Support pdfminer.six 20191020

[ci skip]

.docker/Dockerfile

@@ -1,6 +1,6 @@
 # OCRmyPDF
 #
-FROM ubuntu:19.04 as base
+FROM ubuntu:19.10 as base
 
 FROM base as builder
 
@@ -10,16 +10,11 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
   build-essential autoconf automake libtool \
   libleptonica-dev \
   zlib1g-dev \
-  ocrmypdf \
-  pngquant \
+  python3-setuptools \
   python3-pip \
-  python3-venv \
-  tesseract-ocr \
-  unpaper \
   wget \
   git
 
-
 # Compile and install jbig2
 # Needs libleptonica-dev, zlib1g-dev
 RUN \
@@ -31,15 +26,15 @@ RUN \
   && cd .. \
   && rm -rf jbig2
 
-RUN python3 -m venv /appenv
-
 COPY . /app
 
 WORKDIR /app
 
-RUN . /appenv/bin/activate; \
-  pip install --upgrade pip \
-  && pip install .
+RUN pip3 install --no-cache-dir \
+  -r requirements/main.txt \
+  -r requirements/webservice.txt \
+  -r requirements/test.txt \
+  .
 
 FROM base
 
@@ -53,7 +48,6 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
   zlib1g \
   pngquant \
   python3 \
-  python3-venv \
   qpdf \
   tesseract-ocr \
   tesseract-ocr-chi-sim \
@@ -62,10 +56,13 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
   tesseract-ocr-fra \
   tesseract-ocr-por \
   tesseract-ocr-spa \
-  unpaper \
-  wget
+  unpaper
+
+WORKDIR /app
+
+COPY --from=builder /usr/local/lib/ /usr/local/lib/
+COPY --from=builder /usr/local/bin/ /usr/local/bin/
 
-# Copy
 COPY --from=builder /app/misc/webservice.py /app/
 
 # Copy minimal project files to get the test suite.
@@ -74,7 +71,4 @@ COPY --from=builder /app/requirements /app/requirements
 COPY --from=builder /app/tests /app/tests
 COPY --from=builder /app/src /app/src
 
-COPY --from=builder /appenv /appenv
-COPY --from=builder /usr/local /usr/local
-
-ENTRYPOINT ["/appenv/bin/ocrmypdf"]
+ENTRYPOINT ["/usr/local/bin/ocrmypdf"]

.docker/alpine.dockerfile

@@ -1,91 +0,0 @@
-FROM alpine:3.9 as base
-
-FROM base as builder
-
-ENV LANG=C.UTF-8
-
-# Normally:
-# echo '@testing http://nl.alpinelinux.org/alpine/edge/testing' >> /etc/apk/repositories
-
-RUN \
-  echo -e '@testing http://nl.alpinelinux.org/alpine/edge/testing\n@community http://nl.alpinelinux.org/alpine/edge/community'\
-  >> /etc/apk/repositories \
-  # Add runtime dependencies
-  && apk add --update \
-  python3-dev \
-  py3-setuptools \
-  jbig2enc@testing \
-  ghostscript \
-  qpdf@community \
-  qpdf-dev@community \
-  tesseract-ocr \
-  unpaper \
-  pngquant \
-  libxml2-dev \
-  libxslt-dev \
-  zlib-dev \
-  libffi-dev \
-  leptonica-dev \
-  binutils \
-  && pip3 install --upgrade pip \
-  # Install pybind11 for pikepdf
-  && pip3 install pybind11 \
-  # Install flask for the webservice
-  && pip3 install flask \
-  # Add build dependencies
-  && apk add --virtual build-dependencies \
-  build-base \
-  git
-
-COPY . /app
-
-WORKDIR /app
-
-RUN pip3 install .
-
-FROM base
-
-ENV LANG=C.UTF-8
-
-# Normally:
-# echo '@testing http://nl.alpinelinux.org/alpine/edge/testing' >> /etc/apk/repositories
-
-RUN \
-  echo -e '@testing http://nl.alpinelinux.org/alpine/edge/testing\n@community http://nl.alpinelinux.org/alpine/edge/community'\
-  >> /etc/apk/repositories \
-  # Add runtime dependencies
-  && apk add --update \
-  python3 \
-  jbig2enc@testing \
-  ghostscript \
-  qpdf@community \
-  qpdf-dev@community \
-  tesseract-ocr \
-  tesseract-ocr-data-deu \
-  tesseract-ocr-data-chi_sim \
-  unpaper \
-  pngquant \
-  libxml2 \
-  libxslt \
-  zlib \
-  libffi \
-  leptonica-dev \
-  binutils \
-  && mkdir /app
-
-WORKDIR /app
-
-# Copy build artifacts (python site-packages)
-COPY --from=builder /usr/lib/python3.6/site-packages /usr/lib/python3.6/site-packages
-COPY --from=builder /usr/bin/ocrmypdf /usr/bin/dumppdf.py /usr/bin/latin2ascii.py /usr/bin/pdf2txt.py /usr/bin/img2pdf /usr/bin/chardetect /usr/bin/
-
-# Copy
-COPY --from=builder /app/misc/webservice.py /app/
-
-# Copy minimal project files to get the test suite.
-COPY --from=builder /app/setup.cfg /app/setup.py /app/README.md /app/
-COPY --from=builder /app/requirements /app/requirements
-COPY --from=builder /app/tests /app/tests
-COPY --from=builder /app/src /app/src
-
-ENTRYPOINT ["/usr/bin/ocrmypdf"]

.travis.yml

@@ -1,7 +1,7 @@
 cache:
   pip: true
   directories:
-  - $HOME/Library/Caches/Homebrew
+    - $HOME/Library/Caches/Homebrew
 
 matrix:
   include:
@@ -12,27 +12,56 @@ matrix:
       python: "3.6"
       env:
         - DIST=trusty
-      addons: &trusty_apt
+        - MINIMAL=true
+      addons:
         apt:
           update: true
           sources:
-          - sourceline: 'ppa:alex-p/tesseract-ocr'
-          - sourceline: 'ppa:heyarje/libav-11'
-          - sourceline: 'ppa:vshn/ghostscript'
+            - sourceline: "ppa:alex-p/tesseract-ocr"
+            - sourceline: "ppa:vshn/ghostscript"
           packages:
-          - ghostscript
-          - libavcodec56
-          - libavformat56
-          - libavutil54
-          - libexempi3
-          - libffi-dev
-          - pngquant
-          - poppler-utils
-          - qpdf
-          - tesseract-ocr
-          - tesseract-ocr-deu
-          - tesseract-ocr-eng
-          - tesseract-ocr-fra
+            - ghostscript
+            - libffi-dev
+            - qpdf
+            - tesseract-ocr
+            - tesseract-ocr-deu
+            - tesseract-ocr-eng
+            - tesseract-ocr-fra
+      before_install: |
+        pip3 install --upgrade pip
+        pip3 install --upgrade wheel
+    - os: linux
+      dist: trusty
+      sudo: required
+      language: python
+      python: "3.6"
+      env:
+        - DIST=trusty
+      addons:
+        apt:
+          update: true
+          sources:
+            - sourceline: "ppa:alex-p/tesseract-ocr"
+            - sourceline: "ppa:heyarje/libav-11"
+            - sourceline: "ppa:vshn/ghostscript"
+          packages:
+            - ghostscript
+            - libavcodec56
+            - libavformat56
+            - libavutil54
+            - libffi-dev
+            - qpdf
+            - tesseract-ocr
+            - tesseract-ocr-deu
+            - tesseract-ocr-eng
+            - tesseract-ocr-fra
+            - libexempi3 # --- optional extras from here ---
+            - pngquant
+            - poppler-utils
+      before_install: |
+        mkdir -p bin packages
+        pip3 install --upgrade pip
+        pip3 install --upgrade wheel
     - os: linux
       dist: xenial
       sudo: required
@@ -44,21 +73,44 @@ matrix:
         apt:
           update: true
           sources:
-          - sourceline: 'ppa:alex-p/tesseract-ocr'
+            - sourceline: "ppa:alex-p/tesseract-ocr"
           packages:
-          - ghostscript
-          - libexempi3
-          - libffi-dev
-          - pngquant
-          - poppler-utils
-          - qpdf
-          - tesseract-ocr
-          - tesseract-ocr-deu
-          - tesseract-ocr-eng
-          - tesseract-ocr-fra
-          - unpaper
+            - ghostscript
+            - libexempi3
+            - libffi-dev
+            - pngquant
+            - poppler-utils
+            - qpdf
+            - tesseract-ocr
+            - tesseract-ocr-deu
+            - tesseract-ocr-eng
+            - tesseract-ocr-fra
+            - unpaper
+    # - os: linux
+    #   dist: bionic
+    #   sudo: required
+    #   language: python
+    #   python: "3.8"
+    #   env:
+    #     - DIST=bionic
+    #   addons:
+    #     apt:
+    #       update: true
+    #       sources:
+    #         - sourceline: "ppa:alex-p/tesseract-ocr"
+    #       packages:
+    #         - ghostscript
+    #         - libexempi3
+    #         - libffi-dev
+    #         - pngquant
+    #         - poppler-utils
+    #         - qpdf
+    #         - tesseract-ocr
+    #         - tesseract-ocr-deu
+    #         - tesseract-ocr-eng
+    #         - tesseract-ocr-fra
+    #         - unpaper
     - os: osx
-      osx_image: xcode9.2
       language: generic
       addons:
         homebrew:
@@ -74,73 +126,36 @@ matrix:
             - qpdf
             - tesseract
             - unpaper
-    - os: osx
-      osx_image: xcode9.2
-      language: generic
-      env:
-        - ADD_PDFMINER=1
-      addons:
-        homebrew:
-          update: true
-          packages:
-            - exempi
-            - ghostscript
-            - jbig2enc
-            - leptonica
-            - openjpeg
-            - pngquant
-            - python
-            - qpdf
-            - tesseract
-            - unpaper
+      before_install: |
+        pip3 install --upgrade pip
+        pip3 install wheel
 
 before_cache:
-- rm -f $HOME/.cache/pip/log/debug.log
-
-before_install: |
-  mkdir -p bin
-  if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then
-    pip3 install --upgrade pip
-    pip3 install --upgrade wheel
-    if [[ "$DIST" == "trusty" ]]; then
-      mkdir -p packages
-      wget -q 'https://www.dropbox.com/s/vaq0kbwi6e6au80/unpaper_6.1-1.deb?raw=1' -O packages/unpaper_6.1-1.deb
-      sudo dpkg -i packages/unpaper_6.1-1.deb
-    fi
-  elif [[ "$TRAVIS_OS_NAME" == "osx" ]]; then
-    pip3 install --upgrade pip
-    pip3 install wheel
-  fi
+  - rm -f $HOME/.cache/pip/log/debug.log
 
 install:
-- export PATH=$PWD/bin:$PATH
-- pip3 install pycparser  # py3.7 workaround for https://github.com/eliben/pycparser/issues/251
-- pip3 install -r requirements/main.txt
-- pip3 install --no-deps .
-- |
-  if [[ "$ADD_PDFMINER" == "1" ]]; then
-    pip3 install --no-deps .[pdfminer]
-  fi
-- pip3 install -r requirements/test.txt
+  - mkdir -p bin
+  - export PATH=$PWD/bin:$PATH
+  - pip3 install -r requirements/main.txt -r requirements/test.txt .
 
 script:
-- tesseract --version
-- qpdf --version
-- pytest -n auto
+  - tesseract --version
+  - qpdf --version
+  - pytest -n auto
 
 deploy:
   # release for main pypi
-  # 3.6 is considered the build leader and does the deploy, otherwise there is
+  # 3.7 is considered the build leader and does the deploy, otherwise there is
   # a race and all versions will try to deploy
   # OTOH if we ever need separate binary wheels then each version needs its
   # own deploy
-- provider: pypi
-  user: ocrmypdf-travis
-  password:
-    secure: "DTFOmmNL6olA0+yXvp4u9jXZlZeqrJsJ0526jzqf4a3gZ6jnGTq5UI6WzRsslSyoMMfXKtHQebqHM6ogSgCZinyZ3ufHJo8fn9brxbEc2gsiWkbj5o3bGwdWMT1vNNE7XW0VCpw87rZ1EEwjl4FJHFudMlPR1yfU5+uq0k0PACo="
-  distributions: "sdist bdist_wheel"
-  on:
-    branch: master
-    tags: true
-    condition: $TRAVIS_PYTHON_VERSION == "3.6" && $TRAVIS_OS_NAME == "linux"
-  skip_upload_docs: true
+  - provider: pypi
+    user: ocrmypdf-travis
+    password:
+      secure: "DTFOmmNL6olA0+yXvp4u9jXZlZeqrJsJ0526jzqf4a3gZ6jnGTq5UI6WzRsslSyoMMfXKtHQebqHM6ogSgCZinyZ3ufHJo8fn9brxbEc2gsiWkbj5o3bGwdWMT1vNNE7XW0VCpw87rZ1EEwjl4FJHFudMlPR1yfU5+uq0k0PACo="
+    distributions: "sdist bdist_wheel"
+    on:
+      branch: master
+      tags: true
+      condition: $TRAVIS_PYTHON_VERSION == "3.7" && $TRAVIS_OS_NAME == "linux"
+    skip_upload_docs: true

MANIFEST.in

@@ -1,43 +0,0 @@
-# requirements
-recursive-include requirements *
-
-# git
-include .git_archival.txt
-
-# docker
-include .dockerignore
-recursive-include .docker *
-
-# tests
-include .coveragerc
-recursive-include tests *.bin
-recursive-include tests *.jpg
-recursive-include tests *.jsonl
-recursive-include tests *.png
-recursive-include tests *.pdf
-recursive-include tests *.py
-recursive-include tests *.rst
-recursive-include tests *.txt
-recursive-exclude tests/resources/private *
-
-# documentation
-include LICENSE
-include *.rst
-recursive-exclude .github *
-recursive-include docs *.py
-recursive-include docs *.rst
-recursive-include docs *.svg
-recursive-exclude docs/_build *
-
-
-# support files
-recursive-include src/ocrmypdf/data *
-include *.py
-exclude tasks.py
-recursive-exclude .travis *
-exclude .travis*
-
-
-# code
-exclude src/ocrmypdf/lib/_leptonica.py
-exclude scratch.py

README.md

@@ -1,6 +1,6 @@
 <img src="docs/images/logo.svg" width="240" alt="OCRmyPDF">
 
-[![Travis build status][travis]](https://travis-ci.org/jbarlow83/OCRmyPDF) [![PyPI version][pypi]](https://pypi.org/project/ocrmypdf/) ![Homebrew version][homebrew] ![ReadTheDocs][docs]
+[![Travis build status][travis]](https://travis-ci.org/jbarlow83/OCRmyPDF) [![PyPI version][pypi]](https://pypi.org/project/ocrmypdf/) ![Homebrew version][homebrew] ![ReadTheDocs][docs] ![Python versions][pyversions]
 
 [travis]: https://travis-ci.org/jbarlow83/OCRmyPDF.svg?branch=master "Travis build status"
 
@@ -10,6 +10,8 @@
 
 [docs]: https://readthedocs.org/projects/ocrmypdf/badge/?version=latest "RTD"
 
+[pyversions]: https://img.shields.io/pypi/pyversions/ocrmypdf "Supported Python versions"
+
 OCRmyPDF adds an OCR text layer to scanned PDF files, allowing them to be searched or copy-pasted.
 
 ```bash
@@ -120,7 +122,7 @@ If you detect an issue, please:
 Requirements
 ------------
 
-Runs on CPython 3.5, 3.6 and 3.7. Requires external program installations of Ghostscript, Tesseract OCR, QPDF, and Leptonica. ocrmypdf is pure Python, but uses CFFI to portably generate library bindings.
+In addition to the required Python version (3.6+), OCRmyPDF requires external program installations of Ghostscript, Tesseract OCR, QPDF, and Leptonica. ocrmypdf is pure Python, but uses CFFI to portably generate library bindings.
 
 Press & Media
 -------------

docs/advanced.rst

@@ -322,15 +322,23 @@ working files on a per page basis have the page number as a prefix
 (starting with page 1), an infix indicates the processing stage, and a
 suffix indicates the file type. Some important files include:
 
--  ``.page.png`` - what the input page looks like
--  ``.image`` - the image we will show the user if we are in a mode that
-   changes the final appearance; may be in one of several image formats
--  ``.text.pdf`` - the OCR file; this will load as a blank page but
-   should have visible text if checked with a tool like pdftotext or
-   pdfminder.six
--  ``.ocr.png`` - the file that is sent to Tesseract for OCR; depending
+-  ``_rasterize.png`` - what the input page looks like
+-  ``_ocr.png`` - the file that is sent to Tesseract for OCR; depending
    on arguments this may differ from the presentation image
--  ``layers.rendered.pdf`` - the composite PDF, before metadata repair
-   and optimization
+-  ``_pp_deskew.png`` - the image, after deskewing
+-  ``_pp_clean.png`` - the image, after cleaning with unpaper
+-  ``_ocr_tess.pdf`` - the OCR file; appears as a blank page with invisible
+   text embedded
+-  ``_ocr_tess.txt`` - the OCR text (not necessarily all text on the page,
+   if the page is mixed format)
+-  ``fix_docinfo.pdf`` - a temporary file created to fix the PDF DocumentInfo
+   data structure
+-  ``graft_layers.pdf`` - the rendered PDF with OCR layers grafted on
+-  ``pdfa.pdf`` - ``graft_layers.pdf`` after conversion to PDF/A
+-  ``pdfa.ps`` - a PostScript file used by Ghostscript for PDF/A conversion
+-  ``optimize.pdf`` - the PDF generated before optimization
+-  ``optimize.out.pdf`` - the PDF generated by optimization
+-  ``origin`` - the input file
+-  ``origin.pdf`` - the input file or the input image converted to PDF
 -  ``images/*`` - images extracted during the optimization process; here
    the prefix indicates a PDF object ID not a page number

docs/batch.rst

@@ -47,7 +47,7 @@ where the PDFs are stored):
 
 .. code-block:: bash
 
-   find . -printf '%p' -name '*.pdf' -exec docker run --rm -v <host dir>:<container dir> jbarlow83/ocrmypdf-alpine '<container dir>/{}' '<container dir>/{}' \;
+   find . -printf '%p' -name '*.pdf' -exec docker run --rm -v <host dir>:<container dir> jbarlow83/ocrmypdf '<container dir>/{}' '<container dir>/{}' \;
 
 This only runs one ``ocrmypdf`` process at a time. This variation uses
 ``find`` to create a directory list and ``parallel`` to parallelize runs

docs/contributing.rst

@@ -0,0 +1,38 @@
+=======================
+Contributing guidelines
+=======================
+
+Contributions are welcome!
+
+Big changes
+===========
+
+Please open a new issue to discuss or propose a major change. Not only is it fun
+to discuss big ideas, but we might save each other's time too. Perhaps some of the
+work you're contemplating is already half-done in a development branch.
+
+Code style
+==========
+
+We use PEP8, ``black`` for code formatting and ``isort`` for import sorting. The
+settings for programs are in ``pyproject.toml`` and ``setup.cfg``.
+
+Tests
+=====
+
+New features should come with tests that confirm their correctness.
+
+New Python dependencies
+=======================
+
+If you are proposing a change that will require a new Python dependency, we
+prefer dependencies that are already packaged by Debian or Red Hat. This makes
+life much easier for our downstream package maintainers.
+
+Python dependencies must also be GPLv3 compatible.
+
+New non-Python dependencies
+===========================
+
+OCRmyPDF uses several external programs (Tesseract, Ghostscript and others) for
+its functionality. In general we prefer to avoid adding new external programs.

docs/cookbook.rst

@@ -165,8 +165,8 @@ might remove desirable content, especially from poor quality scans.
 -  ``--deskew`` will correct pages were scanned at a skewed angle by
    rotating them back into place. Skew determination and correction is
    performed using `Postl's variance of line
-   sums <http://www.leptonica.com/skew-measurement.html>`__ algorithm as
-   implemented in `Leptonica <http://www.leptonica.com/index.html>`__.
+   sums <http://www.leptonica.org/skew-measurement.html>`__ algorithm as
+   implemented in `Leptonica <http://www.leptonica.org/index.html>`__.
 -  ``--clean`` uses
    `unpaper <https://www.flameeyes.eu/projects/unpaper>`__ to clean up
    pages before OCR, but does not alter the final output. This makes it
@@ -216,6 +216,15 @@ processing or PDF/A conversion.
 
     ocrmypdf --tesseract-timeout=0 --remove-background input.pdf output.pdf
 
+Optimize images without performing OCR
+--------------------------------------
+
+You can also optimize all images without performing any OCR:
+
+.. code-block:: bash
+
+    ocrmypdf --tesseract-timeout=0 --optimize 3 --skip-text input.pdf output.pdf
+
 Redo existing OCR
 =================
 

docs/docker.rst

@@ -22,21 +22,20 @@ Installing the Docker image
 If you have `Docker <https://docs.docker.com/>`__ installed on your
 system, you can install a Docker image of the latest release.
 
-The recommended OCRmyPDF Docker image is currently named
-``ocrmypdf-alpine``:
-
-.. code-block:: bash
-
-   docker pull jbarlow83/ocrmypdf-alpine
-
-Follow the Docker installation instructions for your platform. If you
-can run this command successfully, your system is ready to download and
+If you can run this command successfully, your system is ready to download and
 execute the image:
 
 .. code-block:: bash
 
    docker run hello-world
 
+The recommended OCRmyPDF Docker image is currently named ``ocrmypdf``:
+
+.. code-block:: bash
+
+   docker pull jbarlow83/ocrmypdf
+
+
 OCRmyPDF will use all available CPU cores. By default, the VirtualBox
 machine instance on Windows and macOS has only a single CPU core
 enabled. Use the VirtualBox Manager to determine the name of your Docker
@@ -51,6 +50,9 @@ CPUs:
    docker-machine start "yourVM"
    eval $(docker-machine env "yourVM")
 
+See the Docker documentation for
+`adjusting memory and CPU on other platforms <https://docs.docker.com/config/containers/resource_constraints/>`__.
+
 Using the Docker image on the command line
 ==========================================
 
@@ -63,7 +65,7 @@ To start a Docker container (instance of the image):
 
 .. code-block:: bash
 
-   docker tag jbarlow83/ocrmypdf-alpine ocrmypdf
+   docker tag jbarlow83/ocrmypdf ocrmypdf
    docker run --rm -i ocrmypdf (... all other arguments here...)
 
 For convenience, create a shell alias to hide the Docker command. It is
@@ -103,7 +105,7 @@ on the public one:
 
 .. code-block:: dockerfile
 
-   FROM jbarlow83/ocrmypdf-alpine
+   FROM jbarlow83/ocrmypdf
 
    # Add French
    RUN apk add tesseract-ocr-data-fra
@@ -117,17 +119,16 @@ The OCRmyPDF test suite is installed with image. To run it:
 
 .. code-block:: bash
 
-   docker run --entrypoint python3 jbarlow83/ocrmypdf-alpine setup.py test
+   docker run --entrypoint python3  jbarlow83/ocrmypdf -m pytest
 
 Accessing the shell
 ===================
 
-``bash`` is not installed in the image. To use the busybox shell in the
-Docker image:
+To use the bash shell in the Docker image:
 
 .. code-block:: bash
 
-   docker run -it --entrypoint busybox  jbarlow83/ocrmypdf-alpine sh
+   docker run -it --entrypoint bash  jbarlow83/ocrmypdf
 
 Using the OCRmyPDF web service wrapper
 ======================================
@@ -137,7 +138,12 @@ service. The webservice may be launched as follows:
 
 .. code-block:: bash
 
-   docker run --entrypoint python3 -p 5000:5000 jbarlow83/ocrmypdf-alpine webservice.py
+   docker run --entrypoint python3 -p 5000:5000  jbarlow83/ocrmypdf webservice.py
+
+This will configure the machine to listen on port 5000. On Linux machines
+this is port 5000 of localhost. On macOS or Windows machines running
+Docker, this is port 5000 of the virtual machine that runs your Docker
+images. You can find its IP address using the command ``docker-machine ip``.
 
 Unlike command line usage this program will open a socket and wait for
 connections.
@@ -162,14 +168,3 @@ also licensed in this way.
 
 In addition to the above, please read our
 :ref:`general remarks on using OCRmyPDF as a service <ocr-service>`.
-
-Ubuntu-based Docker image
-=========================
-
-A Ubuntu-based OCRmyPDF image is also available. The main advantage this
-image offers is that it supports manylinux Python wheels (which are not
-supported on Alpine Linux). This may be useful for plugins.
-
-.. code-block:: bash
-
-   docker pull jbarlow83/ocrmypdf

docs/index.rst

@@ -22,11 +22,16 @@ PDF is the best format for storing and exchanging scanned documents.  Unfortunat
    cookbook
    docker
    advanced
-   api
    batch
    security
    errors
 
+.. toctree::
+   :caption: Developers
+   :maxdepth: 2
+
+   api
+   contributing
 
 Indices and tables
 ==================

docs/installation.rst

@@ -21,7 +21,7 @@ installing the Python binary wheels.
 Installing on Linux
 ===================
 
-Debian and Ubuntu 16.10 or newer
+Debian and Ubuntu 18.04 or newer
 --------------------------------
 
 .. |deb-stable| image:: https://repology.org/badge/version-for-repo/debian_stable/ocrmypdf.svg
@@ -33,27 +33,29 @@ Debian and Ubuntu 16.10 or newer
 .. |deb-unstable| image:: https://repology.org/badge/version-for-repo/debian_unstable/ocrmypdf.svg
     :alt: Debian unstable
 
-.. |ubu-1710| image:: https://repology.org/badge/version-for-repo/ubuntu_17_10/ocrmypdf.svg
-    :alt: Ubuntu 17.10
-
 .. |ubu-1804| image:: https://repology.org/badge/version-for-repo/ubuntu_18_04/ocrmypdf.svg
     :alt: Ubuntu 18.04 LTS
 
 .. |ubu-1810| image:: https://repology.org/badge/version-for-repo/ubuntu_18_10/ocrmypdf.svg
     :alt: Ubuntu 18.10
 
+.. |ubu-1904| image:: https://repology.org/badge/version-for-repo/ubuntu_19_04/ocrmypdf.svg
+    :alt: Ubuntu 19.04
 
-+-------------------------------------------+
-| **OCRmyPDF versions in Debian & Ubuntu**  |
-+-------------------------------------------+
-| |latest|                                  |
-+-------------------------------------------+
-| |deb-stable| |deb-testing| |deb-unstable| |
-+-------------------------------------------+
-| |ubu-1710| |ubu-1804| |ubu-1810|          |
-+-------------------------------------------+
+.. |ubu-1910| image:: https://repology.org/badge/version-for-repo/ubuntu_19_10/ocrmypdf.svg
+    :alt: Ubuntu 19.10
 
-Users of Debian 9 ("stretch") or later or Ubuntu 16.10 or later may
++-----------------------------------------------+
+| **OCRmyPDF versions in Debian & Ubuntu**      |
++-----------------------------------------------+
+| |latest|                                      |
++-----------------------------------------------+
+| |deb-stable| |deb-testing| |deb-unstable|     |
++-----------------------------------------------+
+| |ubu-1804| |ubu-1810| |ubu-1904| |ubu-1910|   |
++-----------------------------------------------+
+
+Users of Debian 9 ("stretch") or later or Ubuntu 18.04 or later may
 simply
 
 .. code-block:: bash
@@ -64,7 +66,8 @@ As indicated in the table above, Debian and Ubuntu releases may lag
 behind the latest version. If the version available for your platform is
 out of date, you could opt to install the latest version from source.
 See `Installing HEAD revision from
-sources <#installing-head-revision-from-sources>`__.
+sources <#installing-head-revision-from-sources>`__. Ubuntu 16.10 to 17.10
+inclusive also had ocrmypdf, but these versions are end of life.
 
 For full details on version availability for your platform, check the
 `Debian Package Tracker <https://tracker.debian.org/pkg/ocrmypdf>`__ or
@@ -81,20 +84,22 @@ For full details on version availability for your platform, check the
 Fedora 29 or newer
 ------------------
 
-.. |fedora-29| image:: https://repology.org/badge/version-for-repo/fedora29/ocrmypdf.svg
+.. |fedora-29| image:: https://repology.org/badge/version-for-repo/fedora_29/ocrmypdf.svg
     :alt: Fedora 29
 
+.. |fedora-30| image:: https://repology.org/badge/version-for-repo/fedora_30/ocrmypdf.svg
+    :alt: Fedora 30
+
 .. |fedora-rawhide| image:: https://repology.org/badge/version-for-repo/fedora_rawhide/ocrmypdf.svg
     :alt: Fedore Rawhide
 
-
-+------------------------------+
-| **OCRmyPDF version**         |
-+------------------------------+
-| |latest|                     |
-+------------------------------+
-| |fedora-29| |fedora-rawhide| |
-+------------------------------+
++-----------------------------------------------+
+| **OCRmyPDF version**                          |
++-----------------------------------------------+
+| |latest|                                      |
++-----------------------------------------------+
+| |fedora-29| |fedora-30| |fedora-rawhide|      |
++-----------------------------------------------+
 
 Users of Fedora 29 later may simply
 
@@ -267,15 +272,11 @@ Now we need to install ``pip`` and let it install ocrmypdf:
     curl https://bootstrap.pypa.io/ez_setup.py -o - | python3.6 && python3.6 -m easy_install pip
     pip3.6 install ocrmypdf
 
-These installation instructions omit the optional dependency
-``unpaper``, which is only available at version 0.4.2 in Ubuntu 14.04.
-The author could not find a backport of ``unpaper``, and created a .deb
-package to do the job of installing unpaper 6.1 (for x86 64-bit only):
-
-.. code-block:: bash
-
-    wget -q 'https://www.dropbox.com/s/vaq0kbwi6e6au80/unpaper_6.1-1.deb?raw=1' -O unpaper_6.1-1.deb
-    sudo dpkg -i unpaper_6.1-1.deb
+The optional dependency ``unpaper`` is only available at 0.4.2 in Ubuntu 14.04,
+and no backports are available. Previously the author maintained a backported
+.deb package for unpaper 6.1, but since Ubuntu 14.04 is now end of life, this is
+not supported. As such, ``unpaper`` is not available on Ubuntu 14.04 or must by
+compiled by hand.
 
 To add JBIG2 encoding, see :ref:`jbig2`.
 
@@ -403,14 +404,19 @@ The command line program should now be available:
 Installing on FreeBSD
 =====================
 
-FreeBSD 11.2 is known to work. Other versions likely work but have not
-been tested.
+.. image:: https://repology.org/badge/version-for-repo/freebsd/python:ocrmypdf.svg
+    :alt: FreeBSD
+    :target: https://repology.org/project/python:ocrmypdf/versions
 
-In general it should work to:
+FreeBSD 11.2, 11.3, 12.0-RELEASE and 13.0-CURRENT are supported. Other
+versions likely work but have not been tested.
 
-#. `Install and build
-   pikepdf <https://pikepdf.readthedocs.io/en/latest/installation.html#installing-on-freebsd-11-2>`__.
-#. Install the equivalent list of dependencies for Linux.
+.. code-block:: bash
+
+    pkg install py36-ocrmypdf
+
+To install a more recent version, you could attempt to first install the system
+version with ``pkg``, then use ``pip install --user ocrmypdf``.
 
 Installing the Docker image
 ===========================
@@ -503,11 +509,11 @@ manager. ``pip`` cannot provide them.
 -  Python 3.6 or newer
 -  Ghostscript 9.15 or newer
 -  qpdf 8.1.0 or newer
--  Tesseract 4.0.0-alpha or newer
+-  Tesseract 4.0.0-beta or newer
 
 As of ocrmypdf 7.2.1, the following versions are recommended:
 
--  Python 3.7
+-  Python 3.7 or 3.8
 -  Ghostscript 9.23 or newer
 -  qpdf 8.2.1
 -  Tesseract 4.0.0 or newer

docs/introduction.rst

@@ -208,7 +208,7 @@ consider one of these similar open source programs:
 Web front-ends
 ==============
 
-The Docker image ``ocrmypdf-alpine`` provides a web service front-end
+The Docker image ``ocrmypdf`` provides a web service front-end
 that allows files to submitted over HTTP and the results "downloaded".
 This is an HTTP server intended to simplify web services deployments; it
 is not intended to be deployed on the public internet and no real

docs/release_notes.rst

@@ -13,6 +13,64 @@ Note that it is licensed under GPLv3, so scripts that
 ``import ocrmypdf`` and are released publicly should probably also be
 licensed under GPLv3.
 
+v9.0.5
+======
+
+-  The Alpine Docker image (jbarlow83/ocrmypdf-alpine) has been dropped due to
+   the difficulties of supporting Alpine Linux.
+-  The primary Docker image (jbarlow83/ocrmypdf) has been improved to take on
+   the extra features that used to be exclusive to the Alpine image.
+-  No changes to application code.
+-  pdfminer.six version 20191020 is now supported.
+
+v9.0.4
+======
+
+-  Fixed compatibility with Python 3.8 (but requires source install for the moment).
+-  Fixed Tesseract settings for ``--user-words`` and ``--user-patterns``.
+-  Changed to pikepdf 1.6.5 (for Python 3.8).
+-  Changed to Pillow 6.2.0 (to mitigate a security vulnerability in earlier Pillow).
+-  A debug message now mentions when English is automatically selected if the locale
+   is not English.
+
+v9.0.3
+======
+
+-  Embed an encoded version of the sRGB ICC profile in the intermediate
+   Postscript file (used for PDF/A conversion). Previously we included the
+   filename, which required Postscript to run with file access enabled. For
+   security, Ghostscript 9.28 enables ``-dSAFER`` and as such, no longer
+   permits access to any file by default. This fix is necessary for
+   compatibility with Ghostscript 9.28.
+-  Exclude a test that sometimes times out and fails in continuous integration
+   from the standard test suite.
+
+v9.0.2
+======
+
+-  The image optimizer now skips optimizing flate (PNG) encoded images in some
+   situations where the optimization effort was likely wasted.
+-  The image optimizer now ignores images that specify arbitrary decode arrays,
+   since these are rare.
+-  Fixed an issue that caused inversion of black and white in monochrome images.
+   We are not certain but the problem seems to be linked to Leptonica 1.76.0 and
+   older.
+-  Fixed some cases where the test suite failed if
+   English or German Tesseract language packs were not installed.
+-  Fixed a runtime error if the Tesseract English language is not installed.
+-  Improved explicit closing of Pillow images after use.
+-  Actually fixed of Alpine Docker image build.
+-  Changed to pikepdf 1.6.3.
+
+v9.0.1
+======
+
+-  Fixed test suite failing when either of optional dependencies unpaper and
+   pngquant were missing.
+-  Attempted fix of Alpine Docker image build.
+-  Documented that FreeBSD ports are now available.
+-  Changed to pikepdf 1.6.1.
+
 v9.0.0
 ======
 
@@ -59,6 +117,8 @@ v9.0.0
 -  Updated Docker images to use newer versions.
 -  Fixed images encoded as JBIG2 with a colorspace other than ``/DeviceGray``
    were not interpreted correctly.
+-  Fixed a OCR text-image registration (i.e. alignment) problem when the page
+   when MediaBox had a nonzero corner.
 
 v8.3.2
 ======

pyproject.toml

@@ -10,7 +10,7 @@ build-backend = "setuptools.build_meta"
 
 [tool.black]
 line-length = 88
-py36 = true
+target-version = ["py36", "py37", "py38"]
 skip-string-normalization = true
 include = '\.pyi?$'
 exclude = '''

requirements/main.txt

@@ -1,13 +1,10 @@
 # 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
-chardet == 3.0.4
-cffi == 1.12.2
+cffi == 1.13.2
 img2pdf == 0.3.3
-pdfminer.six == 20181108
-pikepdf == 1.6.0
-Pillow >= 5.0.0, != 5.1.0 ; sys_platform == "darwin"
-pycparser == 2.19
-python-xmp-toolkit == 2.0.1
-reportlab == 3.5.13
-tqdm == 4.32.1
+pdfminer.six == 20191020
+pikepdf == 1.6.5
+Pillow >= 6.2.0
+reportlab == 3.5.32
+tqdm == 4.37.0

requirements/test.txt

@@ -1,8 +1,8 @@
-pytest >= 4.4.1, < 5
+pytest >= 5.0.0
 pytest-helpers-namespace >= 2019.1.8
-pytest-xdist == 1.28.0
+pytest-xdist >= 1.29.0  # For DumpError fix
 pytest-cov >= 2.6.1
-python-xmp-toolkit   # requires apt-get install libexempi3
+python-xmp-toolkit == 2.0.1  # requires apt-get install libexempi3
                      # or brew install exempi
 PyPDF2 >= 1.26.0
 #PyMuPDF == 1.13.4    # optional

requirements/webservice.txt

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

setup.py

@@ -68,6 +68,7 @@ setup(
     classifiers=[
         "Programming Language :: Python :: 3.6",
         "Programming Language :: Python :: 3.7",
+        "Programming Language :: Python :: 3.8",
         "Development Status :: 5 - Production/Stable",
         "Environment :: Console",
         "Intended Audience :: End Users/Desktop",
@@ -95,11 +96,9 @@ setup(
         'chardet >= 3.0.4, < 4',  # unlisted requirement of pdfminer.six 20181108
         'cffi >= 1.9.1',  # must be a setup and install requirement
         'img2pdf >= 0.3.0, < 0.4',  # pure Python, so track HEAD closely
-        'pdfminer.six == 20181108',
-        'pikepdf >= 1.6.0, < 2',
-        'Pillow >= 4.0.0, != 5.1.0 ; sys_platform == "darwin"',
-        # Pillow < 4 has BytesIO/TIFF bug w/img2pdf 0.2.3
-        # block 5.1.0, broken wheels
+        'pdfminer.six >= 20181108, <= 20191020',
+        'pikepdf >= 1.6.5, < 2',
+        'Pillow >= 6.2.0',
         'reportlab >= 3.3.0',  # oldest released version with sane image handling
         'tqdm >= 4',
     ],

src/ocrmypdf/_pipeline.py

@@ -54,9 +54,9 @@ def triage_image_file(input_file, output_file, options, log):
         # Recover the original filename
         log.error(str(e).replace(input_file, options.input_file))
         raise UnsupportedImageFormatError() from e
-    else:
-        log.info("Input file is an image")
 
+    with im:
+        log.info("Input file is an image")
         if 'dpi' in im.info:
             if im.info['dpi'] <= (96, 96) and not options.image_dpi:
                 log.info("Image size: (%d, %d)" % im.size)
@@ -89,7 +89,6 @@ def triage_image_file(input_file, output_file, options, log):
             elif im.mode == 'CMYK':
                 log.info('Input CMYK image has no ICC profile, not usable')
                 raise UnsupportedImageFormatError()
-        im.close()
 
     try:
         log.info("Image seems valid. Try converting to PDF...")
@@ -129,7 +128,7 @@ def triage(input_file, output_file, options, log):
         if _pdf_guess_version(input_file):
             if options.image_dpi:
                 log.warning(
-                    "Argument --image-dpi ignored because the "
+                    "Argument --image-dpi is being ignored because the "
                     "input file is a PDF, not an image."
                 )
             # Origin file is a pdf create a symlink with pdf extension

src/ocrmypdf/_sync.py

@@ -189,7 +189,7 @@ def worker_init(queue):
     root.addHandler(h)
 
 
-def worker_thread_init(queue):
+def worker_thread_init(_queue):
     pass
 
 
@@ -225,15 +225,15 @@ def exec_concurrent(context):
     if max_workers > 1:
         context.log.info("Start processing %d pages concurrent", max_workers)
 
-    # Tesseract 4.0 is multithreaded, and we also run multiple workers. We want to
-    # avoid the situation where we end up trying to run NxN jobs on N CPU cores,
-    # as that gives poor performance. Performance testing shows we're better off
+    # Tesseract 4.x can be multithreaded, and we also run multiple workers. We want
+    # to manage how many threads it uses to avoid creating total threads than cores.
+    # Performance testing shows we're better off
     # parallelizing ocrmypdf and forcing Tesseract to be single threaded, which we
     # get by setting the envvar OMP_THREAD_LIMIT to 1. But if the page count of the
     # input file is small, then we allow Tesseract to use threads, subject to the
-    # constraint: (ocrmypdf workers) * (tesseract threads) <= max_workers and limiting
-    # Tesseract to 4 threads.
-    tess_threads = min(4, context.options.jobs // max_workers)
+    # constraint: (ocrmypdf workers) * (tesseract threads) <= max_workers.
+    # As of Tesseract 4.1, 3 threads is the most effective on a 4 core/8 thread system.
+    tess_threads = min(3, context.options.jobs // max_workers)
     if context.options.tesseract_env is None:
         context.options.tesseract_env = os.environ.copy()
     context.options.tesseract_env.setdefault('OMP_THREAD_LIMIT', str(tess_threads))
@@ -301,7 +301,7 @@ def exec_concurrent(context):
 class NeverRaise(Exception):
     """An exception that is never raised"""
 
-    pass
+    pass  # pylint: disable=unnecessary-pass
 
 
 def run_pipeline(options, api=False):

src/ocrmypdf/_validation.py

@@ -17,6 +17,7 @@
 # along with OCRmyPDF.  If not, see <http://www.gnu.org/licenses/>.
 
 
+import locale
 import logging
 import os
 import sys
@@ -47,6 +48,7 @@ from .helpers import is_file_writable, is_iterable_notstr, monotonic, re_symlink
 # External dependencies
 
 HOCR_OK_LANGS = frozenset(['eng', 'deu', 'spa', 'ita', 'por'])
+DEFAULT_LANGUAGE = 'eng'  # Enforce English hegemony
 
 log = logging.getLogger(__name__)
 
@@ -58,7 +60,10 @@ verify_python3_env()
 
 def check_options_languages(options):
     if not options.language:
-        options.language = ['eng']  # Enforce English hegemony
+        options.language = [DEFAULT_LANGUAGE]
+        system_lang = locale.getlocale()[0]
+        if system_lang and not system_lang.startswith('en'):
+            log.debug("No language specified; assuming --language %s", DEFAULT_LANGUAGE)
 
     # Support v2.x "eng+deu" language syntax
     if '+' in options.language[0]:
@@ -107,7 +112,7 @@ def check_options_output(options):
         options.pdf_renderer = 'sandwich'
 
     if options.pdf_renderer == 'sandwich' and not tesseract.has_textonly_pdf(
-        options.tesseract_env
+        options.tesseract_env, languages
     ):
         raise MissingDependencyError(
             "You are using an alpha version of Tesseract 4.0 that does not support "

src/ocrmypdf/api.py

@@ -33,7 +33,7 @@ class TqdmConsole:
 
     def __init__(self, file):
         self.file = file
-        self.py36 = sys.version_info >= (3, 6)
+        self.py36 = sys.version_info[0:2] == (3, 6)
 
     def write(self, msg):
         # When no progress bar is active, tqdm.write() routes to print()
@@ -122,13 +122,23 @@ def create_options(*, input_file, output_file, **kwargs):
     for arg, val in kwargs.items():
         if val is None:
             continue
-        if arg == 'tesseract_env':
+
+        # These arguments with special handling for which we bypass
+        # argparse
+        if arg in {'tesseract_env', 'progress_bar'}:
             deferred.append((arg, val))
             continue
+
         cmd_style_arg = arg.replace('_', '-')
-        cmdline.append(f"--{cmd_style_arg}")
+
+        # Booleans are special: add only if True, omit for False
         if isinstance(val, bool):
+            if val:
+                cmdline.append(f"--{cmd_style_arg}")
             continue
+
+        # We have a parameter
+        cmdline.append(f"--{cmd_style_arg}")
         if isinstance(val, (int, float)):
             cmdline.append(str(val))
         elif isinstance(val, str):

src/ocrmypdf/exec/ghostscript.py

@@ -259,6 +259,7 @@ def generate_pdfa(
                 "-dQUIET",
                 "-dBATCH",
                 "-dNOPAUSE",
+                "-dSAFER",
                 "-dCompatibilityLevel=" + str(pdf_version),
                 "-sDEVICE=pdfwrite",
                 "-dAutoRotatePages=/None",

src/ocrmypdf/exec/pngquant.py

@@ -40,8 +40,7 @@ def available():
 
 def quantize(input_file, output_file, quality_min, quality_max):
     if input_file.endswith('.jpg'):
-        im = Image.open(input_file)
-        with NamedTemporaryFile(suffix='.png') as tmp:
+        with Image.open(input_file) as im, NamedTemporaryFile(suffix='.png') as tmp:
             im.save(tmp)
             args = [
                 'pngquant',

src/ocrmypdf/exec/tesseract.py

@@ -17,10 +17,8 @@
 
 import os
 import shutil
-import sys
 from collections import namedtuple
 from contextlib import suppress
-from functools import lru_cache
 from os import fspath
 from subprocess import PIPE, STDOUT, CalledProcessError, TimeoutExpired, run
 
@@ -61,13 +59,13 @@ def v4(tesseract_env=None):
     return version(tesseract_env) >= '4'
 
 
-def has_textonly_pdf(tesseract_env=None):
+def has_textonly_pdf(tesseract_env=None, langs=None):
     """Does Tesseract have textonly_pdf capability?
 
     Available in v4.00.00alpha since January 2017. Best to
-    parse the parameter list
+    parse the parameter list.
     """
-    args_tess = ['tesseract', '--print-parameters', 'pdf']
+    args_tess = tess_base_args(langs, engine_mode=None) + ['--print-parameters', 'pdf']
     params = ''
     try:
         proc = run(
@@ -233,8 +231,8 @@ def _generate_null_hocr(output_hocr, output_sidecar, image):
     the same size as the input image."""
     from PIL import Image
 
-    im = Image.open(image)
-    w, h = im.size
+    with Image.open(image) as im:
+        w, h = im.size
 
     with open(output_hocr, 'w', encoding="utf-8") as f:
         f.write(HOCR_TEMPLATE.format(w, h))
@@ -271,6 +269,9 @@ def generate_hocr(
     if user_patterns:
         args_tesseract.extend(['--user-patterns', user_patterns])
 
+    if user_words or user_patterns:
+        args_tesseract.extend(['-c', 'lstm_use_matrix=1'])
+
     # Reminder: test suite tesseract spoofers will break after any changes
     # to the number of order parameters here
     args_tesseract.extend([input_file, prefix, 'hocr', 'txt'] + tessconfig)
@@ -358,7 +359,7 @@ def generate_pdf(
     if pagesegmode is not None:
         args_tesseract.extend(['--psm', str(pagesegmode)])
 
-    if text_only and has_textonly_pdf(tesseract_env):
+    if text_only and has_textonly_pdf(tesseract_env, language):
         args_tesseract.extend(['-c', 'textonly_pdf=1'])
 
     if user_words:

src/ocrmypdf/exec/unpaper.py

@@ -42,33 +42,30 @@ def run(input_file, output_file, dpi, log, mode_args):
 
     SUFFIXES = {'1': '.pbm', 'L': '.pgm', 'RGB': '.ppm'}
 
-    im = Image.open(input_file)
-    if im.mode not in SUFFIXES.keys():
-        log.info("Converting image to other colorspace")
+    with TemporaryDirectory() as tmpdir, Image.open(input_file) as im:
+        if im.mode not in SUFFIXES.keys():
+            log.info("Converting image to other colorspace")
+            try:
+                if im.mode == 'P' and len(im.getcolors()) == 2:
+                    im = im.convert(mode='1')
+                else:
+                    im = im.convert(mode='RGB')
+            except IOError as e:
+                im.close()
+                raise MissingDependencyError(
+                    "Could not convert image with type " + im.mode
+                ) from e
+
         try:
-            if im.mode == 'P' and len(im.getcolors()) == 2:
-                im = im.convert(mode='1')
-            else:
-                im = im.convert(mode='RGB')
-        except IOError as e:
-            im.close()
+            suffix = SUFFIXES[im.mode]
+        except KeyError:
             raise MissingDependencyError(
-                "Could not convert image with type " + im.mode
+                "Failed to convert image to a supported format."
             ) from e
 
-    try:
-        suffix = SUFFIXES[im.mode]
-    except KeyError:
-        im.close()
-        raise MissingDependencyError(
-            "Failed to convert image to a supported format."
-        ) from e
-
-    with TemporaryDirectory() as tmpdir:
         input_pnm = os.path.join(tmpdir, f'input{suffix}')
         output_pnm = os.path.join(tmpdir, f'output{suffix}')
         im.save(input_pnm, format='PPM')
-        im.close()
 
         # To prevent any shenanigans from accepting arbitrary parameters in
         # --unpaper-args, we:
@@ -95,10 +92,12 @@ def run(input_file, output_file, dpi, log, mode_args):
             log.debug(proc.stdout)
             # unpaper sets dpi to 72; fix this
             try:
-                Image.open(output_pnm).save(output_file, dpi=(dpi, dpi))
+                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: "
+                    "unpaper: failed to produce the expected output file. "
+                    + " Called with: "
                     + str(args_unpaper)
                 ) from None
 

src/ocrmypdf/optimize.py

@@ -73,6 +73,9 @@ def extract_image_filter(pike, root, log, image, xref):
     if filtdp[0] == Name.JPXDecode:
         return None  # Don't do JPEG2000
 
+    if Name.Decode in image:
+        return None  # Don't mess with custom Decode tables
+
     return pim, filtdp
 
 
@@ -104,6 +107,15 @@ def extract_image_generic(*, pike, root, log, image, xref, options):
         return None
     pim, filtdp = result
 
+    # Don't try to PNG-optimize 1bpp images, since JBIG2 does it better.
+    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,
@@ -343,6 +355,7 @@ def transcode_jpegs(pike, jpegs, root, log, options):
 
 
 def transcode_pngs(pike, images, image_name_fn, root, log, options):
+    modified = set()
     if options.optimize >= 2:
         png_quality = (
             max(10, options.png_quality - 10),
@@ -363,6 +376,7 @@ def transcode_pngs(pike, images, image_name_fn, root, log, options):
                         png_quality[1],
                     )
                 )
+                modified.add(xref)
             with tqdm(
                 desc="PNGs",
                 total=len(futures),
@@ -372,10 +386,14 @@ def transcode_pngs(pike, images, image_name_fn, root, log, options):
                 for _future in concurrent.futures.as_completed(futures):
                     pbar.update()
 
-    for xref in images:
+    for xref in modified:
         im_obj = pike.get_object(xref, 0)
         try:
-            compdata = leptonica.CompressedData.open(png_name(root, xref))
+            pix = leptonica.Pix.open(png_name(root, xref))
+            if pix.mode == '1':
+                compdata = pix.generate_pdf_ci_data(leptonica.lept.L_G4_ENCODE, 0)
+            else:
+                compdata = leptonica.CompressedData.open(png_name(root, xref))
         except leptonica.LeptonicaError as e:
             # Most likely this means file not found, i.e. quantize did not
             # produce an improved version
@@ -391,62 +409,83 @@ def transcode_pngs(pike, images, image_name_fn, root, log, options):
                 f"{len(compdata)} > {int(im_obj.stream_dict.Length)}"
             )
             continue
+        if compdata.type == leptonica.lept.L_FLATE_ENCODE:
+            return rewrite_png(pike, im_obj, compdata, log)
+        elif compdata.type == leptonica.lept.L_G4_ENCODE:
+            return rewrite_png_as_g4(pike, im_obj, compdata, log)
 
-        # 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
-        # From there the PNG decoder can infer the rest from the file.
-        # In practice the predictor should be Paeth, 14, so we'll use that.
-        # See:
-        #   - PDF RM 7.4.4.4 Table 10
-        #   - https://github.com/DanBloomberg/leptonica/blob/master/src/pdfio2.c#L757
-        predictor = 14 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
+def rewrite_png_as_g4(pike, im_obj, compdata, log):
+    im_obj.BitsPerComponent = 1
+    im_obj.Width = compdata.w
+    im_obj.Height = compdata.h
 
-        if compdata.ncolors > 0:
-            # .ncolors is the number of colors in the palette, not the number of
-            # colors used in a true color image
-            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:
-            if compdata.spp == 1:
-                # PDF interprets binary-1 as black in 1bpp, but PNG sets
-                # black to 0 for 1bpp. Create a palette that informs the PDF
-                # of the mapping - seems cleaner to go this way but pikepdf
-                # needs to be patched to support it.
-                # palette = [Name.Indexed, Name.DeviceGray, 1, b"\xff\x00"]
-                # cs = palette
-                cs = Name.DeviceGray
-            elif compdata.spp == 3:
-                cs = Name.DeviceRGB
-            elif compdata.spp == 4:
-                cs = Name.DeviceCMYK
-        if compdata.bps == 1:
-            im_obj.Decode = [1, 0]  # Bit of a kludge but this inverts photometric too
-        im_obj.ColorSpace = cs
-        im_obj.write(compdata.read(), filter=Name.FlateDecode, decode_parms=dparms)
+    im_obj.write(compdata.read())
+
+    log.debug(f"PNG to G4 {im_obj.objgen}")
+    if Name.Predictor in im_obj:
+        del im_obj.Predictor
+    if Name.DecodeParms in im_obj:
+        del im_obj.DecodeParms
+    im_obj.DecodeParms = Dictionary(
+        K=-1, BlackIs1=bool(compdata.minisblack), Columns=compdata.w
+    )
+
+    im_obj.Filter = Name.CCITTFaxDecode
+    return
+
+
+def rewrite_png(pike, im_obj, compdata, log):
+    # 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, output_file, context, save_settings):
@@ -510,11 +549,11 @@ def main(infile, outfile, level, jobs=1):
         """Emulate ocrmypdf's options"""
 
         def __init__(
-            self, input_file, jobs, optimize, jpeg_quality, png_quality, jb2lossy
+            self, input_file, jobs, optimize_, jpeg_quality, png_quality, jb2lossy
         ):
             self.input_file = input_file
             self.jobs = jobs
-            self.optimize = optimize
+            self.optimize = optimize_
             self.jpeg_quality = jpeg_quality
             self.png_quality = png_quality
             self.jbig2_page_group_size = 0
@@ -525,7 +564,7 @@ def main(infile, outfile, level, jobs=1):
     options = OptimizeOptions(
         input_file=infile,
         jobs=jobs,
-        optimize=int(level),
+        optimize_=int(level),
         jpeg_quality=0,  # Use default
         png_quality=0,
         jb2lossy=False,

src/ocrmypdf/pdfa.py

@@ -31,6 +31,7 @@ Ghostscript's handling of pdfmark.
 
 """
 
+import base64
 import os
 from binascii import hexlify
 from pathlib import Path
@@ -48,26 +49,13 @@ SRGB_ICC_PROFILE = pkg_resources.resource_filename('ocrmypdf', ICC_PROFILE_RELPA
 # files, from the Ghostscript documentation. Lines beginning with % are
 # comments. Python substitution variables have a '$' prefix.
 pdfa_def_template = u"""%!
-% Define entries in the document Info dictionary :
+% Define an ICC profile :
 /ICCProfile $icc_profile
 def
 
-% Define an ICC profile :
-
 [/_objdef {icc_PDFA} /type /stream /OBJ pdfmark
-[{icc_PDFA}
-<<
-  /N currentpagedevice /ProcessColorModel known {
-    currentpagedevice /ProcessColorModel get dup /DeviceGray eq
-    {pop 1} {
-      /DeviceRGB eq
-      {3}{4} ifelse
-    } ifelse
-  } {
-    (ERROR, unable to determine ProcessColorModel) == flush
-  } ifelse
->> /PUT pdfmark
-[{icc_PDFA} ICCProfile (r) file /PUT pdfmark
+[{icc_PDFA} << /N 3 >> /PUT pdfmark
+[{icc_PDFA} ICCProfile /PUT pdfmark
 
 % Define the output intent dictionary :
 
@@ -104,16 +92,10 @@ def generate_pdfa_ps(target_filename, icc='sRGB'):
     else:
         raise NotImplementedError("Only supporting sRGB")
 
-    # pdfmark must contain the full path to the ICC profile, and pdfmark must be
-    # also encoded in ASCII. ocrmypdf can be installed anywhere, including to
-    # paths that have a non-ASCII character in the filename. Ghostscript
-    # accepts hex-encoded strings and converts them to byte strings, so
-    # we encode the path with fsencode() and use the hex representation.
-    # UTF-16 not accepted here. (Even though ASCII encodable is the usual case,
-    # do this always to avoid making it a rare conditional.)
-    bytes_icc_profile = os.fsencode(icc_profile)
-    hex_icc_profile = hexlify(bytes_icc_profile)
-    icc_profile = '<' + hex_icc_profile.decode('ascii') + '>'
+    # Read the ICC profile, encode as ASCII85 and convert to a string which we
+    # will insert in the .ps file
+    bytes_icc_profile = Path(icc_profile).read_bytes()
+    icc_profile = base64.a85encode(bytes_icc_profile, adobe=True).decode('ascii')
 
     t = Template(pdfa_def_template)
     ps = t.substitute(icc_profile=icc_profile, icc_identifier=icc)

src/ocrmypdf/pdfinfo/layout.py

@@ -20,6 +20,7 @@ from math import copysign
 from pathlib import Path
 from unittest.mock import patch
 
+import pdfminer
 import pdfminer.encodingdb
 import pdfminer.pdfdevice
 import pdfminer.pdfinterp
@@ -36,51 +37,54 @@ from ..exceptions import EncryptedPdfError
 STRIP_NAME = re.compile(r'[0-9]+')
 
 #
-# Unconditional pdfminer patches
+#  pdfminer 20181108 patches
 #
 
+if pdfminer.__version__ == '20181108':
 
-def name2unicode(name):
-    """Fix pdfminer's name2unicode function
+    def name2unicode(name):
+        """Fix pdfminer's name2unicode function
 
-    Font cids that are mapped to names of the form /g123 seem to be, by convention
-    characters with no corresponding Unicode entry. These can be subsetted fonts
-    or symbolic fonts. There seems to be no way to map /g123 fonts to Unicode,
-    barring a ToUnicode data structure.
-    """
-    if name in glyphname2unicode:
-        return glyphname2unicode[name]
-    if name.startswith('g') or name.startswith('a'):
-        raise KeyError(name)
-    if name.startswith('uni'):
-        try:
-            return chr(int(name[3:], 16))
-        except ValueError:  # Not hexadecimal
+        Font cids that are mapped to names of the form /g123 seem to be, by convention
+        characters with no corresponding Unicode entry. These can be subsetted fonts
+        or symbolic fonts. There seems to be no way to map /g123 fonts to Unicode,
+        barring a ToUnicode data structure.
+        """
+        if name in glyphname2unicode:
+            return glyphname2unicode[name]
+        if name.startswith('g') or name.startswith('a'):
             raise KeyError(name)
-    m = STRIP_NAME.search(name)
-    if not m:
-        raise KeyError(name)
-    return chr(int(m.group(0)))
+        if name.startswith('uni'):
+            try:
+                return chr(int(name[3:], 16))
+            except ValueError:  # Not hexadecimal
+                raise KeyError(name)
+        m = STRIP_NAME.search(name)
+        if not m:
+            raise KeyError(name)
+        return chr(int(m.group(0)))
 
+    pdfminer.encodingdb.name2unicode = name2unicode
 
-pdfminer.encodingdb.name2unicode = name2unicode
+    original_PDFFont_init = PDFFont.__init__
 
-original_PDFFont_init = PDFFont.__init__
+    def PDFFont__init__(self, descriptor, widths, default_width=None):
+        original_PDFFont_init(self, descriptor, widths, default_width)
+        # PDF spec says descent should be negative
+        # A font with a positive descent implies it floats entirely above the
+        # baseline, i.e. it's not really a baseline anymore. I have fonts that
+        # claim a positive descent, but treating descent as positive always seems
+        # to misposition text.
+        if self.descent > 0:
+            self.descent = -self.descent
 
+    PDFFont.__init__ = PDFFont__init__
 
-def PDFFont__init__(self, descriptor, widths, default_width=None):
-    original_PDFFont_init(self, descriptor, widths, default_width)
-    # PDF spec says descent should be negative
-    # A font with a positive descent implies it floats entirely above the
-    # baseline, i.e. it's not really a baseline anymore. I have fonts that
-    # claim a positive descent, but treating descent as positive always seems
-    # to misposition text.
-    if self.descent > 0:
-        self.descent = -self.descent
+#
+# end of pdfminer 20181108 patches
+#
 
 
-PDFFont.__init__ = PDFFont__init__
-
 original_PDFSimpleFont_init = PDFSimpleFont.__init__
 
 
@@ -97,6 +101,7 @@ def PDFSimpleFont__init__(self, descriptor, widths, spec):
 
 
 PDFSimpleFont.__init__ = PDFSimpleFont__init__
+
 #
 # pdfminer patches when creator is PScript5.dll
 #

tests/conftest.py

@@ -176,10 +176,9 @@ def no_outpdf(tmp_path):
 def check_ocrmypdf(input_file, output_file, *args, env=None):
     """Run ocrmypdf and confirmed that a valid file was created"""
 
-    # p, out, err = run_ocrmypdf(input_file, output_file, *args, env=env)
-
     options = cli.parser.parse_args(
-        [str(input_file), str(output_file)] + [str(arg) for arg in args]
+        [str(input_file), str(output_file)]
+        + [str(arg) for arg in args if arg is not None]
     )
     api.check_options(options)
     if env:

tests/spoof/tesseract_badutf8.py

@@ -52,7 +52,7 @@ def main():
     elif sys.argv[1] == '--list-langs':
         print('List of available languages (1):\neng', file=sys.stderr)
         sys.exit(0)
-    elif sys.argv[1] == '--print-parameters':
+    elif sys.argv[-2] == '--print-parameters':
         print("Some parameters", file=sys.stderr)
         print("textonly_pdf\t1\tSome help text")
         sys.exit(0)

tests/spoof/tesseract_big_image_error.py

@@ -44,7 +44,7 @@ def main():
     elif sys.argv[1] == '--list-langs':
         print('List of available languages (1):\neng\n', file=sys.stderr)
         sys.exit(0)
-    elif sys.argv[1] == '--print-parameters':
+    elif sys.argv[-2] == '--print-parameters':
         print('A parameter list would go here\ntextonly_pdf 0\n', file=sys.stderr)
         sys.exit(0)
     elif sys.argv[-2] == 'hocr':

tests/spoof/tesseract_cache.py

@@ -100,6 +100,8 @@ def main():
     # Convert non-standard but supported -psm to --psm
     sys.argv = ['--psm' if arg == '-psm' else arg for arg in sys.argv]
 
+    if '_OCRMYPDF_TEST_INFILE' not in os.environ:
+        real_tesseract()  # test not properly set up
     source = os.environ['_OCRMYPDF_TEST_INFILE']  # required
     args = parser.parse_args()
 

tests/spoof/tesseract_crash.py

@@ -50,7 +50,7 @@ def main():
     elif sys.argv[1] == '--list-langs':
         print('List of available languages (1):\neng', file=sys.stderr)
         sys.exit(0)
-    elif sys.argv[1] == '--print-parameters':
+    elif sys.argv[-2] == '--print-parameters':
         print('A parameter list would go here\ntextonly_pdf 0\n', file=sys.stderr)
         sys.exit(0)
     elif sys.argv[-2] == 'hocr':

tests/spoof/tesseract_noop.py

@@ -76,7 +76,7 @@ def main():
     elif sys.argv[1] == '--list-langs':
         print('List of available languages (1):\neng', file=sys.stderr)
         sys.exit(0)
-    elif sys.argv[1] == '--print-parameters':
+    elif sys.argv[-2] == '--print-parameters':
         print("Some parameters", file=sys.stderr)
         print("textonly_pdf\t1\tSome help text")
         sys.exit(0)

tests/test_completion.py

@@ -19,6 +19,11 @@ from subprocess import run, PIPE
 
 import pytest
 
+pytestmark = pytest.mark.skipif(
+    pytest.helpers.running_in_docker(),  # pylint: disable=no-member
+    reason="docker can't complete",
+)
+
 
 def test_fish():
     try:

tests/test_lept.py

@@ -38,7 +38,8 @@ def test_colormap_backgroundnorm(resources):
 def crom_pix(resources):
     pix = lept.Pix.open(resources / 'crom.png')
     im = Image.open(resources / 'crom.png')
-    return pix, im
+    yield pix, im
+    im.close()
 
 
 def test_pix_basic(crom_pix):

tests/test_main.py

@@ -1,4 +1,4 @@
-# © 2015-17 James R. Barlow: github.com/jbarlow83
+# © 2015-19 James R. Barlow: github.com/jbarlow83
 #
 # This file is part of OCRmyPDF.
 #
@@ -18,10 +18,9 @@
 import logging
 import os
 import shutil
-import sys
 from math import isclose
 from pathlib import Path
-from subprocess import DEVNULL, PIPE, run, Popen
+from subprocess import PIPE, run
 
 import PIL
 import pytest
@@ -84,11 +83,6 @@ def spoof_no_tess_gs_raster_fail(tmp_path_factory):
     )
 
 
-@pytest.fixture(scope='session')
-def spoof_tess_bad_utf8(tmp_path_factory):
-    return spoof(tmp_path_factory, tesseract='tesseract_badutf8.py')
-
-
 def test_quick(spoof_tesseract_cache, resources, outpdf):
     check_ocrmypdf(resources / 'ccitt.pdf', outpdf, env=spoof_tesseract_cache)
 
@@ -124,8 +118,8 @@ def test_deskew(spoof_tesseract_noop, resources, outdir):
 
 def test_remove_background(spoof_tesseract_noop, resources, outdir):
     # Ensure the input image does not contain pure white/black
-    im = Image.open(resources / 'congress.jpg')
-    assert im.getextrema() != ((0, 255), (0, 255), (0, 255))
+    with Image.open(resources / 'congress.jpg') as im:
+        assert im.getextrema() != ((0, 255), (0, 255), (0, 255))
 
     output_pdf = check_ocrmypdf(
         resources / 'congress.jpg',
@@ -151,8 +145,8 @@ def test_remove_background(spoof_tesseract_noop, resources, outdir):
     )
 
     # The output image should contain pure white and black
-    im = Image.open(output_png)
-    assert im.getextrema() == ((0, 255), (0, 255), (0, 255))
+    with Image.open(output_png) as im:
+        assert im.getextrema() == ((0, 255), (0, 255), (0, 255))
 
 
 # This will run 5 * 2 * 2 = 20 test cases
@@ -223,7 +217,7 @@ def test_skip_ocr(spoof_tesseract_cache, resources, outpdf):
     assert pdfinfo[0].has_text
 
 
-def test_redo_ocr(spoof_tesseract_cache, resources, outpdf):
+def test_redo_ocr(resources, outpdf):
     in_ = resources / 'graph_ocred.pdf'
     before = PdfInfo(in_, detailed_page_analysis=True)
     out = outpdf
@@ -355,10 +349,9 @@ def test_german(spoof_tesseract_cache, resources, outdir):
         sidecar,
         env=spoof_tesseract_cache,
     )
-    print(os.environ)
-    assert (
-        p.returncode == ExitCode.ok
-    ), "This test may fail if Tesseract language packs are missing"
+    if 'deu' not in tesseract.languages():
+        pytest.xfail(reason="tesseract-deu language pack not installed")
+    assert p.returncode == ExitCode.ok, "Requires tesseract deu language pack"
 
 
 def test_klingon(resources, outpdf):
@@ -550,68 +543,6 @@ def test_jbig2_passthrough(spoof_tesseract_cache, resources, outpdf):
     assert out_pageinfo[0].images[0].enc == Encoding.jbig2
 
 
-def test_stdin(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf):
-    input_file = str(resources / 'francais.pdf')
-    output_file = str(outpdf)
-
-    # Runs: ocrmypdf - output.pdf < testfile.pdf
-    with open(input_file, 'rb') as input_stream:
-        p_args = ocrmypdf_exec + ['-', output_file]
-        p = run(
-            p_args,
-            stdout=PIPE,
-            stderr=PIPE,
-            stdin=input_stream,
-            env=spoof_tesseract_noop,
-        )
-        assert p.returncode == ExitCode.ok
-
-
-def test_stdout(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf):
-    input_file = str(resources / 'francais.pdf')
-    output_file = str(outpdf)
-
-    # Runs: ocrmypdf francais.pdf - > test_stdout.pdf
-    with open(output_file, 'wb') as output_stream:
-        p_args = ocrmypdf_exec + [input_file, '-']
-        p = run(
-            p_args,
-            stdout=output_stream,
-            stderr=PIPE,
-            stdin=DEVNULL,
-            env=spoof_tesseract_noop,
-        )
-        assert p.returncode == ExitCode.ok
-
-    assert qpdf.check(output_file, log=None)
-
-
-@pytest.mark.skipif(
-    sys.version_info[0:3] >= (3, 6, 4), reason="issue fixed in Python 3.6.4"
-)
-def test_closed_streams(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf):
-    input_file = str(resources / 'francais.pdf')
-    output_file = str(outpdf)
-
-    def evil_closer():
-        os.close(0)
-        os.close(1)
-
-    p_args = ocrmypdf_exec + [input_file, output_file]
-    p = Popen(  # pylint: disable=subprocess-popen-preexec-fn
-        p_args,
-        close_fds=True,
-        stdout=None,
-        stderr=PIPE,
-        stdin=None,
-        env=spoof_tesseract_noop,
-        preexec_fn=evil_closer,
-    )
-    out, err = p.communicate()
-    print(err.decode())
-    assert p.returncode == ExitCode.ok
-
-
 def test_masks(spoof_tesseract_noop, resources, outpdf):
     assert (
         ocrmypdf.ocr(
@@ -682,6 +613,7 @@ language_model_penalty_non_freq_dict_word 0
     )
 
 
+@pytest.mark.slow  # This test sometimes times out in CI
 @pytest.mark.parametrize('renderer', RENDERERS)
 def test_tesseract_config_notfound(renderer, resources, outdir):
     cfg_file = outdir / 'nofile.cfg'
@@ -698,6 +630,7 @@ def test_tesseract_config_notfound(renderer, resources, outdir):
     assert p.returncode == ExitCode.ok, err
 
 
+@pytest.mark.slow  # This test sometimes times out in CI
 @pytest.mark.parametrize('renderer', RENDERERS)
 def test_tesseract_config_invalid(renderer, resources, outdir):
     cfg_file = outdir / 'test.cfg'
@@ -720,26 +653,11 @@ THIS FILE IS INVALID
     assert p.returncode == ExitCode.invalid_config
 
 
-@pytest.mark.skipif(tesseract.v4(), reason='arg has no effect in 4.0-beta1')
-def test_user_words(resources, outdir):
+@pytest.mark.skipif(not tesseract.has_user_words(), reason='not functional until 4.1.0')
+def test_user_words_ocr(resources, outdir):
+    # Does not actually test if --user-words causes output to differ
     word_list = outdir / 'wordlist.txt'
-    sidecar_before = outdir / 'sidecar_before.txt'
-    sidecar_after = outdir / 'sidecar_after.txt'
-
-    # Don't know how to make this test pass on various versions and platforms
-    # so weaken to merely testing that the argument is accepted
-    consistent = False
-
-    if consistent:
-        check_ocrmypdf(
-            resources / 'crom.png',
-            outdir / 'out.pdf',
-            '--image-dpi',
-            150,
-            '--sidecar',
-            sidecar_before,
-        )
-        assert 'cromulent' not in sidecar_before.open().read()
+    sidecar_after = outdir / 'sidecar.txt'
 
     with word_list.open('w') as f:
         f.write('cromulent\n')  # a perfectly cromulent word
@@ -755,9 +673,6 @@ def test_user_words(resources, outdir):
         word_list,
     )
 
-    if consistent:
-        assert 'cromulent' in sidecar_after.open().read()
-
 
 def test_form_xobject(spoof_tesseract_noop, resources, outpdf):
     check_ocrmypdf(
@@ -878,6 +793,7 @@ def test_compression_preserved(
         assert pdfimage.color == Colorspace.rgb, "Colorspace changed"
     elif im.mode.startswith('L'):
         assert pdfimage.color == Colorspace.gray, "Colorspace changed"
+    im.close()
 
 
 @pytest.mark.parametrize(
@@ -939,6 +855,7 @@ def test_compression_changed(
         assert pdfimage.color == Colorspace.rgb, "Colorspace changed"
     elif im.mode.startswith('L'):
         assert pdfimage.color == Colorspace.gray, "Colorspace changed"
+    im.close()
 
 
 def test_sidecar_pagecount(spoof_tesseract_cache, resources, outpdf):
@@ -993,36 +910,6 @@ def test_pdfa_n(spoof_tesseract_cache, pdfa_level, resources, outpdf):
     assert pdfa_info['conformance'] == f'PDF/A-{pdfa_level}B'
 
 
-@pytest.mark.skipif(sys.version_info >= (3, 7, 0), reason='better utf-8')
-@pytest.mark.skipif(
-    Path('/etc/alpine-release').exists(), reason="invalid test on alpine"
-)
-def test_bad_locale():
-    env = os.environ.copy()
-    env['LC_ALL'] = 'C'
-
-    p, out, err = run_ocrmypdf('a', 'b', env=env)
-    assert out == '', "stdout not clean"
-    assert p.returncode != 0
-    assert 'configured to use ASCII as encoding' in err, "should whine"
-
-
-@pytest.mark.parametrize('renderer', RENDERERS)
-def test_bad_utf8(spoof_tess_bad_utf8, renderer, resources, no_outpdf):
-    p, out, err = run_ocrmypdf(
-        resources / 'ccitt.pdf',
-        no_outpdf,
-        '--pdf-renderer',
-        renderer,
-        env=spoof_tess_bad_utf8,
-    )
-
-    assert out == '', "stdout not clean"
-    assert p.returncode != 0
-    assert 'not utf-8' in err, "should whine about utf-8"
-    assert '\\x96' in err, 'should repeat backslash encoded output'
-
-
 @pytest.mark.skipif(
     PIL.__version__ < '5.0.0', reason="Pillow < 5.0.0 doesn't raise the exception"
 )
@@ -1051,14 +938,6 @@ def test_text_curves(spoof_tesseract_noop, resources, outpdf):
     assert len(info.pages[0].images) != 0, "force did not rasterize"
 
 
-def test_dev_null(spoof_tesseract_noop, resources):
-    p, out, err = run_ocrmypdf(
-        resources / 'trivial.pdf', os.devnull, '--force-ocr', env=spoof_tesseract_noop
-    )
-    assert p.returncode == 0, "could not send output to /dev/null"
-    assert len(out) == 0, "wrote to stdout"
-
-
 def test_output_is_dir(spoof_tesseract_noop, resources, outdir):
     p, out, err = run_ocrmypdf(
         resources / 'trivial.pdf', outdir, '--force-ocr', env=spoof_tesseract_noop
@@ -1125,3 +1004,14 @@ def test_fast_web_view(
     )
     with pikepdf.open(outpdf) as pdf:
         assert pdf.is_linearized == expected
+
+
+def test_image_dpi_not_image(caplog, spoof_tesseract_noop, resources, outpdf):
+    check_ocrmypdf(
+        resources / 'trivial.pdf',
+        outpdf,
+        '--image-dpi',
+        '100',
+        env=spoof_tesseract_noop,
+    )
+    assert '--image-dpi is being ignored' in caplog.text

tests/test_optimize.py

@@ -48,13 +48,14 @@ def test_mono_not_inverted(resources, outdir):
         xres=10,
         yres=10,
         raster_device='pnggray',
-        log=logging.getLogger(name='test_mono_flip'),
+        log=logging.getLogger(name='test_mono_not_inverted'),
     )
 
-    im = Image.open(fspath(outdir / 'im.png'))
-    assert im.getpixel((0, 0)) == 255, "Expected white background"
+    with Image.open(fspath(outdir / 'im.png')) as im:
+        assert im.getpixel((0, 0)) == 255, "Expected white background"
 
 
+@pytest.mark.skipif(not pngquant.available(), reason='need pngquant')
 def test_jpg_png_params(resources, outpdf, spoof_tesseract_noop):
     check_ocrmypdf(
         resources / 'crom.png',
@@ -109,10 +110,10 @@ def test_flate_to_jbig2(resources, outdir, spoof_tesseract_noop):
     # This test requires an image that pngquant is capable of converting to
     # to 1bpp - so use an existing 1bpp image, convert up, confirm it can
     # convert down
-    im = Image.open(fspath(resources / 'typewriter.png'))
-    assert im.mode in ('1', 'P')
-    im = im.convert('L')
-    im.save(fspath(outdir / 'type8.png'))
+    with Image.open(fspath(resources / 'typewriter.png')) as im:
+        assert im.mode in ('1', 'P')
+        im = im.convert('L')
+        im.save(fspath(outdir / 'type8.png'))
 
     check_ocrmypdf(
         outdir / 'type8.png',

tests/test_rotation.py

@@ -224,12 +224,12 @@ def test_rotate_deskew_timeout(resources, outdir):
 @pytest.mark.parametrize('image_angle', (0, 90, 180, 270))
 def test_rotate_page_level(image_angle, page_angle, resources, outdir):
     def make_rotate_test(prefix, image_angle, page_angle):
-        im = Image.open(fspath(resources / 'typewriter.png'))
-        if image_angle != 0:
-            ccw_angle = -image_angle % 360
-            im = im.transpose(getattr(Image, f'ROTATE_{ccw_angle}'))
         memimg = BytesIO()
-        im.save(memimg, format='PNG')
+        with Image.open(fspath(resources / 'typewriter.png')) as im:
+            if image_angle != 0:
+                ccw_angle = -image_angle % 360
+                im = im.transpose(getattr(Image, f'ROTATE_{ccw_angle}'))
+            im.save(memimg, format='PNG')
         memimg.seek(0)
         mempdf = BytesIO()
         img2pdf.convert(

tests/test_stdio.py

@@ -0,0 +1,137 @@
+# © 2019 James R. Barlow: github.com/jbarlow83
+#
+# This file is part of OCRmyPDF.
+#
+# OCRmyPDF is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# OCRmyPDF is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with OCRmyPDF.  If not, see <http://www.gnu.org/licenses/>.
+
+import os
+import sys
+from pathlib import Path
+from subprocess import DEVNULL, PIPE, run, Popen
+
+import pytest
+
+from ocrmypdf.exceptions import ExitCode
+from ocrmypdf.exec import qpdf
+
+# pytest.helpers is dynamic
+# pylint: disable=no-member,redefined-outer-name
+
+run_ocrmypdf = pytest.helpers.run_ocrmypdf
+spoof = pytest.helpers.spoof
+
+
+@pytest.fixture(scope='session')
+def spoof_tess_bad_utf8(tmp_path_factory):
+    return spoof(tmp_path_factory, tesseract='tesseract_badutf8.py')
+
+
+def test_stdin(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf):
+    input_file = str(resources / 'francais.pdf')
+    output_file = str(outpdf)
+
+    # Runs: ocrmypdf - output.pdf < testfile.pdf
+    with open(input_file, 'rb') as input_stream:
+        p_args = ocrmypdf_exec + ['-', output_file]
+        p = run(
+            p_args,
+            stdout=PIPE,
+            stderr=PIPE,
+            stdin=input_stream,
+            env=spoof_tesseract_noop,
+        )
+        assert p.returncode == ExitCode.ok
+
+
+def test_stdout(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf):
+    input_file = str(resources / 'francais.pdf')
+    output_file = str(outpdf)
+
+    # Runs: ocrmypdf francais.pdf - > test_stdout.pdf
+    with open(output_file, 'wb') as output_stream:
+        p_args = ocrmypdf_exec + [input_file, '-']
+        p = run(
+            p_args,
+            stdout=output_stream,
+            stderr=PIPE,
+            stdin=DEVNULL,
+            env=spoof_tesseract_noop,
+        )
+        assert p.returncode == ExitCode.ok
+
+    assert qpdf.check(output_file, log=None)
+
+
+@pytest.mark.skipif(
+    sys.version_info[0:3] >= (3, 6, 4), reason="issue fixed in Python 3.6.4"
+)
+def test_closed_streams(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf):
+    input_file = str(resources / 'francais.pdf')
+    output_file = str(outpdf)
+
+    def evil_closer():
+        os.close(0)
+        os.close(1)
+
+    p_args = ocrmypdf_exec + [input_file, output_file]
+    p = Popen(  # pylint: disable=subprocess-popen-preexec-fn
+        p_args,
+        close_fds=True,
+        stdout=None,
+        stderr=PIPE,
+        stdin=None,
+        env=spoof_tesseract_noop,
+        preexec_fn=evil_closer,
+    )
+    out, err = p.communicate()
+    print(err.decode())
+    assert p.returncode == ExitCode.ok
+
+
+@pytest.mark.skipif(sys.version_info >= (3, 7, 0), reason='better utf-8')
+@pytest.mark.skipif(
+    Path('/etc/alpine-release').exists(), reason="invalid test on alpine"
+)
+def test_bad_locale():
+    env = os.environ.copy()
+    env['LC_ALL'] = 'C'
+
+    p, out, err = run_ocrmypdf('a', 'b', env=env)
+    assert out == '', "stdout not clean"
+    assert p.returncode != 0
+    assert 'configured to use ASCII as encoding' in err, "should whine"
+
+
+@pytest.mark.parametrize('renderer', ['hocr', 'sandwich'])
+def test_bad_utf8(spoof_tess_bad_utf8, renderer, resources, no_outpdf):
+    p, out, err = run_ocrmypdf(
+        resources / 'ccitt.pdf',
+        no_outpdf,
+        '--pdf-renderer',
+        renderer,
+        env=spoof_tess_bad_utf8,
+    )
+
+    assert out == '', "stdout not clean"
+    assert p.returncode != 0
+    assert 'not utf-8' in err, "should whine about utf-8"
+    assert '\\x96' in err, 'should repeat backslash encoded output'
+
+
+def test_dev_null(spoof_tesseract_noop, resources):
+    p, out, err = run_ocrmypdf(
+        resources / 'trivial.pdf', os.devnull, '--force-ocr', env=spoof_tesseract_noop
+    )
+    assert p.returncode == 0, "could not send output to /dev/null"
+    assert len(out) == 0, "wrote to stdout"

tests/test_validation.py

@@ -15,20 +15,23 @@
 # You should have received a copy of the GNU General Public License
 # along with OCRmyPDF.  If not, see <http://www.gnu.org/licenses/>.
 
+import locale
+import logging
 import os
-from unittest.mock import MagicMock, patch
+from unittest.mock import patch
 
 import pytest
 
 import ocrmypdf._validation as vd
 from ocrmypdf.api import create_options
 from ocrmypdf.exceptions import MissingDependencyError, BadArgsError
+from ocrmypdf.pdfinfo import PdfInfo
 
 
 def make_opts(input_file='a.pdf', output_file='b.pdf', language='eng', **kwargs):
-    return create_options(
-        input_file=input_file, output_file=output_file, language=language, **kwargs
-    )
+    if language is not None:
+        kwargs['language'] = language
+    return create_options(input_file=input_file, output_file=output_file, **kwargs)
 
 
 def test_hocr_notlatin_warning(caplog):
@@ -37,15 +40,21 @@ def test_hocr_notlatin_warning(caplog):
 
 
 def test_old_ghostscript(caplog):
-    with patch('ocrmypdf.exec.ghostscript.version', return_value='9.19'):
+    with patch('ocrmypdf.exec.ghostscript.version', return_value='9.19'), patch(
+        'ocrmypdf.exec.tesseract.has_textonly_pdf', return_value=True
+    ):
         vd.check_options_output(make_opts(language='chi_sim', output_type='pdfa'))
         assert 'Ghostscript does not work correctly' in caplog.text
 
-    with patch('ocrmypdf.exec.ghostscript.version', return_value='9.18'):
+    with patch('ocrmypdf.exec.ghostscript.version', return_value='9.18'), patch(
+        'ocrmypdf.exec.tesseract.has_textonly_pdf', return_value=True
+    ):
         with pytest.raises(MissingDependencyError):
             vd.check_options_output(make_opts(output_type='pdfa-3'))
 
-    with patch('ocrmypdf.exec.ghostscript.version', return_value='9.24'):
+    with patch('ocrmypdf.exec.ghostscript.version', return_value='9.24'), patch(
+        'ocrmypdf.exec.tesseract.has_textonly_pdf', return_value=True
+    ):
         with pytest.raises(MissingDependencyError):
             vd.check_dependency_versions(make_opts())
 
@@ -113,3 +122,41 @@ def test_report_file_size(tmp_path, caplog):
     os.truncate(out, 50000)
     vd.report_output_file_size(opts, in_, out)
     assert 'No reason' in caplog.text
+
+
+def test_false_action_store_true():
+    opts = make_opts(keep_temporary_files=True)
+    assert opts.keep_temporary_files == True
+    opts = make_opts(keep_temporary_files=False)
+    assert opts.keep_temporary_files == False
+
+
+@pytest.mark.parametrize('progress_bar', [True, False])
+def test_no_progress_bar(progress_bar, resources):
+    opts = make_opts(progress_bar=progress_bar, input_file=(resources / 'trivial.pdf'))
+    with patch('ocrmypdf.pdfinfo.info.tqdm', autospec=True) as tqdmpatch:
+        vd.check_options(opts)
+        pdfinfo = PdfInfo(opts.input_file, progbar=opts.progress_bar)
+        assert pdfinfo is not None
+        assert tqdmpatch.called
+        _args, kwargs = tqdmpatch.call_args
+        assert kwargs['disable'] != progress_bar
+
+
+def test_language_warning(caplog):
+    opts = make_opts(language=None)
+    caplog.set_level(logging.DEBUG)
+    with patch(
+        'ocrmypdf._validation.locale.getlocale', return_value=('en_US', 'UTF-8')
+    ):
+        vd.check_options_languages(opts)
+        assert opts.language == ['eng']
+        assert '' in caplog.text
+
+    opts = make_opts(language=None)
+    with patch(
+        'ocrmypdf._validation.locale.getlocale', return_value=('fr_FR', 'UTF-8')
+    ):
+        vd.check_options_languages(opts)
+        assert opts.language == ['eng']
+        assert 'assuming --language' in caplog.text