block_1_2' title="bbox 388 244 597 269">
-    <p class='ocr_par' id='par_1_2' lang='eng' title="bbox 388 244 597 269">
-     <span class='ocr_line' id='line_1_3' title="bbox 388 244 597 269; baseline 0.005 -1; x_size 32.400002; x_descenders 8.1000004; x_ascenders 8.1000004">
-      <span class='ocrx_word' id='word_1_8' title='bbox 388 244 597 269; x_wconf 96'>MESSAGE</span>
+   <div class='ocr_carea' id='block_1_2' title="bbox 384 245 594 269">
+    <p class='ocr_par' id='par_1_2' lang='eng' title="bbox 384 245 594 269">
+     <span class='ocr_line' id='line_1_3' title="bbox 384 245 594 269; baseline 0 0; x_size 32.666668; x_descenders 8.166667; x_ascenders 8.166667">
+      <span class='ocrx_word' id='word_1_12' title='bbox 384 245 594 269; x_wconf 94'>MESSAGE</span>
      </span>
     </p>
    </div>
-   <div class='ocr_carea' id='block_1_3' title="bbox 462 302 526 313">
-    <p class='ocr_par' id='par_1_3' lang='eng' title="bbox 462 302 526 313">
-     <span class='ocr_line' id='line_1_4' title="bbox 462 302 526 313; baseline 0 0; x_size 20; x_descenders 5; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_9' title='bbox 462 302 484 313; x_wconf 87'>OF</span>
-      <span class='ocrx_word' id='word_1_10' title='bbox 493 302 526 313; x_wconf 96'>THE</span>
+   <div class='ocr_carea' id='block_1_3' title="bbox 459 302 523 313">
+    <p class='ocr_par' id='par_1_3' lang='eng' title="bbox 459 302 523 313">
+     <span class='ocr_line' id='line_1_4' title="bbox 459 302 523 313; baseline 0 0; x_size 20; x_descenders 5; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_13' title='bbox 459 302 481 313; x_wconf 87'>OF</span>
+      <span class='ocrx_word' id='word_1_14' title='bbox 490 302 523 313; x_wconf 96'>THE</span>
      </span>
     </p>
    </div>
-   <div class='ocr_carea' id='block_1_4' title="bbox 151 337 842 377">
-    <p class='ocr_par' id='par_1_4' lang='eng' title="bbox 151 337 842 377">
-     <span class='ocr_line' id='line_1_5' title="bbox 151 337 842 377; baseline 0.009 -8; x_size 43.365078; x_descenders 10.841269; x_ascenders 10.841269">
-      <span class='ocrx_word' id='word_1_11' title='bbox 151 337 361 370; x_wconf 96'>PRESIDENT</span>
-      <span class='ocrx_word' id='word_1_12' title='bbox 382 338 425 370; x_wconf 96'>OF</span>
-      <span class='ocrx_word' id='word_1_13' title='bbox 447 338 521 371; x_wconf 96'>THE</span>
-      <span class='ocrx_word' id='word_1_14' title='bbox 542 339 676 372; x_wconf 96'>UNITED</span>
-      <span class='ocrx_word' id='word_1_15' title='bbox 697 340 842 377; x_wconf 95'>STATES,</span>
+   <div class='ocr_carea' id='block_1_4' title="bbox 148 338 839 375">
+    <p class='ocr_par' id='par_1_4' lang='eng' title="bbox 148 338 839 375">
+     <span class='ocr_line' id='line_1_5' title="bbox 148 338 839 375; baseline 0.003 -3; x_size 43.490196; x_descenders 10.872549; x_ascenders 10.872549">
+      <span class='ocrx_word' id='word_1_15' title='bbox 148 339 358 372; x_wconf 96'>PRESIDENT</span>
+      <span class='ocrx_word' id='word_1_16' title='bbox 379 339 422 371; x_wconf 95'>OF</span>
+      <span class='ocrx_word' id='word_1_17' title='bbox 444 338 518 371; x_wconf 96'>THE</span>
+      <span class='ocrx_word' id='word_1_18' title='bbox 539 338 673 371; x_wconf 96'>UNITED</span>
+      <span class='ocrx_word' id='word_1_19' title='bbox 694 338 839 375; x_wconf 96'>STATES,</span>
      </span>
     </p>
    </div>
-   <div class='ocr_carea' id='block_1_5' title="bbox 425 406 565 417">
-    <p class='ocr_par' id='par_1_5' lang='eng' title="bbox 425 406 565 417">
-     <span class='ocr_line' id='line_1_6' title="bbox 425 406 565 417; baseline 0 -1; x_size 20; x_descenders 5; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_16' title='bbox 425 406 565 417; x_wconf 94'>COMMUNICATING</span>
+   <div class='ocr_carea' id='block_1_5' title="bbox 423 406 563 417">
+    <p class='ocr_par' id='par_1_5' lang='eng' title="bbox 423 406 563 417">
+     <span class='ocr_line' id='line_1_6' title="bbox 423 406 563 417; baseline 0 -1; x_size 20; x_descenders 5; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_20' title='bbox 423 406 563 417; x_wconf 95'>COMMUNICATING</span>
      </span>
     </p>
    </div>
-   <div class='ocr_carea' id='block_1_6' title="bbox 125 445 861 494">
-    <p class='ocr_par' id='par_1_6' lang='eng' title="bbox 125 445 861 494">
-     <span class='ocr_line' id='line_1_7' title="bbox 125 445 861 472; baseline 0.008 -11; x_size 22; x_descenders 5; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_17' title='bbox 125 445 143 462; x_wconf 93'>A</span>
-      <span class='ocrx_word' id='word_1_18' title='bbox 149 450 191 467; x_wconf 93'>copy</span>
-      <span class='ocrx_word' id='word_1_19' title='bbox 195 445 217 467; x_wconf 95'>of</span>
-      <span class='ocrx_word' id='word_1_20' title='bbox 222 446 331 467; x_wconf 96'>regulations</span>
-      <span class='ocrx_word' id='word_1_21' title='bbox 337 446 371 467; x_wconf 95'>for</span>
-      <span class='ocrx_word' id='word_1_22' title='bbox 376 446 403 463; x_wconf 95'>the</span>
-      <span class='ocrx_word' id='word_1_23' title='bbox 412 447 494 463; x_wconf 95'>consular</span>
-      <span class='ocrx_word' id='word_1_24' title='bbox 499 449 557 464; x_wconf 96'>courts</span>
-      <span class='ocrx_word' id='word_1_25' title='bbox 565 448 587 469; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_26' title='bbox 592 448 619 465; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_27' title='bbox 630 448 694 466; x_wconf 90'>United</span>
-      <span class='ocrx_word' id='word_1_28' title='bbox 702 448 760 466; x_wconf 93'>States</span>
-      <span class='ocrx_word' id='word_1_29' title='bbox 768 450 787 466; x_wconf 93'>in</span>
-      <span class='ocrx_word' id='word_1_30' title='bbox 795 449 861 472; x_wconf 95'>Japan,</span>
+   <div class='ocr_carea' id='block_1_6' title="bbox 123 447 859 492">
+    <p class='ocr_par' id='par_1_6' lang='eng' title="bbox 123 447 859 492">
+     <span class='ocr_line' id='line_1_7' title="bbox 123 447 859 470; baseline 0.001 -6; x_size 22; x_descenders 5; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_21' title='bbox 123 447 141 465; x_wconf 95'>A</span>
+      <span class='ocrx_word' id='word_1_22' title='bbox 147 452 189 469; x_wconf 95'>copy</span>
+      <span class='ocrx_word' id='word_1_23' title='bbox 193 447 215 469; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_24' title='bbox 220 447 329 469; x_wconf 96'>regulations</span>
+      <span class='ocrx_word' id='word_1_25' title='bbox 335 447 369 468; x_wconf 95'>for</span>
+      <span class='ocrx_word' id='word_1_26' title='bbox 374 447 401 464; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_27' title='bbox 410 447 492 464; x_wconf 96'>consular</span>
+      <span class='ocrx_word' id='word_1_28' title='bbox 497 449 555 464; x_wconf 79'>courts</span>
+      <span class='ocrx_word' id='word_1_29' title='bbox 563 447 585 469; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_30' title='bbox 590 447 617 464; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_31' title='bbox 628 447 692 465; x_wconf 96'>United</span>
+      <span class='ocrx_word' id='word_1_32' title='bbox 700 447 758 465; x_wconf 93'>States</span>
+      <span class='ocrx_word' id='word_1_33' title='bbox 766 448 785 464; x_wconf 93'>in</span>
+      <span class='ocrx_word' id='word_1_34' title='bbox 793 447 859 470; x_wconf 96'>Japan,</span>
      </span>
-     <span class='ocr_line' id='line_1_8' title="bbox 147 468 834 494; baseline 0.006 -10; x_size 22; x_descenders 5; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_31' title='bbox 147 468 215 484; x_wconf 96'>decreed</span>
-      <span class='ocrx_word' id='word_1_32' title='bbox 222 468 258 485; x_wconf 96'>and</span>
-      <span class='ocrx_word' id='word_1_33' title='bbox 265 468 323 485; x_wconf 96'>issued</span>
-      <span class='ocrx_word' id='word_1_34' title='bbox 331 468 353 489; x_wconf 96'>by</span>
-      <span class='ocrx_word' id='word_1_35' title='bbox 361 468 388 485; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_36' title='bbox 395 468 476 486; x_wconf 95'>minister</span>
-      <span class='ocrx_word' id='word_1_37' title='bbox 482 469 504 491; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_38' title='bbox 509 470 535 487; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_39' title='bbox 546 470 610 487; x_wconf 96'>United</span>
-      <span class='ocrx_word' id='word_1_40' title='bbox 618 471 676 488; x_wconf 96'>States</span>
-      <span class='ocrx_word' id='word_1_41' title='bbox 682 471 702 488; x_wconf 96'>in</span>
-      <span class='ocrx_word' id='word_1_42' title='bbox 710 471 748 488; x_wconf 95'>that</span>
-      <span class='ocrx_word' id='word_1_43' title='bbox 755 474 834 494; x_wconf 77'>country.</span>
+     <span class='ocr_line' id='line_1_8' title="bbox 145 468 832 492; baseline 0 -6; x_size 23; x_descenders 5; x_ascenders 6">
+      <span class='ocrx_word' id='word_1_35' title='bbox 145 470 213 486; x_wconf 96'>decreed</span>
+      <span class='ocrx_word' id='word_1_36' title='bbox 220 470 256 487; x_wconf 95'>and</span>
+      <span class='ocrx_word' id='word_1_37' title='bbox 263 469 321 487; x_wconf 95'>issued</span>
+      <span class='ocrx_word' id='word_1_38' title='bbox 329 469 351 490; x_wconf 96'>by</span>
+      <span class='ocrx_word' id='word_1_39' title='bbox 359 469 386 486; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_40' title='bbox 393 468 474 486; x_wconf 96'>minister</span>
+      <span class='ocrx_word' id='word_1_41' title='bbox 480 469 502 491; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_42' title='bbox 507 470 533 487; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_43' title='bbox 544 469 608 487; x_wconf 96'>United</span>
+      <span class='ocrx_word' id='word_1_44' title='bbox 616 470 674 487; x_wconf 96'>States</span>
+      <span class='ocrx_word' id='word_1_45' title='bbox 680 470 700 487; x_wconf 96'>in</span>
+      <span class='ocrx_word' id='word_1_46' title='bbox 708 470 746 487; x_wconf 96'>that</span>
+      <span class='ocrx_word' id='word_1_47' title='bbox 753 472 832 492; x_wconf 83'>country.</span>
      </span>
     </p>
    </div>
-   <div class='ocr_carea' id='block_1_7' title="bbox 122 539 861 579">
-    <p class='ocr_par' id='par_1_7' lang='eng' title="bbox 122 539 861 579">
-     <span class='ocr_caption' id='line_1_9' title="bbox 122 539 861 579; baseline 0.008 -24; x_size 18; x_descenders 4; x_ascenders 4">
-      <span class='ocrx_word' id='word_1_44' title='bbox 122 541 201 556; x_wconf 81'>January</span>
-      <span class='ocrx_word' id='word_1_45' title='bbox 207 542 230 560; x_wconf 93'>27,</span>
-      <span class='ocrx_word' id='word_1_46' title='bbox 237 542 343 560; x_wconf 39'>1871,—Read,</span>
-      <span class='ocrx_word' id='word_1_47' title='bbox 347 542 414 557; x_wconf 96'>referred</span>
-      <span class='ocrx_word' id='word_1_48' title='bbox 420 544 436 557; x_wconf 96'>to</span>
-      <span class='ocrx_word' id='word_1_49' title='bbox 442 543 470 557; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_50' title='bbox 479 539 571 579; x_wconf 66'>Committee</span>
-      <span class='ocrx_word' id='word_1_51' title='bbox 582 548 602 558; x_wconf 93'>on</span>
-      <span class='ocrx_word' id='word_1_52' title='bbox 613 544 704 563; x_wconf 96'>Commerce,</span>
-      <span class='ocrx_word' id='word_1_53' title='bbox 711 545 742 560; x_wconf 96'>and</span>
-      <span class='ocrx_word' id='word_1_54' title='bbox 749 546 813 560; x_wconf 96'>ordered</span>
-      <span class='ocrx_word' id='word_1_55' title='bbox 818 547 835 560; x_wconf 96'>to</span>
-      <span class='ocrx_word' id='word_1_56' title='bbox 841 546 861 560; x_wconf 96'>be</span>
+   <div class='ocr_carea' id='block_1_7' title="bbox 121 539 860 579">
+    <p class='ocr_par' id='par_1_7' lang='eng' title="bbox 121 539 860 579">
+     <span class='ocr_caption' id='line_1_9' title="bbox 121 539 860 579; baseline 0 -21; x_size 19; x_descenders 4; x_ascenders 4">
+      <span class='ocrx_word' id='word_1_48' title='bbox 121 544 200 559; x_wconf 95'>JANUARY</span>
+      <span class='ocrx_word' id='word_1_49' title='bbox 206 544 229 562; x_wconf 85'>27,</span>
+      <span class='ocrx_word' id='word_1_50' title='bbox 236 543 342 561; x_wconf 58'>1871,—Read,</span>
+      <span class='ocrx_word' id='word_1_51' title='bbox 346 543 413 558; x_wconf 95'>referred</span>
+      <span class='ocrx_word' id='word_1_52' title='bbox 419 545 435 558; x_wconf 96'>to</span>
+      <span class='ocrx_word' id='word_1_53' title='bbox 441 543 469 557; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_54' title='bbox 478 539 570 579; x_wconf 66'>Committee</span>
+      <span class='ocrx_word' id='word_1_55' title='bbox 581 547 601 557; x_wconf 96'>on</span>
+      <span class='ocrx_word' id='word_1_56' title='bbox 612 543 703 562; x_wconf 96'>Commerce,</span>
+      <span class='ocrx_word' id='word_1_57' title='bbox 710 543 741 558; x_wconf 96'>and</span>
+      <span class='ocrx_word' id='word_1_58' title='bbox 748 544 812 558; x_wconf 96'>ordered</span>
+      <span class='ocrx_word' id='word_1_59' title='bbox 817 545 834 558; x_wconf 96'>to</span>
+      <span class='ocrx_word' id='word_1_60' title='bbox 840 544 860 558; x_wconf 96'>be</span>
      </span>
-     <span class='ocr_caption' id='line_1_10' title="bbox 459 565 525 579; baseline 0 -4; x_size 21.5; x_descenders 5.5; x_ascenders 5.5">
-      <span class='ocrx_word' id='word_1_57' title='bbox 459 565 525 579; x_wconf 96'>printed.</span>
+     <span class='ocr_caption' id='line_1_10' title="bbox 458 565 524 579; baseline 0 -4; x_size 18.782608; x_descenders 4; x_ascenders 4.782609">
+      <span class='ocrx_word' id='word_1_61' title='bbox 458 565 524 579; x_wconf 93'>printed.</span>
      </span>
     </p>
    </div>
-   <div class='ocr_carea' id='block_1_8' title="bbox 122 629 858 769">
-    <p class='ocr_par' id='par_1_8' lang='eng' title="bbox 124 629 566 652">
-     <span class='ocr_line' id='line_1_11' title="bbox 124 629 566 652; baseline 0.005 -6; x_size 22.268028; x_descenders 5.2680273; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_58' title='bbox 124 629 150 646; x_wconf 92'>To</span>
-      <span class='ocrx_word' id='word_1_59' title='bbox 158 629 186 646; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_60' title='bbox 194 629 256 646; x_wconf 96'>Senate</span>
-      <span class='ocrx_word' id='word_1_61' title='bbox 264 630 301 647; x_wconf 96'>and</span>
-      <span class='ocrx_word' id='word_1_62' title='bbox 307 630 368 647; x_wconf 96'>House</span>
-      <span class='ocrx_word' id='word_1_63' title='bbox 376 630 398 652; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_64' title='bbox 405 630 555 652; x_wconf 93'>Representatives</span>
-      <span class='ocrx_word' id='word_1_65' title='bbox 560 638 566 648; x_wconf 92'>:</span>
+   <div class='ocr_carea' id='block_1_8' title="bbox 121 631 858 770">
+    <p class='ocr_par' id='par_1_8' lang='eng' title="bbox 123 631 565 653">
+     <span class='ocr_line' id='line_1_11' title="bbox 123 631 565 653; baseline 0 -5; x_size 22; x_descenders 5; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_62' title='bbox 123 631 149 649; x_wconf 96'>To</span>
+      <span class='ocrx_word' id='word_1_63' title='bbox 157 631 185 648; x_wconf 95'>the</span>
+      <span class='ocrx_word' id='word_1_64' title='bbox 193 631 255 648; x_wconf 95'>Senate</span>
+      <span class='ocrx_word' id='word_1_65' title='bbox 263 631 300 649; x_wconf 95'>and</span>
+      <span class='ocrx_word' id='word_1_66' title='bbox 306 631 367 648; x_wconf 95'>House</span>
+      <span class='ocrx_word' id='word_1_67' title='bbox 375 631 397 653; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_68' title='bbox 404 631 554 652; x_wconf 74'>Representatives</span>
+      <span class='ocrx_word' id='word_1_69' title='bbox 559 638 565 648; x_wconf 86'>:</span>
      </span>
     </p>
 
-    <p class='ocr_par' id='par_1_9' lang='eng' title="bbox 122 656 858 726">
-     <span class='ocr_line' id='line_1_12' title="bbox 146 656 858 681; baseline 0.008 -9; x_size 21; x_descenders 4; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_66' title='bbox 146 656 156 672; x_wconf 80'>I</span>
-      <span class='ocrx_word' id='word_1_67' title='bbox 163 657 251 673; x_wconf 80'>transmit</span>
-      <span class='ocrx_word' id='word_1_68' title='bbox 262 656 358 680; x_wconf 96'>herewith,</span>
-      <span class='ocrx_word' id='word_1_69' title='bbox 366 657 395 674; x_wconf 96'>for</span>
-      <span class='ocrx_word' id='word_1_70' title='bbox 402 658 435 674; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_71' title='bbox 442 658 578 676; x_wconf 96'>consideration</span>
-      <span class='ocrx_word' id='word_1_72' title='bbox 587 659 608 676; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_73' title='bbox 617 659 715 681; x_wconf 96'>Congress,</span>
-      <span class='ocrx_word' id='word_1_74' title='bbox 724 665 735 677; x_wconf 96'>a</span>
-      <span class='ocrx_word' id='word_1_75' title='bbox 743 661 806 681; x_wconf 96'>report</span>
-      <span class='ocrx_word' id='word_1_76' title='bbox 811 660 858 678; x_wconf 96'>from</span>
+    <p class='ocr_par' id='par_1_9' lang='eng' title="bbox 121 658 858 724">
+     <span class='ocr_line' id='line_1_12' title="bbox 145 658 857 681; baseline 0.003 -7; x_size 22; x_descenders 5; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_70' title='bbox 145 658 155 674; x_wconf 92'>I</span>
+      <span class='ocrx_word' id='word_1_71' title='bbox 162 659 250 675; x_wconf 92'>transmit</span>
+      <span class='ocrx_word' id='word_1_72' title='bbox 261 658 357 681; x_wconf 94'>herewith,</span>
+      <span class='ocrx_word' id='word_1_73' title='bbox 365 658 394 675; x_wconf 96'>for</span>
+      <span class='ocrx_word' id='word_1_74' title='bbox 401 659 434 675; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_75' title='bbox 441 658 577 675; x_wconf 96'>consideration</span>
+      <span class='ocrx_word' id='word_1_76' title='bbox 586 658 607 675; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_77' title='bbox 616 658 714 680; x_wconf 95'>Congress,</span>
+      <span class='ocrx_word' id='word_1_78' title='bbox 723 663 734 675; x_wconf 94'>a</span>
+      <span class='ocrx_word' id='word_1_79' title='bbox 742 659 805 679; x_wconf 96'>report</span>
+      <span class='ocrx_word' id='word_1_80' title='bbox 810 658 857 676; x_wconf 96'>from</span>
      </span>
-     <span class='ocr_line' id='line_1_13' title="bbox 122 678 858 704; baseline 0.008 -10; x_size 22; x_descenders 5; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_77' title='bbox 122 678 155 694; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_78' title='bbox 163 678 260 700; x_wconf 0'>Secretary</span>
-      <span class='ocrx_word' id='word_1_79' title='bbox 268 679 289 695; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_80' title='bbox 299 679 357 702; x_wconf 96'>State,</span>
-      <span class='ocrx_word' id='word_1_81' title='bbox 366 680 403 696; x_wconf 95'>and</span>
-      <span class='ocrx_word' id='word_1_82' title='bbox 411 680 444 697; x_wconf 95'>the</span>
-      <span class='ocrx_word' id='word_1_83' title='bbox 457 681 524 701; x_wconf 96'>papers</span>
-      <span class='ocrx_word' id='word_1_84' title='bbox 532 681 592 698; x_wconf 95'>which</span>
-      <span class='ocrx_word' id='word_1_85' title='bbox 605 682 736 703; x_wconf 84'>accompanied</span>
-      <span class='ocrx_word' id='word_1_86' title='bbox 744 682 765 704; x_wconf 93'>it,</span>
-      <span class='ocrx_word' id='word_1_87' title='bbox 774 687 858 700; x_wconf 92'>concern-</span>
+     <span class='ocr_line' id='line_1_13' title="bbox 121 680 858 703; baseline 0 -6; x_size 22; x_descenders 5; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_81' title='bbox 121 680 155 697; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_82' title='bbox 163 680 259 702; x_wconf 94'>Secretary</span>
+      <span class='ocrx_word' id='word_1_83' title='bbox 268 680 288 697; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_84' title='bbox 299 680 357 703; x_wconf 96'>State,</span>
+      <span class='ocrx_word' id='word_1_85' title='bbox 366 681 403 697; x_wconf 96'>and</span>
+      <span class='ocrx_word' id='word_1_86' title='bbox 410 681 444 698; x_wconf 95'>the</span>
+      <span class='ocrx_word' id='word_1_87' title='bbox 456 681 524 701; x_wconf 95'>papers</span>
+      <span class='ocrx_word' id='word_1_88' title='bbox 531 680 592 697; x_wconf 94'>which</span>
+      <span class='ocrx_word' id='word_1_89' title='bbox 605 680 736 702; x_wconf 94'>accompanied</span>
+      <span class='ocrx_word' id='word_1_90' title='bbox 744 680 765 702; x_wconf 93'>it,</span>
+      <span class='ocrx_word' id='word_1_91' title='bbox 774 685 858 698; x_wconf 92'>concern-</span>
      </span>
-     <span class='ocr_line' id='line_1_14' title="bbox 122 700 835 726; baseline 0.007 -9; x_size 21; x_descenders 4; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_88' title='bbox 122 700 155 721; x_wconf 96'>ing</span>
-      <span class='ocrx_word' id='word_1_89' title='bbox 163 700 277 721; x_wconf 96'>regulations</span>
-      <span class='ocrx_word' id='word_1_90' title='bbox 285 701 314 718; x_wconf 96'>for</span>
-      <span class='ocrx_word' id='word_1_91' title='bbox 321 702 354 719; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_92' title='bbox 361 702 446 719; x_wconf 96'>consular</span>
-      <span class='ocrx_word' id='word_1_93' title='bbox 453 703 516 720; x_wconf 96'>courts</span>
-      <span class='ocrx_word' id='word_1_94' title='bbox 524 703 546 720; x_wconf 95'>of</span>
-      <span class='ocrx_word' id='word_1_95' title='bbox 552 704 583 720; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_96' title='bbox 591 704 661 721; x_wconf 96'>United</span>
-      <span class='ocrx_word' id='word_1_97' title='bbox 669 704 732 721; x_wconf 96'>States</span>
-      <span class='ocrx_word' id='word_1_98' title='bbox 740 704 759 721; x_wconf 96'>in</span>
-      <span class='ocrx_word' id='word_1_99' title='bbox 768 705 835 726; x_wconf 94'>Japan.</span>
+     <span class='ocr_line' id='line_1_14' title="bbox 122 702 835 724; baseline 0.001 -5; x_size 21; x_descenders 4; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_92' title='bbox 122 703 155 723; x_wconf 96'>ing</span>
+      <span class='ocrx_word' id='word_1_93' title='bbox 163 702 277 723; x_wconf 96'>regulations</span>
+      <span class='ocrx_word' id='word_1_94' title='bbox 285 702 314 719; x_wconf 96'>for</span>
+      <span class='ocrx_word' id='word_1_95' title='bbox 321 703 354 720; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_96' title='bbox 361 703 446 720; x_wconf 96'>consular</span>
+      <span class='ocrx_word' id='word_1_97' title='bbox 453 703 516 720; x_wconf 95'>courts</span>
+      <span class='ocrx_word' id='word_1_98' title='bbox 524 703 546 720; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_99' title='bbox 552 704 583 720; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_100' title='bbox 591 703 661 720; x_wconf 96'>United</span>
+      <span class='ocrx_word' id='word_1_101' title='bbox 669 703 732 720; x_wconf 96'>States</span>
+      <span class='ocrx_word' id='word_1_102' title='bbox 740 702 759 719; x_wconf 96'>in</span>
+      <span class='ocrx_word' id='word_1_103' title='bbox 768 703 835 724; x_wconf 90'>Japan.</span>
      </span>
     </p>
 
-    <p class='ocr_par' id='par_1_10' lang='eng' title="bbox 682 726 835 745">
-     <span class='ocr_line' id='line_1_15' title="bbox 682 726 835 745; baseline 0.007 -1; x_size 23.528408; x_descenders 5.5; x_ascenders 5.5">
-      <span class='ocrx_word' id='word_1_100' title='bbox 682 726 705 744; x_wconf 85'>U.</span>
-      <span class='ocrx_word' id='word_1_101' title='bbox 715 726 734 744; x_wconf 89'>8.</span>
-      <span class='ocrx_word' id='word_1_102' title='bbox 743 727 835 745; x_wconf 80'>GRANT.</span>
+    <p class='ocr_par' id='par_1_10' lang='eng' title="bbox 682 724 835 743">
+     <span class='ocr_line' id='line_1_15' title="bbox 682 724 835 743; baseline 0 -1; x_size 23.573172; x_descenders 5.5365853; x_ascenders 5.5">
+      <span class='ocrx_word' id='word_1_104' title='bbox 682 725 705 743; x_wconf 85'>U.</span>
+      <span class='ocrx_word' id='word_1_105' title='bbox 715 724 734 743; x_wconf 89'>8.</span>
+      <span class='ocrx_word' id='word_1_106' title='bbox 743 725 835 743; x_wconf 83'>GRANT.</span>
      </span>
     </p>
 
-    <p class='ocr_par' id='par_1_11' lang='eng' title="bbox 144 744 482 769">
-     <span class='ocr_line' id='line_1_16' title="bbox 144 744 482 769; baseline 0.006 -7; x_size 23; x_descenders 5; x_ascenders 4">
-      <span class='ocrx_word' id='word_1_103' title='bbox 144 744 293 768; x_wconf 80'>‘WASHINGTON,</span>
-      <span class='ocrx_word' id='word_1_104' title='bbox 302 745 385 767; x_wconf 96'>January</span>
-      <span class='ocrx_word' id='word_1_105' title='bbox 393 747 422 769; x_wconf 93'>27,</span>
-      <span class='ocrx_word' id='word_1_106' title='bbox 431 747 482 764; x_wconf 93'>1871.</span>
+    <p class='ocr_par' id='par_1_11' lang='eng' title="bbox 144 746 482 770">
+     <span class='ocr_line' id='line_1_16' title="bbox 144 746 482 770; baseline 0 -6; x_size 24; x_descenders 6; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_107' title='bbox 144 746 293 769; x_wconf 78'>‘WASHINGTON,</span>
+      <span class='ocrx_word' id='word_1_108' title='bbox 302 746 385 768; x_wconf 96'>January</span>
+      <span class='ocrx_word' id='word_1_109' title='bbox 393 748 422 770; x_wconf 92'>27,</span>
+      <span class='ocrx_word' id='word_1_110' title='bbox 431 747 482 764; x_wconf 92'>1871.</span>
      </span>
     </p>
    </div>
-   <div class='ocr_carea' id='block_1_9' title="bbox 118 846 863 1226">
-    <p class='ocr_par' id='par_1_12' lang='eng' title="bbox 442 846 833 893">
-     <span class='ocr_line' id='line_1_17' title="bbox 442 846 709 870; baseline 0.007 -8; x_size 23; x_descenders 6; x_ascenders 4">
-      <span class='ocrx_word' id='word_1_107' title='bbox 442 846 585 863; x_wconf 96'>DEPARTMENT</span>
-      <span class='ocrx_word' id='word_1_108' title='bbox 598 850 623 864; x_wconf 96'>OF</span>
-      <span class='ocrx_word' id='word_1_109' title='bbox 635 847 709 870; x_wconf 95'>STATE,</span>
+   <div class='ocr_carea' id='block_1_9' title="bbox 120 846 865 1225">
+    <p class='ocr_par' id='par_1_12' lang='eng' title="bbox 443 846 834 891">
+     <span class='ocr_line' id='line_1_17' title="bbox 443 846 710 869; baseline 0.004 -7; x_size 23; x_descenders 6; x_ascenders 4">
+      <span class='ocrx_word' id='word_1_111' title='bbox 443 846 586 863; x_wconf 95'>DEPARTMENT</span>
+      <span class='ocrx_word' id='word_1_112' title='bbox 599 849 624 863; x_wconf 96'>OF</span>
+      <span class='ocrx_word' id='word_1_113' title='bbox 636 846 710 869; x_wconf 95'>STATE,</span>
      </span>
-     <span class='ocr_line' id='line_1_18' title="bbox 445 868 833 893; baseline 0.008 -8; x_size 24; x_descenders 6; x_ascenders 6">
-      <span class='ocrx_word' id='word_1_110' title='bbox 445 890 447 892; x_wconf 1'>.</span>
-      <span class='ocrx_word' id='word_1_111' title='bbox 528 868 648 892; x_wconf 96'>Washington,</span>
-      <span class='ocrx_word' id='word_1_112' title='bbox 656 869 739 891; x_wconf 96'>January</span>
-      <span class='ocrx_word' id='word_1_113' title='bbox 747 870 774 893; x_wconf 95'>26,</span>
-      <span class='ocrx_word' id='word_1_114' title='bbox 783 870 833 888; x_wconf 85'>1870,</span>
+     <span class='ocr_line' id='line_1_18' title="bbox 529 868 834 891; baseline 0 -5; x_size 23.255997; x_descenders 5.2559972; x_ascenders 6">
+      <span class='ocrx_word' id='word_1_114' title='bbox 529 868 649 891; x_wconf 95'>Washington,</span>
+      <span class='ocrx_word' id='word_1_115' title='bbox 657 868 740 889; x_wconf 96'>January</span>
+      <span class='ocrx_word' id='word_1_116' title='bbox 748 868 775 891; x_wconf 94'>26,</span>
+      <span class='ocrx_word' id='word_1_117' title='bbox 784 868 834 886; x_wconf 85'>1870,</span>
      </span>
     </p>
 
-    <p class='ocr_par' id='par_1_13' lang='eng' title="bbox 119 893 863 1050">
-     <span class='ocr_line' id='line_1_19' title="bbox 145 893 856 915; baseline 0.006 -5; x_size 21; x_descenders 4; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_115' title='bbox 145 893 183 910; x_wconf 96'>The</span>
-      <span class='ocrx_word' id='word_1_116' title='bbox 192 893 289 915; x_wconf 96'>Secretary</span>
-      <span class='ocrx_word' id='word_1_117' title='bbox 296 894 318 911; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_118' title='bbox 327 894 380 912; x_wconf 96'>State</span>
-      <span class='ocrx_word' id='word_1_119' title='bbox 388 895 421 912; x_wconf 96'>has</span>
-      <span class='ocrx_word' id='word_1_120' title='bbox 429 894 461 912; x_wconf 97'>the</span>
-      <span class='ocrx_word' id='word_1_121' title='bbox 468 895 527 912; x_wconf 96'>honor</span>
-      <span class='ocrx_word' id='word_1_122' title='bbox 534 897 554 912; x_wconf 96'>to</span>
-      <span class='ocrx_word' id='word_1_123' title='bbox 563 896 632 913; x_wconf 96'>submit</span>
-      <span class='ocrx_word' id='word_1_124' title='bbox 639 896 729 914; x_wconf 91'>herewith,</span>
-      <span class='ocrx_word' id='word_1_125' title='bbox 738 897 767 914; x_wconf 91'>for</span>
-      <span class='ocrx_word' id='word_1_126' title='bbox 775 897 856 914; x_wconf 96'>revision</span>
+    <p class='ocr_par' id='par_1_13' lang='eng' title="bbox 121 894 865 1051">
+     <span class='ocr_line' id='line_1_19' title="bbox 146 894 857 917; baseline 0 -5; x_size 22; x_descenders 5; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_118' title='bbox 146 895 184 912; x_wconf 96'>The</span>
+      <span class='ocrx_word' id='word_1_119' title='bbox 193 895 290 917; x_wconf 96'>Secretary</span>
+      <span class='ocrx_word' id='word_1_120' title='bbox 297 895 319 912; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_121' title='bbox 328 895 381 913; x_wconf 96'>State</span>
+      <span class='ocrx_word' id='word_1_122' title='bbox 389 896 422 913; x_wconf 96'>has</span>
+      <span class='ocrx_word' id='word_1_123' title='bbox 430 894 462 912; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_124' title='bbox 469 895 528 912; x_wconf 96'>honor</span>
+      <span class='ocrx_word' id='word_1_125' title='bbox 535 897 555 912; x_wconf 96'>to</span>
+      <span class='ocrx_word' id='word_1_126' title='bbox 564 895 633 912; x_wconf 96'>submit</span>
+      <span class='ocrx_word' id='word_1_127' title='bbox 640 895 730 913; x_wconf 90'>herewith,</span>
+      <span class='ocrx_word' id='word_1_128' title='bbox 739 895 768 912; x_wconf 90'>for</span>
+      <span class='ocrx_word' id='word_1_129' title='bbox 776 895 857 912; x_wconf 95'>revision</span>
      </span>
-     <span class='ocr_line' id='line_1_20' title="bbox 121 909 856 939; baseline 0.007 -7; x_size 22; x_descenders 5; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_127' title='bbox 121 915 147 937; x_wconf 96'>by</span>
-      <span class='ocrx_word' id='word_1_128' title='bbox 159 915 257 938; x_wconf 96'>Congress,</span>
-      <span class='ocrx_word' id='word_1_129' title='bbox 265 916 284 933; x_wconf 96'>in</span>
-      <span class='ocrx_word' id='word_1_130' title='bbox 293 917 402 938; x_wconf 96'>conformity</span>
-      <span class='ocrx_word' id='word_1_131' title='bbox 411 917 455 934; x_wconf 96'>with</span>
-      <span class='ocrx_word' id='word_1_132' title='bbox 463 917 495 934; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_133' title='bbox 504 918 606 939; x_wconf 96'>provisions</span>
-      <span class='ocrx_word' id='word_1_134' title='bbox 614 918 635 935; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_135' title='bbox 645 919 714 936; x_wconf 96'>section</span>
-      <span class='ocrx_word' id='word_1_136' title='bbox 723 909 735 936; x_wconf 91'>6</span>
-      <span class='ocrx_word' id='word_1_137' title='bbox 748 919 769 936; x_wconf 95'>of</span>
-      <span class='ocrx_word' id='word_1_138' title='bbox 778 920 811 937; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_139' title='bbox 825 921 856 937; x_wconf 96'>act</span>
+     <span class='ocr_line' id='line_1_20' title="bbox 122 907 857 940; baseline 0 -5; x_size 23; x_descenders 5; x_ascenders 6">
+      <span class='ocrx_word' id='word_1_130' title='bbox 122 918 148 939; x_wconf 95'>by</span>
+      <span class='ocrx_word' id='word_1_131' title='bbox 160 917 258 940; x_wconf 96'>Congress,</span>
+      <span class='ocrx_word' id='word_1_132' title='bbox 266 918 285 935; x_wconf 96'>in</span>
+      <span class='ocrx_word' id='word_1_133' title='bbox 294 918 403 939; x_wconf 96'>conformity</span>
+      <span class='ocrx_word' id='word_1_134' title='bbox 412 917 456 935; x_wconf 96'>with</span>
+      <span class='ocrx_word' id='word_1_135' title='bbox 464 917 496 934; x_wconf 97'>the</span>
+      <span class='ocrx_word' id='word_1_136' title='bbox 505 918 607 939; x_wconf 96'>provisions</span>
+      <span class='ocrx_word' id='word_1_137' title='bbox 615 917 636 934; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_138' title='bbox 646 918 715 935; x_wconf 97'>section</span>
+      <span class='ocrx_word' id='word_1_139' title='bbox 724 907 736 934; x_wconf 92'>6</span>
+      <span class='ocrx_word' id='word_1_140' title='bbox 749 917 770 934; x_wconf 94'>of</span>
+      <span class='ocrx_word' id='word_1_141' title='bbox 779 918 812 935; x_wconf 94'>the</span>
+      <span class='ocrx_word' id='word_1_142' title='bbox 826 919 857 935; x_wconf 96'>act</span>
      </span>
-     <span class='ocr_line' id='line_1_21' title="bbox 120 938 856 962; baseline 0.007 -8; x_size 21; x_descenders 4; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_140' title='bbox 120 938 216 959; x_wconf 95'>approved</span>
-      <span class='ocrx_word' id='word_1_141' title='bbox 228 938 264 955; x_wconf 95'>22d</span>
-      <span class='ocrx_word' id='word_1_142' title='bbox 277 938 298 956; x_wconf 95'>of</span>
-      <span class='ocrx_word' id='word_1_143' title='bbox 312 939 367 961; x_wconf 96'>June,</span>
-      <span class='ocrx_word' id='word_1_144' title='bbox 382 939 431 962; x_wconf 72'>1860,</span>
-      <span class='ocrx_word' id='word_1_145' title='bbox 440 944 452 956; x_wconf 72'>a</span>
-      <span class='ocrx_word' id='word_1_146' title='bbox 465 944 512 961; x_wconf 96'>copy</span>
-      <span class='ocrx_word' id='word_1_147' title='bbox 524 940 545 957; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_148' title='bbox 554 940 685 962; x_wconf 80'>“regulations</span>
-      <span class='ocrx_word' id='word_1_149' title='bbox 692 941 721 958; x_wconf 96'>for</span>
-      <span class='ocrx_word' id='word_1_150' title='bbox 729 942 761 958; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_151' title='bbox 769 942 856 959; x_wconf 96'>consular</span>
+     <span class='ocr_line' id='line_1_21' title="bbox 121 939 857 962; baseline -0.001 -5; x_size 21; x_descenders 4; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_143' title='bbox 121 940 217 961; x_wconf 96'>approved</span>
+      <span class='ocrx_word' id='word_1_144' title='bbox 229 940 265 957; x_wconf 95'>22d</span>
+      <span class='ocrx_word' id='word_1_145' title='bbox 278 939 299 957; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_146' title='bbox 313 940 368 962; x_wconf 96'>June,</span>
+      <span class='ocrx_word' id='word_1_147' title='bbox 383 940 432 962; x_wconf 57'>1860,</span>
+      <span class='ocrx_word' id='word_1_148' title='bbox 441 944 453 956; x_wconf 57'>a</span>
+      <span class='ocrx_word' id='word_1_149' title='bbox 466 944 513 961; x_wconf 96'>copy</span>
+      <span class='ocrx_word' id='word_1_150' title='bbox 525 940 546 957; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_151' title='bbox 555 940 686 961; x_wconf 82'>“regulations</span>
+      <span class='ocrx_word' id='word_1_152' title='bbox 693 940 722 957; x_wconf 95'>for</span>
+      <span class='ocrx_word' id='word_1_153' title='bbox 730 940 762 956; x_wconf 97'>the</span>
+      <span class='ocrx_word' id='word_1_154' title='bbox 770 940 857 957; x_wconf 95'>consular</span>
      </span>
-     <span class='ocr_line' id='line_1_22' title="bbox 120 960 856 985; baseline 0.005 -8; x_size 21; x_descenders 4; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_152' title='bbox 120 961 183 977; x_wconf 96'>courts</span>
-      <span class='ocrx_word' id='word_1_153' title='bbox 197 960 218 977; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_154' title='bbox 227 961 260 977; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_155' title='bbox 271 961 341 978; x_wconf 96'>United</span>
-      <span class='ocrx_word' id='word_1_156' title='bbox 355 961 418 979; x_wconf 96'>States</span>
-      <span class='ocrx_word' id='word_1_157' title='bbox 429 962 449 979; x_wconf 94'>in</span>
-      <span class='ocrx_word' id='word_1_158' title='bbox 464 962 543 983; x_wconf 96'>Japan,”</span>
-      <span class='ocrx_word' id='word_1_159' title='bbox 555 963 632 980; x_wconf 96'>decreed</span>
-      <span class='ocrx_word' id='word_1_160' title='bbox 644 964 682 980; x_wconf 96'>and</span>
-      <span class='ocrx_word' id='word_1_161' title='bbox 694 963 758 981; x_wconf 96'>issued</span>
-      <span class='ocrx_word' id='word_1_162' title='bbox 771 964 796 985; x_wconf 93'>by</span>
-      <span class='ocrx_word' id='word_1_163' title='bbox 805 964 825 981; x_wconf 88'>C.</span>
-      <span class='ocrx_word' id='word_1_164' title='bbox 834 964 856 981; x_wconf 70'>E.</span>
+     <span class='ocr_line' id='line_1_22' title="bbox 121 962 857 983; baseline -0.001 -3; x_size 21; x_descenders 4; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_155' title='bbox 121 963 184 980; x_wconf 96'>courts</span>
+      <span class='ocrx_word' id='word_1_156' title='bbox 198 962 219 979; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_157' title='bbox 228 963 261 979; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_158' title='bbox 272 962 342 980; x_wconf 96'>United</span>
+      <span class='ocrx_word' id='word_1_159' title='bbox 356 962 419 980; x_wconf 96'>States</span>
+      <span class='ocrx_word' id='word_1_160' title='bbox 430 962 450 979; x_wconf 95'>in</span>
+      <span class='ocrx_word' id='word_1_161' title='bbox 465 962 544 983; x_wconf 95'>Japan,”</span>
+      <span class='ocrx_word' id='word_1_162' title='bbox 556 962 633 979; x_wconf 96'>decreed</span>
+      <span class='ocrx_word' id='word_1_163' title='bbox 645 963 683 979; x_wconf 96'>and</span>
+      <span class='ocrx_word' id='word_1_164' title='bbox 695 962 759 980; x_wconf 93'>issued</span>
+      <span class='ocrx_word' id='word_1_165' title='bbox 772 962 797 983; x_wconf 92'>by</span>
+      <span class='ocrx_word' id='word_1_166' title='bbox 806 962 826 979; x_wconf 81'>C.</span>
+      <span class='ocrx_word' id='word_1_167' title='bbox 835 962 857 979; x_wconf 47'>BE.</span>
      </span>
-     <span class='ocr_line' id='line_1_23' title="bbox 120 982 856 1008; baseline 0.007 -9; x_size 22; x_descenders 5; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_165' title='bbox 120 982 150 999; x_wconf 96'>De</span>
-      <span class='ocrx_word' id='word_1_166' title='bbox 156 982 215 1005; x_wconf 96'>Long,</span>
-      <span class='ocrx_word' id='word_1_167' title='bbox 224 983 256 1000; x_wconf 93'>the</span>
-      <span class='ocrx_word' id='word_1_168' title='bbox 264 984 348 1001; x_wconf 91'>miniater</span>
-      <span class='ocrx_word' id='word_1_169' title='bbox 359 984 381 1001; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_170' title='bbox 389 984 421 1001; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_171' title='bbox 429 984 499 1001; x_wconf 96'>United</span>
-      <span class='ocrx_word' id='word_1_172' title='bbox 507 985 570 1002; x_wconf 96'>States</span>
-      <span class='ocrx_word' id='word_1_173' title='bbox 577 985 595 1002; x_wconf 96'>in</span>
-      <span class='ocrx_word' id='word_1_174' title='bbox 604 985 646 1003; x_wconf 96'>that</span>
-      <span class='ocrx_word' id='word_1_175' title='bbox 654 987 738 1008; x_wconf 96'>country,</span>
-      <span class='ocrx_word' id='word_1_176' title='bbox 747 986 766 1003; x_wconf 93'>in</span>
-      <span class='ocrx_word' id='word_1_177' title='bbox 775 986 856 1007; x_wconf 93'>Septem-</span>
+     <span class='ocr_line' id='line_1_23' title="bbox 122 984 858 1007; baseline 0 -6; x_size 22; x_descenders 5; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_168' title='bbox 122 985 152 1001; x_wconf 96'>De</span>
+      <span class='ocrx_word' id='word_1_169' title='bbox 158 984 217 1007; x_wconf 96'>Long,</span>
+      <span class='ocrx_word' id='word_1_170' title='bbox 226 985 258 1002; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_171' title='bbox 266 985 350 1002; x_wconf 96'>minister</span>
+      <span class='ocrx_word' id='word_1_172' title='bbox 361 985 383 1002; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_173' title='bbox 391 985 423 1002; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_174' title='bbox 431 984 501 1001; x_wconf 97'>United</span>
+      <span class='ocrx_word' id='word_1_175' title='bbox 509 985 572 1002; x_wconf 96'>States</span>
+      <span class='ocrx_word' id='word_1_176' title='bbox 579 984 597 1001; x_wconf 96'>in</span>
+      <span class='ocrx_word' id='word_1_177' title='bbox 606 984 648 1002; x_wconf 96'>that</span>
+      <span class='ocrx_word' id='word_1_178' title='bbox 656 986 740 1006; x_wconf 96'>country,</span>
+      <span class='ocrx_word' id='word_1_179' title='bbox 749 984 768 1001; x_wconf 92'>in</span>
+      <span class='ocrx_word' id='word_1_180' title='bbox 777 984 858 1005; x_wconf 91'>Septem-</span>
      </span>
-     <span class='ocr_line' id='line_1_24' title="bbox 119 1004 863 1040; baseline 0.007 -19; x_size 22; x_descenders 5; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_178' title='bbox 119 1004 158 1026; x_wconf 93'>ber,</span>
-      <span class='ocrx_word' id='word_1_179' title='bbox 167 1005 221 1027; x_wconf 88'>1870;</span>
-      <span class='ocrx_word' id='word_1_180' title='bbox 234 1006 271 1022; x_wconf 87'>and</span>
-      <span class='ocrx_word' id='word_1_181' title='bbox 285 1006 324 1023; x_wconf 87'>also</span>
-      <span class='ocrx_word' id='word_1_182' title='bbox 337 1006 369 1023; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_183' title='bbox 381 1011 448 1028; x_wconf 96'>papers</span>
-      <span class='ocrx_word' id='word_1_184' title='bbox 459 1007 566 1024; x_wconf 96'>mentioned</span>
-      <span class='ocrx_word' id='word_1_185' title='bbox 574 1007 593 1024; x_wconf 94'>in</span>
-      <span class='ocrx_word' id='word_1_186' title='bbox 602 1008 633 1025; x_wconf 93'>the</span>
-      <span class='ocrx_word' id='word_1_187' title='bbox 641 1008 739 1030; x_wconf 92'>subjoined</span>
-      <span class='ocrx_word' id='word_1_188' title='bbox 746 1008 784 1030; x_wconf 97'>list,</span>
-      <span class='ocrx_word' id='word_1_189' title='bbox 782 1005 863 1040; x_wconf 86'>which,</span>
+     <span class='ocr_line' id='line_1_24' title="bbox 121 1006 865 1034; baseline 0 -11; x_size 22; x_descenders 5; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_181' title='bbox 121 1007 160 1028; x_wconf 91'>ber,</span>
+      <span class='ocrx_word' id='word_1_182' title='bbox 169 1007 223 1029; x_wconf 95'>1870;</span>
+      <span class='ocrx_word' id='word_1_183' title='bbox 236 1008 273 1024; x_wconf 84'>and</span>
+      <span class='ocrx_word' id='word_1_184' title='bbox 287 1007 326 1024; x_wconf 95'>also</span>
+      <span class='ocrx_word' id='word_1_185' title='bbox 339 1007 371 1024; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_186' title='bbox 383 1011 450 1029; x_wconf 96'>papers</span>
+      <span class='ocrx_word' id='word_1_187' title='bbox 461 1007 568 1024; x_wconf 96'>mentioned</span>
+      <span class='ocrx_word' id='word_1_188' title='bbox 576 1006 595 1023; x_wconf 96'>in</span>
+      <span class='ocrx_word' id='word_1_189' title='bbox 604 1007 635 1024; x_wconf 93'>the</span>
+      <span class='ocrx_word' id='word_1_190' title='bbox 643 1007 741 1029; x_wconf 92'>subjoined</span>
+      <span class='ocrx_word' id='word_1_191' title='bbox 748 1006 786 1028; x_wconf 96'>list,</span>
+      <span class='ocrx_word' id='word_1_192' title='bbox 794 1007 865 1034; x_wconf 86'>which,</span>
      </span>
-     <span class='ocr_line' id='line_1_25' title="bbox 120 1027 555 1050; baseline 0.007 -7; x_size 22; x_descenders 5; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_190' title='bbox 120 1027 195 1044; x_wconf 96'>contain</span>
-      <span class='ocrx_word' id='word_1_191' title='bbox 203 1028 322 1049; x_wconf 96'>suggestions</span>
-      <span class='ocrx_word' id='word_1_192' title='bbox 330 1033 353 1045; x_wconf 95'>on</span>
-      <span class='ocrx_word' id='word_1_193' title='bbox 361 1028 392 1045; x_wconf 95'>the</span>
-      <span class='ocrx_word' id='word_1_194' title='bbox 400 1029 471 1050; x_wconf 96'>subject</span>
-      <span class='ocrx_word' id='word_1_195' title='bbox 478 1029 555 1046; x_wconf 86'>thereof.</span>
+     <span class='ocr_line' id='line_1_25' title="bbox 122 1029 557 1051; baseline 0 -5; x_size 22; x_descenders 5; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_193' title='bbox 122 1029 197 1046; x_wconf 96'>contain</span>
+      <span class='ocrx_word' id='word_1_194' title='bbox 205 1029 324 1051; x_wconf 96'>suggestions</span>
+      <span class='ocrx_word' id='word_1_195' title='bbox 332 1034 355 1046; x_wconf 96'>on</span>
+      <span class='ocrx_word' id='word_1_196' title='bbox 363 1029 394 1046; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_197' title='bbox 402 1029 473 1050; x_wconf 93'>subject</span>
+      <span class='ocrx_word' id='word_1_198' title='bbox 480 1029 557 1046; x_wconf 85'>thereof.</span>
      </span>
     </p>
 
-    <p class='ocr_par' id='par_1_14' lang='eng' title="bbox 118 1050 854 1185">
-     <span class='ocr_line' id='line_1_26' title="bbox 142 1050 853 1076; baseline 0.007 -11; x_size 21; x_descenders 4; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_196' title='bbox 142 1050 162 1065; x_wconf 96'>A</span>
-      <span class='ocrx_word' id='word_1_197' title='bbox 169 1054 216 1071; x_wconf 96'>copy</span>
-      <span class='ocrx_word' id='word_1_198' title='bbox 224 1050 245 1066; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_199' title='bbox 251 1050 323 1067; x_wconf 81'>Article</span>
-      <span class='ocrx_word' id='word_1_200' title='bbox 330 1050 397 1067; x_wconf 77'>XXVI</span>
-      <span class='ocrx_word' id='word_1_201' title='bbox 405 1051 425 1068; x_wconf 95'>of</span>
-      <span class='ocrx_word' id='word_1_202' title='bbox 431 1051 462 1068; x_wconf 95'>the</span>
-      <span class='ocrx_word' id='word_1_203' title='bbox 470 1052 556 1068; x_wconf 94'>consular</span>
-      <span class='ocrx_word' id='word_1_204' title='bbox 560 1052 674 1073; x_wconf 96'>regulations</span>
-      <span class='ocrx_word' id='word_1_205' title='bbox 680 1053 696 1069; x_wconf 94'>is</span>
-      <span class='ocrx_word' id='word_1_206' title='bbox 701 1053 740 1070; x_wconf 96'>also</span>
-      <span class='ocrx_word' id='word_1_207' title='bbox 747 1053 853 1076; x_wconf 96'>submitted,</span>
+    <p class='ocr_par' id='par_1_14' lang='eng' title="bbox 120 1051 857 1185">
+     <span class='ocr_line' id='line_1_26' title="bbox 144 1051 855 1074; baseline -0.001 -6; x_size 21; x_descenders 4; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_199' title='bbox 144 1052 164 1067; x_wconf 95'>A</span>
+      <span class='ocrx_word' id='word_1_200' title='bbox 171 1056 218 1073; x_wconf 95'>copy</span>
+      <span class='ocrx_word' id='word_1_201' title='bbox 226 1052 247 1068; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_202' title='bbox 253 1051 325 1068; x_wconf 93'>Article</span>
+      <span class='ocrx_word' id='word_1_203' title='bbox 332 1051 399 1068; x_wconf 86'>XXVI</span>
+      <span class='ocrx_word' id='word_1_204' title='bbox 407 1052 427 1069; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_205' title='bbox 433 1051 464 1068; x_wconf 95'>the</span>
+      <span class='ocrx_word' id='word_1_206' title='bbox 472 1052 558 1068; x_wconf 95'>consular</span>
+      <span class='ocrx_word' id='word_1_207' title='bbox 562 1051 676 1072; x_wconf 95'>regulations</span>
+      <span class='ocrx_word' id='word_1_208' title='bbox 682 1052 698 1068; x_wconf 96'>is</span>
+      <span class='ocrx_word' id='word_1_209' title='bbox 703 1052 742 1068; x_wconf 94'>also</span>
+      <span class='ocrx_word' id='word_1_210' title='bbox 749 1051 855 1074; x_wconf 95'>submitted,</span>
      </span>
-     <span class='ocr_line' id='line_1_27' title="bbox 119 1071 854 1097; baseline 0.007 -9; x_size 21; x_descenders 4; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_208' title='bbox 119 1071 156 1088; x_wconf 96'>and</span>
-      <span class='ocrx_word' id='word_1_209' title='bbox 169 1071 202 1088; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_210' title='bbox 216 1072 313 1094; x_wconf 96'>Secretary</span>
-      <span class='ocrx_word' id='word_1_211' title='bbox 319 1073 340 1089; x_wconf 90'>of</span>
-      <span class='ocrx_word' id='word_1_212' title='bbox 349 1073 402 1090; x_wconf 96'>State</span>
-      <span class='ocrx_word' id='word_1_213' title='bbox 413 1073 532 1095; x_wconf 96'>respectfully</span>
-      <span class='ocrx_word' id='word_1_214' title='bbox 540 1075 632 1097; x_wconf 96'>suggests,</span>
-      <span class='ocrx_word' id='word_1_215' title='bbox 641 1075 670 1091; x_wconf 96'>for</span>
-      <span class='ocrx_word' id='word_1_216' title='bbox 677 1075 710 1092; x_wconf 97'>the</span>
-      <span class='ocrx_word' id='word_1_217' title='bbox 718 1074 854 1093; x_wconf 96'>consideration</span>
+     <span class='ocr_line' id='line_1_27' title="bbox 121 1072 856 1096; baseline 0 -5; x_size 22; x_descenders 5; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_211' title='bbox 121 1073 158 1091; x_wconf 95'>and</span>
+      <span class='ocrx_word' id='word_1_212' title='bbox 171 1073 204 1090; x_wconf 95'>the</span>
+      <span class='ocrx_word' id='word_1_213' title='bbox 218 1074 315 1095; x_wconf 96'>Secretary</span>
+      <span class='ocrx_word' id='word_1_214' title='bbox 321 1074 342 1090; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_215' title='bbox 351 1074 404 1091; x_wconf 96'>State</span>
+      <span class='ocrx_word' id='word_1_216' title='bbox 415 1073 534 1095; x_wconf 96'>respectfully</span>
+      <span class='ocrx_word' id='word_1_217' title='bbox 542 1074 634 1096; x_wconf 96'>suggests,</span>
+      <span class='ocrx_word' id='word_1_218' title='bbox 643 1074 672 1090; x_wconf 96'>for</span>
+      <span class='ocrx_word' id='word_1_219' title='bbox 679 1074 712 1091; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_220' title='bbox 720 1072 856 1091; x_wconf 96'>consideration</span>
      </span>
-     <span class='ocr_line' id='line_1_28' title="bbox 118 1093 852 1118; baseline 0.005 -8; x_size 22; x_descenders 5; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_218' title='bbox 118 1093 140 1110; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_219' title='bbox 149 1093 246 1116; x_wconf 94'>Congress,</span>
-      <span class='ocrx_word' id='word_1_220' title='bbox 255 1095 287 1111; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_221' title='bbox 300 1095 395 1117; x_wconf 96'>propriety</span>
-      <span class='ocrx_word' id='word_1_222' title='bbox 406 1096 427 1112; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_223' title='bbox 435 1096 516 1117; x_wconf 96'>limiting</span>
-      <span class='ocrx_word' id='word_1_224' title='bbox 528 1095 560 1113; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_225' title='bbox 571 1101 633 1118; x_wconf 96'>power</span>
-      <span class='ocrx_word' id='word_1_226' title='bbox 641 1097 662 1114; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_227' title='bbox 671 1097 764 1114; x_wconf 96'>ministers</span>
-      <span class='ocrx_word' id='word_1_228' title='bbox 772 1089 799 1122; x_wconf 95'>to</span>
-      <span class='ocrx_word' id='word_1_229' title='bbox 798 1097 851 1115; x_wconf 95'>make</span>
+     <span class='ocr_line' id='line_1_28' title="bbox 120 1095 854 1118; baseline -0.001 -5; x_size 23; x_descenders 5; x_ascenders 6">
+      <span class='ocrx_word' id='word_1_221' title='bbox 120 1095 142 1113; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_222' title='bbox 151 1095 248 1118; x_wconf 96'>Congress,</span>
+      <span class='ocrx_word' id='word_1_223' title='bbox 257 1097 289 1113; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_224' title='bbox 302 1096 397 1118; x_wconf 96'>propriety</span>
+      <span class='ocrx_word' id='word_1_225' title='bbox 408 1096 429 1113; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_226' title='bbox 437 1096 518 1117; x_wconf 96'>limiting</span>
+      <span class='ocrx_word' id='word_1_227' title='bbox 530 1095 562 1113; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_228' title='bbox 573 1100 635 1117; x_wconf 96'>power</span>
+      <span class='ocrx_word' id='word_1_229' title='bbox 643 1096 664 1113; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_230' title='bbox 673 1096 766 1113; x_wconf 96'>ministers</span>
+      <span class='ocrx_word' id='word_1_231' title='bbox 774 1091 801 1122; x_wconf 96'>to</span>
+      <span class='ocrx_word' id='word_1_232' title='bbox 800 1095 854 1113; x_wconf 96'>make</span>
      </span>
-     <span class='ocr_line' id='line_1_29' title="bbox 119 1115 853 1141; baseline 0.007 -9; x_size 23; x_descenders 5; x_ascenders 6">
-      <span class='ocrx_word' id='word_1_230' title='bbox 119 1115 194 1133; x_wconf 88'>decrees</span>
-      <span class='ocrx_word' id='word_1_231' title='bbox 206 1116 243 1133; x_wconf 88'>and</span>
-      <span class='ocrx_word' id='word_1_232' title='bbox 248 1116 358 1139; x_wconf 96'>regulation,</span>
-      <span class='ocrx_word' id='word_1_233' title='bbox 367 1117 385 1134; x_wconf 96'>in</span>
-      <span class='ocrx_word' id='word_1_234' title='bbox 393 1118 425 1135; x_wconf 95'>the</span>
-      <span class='ocrx_word' id='word_1_235' title='bbox 433 1122 486 1135; x_wconf 96'>sense</span>
-      <span class='ocrx_word' id='word_1_236' title='bbox 493 1118 513 1135; x_wconf 96'>in</span>
-      <span class='ocrx_word' id='word_1_237' title='bbox 521 1118 581 1136; x_wconf 96'>which</span>
-      <span class='ocrx_word' id='word_1_238' title='bbox 589 1119 605 1136; x_wconf 96'>it</span>
-      <span class='ocrx_word' id='word_1_239' title='bbox 612 1119 629 1136; x_wconf 96'>is</span>
-      <span class='ocrx_word' id='word_1_240' title='bbox 636 1118 708 1136; x_wconf 95'>limited</span>
-      <span class='ocrx_word' id='word_1_241' title='bbox 716 1120 741 1141; x_wconf 96'>by</span>
-      <span class='ocrx_word' id='word_1_242' title='bbox 747 1120 853 1141; x_wconf 96'>paragraph</span>
+     <span class='ocr_line' id='line_1_29' title="bbox 121 1117 856 1140; baseline 0 -6; x_size 22; x_descenders 4; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_233' title='bbox 121 1118 197 1135; x_wconf 95'>decrees</span>
+      <span class='ocrx_word' id='word_1_234' title='bbox 208 1118 246 1135; x_wconf 94'>and</span>
+      <span class='ocrx_word' id='word_1_235' title='bbox 250 1117 361 1140; x_wconf 94'>regulation,</span>
+      <span class='ocrx_word' id='word_1_236' title='bbox 369 1118 388 1135; x_wconf 96'>in</span>
+      <span class='ocrx_word' id='word_1_237' title='bbox 395 1119 428 1136; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_238' title='bbox 435 1122 489 1135; x_wconf 96'>sense</span>
+      <span class='ocrx_word' id='word_1_239' title='bbox 495 1118 516 1135; x_wconf 96'>in</span>
+      <span class='ocrx_word' id='word_1_240' title='bbox 523 1118 584 1136; x_wconf 96'>which</span>
+      <span class='ocrx_word' id='word_1_241' title='bbox 591 1118 608 1135; x_wconf 96'>it</span>
+      <span class='ocrx_word' id='word_1_242' title='bbox 614 1118 632 1135; x_wconf 96'>is</span>
+      <span class='ocrx_word' id='word_1_243' title='bbox 638 1117 711 1135; x_wconf 95'>limited</span>
+      <span class='ocrx_word' id='word_1_244' title='bbox 718 1118 743 1139; x_wconf 96'>by</span>
+      <span class='ocrx_word' id='word_1_245' title='bbox 750 1118 856 1139; x_wconf 96'>paragraph</span>
      </span>
-     <span class='ocr_line' id='line_1_30' title="bbox 119 1138 854 1164; baseline 0.007 -10; x_size 22; x_descenders 5; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_243' title='bbox 119 1138 153 1154; x_wconf 94'>431</span>
-      <span class='ocrx_word' id='word_1_244' title='bbox 161 1138 183 1155; x_wconf 95'>of</span>
-      <span class='ocrx_word' id='word_1_245' title='bbox 192 1139 224 1155; x_wconf 95'>the</span>
-      <span class='ocrx_word' id='word_1_246' title='bbox 232 1139 297 1156; x_wconf 96'>article</span>
-      <span class='ocrx_word' id='word_1_247' title='bbox 304 1139 366 1157; x_wconf 93'>before</span>
-      <span class='ocrx_word' id='word_1_248' title='bbox 374 1140 505 1157; x_wconf 91'>named—that</span>
-      <span class='ocrx_word' id='word_1_249' title='bbox 513 1141 534 1161; x_wconf 96'>is,</span>
-      <span class='ocrx_word' id='word_1_250' title='bbox 544 1141 579 1158; x_wconf 87'>“to</span>
-      <span class='ocrx_word' id='word_1_251' title='bbox 588 1142 628 1158; x_wconf 96'>acts</span>
-      <span class='ocrx_word' id='word_1_252' title='bbox 635 1143 757 1164; x_wconf 95'>necessary</span>
-      <span class='ocrx_word' id='word_1_253' title='bbox 732 1134 761 1168; x_wconf 95'>to</span>
-      <span class='ocrx_word' id='word_1_254' title='bbox 767 1142 854 1164; x_wconf 96'>organize</span>
+     <span class='ocr_line' id='line_1_30' title="bbox 122 1140 857 1162; baseline 0 -5; x_size 22; x_descenders 5; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_246' title='bbox 122 1140 156 1157; x_wconf 96'>431</span>
+      <span class='ocrx_word' id='word_1_247' title='bbox 164 1140 186 1157; x_wconf 97'>of</span>
+      <span class='ocrx_word' id='word_1_248' title='bbox 195 1141 227 1157; x_wconf 95'>the</span>
+      <span class='ocrx_word' id='word_1_249' title='bbox 235 1140 300 1158; x_wconf 95'>article</span>
+      <span class='ocrx_word' id='word_1_250' title='bbox 307 1140 369 1158; x_wconf 93'>before</span>
+      <span class='ocrx_word' id='word_1_251' title='bbox 377 1140 508 1158; x_wconf 92'>named—that</span>
+      <span class='ocrx_word' id='word_1_252' title='bbox 516 1141 537 1161; x_wconf 96'>is,</span>
+      <span class='ocrx_word' id='word_1_253' title='bbox 547 1141 582 1158; x_wconf 85'>“to</span>
+      <span class='ocrx_word' id='word_1_254' title='bbox 591 1141 631 1157; x_wconf 96'>acts</span>
+      <span class='ocrx_word' id='word_1_255' title='bbox 638 1141 760 1162; x_wconf 95'>necessary</span>
+      <span class='ocrx_word' id='word_1_256' title='bbox 738 1136 762 1167; x_wconf 96'>to</span>
+      <span class='ocrx_word' id='word_1_257' title='bbox 770 1140 857 1162; x_wconf 96'>organize</span>
      </span>
-     <span class='ocr_line' id='line_1_31' title="bbox 118 1160 659 1185; baseline 0.007 -8; x_size 23; x_descenders 5; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_255' title='bbox 118 1160 156 1177; x_wconf 96'>and</span>
-      <span class='ocrx_word' id='word_1_256' title='bbox 165 1160 207 1182; x_wconf 96'>give</span>
-      <span class='ocrx_word' id='word_1_257' title='bbox 215 1161 309 1183; x_wconf 95'>efficiency</span>
-      <span class='ocrx_word' id='word_1_258' title='bbox 317 1163 337 1179; x_wconf 95'>to</span>
-      <span class='ocrx_word' id='word_1_259' title='bbox 345 1162 377 1179; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_260' title='bbox 385 1164 446 1180; x_wconf 96'>courts</span>
-      <span class='ocrx_word' id='word_1_261' title='bbox 455 1163 530 1180; x_wconf 96'>created</span>
-      <span class='ocrx_word' id='word_1_262' title='bbox 536 1163 561 1185; x_wconf 96'>by</span>
-      <span class='ocrx_word' id='word_1_263' title='bbox 571 1163 604 1181; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_264' title='bbox 612 1164 659 1181; x_wconf 94'>act.”</span>
+     <span class='ocr_line' id='line_1_31' title="bbox 121 1162 662 1185; baseline 0.002 -6; x_size 22; x_descenders 4; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_258' title='bbox 121 1162 159 1180; x_wconf 96'>and</span>
+      <span class='ocrx_word' id='word_1_259' title='bbox 168 1162 210 1184; x_wconf 97'>give</span>
+      <span class='ocrx_word' id='word_1_260' title='bbox 218 1163 312 1184; x_wconf 96'>efficiency</span>
+      <span class='ocrx_word' id='word_1_261' title='bbox 320 1164 340 1180; x_wconf 96'>to</span>
+      <span class='ocrx_word' id='word_1_262' title='bbox 348 1163 380 1180; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_263' title='bbox 388 1164 449 1181; x_wconf 96'>courts</span>
+      <span class='ocrx_word' id='word_1_264' title='bbox 458 1163 533 1180; x_wconf 96'>created</span>
+      <span class='ocrx_word' id='word_1_265' title='bbox 539 1163 564 1185; x_wconf 96'>by</span>
+      <span class='ocrx_word' id='word_1_266' title='bbox 574 1162 607 1180; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_267' title='bbox 615 1163 662 1180; x_wconf 96'>act.”</span>
      </span>
     </p>
 
-    <p class='ocr_par' id='par_1_15' lang='eng' title="bbox 141 1183 555 1205">
-     <span class='ocr_line' id='line_1_32' title="bbox 141 1183 555 1205; baseline 0.007 -6; x_size 23; x_descenders 5; x_ascenders 6">
-      <span class='ocrx_word' id='word_1_265' title='bbox 141 1183 268 1205; x_wconf 34'>Respectfully</span>
-      <span class='ocrx_word' id='word_1_266' title='bbox 276 1183 383 1201; x_wconf 93'>submitted.</span>
-      <span class='ocrx_word' id='word_1_267' title='bbox 553 1194 555 1196; x_wconf 0'>.</span>
+    <p class='ocr_par' id='par_1_15' lang='eng' title="bbox 144 1184 386 1207">
+     <span class='ocr_line' id='line_1_32' title="bbox 144 1184 386 1207; baseline 0 -5; x_size 23; x_descenders 5; x_ascenders 5">
+      <span class='ocrx_word' id='word_1_268' title='bbox 144 1185 271 1207; x_wconf 86'>Respectfully</span>
+      <span class='ocrx_word' id='word_1_269' title='bbox 279 1184 386 1202; x_wconf 93'>submitted.</span>
      </span>
     </p>
 
-    <p class='ocr_par' id='par_1_16' lang='eng' title="bbox 622 1208 831 1226">
-     <span class='ocr_line' id='line_1_33' title="bbox 622 1208 831 1226; baseline 0.005 -1; x_size 23.623137; x_descenders 5.52807; x_ascenders 5.5594797">
-      <span class='ocrx_word' id='word_1_268' title='bbox 622 1208 757 1226; x_wconf 96'>HAMILTON</span>
-      <span class='ocrx_word' id='word_1_269' title='bbox 765 1209 831 1226; x_wconf 96'>FISH.</span>
+    <p class='ocr_par' id='par_1_16' lang='eng' title="bbox 625 1206 834 1225">
+     <span class='ocr_line' id='line_1_33' title="bbox 625 1206 834 1225; baseline -0.005 0; x_size 23.69949; x_descenders 5.5158105; x_ascenders 5.590498">
+      <span class='ocrx_word' id='word_1_270' title='bbox 625 1206 760 1225; x_wconf 96'>HAMILTON</span>
+      <span class='ocrx_word' id='word_1_271' title='bbox 768 1207 834 1224; x_wconf 95'>FISH.</span>
      </span>
     </p>
    </div>
-   <div class='ocr_carea' id='block_1_10' title="bbox 140 1227 314 1246">
-    <p class='ocr_par' id='par_1_17' lang='eng' title="bbox 140 1227 314 1246">
-     <span class='ocr_line' id='line_1_34' title="bbox 140 1227 314 1246; baseline 0.011 -2; x_size 22.296295; x_descenders 5.2962961; x_ascenders 4">
-      <span class='ocrx_word' id='word_1_270' title='bbox 140 1227 180 1244; x_wconf 96'>The</span>
-      <span class='ocrx_word' id='word_1_271' title='bbox 188 1228 314 1246; x_wconf 81'>PRESIDENT,</span>
+   <div class='ocr_carea' id='block_1_10' title="bbox 143 1229 317 1247">
+    <p class='ocr_par' id='par_1_17' lang='eng' title="bbox 143 1229 317 1247">
+     <span class='ocr_line' id='line_1_34' title="bbox 143 1229 317 1247; baseline 0.006 -1; x_size 22.296295; x_descenders 5.2962961; x_ascenders 4">
+      <span class='ocrx_word' id='word_1_272' title='bbox 143 1229 183 1246; x_wconf 96'>The</span>
+      <span class='ocrx_word' id='word_1_273' title='bbox 191 1230 317 1247; x_wconf 74'>PRESIDENT,</span>
      </span>
     </p>
    </div>
-   <div class='ocr_carea' id='block_1_11' title="bbox 341 1307 628 1331">
-    <p class='ocr_par' id='par_1_18' lang='eng' title="bbox 341 1307 628 1331">
-     <span class='ocr_line' id='line_1_35' title="bbox 341 1307 628 1331; baseline 0.007 -7; x_size 23; x_descenders 5; x_ascenders 5">
-      <span class='ocrx_word' id='word_1_272' title='bbox 341 1307 383 1324; x_wconf 95'>List</span>
-      <span class='ocrx_word' id='word_1_273' title='bbox 389 1307 411 1330; x_wconf 92'>of</span>
-      <span class='ocrx_word' id='word_1_274' title='bbox 419 1308 553 1330; x_wconf 90'>accompanying</span>
-      <span class='ocrx_word' id='word_1_275' title='bbox 557 1313 628 1331; x_wconf 89'>papers.</span>
+   <div class='ocr_carea' id='block_1_11' title="bbox 345 1308 632 1331">
+    <p class='ocr_par' id='par_1_18' lang='eng' title="bbox 345 1308 632 1331">
+     <span class='ocr_line' id='line_1_35' title="bbox 345 1308 632 1331; baseline 0 -6; x_size 22; x_descenders 5; x_ascenders 4">
+      <span class='ocrx_word' id='word_1_274' title='bbox 345 1308 387 1325; x_wconf 91'>List</span>
+      <span class='ocrx_word' id='word_1_275' title='bbox 393 1308 415 1331; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_276' title='bbox 423 1308 557 1330; x_wconf 96'>accompanying</span>
+      <span class='ocrx_word' id='word_1_277' title='bbox 561 1312 632 1331; x_wconf 87'>papers.</span>
      </span>
     </p>
    </div>
-   <div class='ocr_carea' id='block_1_12' title="bbox 138 1337 853 1395">
-    <p class='ocr_par' id='par_1_19' lang='eng' title="bbox 138 1337 853 1395">
-     <span class='ocr_line' id='line_1_36' title="bbox 138 1337 853 1372; baseline 0.007 -10; x_size 23; x_descenders 5; x_ascenders 6">
-      <span class='ocrx_word' id='word_1_276' title='bbox 138 1346 155 1363; x_wconf 96'>1,</span>
-      <span class='ocrx_word' id='word_1_277' title='bbox 166 1346 289 1368; x_wconf 96'>Regulations</span>
-      <span class='ocrx_word' id='word_1_278' title='bbox 296 1337 325 1364; x_wconf 96'>for</span>
-      <span class='ocrx_word' id='word_1_279' title='bbox 334 1339 366 1365; x_wconf 97'>the</span>
-      <span class='ocrx_word' id='word_1_280' title='bbox 374 1348 459 1366; x_wconf 96'>consular</span>
-      <span class='ocrx_word' id='word_1_281' title='bbox 466 1350 529 1366; x_wconf 96'>courts</span>
-      <span class='ocrx_word' id='word_1_282' title='bbox 537 1349 558 1366; x_wconf 96'>of</span>
-      <span class='ocrx_word' id='word_1_283' title='bbox 566 1349 599 1366; x_wconf 96'>the</span>
-      <span class='ocrx_word' id='word_1_284' title='bbox 607 1349 678 1367; x_wconf 96'>United</span>
-      <span class='ocrx_word' id='word_1_285' title='bbox 686 1349 749 1367; x_wconf 96'>States</span>
-      <span class='ocrx_word' id='word_1_286' title='bbox 757 1349 777 1367; x_wconf 96'>in</span>
-      <span class='ocrx_word' id='word_1_287' title='bbox 784 1350 853 1372; x_wconf 96'>Japan.</span>
+   <div class='ocr_carea' id='block_1_12' title="bbox 142 1340 857 1393">
+    <p class='ocr_par' id='par_1_19' lang='eng' title="bbox 142 1340 857 1393">
+     <span class='ocr_line' id='line_1_36' title="bbox 142 1340 857 1370; baseline 0 -5; x_size 23; x_descenders 5; x_ascenders 6">
+      <span class='ocrx_word' id='word_1_278' title='bbox 142 1348 159 1365; x_wconf 96'>1,</span>
+      <span class='ocrx_word' id='word_1_279' title='bbox 170 1348 293 1370; x_wconf 96'>Regulations</span>
+      <span class='ocrx_word' id='word_1_280' title='bbox 300 1348 329 1365; x_wconf 96'>for</span>
+      <span class='ocrx_word' id='word_1_281' title='bbox 338 1340 370 1366; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_282' title='bbox 378 1348 463 1366; x_wconf 96'>consular</span>
+      <span class='ocrx_word' id='word_1_283' title='bbox 470 1350 533 1366; x_wconf 96'>courts</span>
+      <span class='ocrx_word' id='word_1_284' title='bbox 541 1349 562 1366; x_wconf 96'>of</span>
+      <span class='ocrx_word' id='word_1_285' title='bbox 570 1348 603 1365; x_wconf 96'>the</span>
+      <span class='ocrx_word' id='word_1_286' title='bbox 611 1348 682 1366; x_wconf 95'>United</span>
+      <span class='ocrx_word' id='word_1_287' title='bbox 690 1348 753 1366; x_wconf 96'>States</span>
+      <span class='ocrx_word' id='word_1_288' title='bbox 761 1347 781 1365; x_wconf 95'>in</span>
+      <span class='ocrx_word' id='word_1_289' title='bbox 788 1348 857 1370; x_wconf 93'>Japan.</span>
      </span>
-     <span class='ocr_line' id='line_1_37' title="bbox 138 1369 812 1395; baseline 0.007 -10; x_size 23; x_descenders 5; x_ascenders 6">
-      <span class='ocrx_word' id='word_1_288' title='bbox 138 1369 155 1385; x_wconf 90'>2.</span>
-      <span class='ocrx_word' id='word_1_289' title='bbox 166 1369 201 1385; x_wconf 95'>Mr.</span>
-      <span class='ocrx_word' id='word_1_290' title='bbox 210 1369 256 1386; x_wconf 96'>Fish</span>
-      <span class='ocrx_word' id='word_1_291' title='bbox 264 1370 284 1387; x_wconf 96'>to</span>
-      <span class='ocrx_word' id='word_1_292' title='bbox 291 1370 326 1387; x_wconf 86'>Mr.</span>
-      <span class='ocrx_word' id='word_1_293' title='bbox 334 1370 364 1387; x_wconf 96'>De</span>
-      <span class='ocrx_word' id='word_1_294' title='bbox 371 1370 429 1393; x_wconf 96'>Long,</span>
-      <span class='ocrx_word' id='word_1_295' title='bbox 438 1371 546 1392; x_wconf 96'>September</span>
-      <span class='ocrx_word' id='word_1_296' title='bbox 553 1372 581 1394; x_wconf 96'>10,</span>
-      <span class='ocrx_word' id='word_1_297' title='bbox 590 1372 640 1389; x_wconf 96'>1870.</span>
-      <span class='ocrx_word' id='word_1_298' title='bbox 810 1393 812 1395; x_wconf 0'>.</span>
+     <span class='ocr_line' id='line_1_37' title="bbox 142 1371 644 1393; baseline 0.004 -7; x_size 30.84; x_descenders 5; x_ascenders 8.8400002">
+      <span class='ocrx_word' id='word_1_290' title='bbox 142 1371 159 1387; x_wconf 90'>2,</span>
+      <span class='ocrx_word' id='word_1_291' title='bbox 170 1371 205 1387; x_wconf 96'>Mr.</span>
+      <span class='ocrx_word' id='word_1_292' title='bbox 214 1371 260 1388; x_wconf 95'>Fish</span>
+      <span class='ocrx_word' id='word_1_293' title='bbox 268 1372 288 1389; x_wconf 95'>to</span>
+      <span class='ocrx_word' id='word_1_294' title='bbox 295 1371 330 1388; x_wconf 95'>Mr.</span>
+      <span class='ocrx_word' id='word_1_295' title='bbox 338 1371 368 1388; x_wconf 96'>De</span>
+      <span class='ocrx_word' id='word_1_296' title='bbox 375 1371 433 1393; x_wconf 96'>Long,</span>
+      <span class='ocrx_word' id='word_1_297' title='bbox 442 1371 550 1392; x_wconf 95'>September</span>
+      <span class='ocrx_word' id='word_1_298' title='bbox 557 1371 585 1393; x_wconf 95'>10,</span>
+      <span class='ocrx_word' id='word_1_299' title='bbox 594 1371 644 1388; x_wconf 79'>1870,</span>
      </span>
     </p>
    </div>
    <div class='ocr_carea' id='block_1_13' title="bbox 0 0 1000 1520">
     <p class='ocr_par' id='par_1_20' lang='eng' title="bbox 0 0 1000 1520">
-     <span class='ocr_line' id='line_1_38' title="bbox 0 0 1000 1520; baseline 0 9; x_size 770; x_descenders -385; x_ascenders 385">
-      <span class='ocrx_word' id='word_1_299' title='bbox 0 0 1000 1520; x_wconf 95'> </span>
+     <span class='ocr_line' id='line_1_38' title="bbox 0 0 1000 1520; baseline 0 2; x_size 763; x_descenders -381.5; x_ascenders 381.5">
+      <span class='ocrx_word' id='word_1_300' title='bbox 0 0 1000 1520; x_wconf 95'> </span>
      </span>
     </p>
    </div>
-   <div class='ocr_carea' id='block_1_14' title="bbox 19 1496 978 1520">
-    <p class='ocr_par' id='par_1_21' lang='eng' title="bbox 19 1496 978 1520">
-     <span class='ocr_line' id='line_1_39' title="bbox 19 1496 978 1520; baseline 0 7; x_size 16; x_descenders -8; x_ascenders 8">
-      <span class='ocrx_word' id='word_1_300' title='bbox 19 1496 978 1520; x_wconf 95'> </span>
+   <div class='ocr_carea' id='block_1_14' title="bbox 24 1504 845 1520">
+    <p class='ocr_par' id='par_1_21' lang='eng' title="bbox 24 1504 845 1520">
+     <span class='ocr_line' id='line_1_39' title="bbox 24 1504 845 1520; baseline 0 0; x_size 8; x_descenders -4; x_ascenders 4">
+      <span class='ocrx_word' id='word_1_301' title='bbox 24 1504 845 1520; x_wconf 95'> </span>
+     </span>
+    </p>
+   </div>
+   <div class='ocr_carea' id='block_1_15' title="bbox 982 987 1000 1500">
+    <p class='ocr_par' id='par_1_22' lang='eng' title="bbox 982 987 1000 1500">
+     <span class='ocr_line' id='line_1_40' title="bbox 982 987 1000 1500; baseline 0 0; x_size 256.5; x_descenders -128.25; x_ascenders 128.25">
+      <span class='ocrx_word' id='word_1_302' title='bbox 982 987 1000 1500; x_wconf 95'> </span>
      </span>
     </p>
    </div>

tests/cache/jbig2/__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt/txt.bin

@@ -1,5 +1,5 @@
-41st CONGRESS; | SENATE.
-3d Session. }
+4ist ConGREss, } SENATE. { Ex. Doc,
+3d Session. No. 25.
 
 MESSAGE
 
@@ -12,7 +12,7 @@ COMMUNICATING
 A copy of regulations for the consular courts of the United States in Japan,
 decreed and issued by the minister of the United States in that country.
 
-January 27, 1871,—Read, referred to the Committee on Commerce, and ordered to be
+JANUARY 27, 1871,—Read, referred to the Committee on Commerce, and ordered to be
 printed.
 
 To the Senate and House of Representatives :
@@ -26,13 +26,13 @@ U. 8. GRANT.
 ‘WASHINGTON, January 27, 1871.
 
 DEPARTMENT OF STATE,
-. Washington, January 26, 1870,
+Washington, January 26, 1870,
 
 The Secretary of State has the honor to submit herewith, for revision
 by Congress, in conformity with the provisions of section 6 of the act
 approved 22d of June, 1860, a copy of “regulations for the consular
-courts of the United States in Japan,” decreed and issued by C. E.
-De Long, the miniater of the United States in that country, in Septem-
+courts of the United States in Japan,” decreed and issued by C. BE.
+De Long, the minister of the United States in that country, in Septem-
 ber, 1870; and also the papers mentioned in the subjoined list, which,
 contain suggestions on the subject thereof.
 
@@ -43,7 +43,7 @@ decrees and regulation, in the sense in which it is limited by paragraph
 431 of the article before named—that is, “to acts necessary to organize
 and give efficiency to the courts created by the act.”
 
-Respectfully submitted. .
+Respectfully submitted.
 
 HAMILTON FISH.
 
@@ -52,7 +52,9 @@ The PRESIDENT,
 List of accompanying papers.
 
 1, Regulations for the consular courts of the United States in Japan.
-2. Mr. Fish to Mr. De Long, September 10, 1870. .
+2, Mr. Fish to Mr. De Long, September 10, 1870,
+
+ 
 
  
 

tests/cache/lichtenstein/__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt/hocr.bin

@@ -9,7 +9,7 @@
   <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 "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/com.github.ocrmypdf.wyikj5nw/000001_ocr.png"; bbox 0 0 512 512; ppageno 0'>
+  <div class='ocr_page' id='page_1' title='image "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/ocrmypdf.io.tyuf6jbx/000001_ocr.png"; bbox 0 0 512 512; ppageno 0'>
    <div class='ocr_carea' id='block_1_1' title="bbox 0 0 477 512">
     <p class='ocr_par' id='par_1_1' lang='eng' title="bbox 0 0 477 512">
      <span class='ocr_line' id='line_1_1' title="bbox 0 0 477 512; baseline 0 0; x_size 684; x_descenders 171; x_ascenders 171">

tests/cache/manifest.jsonl

@@ -1,64 +1,44 @@
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__deu__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/francais.pdf", "args": ["-l", "deu", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000004_ocr.png__000004_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000004_ocr.png", "$TMPDIR/000004_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000004_ocr.png__000004_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000004_ocr.png", "$TMPDIR/000004_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000004_ocr.png__000004_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000004_ocr.png", "$TMPDIR/000004_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000004_ocr.png__000004_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000004_ocr.png", "$TMPDIR/000004_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/2400dpi.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/jbig2.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/graph_ocred.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/skew.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000002_ocr.png__000002_ocr_tess__pdf__txt", "sourcefile": "resources/3small.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000002_ocr.png", "$TMPDIR/000002_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/3small.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000005_ocr.png__000005_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000005_ocr.png", "$TMPDIR/000005_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000005_ocr.png__000005_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000005_ocr.png", "$TMPDIR/000005_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000005_ocr.png__000005_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000005_ocr.png", "$TMPDIR/000005_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000005_ocr.png__000005_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000005_ocr.png", "$TMPDIR/000005_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/ccitt.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/skew.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/ccitt.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_tess__pdf__txt", "sourcefile": "resources/3small.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000006_ocr.png__000006_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000006_ocr.png", "$TMPDIR/000006_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000006_ocr.png__000006_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000006_ocr.png", "$TMPDIR/000006_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000006_ocr.png__000006_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000006_ocr.png", "$TMPDIR/000006_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000006_ocr.png__000006_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000006_ocr.png", "$TMPDIR/000006_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/lichtenstein.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/lichtenstein.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000006_ocr.png__000006_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000006_ocr.png", "$TMPDIR/000006_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/aspect.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/palette.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__osd__--psm__0__000001_rasterize_preview.jpg__stdout", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000001_rasterize_preview.jpg", "stdout"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__osd__--psm__0__000003_rasterize_preview.jpg__stdout", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000003_rasterize_preview.jpg", "stdout"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__osd__--psm__0__000004_rasterize_preview.jpg__stdout", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000004_rasterize_preview.jpg", "stdout"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__osd__--psm__0__000002_rasterize_preview.jpg__stdout", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000002_rasterize_preview.jpg", "stdout"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/aspect.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/jbig2.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__osd__--psm__0__000004_rasterize_preview.jpg__stdout", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000004_rasterize_preview.jpg", "stdout"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__osd__--psm__0__000001_rasterize_preview.jpg__stdout", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000001_rasterize_preview.jpg", "stdout"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__osd__--psm__0__000003_rasterize_preview.jpg__stdout", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000003_rasterize_preview.jpg", "stdout"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__osd__--psm__0__000002_rasterize_preview.jpg__stdout", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000002_rasterize_preview.jpg", "stdout"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/palette.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/palette.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__--psm__7__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/skew.pdf", "args": ["-l", "eng", "--psm", "7", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "$TMPDIR/000002_ocr.png", "$TMPDIR/000002_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000002_ocr.png__000002_ocr_tess__pdf__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000002_ocr.png", "$TMPDIR/000002_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_hocr__hocr__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000004_ocr.png__000004_ocr_hocr__hocr__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "$TMPDIR/000004_ocr.png", "$TMPDIR/000004_ocr_hocr", "hocr", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__--psm__7__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/skew.pdf", "args": ["-l", "eng", "--psm", "7", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000002_ocr.png__000002_ocr_tess__pdf__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000002_ocr.png", "$TMPDIR/000002_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_tess__pdf__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000004_ocr.png__000004_ocr_tess__pdf__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000004_ocr.png", "$TMPDIR/000004_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_tess__pdf__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000004_ocr.png__000004_ocr_tess__pdf__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000004_ocr.png", "$TMPDIR/000004_ocr_tess", "pdf", "txt"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__osd__--psm__0__000001_rasterize_preview.jpg__stdout", "sourcefile": "resources/poster.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000001_rasterize_preview.jpg", "stdout"]}
-{"tesseract_version": "4.1.1", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/poster.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__deu__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/francais.pdf", "args": ["-l", "deu", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__--psm__7__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/skew.pdf", "args": ["-l", "eng", "--psm", "7", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__--psm__7__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/skew.pdf", "args": ["-l", "eng", "--psm", "7", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/aspect.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/ccitt.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/jbig2.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/lichtenstein.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/palette.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/skew.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/2400dpi.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/3small.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/aspect.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/ccitt.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/graph_ocred.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/jbig2.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/lichtenstein.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/palette.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/poster.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/skew.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "$TMPDIR/000002_ocr.png", "$TMPDIR/000002_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000002_ocr.png__000002_ocr_tess__pdf__txt", "sourcefile": "resources/3small.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000002_ocr.png", "$TMPDIR/000002_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000002_ocr.png__000002_ocr_tess__pdf__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000002_ocr.png", "$TMPDIR/000002_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_hocr__hocr__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_tess__pdf__txt", "sourcefile": "resources/3small.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_tess__pdf__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000004_ocr.png__000004_ocr_hocr__hocr__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "$TMPDIR/000004_ocr.png", "$TMPDIR/000004_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000004_ocr.png__000004_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000004_ocr.png", "$TMPDIR/000004_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000004_ocr.png__000004_ocr_tess__pdf__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000004_ocr.png", "$TMPDIR/000004_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000004_ocr.png__000004_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000004_ocr.png", "$TMPDIR/000004_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000005_ocr.png__000005_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000005_ocr.png", "$TMPDIR/000005_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000005_ocr.png__000005_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000005_ocr.png", "$TMPDIR/000005_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000006_ocr.png__000006_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000006_ocr.png", "$TMPDIR/000006_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__eng__000006_ocr.png__000006_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000006_ocr.png", "$TMPDIR/000006_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__osd__--psm__0__000001_rasterize_preview.jpg__stdout", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000001_rasterize_preview.jpg", "stdout"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__osd__--psm__0__000001_rasterize_preview.jpg__stdout", "sourcefile": "resources/poster.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000001_rasterize_preview.jpg", "stdout"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__osd__--psm__0__000002_rasterize_preview.jpg__stdout", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000002_rasterize_preview.jpg", "stdout"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__osd__--psm__0__000003_rasterize_preview.jpg__stdout", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000003_rasterize_preview.jpg", "stdout"]}
+{"tesseract_version": "4.1.1", "platform": "macOS-10.14.6-x86_64-i386-64bit", "python": "3.9.0", "argv_slug": "__-l__osd__--psm__0__000004_rasterize_preview.jpg__stdout", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "osd", "--psm", "0", "$TMPDIR/000004_rasterize_preview.jpg", "stdout"]}

tests/cache/multipage/__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt/hocr.bin

@@ -9,7 +9,7 @@
   <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 "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/com.github.ocrmypdf.asx0dv8k/000001_ocr.png"; bbox 0 0 1600 1962; ppageno 0'>
+  <div class='ocr_page' id='page_1' title='image "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/ocrmypdf.io.sh2wg519/000001_ocr.png"; bbox 0 0 1600 1962; ppageno 0'>
    <div class='ocr_carea' id='block_1_1' title="bbox 84 216 786 1605">
     <p class='ocr_par' id='par_1_1' lang='eng' title="bbox 84 216 786 1605">
      <span class='ocr_line' id='line_1_1' title="bbox 84 216 786 1605; baseline 0 0; x_size 694.5; x_descenders -347.25; x_ascenders 347.25">

tests/cache/multipage/__-l__eng__000003_ocr.png__000003_ocr_hocr__hocr__txt/hocr.bin

@@ -9,7 +9,7 @@
   <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 "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/com.github.ocrmypdf.asx0dv8k/000003_ocr.png"; bbox 0 0 2272 1848; ppageno 0'>
+  <div class='ocr_page' id='page_1' title='image "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/ocrmypdf.io.sh2wg519/000003_ocr.png"; bbox 0 0 2272 1848; ppageno 0'>
    <div class='ocr_carea' id='block_1_1' title="bbox 201 38 2167 118">
     <p class='ocr_par' id='par_1_1' lang='eng' title="bbox 201 38 2167 118">
      <span class='ocr_line' id='line_1_1' title="bbox 201 38 2167 118; baseline -0.001 -17; x_size 79; x_descenders 17; x_ascenders 16">

tests/cache/multipage/__-l__eng__000004_ocr.png__000004_ocr_hocr__hocr__txt/hocr.bin

@@ -9,7 +9,7 @@
   <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 "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/com.github.ocrmypdf.asx0dv8k/000004_ocr.png"; bbox 0 0 2272 1848; ppageno 0'>
+  <div class='ocr_page' id='page_1' title='image "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/ocrmypdf.io.sh2wg519/000004_ocr.png"; bbox 0 0 2272 1848; ppageno 0'>
    <div class='ocr_carea' id='block_1_1' title="bbox 200 38 2168 120">
     <p class='ocr_par' id='par_1_1' lang='eng' title="bbox 200 38 2168 120">
      <span class='ocr_line' id='line_1_1' title="bbox 200 38 2168 120; baseline 0 -18; x_size 80; x_descenders 18; x_ascenders 16">

tests/cache/multipage/__-l__eng__000005_ocr.png__000005_ocr_hocr__hocr__txt/hocr.bin

@@ -9,7 +9,7 @@
   <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 "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/com.github.ocrmypdf.asx0dv8k/000005_ocr.png"; bbox 0 0 2550 3300; ppageno 0'>
+  <div class='ocr_page' id='page_1' title='image "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/ocrmypdf.io.sh2wg519/000005_ocr.png"; bbox 0 0 2550 3300; ppageno 0'>
    <div class='ocr_carea' id='block_1_1' title="bbox 70 403 1182 833">
     <p class='ocr_par' id='par_1_1' lang='eng' title="bbox 70 403 1182 833">
      <span class='ocr_line' id='line_1_1' title="bbox 70 403 1180 461; baseline 0.002 -14; x_size 59; x_descenders 13; x_ascenders 15">

tests/cache/multipage/__-l__eng__000006_ocr.png__000006_ocr_hocr__hocr__txt/hocr.bin

@@ -9,7 +9,7 @@
   <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 "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/com.github.ocrmypdf.asx0dv8k/000006_ocr.png"; bbox 0 0 3300 2550; ppageno 0'>
+  <div class='ocr_page' id='page_1' title='image "/var/folders/2s/7t022mgj0h5cprbq0dtb1ksm0000gn/T/ocrmypdf.io.sh2wg519/000006_ocr.png"; bbox 0 0 3300 2550; ppageno 0'>
    <div class='ocr_carea' id='block_1_1' title="bbox 438 20 1559 450">
     <p class='ocr_par' id='par_1_1' lang='eng' title="bbox 438 20 1559 450">
      <span class='ocr_line' id='line_1_1' title="bbox 438 20 1556 74; baseline 0 -12; x_size 54; x_descenders 12; x_ascenders 14">