v9.8.2 release notes

Use more realistic test data

.coveragerc

@@ -1,5 +1,3 @@
-# Coverage isn't really compatible with subprocesses so results are unreliable
-
 [paths]
 source =
 	src
@@ -8,9 +6,11 @@ source =
 [run]
 branch = true
 parallel = true
+concurrency =
+	thread
+	multiprocessing
 source =
 	src/ocrmypdf
-	tests
 omit =
 	tests/spoof/*
 

.github/ISSUE_TEMPLATE/bug_report.md

@@ -0,0 +1,40 @@
+---
+name: Bug report
+about: Create a report to help us improve
+title: ''
+labels: ''
+assignees: ''
+
+---
+
+**Describe the bug**
+A clear and concise description of what the bug is.
+
+**To Reproduce**
+What command line or API call were you trying to run?
+
+```bash
+ocrmypdf  ...arguments... input.pdf output.pdf
+```
+
+Run with verbosity or higher `-v1` to see more detailed logging. This information may be helpful.
+
+**Example file**
+Include an input PDF or image that demonstrates your issue.
+
+Please provide an input file with no personal or confidential information. At your option you may `GPG-encrypt the file <https://github.com/jbarlow83/OCRmyPDF/wiki>` for OCRmyPDF's author only.
+
+Links to files hosted elsewhere are perfectly acceptable. You could also look in ``tests/resources`` and see if any of those files reproduce your issue.
+
+(Exceptions: Issues with installation, command line argument parsing, test suite failures.Issues without example files usually cannot be resolved.)
+
+**Expected behavior**
+A clear and concise description of what you expected to happen.
+
+**Screenshots**
+If applicable, add screenshots to help explain your problem.
+
+**System**
+ - OS: [e.g. Linux, Windows, macOS]
+ - OCRmyPDF Version: ``ocrmypdf --version``
+ - How did you install ocrmypdf? Did you use a system package manager, `pip`, or a Docker image?

.github/ISSUE_TEMPLATE/feature_request.md

@@ -0,0 +1,17 @@
+---
+name: Feature request
+about: Suggest an idea for this project
+title: ''
+labels: enhancement
+assignees: ''
+
+---
+
+**Is your feature request related to a problem? Please describe.**
+A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
+
+**Describe the solution you'd like**
+A clear and concise description of what you want to happen.
+
+**Additional context**
+Add any other context or screenshots about the feature request here.

.travis.yml

@@ -1,159 +0,0 @@
-branches:
-  except:
-    - azure
-
-cache:
-  pip: true
-  directories:
-    - $HOME/Library/Caches/Homebrew
-
-matrix:
-  include:
-    - os: linux
-      dist: trusty
-      sudo: required
-      language: python
-      python: "3.6"
-      env:
-        - DIST=trusty
-        - MINIMAL=true
-      addons:
-        apt:
-          update: true
-          sources:
-            - sourceline: "ppa:alex-p/tesseract-ocr"
-            - sourceline: "ppa:vshn/ghostscript"
-          packages:
-            - ghostscript
-            - libffi-dev
-            - 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
-            - 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
-      language: python
-      python: "3.7"
-      env:
-        - DIST=xenial
-      addons:
-        apt:
-          update: true
-          sources:
-            - sourceline: "ppa:alex-p/tesseract-ocr"
-          packages:
-            - ghostscript
-            - libexempi3
-            - libffi-dev
-            - pngquant
-            - poppler-utils
-            - 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
-            - tesseract-ocr
-            - tesseract-ocr-deu
-            - tesseract-ocr-eng
-            - tesseract-ocr-fra
-            - unpaper
-    - os: osx
-      language: generic
-      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 $HOME/.cache/pip/log/debug.log
-
-install:
-  - mkdir -p bin
-  - export PATH=$PWD/bin:$PATH
-  - pip3 install -r requirements/main.txt -r requirements/test.txt .
-
-script:
-  - tesseract --version
-  - pytest -n auto
-# deploy:
-#   # release for main pypi
-#   # 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 == "3.7" && $TRAVIS_OS_NAME == "linux"
-#     skip_upload_docs: true

README.md

@@ -38,7 +38,7 @@ Main features
 - Keeps the exact resolution of the original embedded images
 - When possible, inserts OCR information as a "lossless" operation without disrupting any other content
 - Optimizes PDF images, often producing files smaller than the input file
-- If requested deskews and/or cleans the image before performing OCR
+- If requested, deskews and/or cleans the image before performing OCR
 - Validates input and output files
 - Distributes work across all available CPU cores
 - Uses [Tesseract OCR](https://github.com/tesseract-ocr/tesseract) engine to recognize more than [100 languages](https://github.com/tesseract-ocr/tessdata)
@@ -50,7 +50,7 @@ For details: please consult the [documentation](https://ocrmypdf.readthedocs.io/
 Motivation
 ----------
 
-I searched the web for a free command line tool to OCR PDF files: I found many, but none of them were really satisfying.
+I searched the web for a free command line tool to OCR PDF files: I found many, but none of them were really satisfying:
 
 - Either they produced PDF files with misplaced text under the image (making copy/paste impossible)
 - Or they did not handle accents and multilingual characters
@@ -97,7 +97,10 @@ OCRmyPDF uses Tesseract for OCR, and relies on its language packs. For Linux use
 apt-cache search tesseract-ocr
 
 # Debian/Ubuntu users
-apt-get install tesseract-ocr-chi-sim  # Example: Install Chinese Simplified language back
+apt-get install tesseract-ocr-chi-sim  # Example: Install Chinese Simplified language pack
+
+# Arch Linux users
+pacman -S tesseract-data-eng tesseract-data-deu # Example: Install the English and German language packs
 ```
 
 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.
@@ -105,7 +108,7 @@ You can then pass the `-l LANG` argument to OCRmyPDF to give a hint as to what l
 Documentation and support
 -------------------------
 
-Once ocrmypdf is installed, the built-in help which explains the command syntax and options can be accessed via:
+Once OCRmyPDF is installed, the built-in help which explains the command syntax and options can be accessed via:
 
 ```bash
 ocrmypdf --help
@@ -118,20 +121,21 @@ Please report issues on our [GitHub issues](https://github.com/jbarlow83/OCRmyPD
 Requirements
 ------------
 
-In addition to the required Python version (3.6+), OCRmyPDF requires external program installations of Ghostscript, Tesseract OCR, QPDF, and Leptonica. ocrmypdf is pure Python, but uses CFFI to portably generate library bindings. OCRmyPDF works on pretty much  everything: Linux, macOS, Windows and FreeBSD.
+In addition to the required Python version (3.6+), OCRmyPDF requires external program installations of Ghostscript, Tesseract OCR, QPDF, and Leptonica. OCRmyPDF is pure Python, but uses CFFI to portably generate library bindings. OCRmyPDF works on pretty much everything: Linux, macOS, Windows and FreeBSD.
 
 Press & Media
 -------------
 
 - [Going paperless with OCRmyPDF](https://medium.com/@ikirichenko/going-paperless-with-ocrmypdf-e2f36143f46a)
 - [Converting a scanned document into a compressed searchable PDF with redactions](https://medium.com/@treyharris/converting-a-scanned-document-into-a-compressed-searchable-pdf-with-redactions-63f61c34fe4c)
-- [c't 1-2014, page 59](http://heise.de/-2279695): Detailed presentation of OCRmyPDF v1.0 in the leading German IT magazine c't
-- [heise Open Source, 09/2014: Texterkennung mit OCRmyPDF](http://heise.de/-2356670)
+- [c't 1-2014, page 59](https://heise.de/-2279695): Detailed presentation of OCRmyPDF v1.0 in the leading German IT magazine c't
+- [heise Open Source, 09/2014: Texterkennung mit OCRmyPDF](https://heise.de/-2356670)
+- [heise Durchsuchbare PDF-Dokumente mit OCRmyPDF erstellen](https://www.heise.de/ratgeber/Durchsuchbare-PDF-Dokumente-mit-OCRmyPDF-erstellen-4607592.html)
 
 Business enquiries
 ------------------
 
-OCRmyPDF would not be the software that it is today is without companies and users choosing to provide support for feature development and consulting enquiries. We are happy to discuss all enquiries, whether for extending the existing feature set, or integrating OCRmyPDF into a larger system.
+OCRmyPDF would not be the software that it is today without companies and users choosing to provide support for feature development and consulting enquiries. We are happy to discuss all enquiries, whether for extending the existing feature set, or integrating OCRmyPDF into a larger system.
 
 License
 -------

azure-pipelines.yml

@@ -153,12 +153,13 @@ stages:
           matrix:
             Python37:
               python.version: "3.7"
-            Python38:
-              python.version: "3.8"
+            # Python38:
+            #   python.version: "3.8"
         steps:
-          - task: UsePythonVersion@0
-            inputs:
-              versionSpec: "$(python.version)"
+          # https://github.com/actions/virtual-environments/issues/664
+          # - task: UsePythonVersion@0
+          #   inputs:
+          #     versionSpec: "$(python.version)"
           - bash: |
               brew update
               brew unlink python@2
@@ -169,6 +170,7 @@ stages:
                 leptonica \
                 openjpeg \
                 pngquant \
+                python \
                 tesseract \
                 unpaper
             displayName: "Install system packages"

debian/copyright

@@ -95,12 +95,6 @@ Files: tests/resources/vector.pdf
 Copyright: (C) 2018 Catscratch
 License: Expat
 
-Files: test/resources/enron*.pdf
-Copyright: EnronData.org
-License: CC-BY-3.0
- See: https://enrondata.readthedocs.io/en/latest/data/edo-enron-email-pst-dataset/
-Comment: Unprocessed.
-
 Files: src/ocrmypdf/data/sRGB.icc
 Copyright:  Kai-Uwe Behrmann <www.behrmann.name>
             Marti Maria <www.littlecms.com>

docs/api.rst

@@ -51,6 +51,14 @@ Forking a child process to call ``ocrmypdf.ocr()`` is suggested. That
 way your application will survive and remain interactive even if
 OCRmyPDF does not.
 
+.. 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 fork semantics will prevent OCRmyPDF from working
+    correct.
+
 Logging
 -------
 

docs/batch.rst

@@ -69,53 +69,8 @@ Sample script
 This user contributed script also provides an example of batch
 processing.
 
-.. code-block:: python
-
-   #!/usr/bin/env python3
-   # Walk through directory tree, replacing all files with OCR'd version
-   # Original version by DeliciousPickle@github; modified
-
-   import logging
-   import os
-   import subprocess
-   import sys
-
-   import ocrmypdf
-
-   script_dir = os.path.dirname(os.path.realpath(__file__))
-   print(script_dir + '/ocr-tree.py: Start')
-
-   if len(sys.argv) > 1:
-       start_dir = sys.argv[1]
-   else:
-       start_dir = '.'
-
-   if len(sys.argv) > 2:
-       log_file = sys.argv[2]
-   else:
-       log_file = script_dir + '/ocr-tree.log'
-
-   logging.basicConfig(
-           level=logging.INFO, format='%(asctime)s %(message)s',
-           filename=log_file, filemode='w')
-
-   ocrmypdf.configure_logging(ocrmypdf.Verbosity.default)
-
-   for dir_name, subdirs, file_list in os.walk(start_dir):
-       logging.info('\n')
-       logging.info(dir_name + '\n')
-       os.chdir(dir_name)
-       for filename in file_list:
-           file_ext = os.path.splitext(filename)[1]
-           if file_ext == '.pdf':
-               full_path = dir_name + '/' + filename
-               print(full_path)
-               result = ocrmypdf.ocr(filename, filename, deskew=True)
-               if result == ocrmypdf.ExitCode.already_done_ocr:
-                   print("Skipped document because it already contained text")
-               elif result == ocrmypdf.ExitCode.ok:
-                   print("OCR complete")
-               logging.info(result)
+.. literalinclude:: ../misc/batch.py
+    :caption: misc/batch.py
 
 Synology DiskStations
 ---------------------
@@ -131,62 +86,8 @@ products use ARM or Power processors and do not support Docker. Further
 adjustments might be needed to deal with the Synology's relatively
 limited CPU and RAM.
 
-.. code-block:: python
-
-   #!/bin/env python3
-   # Contributed by github.com/Enantiomerie
-
-   # script needs 2 arguments
-   # 1. source dir with *.pdf - default is location of script
-   # 2. move dir where *.pdf and *_OCR.pdf are moved to
-
-   import logging
-   import os
-   import subprocess
-   import sys
-   import time
-   import shutil
-
-   script_dir = os.path.dirname(os.path.realpath(__file__))
-   timestamp = time.strftime("%Y-%m-%d-%H%M_")
-   log_file = script_dir + '/' + timestamp + 'ocrmypdf.log'
-   logging.basicConfig(level=logging.INFO, format='%(asctime)s %(message)s', filename=log_file, filemode='w')
-
-   if len(sys.argv) > 1:
-       start_dir = sys.argv[1]
-   else:
-       start_dir = '.'
-
-   for dir_name, subdirs, file_list in os.walk(start_dir):
-       logging.info('\n')
-       logging.info(dir_name + '\n')
-       os.chdir(dir_name)
-       for filename in file_list:
-           file_ext = os.path.splitext(filename)[1]
-           if file_ext == '.pdf':
-               full_path = dir_name + '/' + filename
-               file_noext = os.path.splitext(filename)[0]
-               timestamp_OCR = time.strftime("%Y-%m-%d-%H%M_OCR_")
-               filename_OCR = timestamp_OCR + file_noext + '.pdf'
-               docker_mount = dir_name + ':/home/docker'
-   # create string for pdf processing
-   # diskstation needs a user:group docker:docker. find uid:gid of your diskstation docker:docker with id docker.
-   # use this uid:gid in -u flag
-   # rw rights for docker:docker at source dir are also necessary
-   # the script is processed as root user via chron
-               cmd = ['docker', 'run', '--rm', '-v', docker_mount, '-u=1030:65538', 'jbarlow83/ocrmypdf', , '--deskew' , filename, filename_OCR]
-               logging.info(cmd)
-               proc = subprocess.run(cmd, stdout=subprocess.PIPE, stderr=subprocess.STDOUT)
-               result = proc.stdout.read()
-               logging.info(result)
-               full_path_OCR = dir_name + '/' + filename_OCR
-               os.chmod(full_path_OCR, 0o666)
-               os.chmod(full_path, 0o666)
-               full_path_OCR_archive = sys.argv[2]
-               full_path_archive = sys.argv[2] + '/no_ocr'
-               shutil.move(full_path_OCR,full_path_OCR_archive)
-               shutil.move(full_path, full_path_archive)
-   logging.info('Finished.\n')
+.. literalinclude:: ../misc/synology.py
+    :caption: misc/synology.py - Sample script for Synology DiskStations
 
 Huge batch jobs
 ---------------
@@ -198,11 +99,45 @@ and all inquiries are appreciated.
 Hot (watched) folders
 =====================
 
+Watched folders with watcher.py
+-------------------------------
+
+OCRmyPDF has a folder watcher called watcher.py, which is currently included in source
+distributions but not part of the main program. It may be used natively or may run
+in a Docker container. Native instances tend to give better performance. watcher.py
+works on all platforms.
+
+Users may need to customize the script to meet their requirements.
+
+.. code-block:: bash
+
+    pip3 install -r requirements/watcher.txt
+
+    env OCR_INPUT_DIRECTORY=/mnt/input-pdfs \
+        OCR_OUTPUT_DIRECTORY=/mnt/output-pdfs \
+        OCR_OUTPUT_DIRECTORY_YEAR_MONTH=1 \
+        python3 watcher.py
+
+.. csv-table:: watcher.py environment variables
+    :header: "Environment variable", "Description"
+    :widths: 50, 50
+
+    "OCR_INPUT_DIRECTORY", "Set input directory to monitor (recursive)"
+    "OCR_OUTPUT_DIRECTORY", "Set output directory (should not be under input)"
+    "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``"
+    "OCR_POLL_NEW_FILE_SECONDS", "Polling interval"
+    "OCR_LOGLEVEL", "Level of log messages to report"
+
+One could configure a networked scanner or scanning computer to drop files in the
+watched folder.
+
 Watched folders with Docker
 ---------------------------
 
-The OCRmyPDF Docker image includes a watcher service. This service can
-be launched as follows:
+The watcher service is included in the OCRmyPDF Docker image. To run it:
 
 .. code-block:: bash
 
@@ -210,6 +145,9 @@ be launched as follows:
         -v <path to files to convert>:/input \
         -v <path to store results>:/output \
         -e OCR_OUTPUT_DIRECTORY_YEAR_MONTH=1 \
+        -e OCR_ON_SUCCESS_DELETE=1 \
+        -e OCR_DESKEW=1 \
+        -e PYTHONUNBUFFERED=1 \
         -it --entrypoint python3 \
         jbarlow83/ocrmypdf \
         watcher.py
@@ -223,61 +161,21 @@ convert it to a OCRed PDF in ``/output/``. The parameters to this image are:
 
     "``-v <path to files to convert>:/input``", "Files placed in this location will be OCRed"
     "``-v <path to store results>:/output``", "This is where OCRed files will be stored"
-    "``-e OCR_OUTPUT_DIRECTORY_YEAR_MONTH=1``", "This will place files in the output in {output}/{year}/{month}/{filename}"
+    "``-e OCR_OUTPUT_DIRECTORY_YEAR_MONTH=1``", "Define environment variable OCR_OUTPUT_DIRECTORY_YEAR_MONTH=1"
+    "``-e OCR_ON_SUCCESS_DELETE=1``", "Define environment variable"
+    "``-e OCR_DESKEW=1``", "Define environment variable"
+    "``-e PYTHONBUFFERED=1``", "This will force STDOUT to be unbuffered and allow you to see messages in docker logs"
 
 This service relies on polling to check for changes to the filesystem. It
 may not be suitable for some environments, such as filesystems shared on a
 slow network.
 
-Watched folders with watcher.py
--------------------------------
+A configuration manager such as Docker Compose could be used to ensure that the
+service is always available.
 
-The watcher service may also be run natively.
-
-.. code-block:: bash
-
-    pip3 install -r reqs/watcher.txt
-
-    env OCR_INPUT_DIRECTORY=/mnt/input-pdfs \
-        OCR_OUTPUT_DIRECTORY=/mnt/output-pdfs \
-        OCR_OUTPUT_DIRECTORY_YEAR_MONTH=1 \
-        python3 watcher.py
-
-Watched folders with CLI
-------------------------
-
-To set up a "hot folder" that will trigger OCR for every file inserted,
-use a program like Python
-`watchdog <https://pypi.python.org/pypi/watchdog>`__ (supports all major
-OS).
-
-One could then configure a scanner to automatically place scanned files
-in a hot folder, so that they will be queued for OCR and copied to the
-destination.
-
-.. code-block:: bash
-
-   pip install watchdog
-
-watchdog installs the command line program ``watchmedo``, which can be
-told to run ``ocrmypdf`` on any .pdf added to the current directory
-(``.``) and place the result in the previously created ``out/`` folder.
-
-.. code-block:: bash
-
-   cd hot-folder
-   mkdir out
-   watchmedo shell-command \
-       --patterns="*.pdf" \
-       --ignore-directories \
-       --command='ocrmypdf "${watch_src_path}" "out/${watch_src_path}" ' \
-       .  # don't forget the final dot
-
-On file servers, you could configure watchmedo as a system service so it
-will run all the time.
-
-For more complex behavior you can write a Python script around to use
-the watchdog API. You can refer to the watcher.py script as an example.
+.. literalinclude:: ../misc/docker-compose.example.yml
+    :language: yaml
+    :caption: misc/docker-compose.example.yml
 
 Caveats
 -------
@@ -304,6 +202,13 @@ Alternatives
 -  `Watchman <https://facebook.github.io/watchman/>`__ is a more
    powerful alternative to ``watchmedo``.
 
+AWS Lambda is not viable
+------------------------
+
+AWS Lambda and its equivalents have low limits on execution time and payload
+size, relative to OCRmyPDF's needs. As of this writing, the request/response
+payload for AWS Lambda was 6 MB, which means many PDFs will not fit.
+
 macOS Automator
 ===============
 
@@ -313,8 +218,7 @@ of Automator, the ``PATH`` may be set differently your Terminal's
 ``PATH``; you may need to explicitly set the PATH to include
 ``ocrmypdf``. The following example may serve as a starting point:
 
-|Example macOS Automator script|
+.. figure:: images/macos-workflow.png
+    :alt: Example macOS Automator workflow
 
 You may customize the command sent to ocrmypdf.
-
-.. |Example macOS Automator script| image:: images/macos-workflow.png

docs/cookbook.rst

@@ -89,6 +89,18 @@ This produces a file named "output.pdf" and a companion text file named
 
     ocrmypdf --sidecar output.txt input.pdf output.pdf
 
+.. note::
+
+    The sidecar file contains the **OCR text** found by OCRmyPDF. If the document
+    contains pages that already have text, that text will not appear in the
+    sidecar. If the option ``--pages`` is used, only those pages on which OCR
+    was performed will be included in the sidecar. If certain pages were skipped
+    because of options like ``--skip-big`` or ``--tesseract-timeout``, those pages
+    will not be in the sidecar.
+
+    To extract all text from a PDF, whether generated from OCR or otherwise,
+    use a program like Poppler's ``pdftotext`` or ``pdfgrep``.
+
 OCR images, not PDFs
 --------------------
 
@@ -225,6 +237,32 @@ You can also optimize all images without performing any OCR:
 
     ocrmypdf --tesseract-timeout=0 --optimize 3 --skip-text input.pdf output.pdf
 
+Perform OCR only certain pages
+------------------------------
+
+You can ask OCRmyPDF to only apply OCR to certain pages.
+
+.. code-block:: bash
+
+    ocrmypdf --pages 2,3,13-17 input.pdf output.pdf
+
+Hyphens denote a range of pages and commas separate page numbers. If you prefer
+to use spaces, quote all of the page numbers: ``--pages '2, 3, 5, 7'``.
+
+OCRmyPDF will warn if your list of page numbers contains duplicates or
+overlap pages. OCRmyPDF does not currently account for document page numbers,
+such as an introduction section of a book that uses Roman numerals. It simply
+counts the number of virtual pieces of paper since the start.
+
+Regardless of the argument to ``--pages``, OCRmyPDF will optimize all pages in
+the file and convert it to PDF/A, unless you disable those options. In this
+example, we want to OCR only the title and otherwise change the PDF as little
+as possible:
+
+.. code-block:: bash
+
+    ocrmypdf --pages 1 --output-type pdf --optimize 0 input.pdf output.pdf
+
 Redo existing OCR
 =================
 

docs/docker.rst

@@ -56,52 +56,52 @@ See the Docker documentation for
 Using the Docker image on the command line
 ==========================================
 
-**Unlike typical Docker containers**, in this mode we are using the
-OCRmyPDF Docker container is intended to be emphemeral – it runs for one
-OCR job and then terminates, just like a command line program. We are
-using Docker as a way of delivering an application, not a server.
+**Unlike typical Docker containers**, in this section the OCRmyPDF Docker
+container is emphemeral – it runs for one OCR job and terminates, just like a
+command line program. We are using Docker to deliver an application (as opposed
+to the more conventional case, where a Docker container runs as a server).
 
 To start a Docker container (instance of the image):
 
 .. code-block:: bash
 
    docker tag jbarlow83/ocrmypdf ocrmypdf
-   docker run --rm -i ocrmypdf (... all other arguments here...)
+   docker run --rm -i ocrmypdf (... all other arguments here...) - -
 
 For convenience, create a shell alias to hide the Docker command. It is
-easier to send the input file to file stdin and read the output from
-stdout – this avoids the occasionally messy permission issues with
-Docker entirely.
+easier to send the input file as stdin and read the output from
+stdout – **this avoids the messy permission issues with Docker entirely**.
 
 .. code-block:: bash
 
-   alias ocrmypdf='docker run --rm -i ocrmypdf'
-   ocrmypdf --version  # runs docker version
-   ocrmypdf <input.pdf >output.pdf
+   alias docker_ocrmypdf='docker run --rm -i ocrmypdf'
+   docker_ocrmypdf --version  # runs docker version
+   docker_ocrmypdf - - <input.pdf >output.pdf
 
 Or in the wonderful `fish shell <https://fishshell.com/>`__:
 
 .. code-block:: fish
 
-   alias ocrmypdf 'docker run --rm ocrmypdf'
-   funcsave ocrmypdf
+   alias docker_ocrmypdf 'docker run --rm ocrmypdf'
+   funcsave docker_ocrmypdf
 
 Alternately, you could mount the local current working directory as a
 Docker volume:
 
 .. code-block:: bash
 
-   docker run --rm -v $(pwd):/data ocrmypdf /data/input.pdf /data/output.pdf
+   alias docker_ocrmypdf='docker run --rm  -i --user "$(id -u):$(id -g)" --workdir /data -v "$PWD:/data" ocrmypdf'
+   docker_ocrmypdf /data/input.pdf /data/output.pdf
 
 .. _docker-lang-packs:
 
 Adding languages to the Docker image
 ====================================
 
-By default the Docker image includes English, German and Simplified
-Chinese, the most popular languages for OCRmyPDF users based on
-feedback. You may add other languages by creating a new Dockerfile based
-on the public one:
+By default the Docker image includes English, German, Simplified Chinese,
+French, Portuguese and Spanish, the most popular languages for OCRmyPDF
+users based on feedback. You may add other languages by creating a new
+Dockerfile based on the public one:
 
 .. code-block:: dockerfile
 

docs/index.rst

@@ -1,8 +1,8 @@
 OCRmyPDF documentation
 ======================
 
-OCRmyPDF adds an OCR text layer to scanned PDF files, allowing them to
-be searched.
+OCRmyPDF adds an optical charcter recognition (OCR) text layer to scanned PDF
+files, allowing them to be searched.
 
 PDF is the best format for storing and exchanging scanned documents.
 Unfortunately, PDFs can be difficult to modify. OCRmyPDF makes it easy to apply
@@ -27,6 +27,7 @@ image processing and OCR to existing PDFs.
    advanced
    batch
    security
+   performance
    errors
 
 .. toctree::

docs/installation.rst

@@ -110,7 +110,7 @@ Fedora 29 or newer
 | |fedora-29| |fedora-30| |fedora-rawhide|      |
 +-----------------------------------------------+
 
-Users of Fedora 29 later may simply
+Users of Fedora 29 or later may simply
 
 .. code-block:: bash
 
@@ -136,23 +136,27 @@ from sources <#installing-head-revision-from-sources>`__.
 Installing the latest version on Ubuntu 18.04 LTS
 -------------------------------------------------
 
-Ubuntu 18.04 includes ocrmypdf 6.1.2. To install a more recent version,
-first install the system version to get most of the dependencies:
+Ubuntu 18.04 includes ocrmypdf 6.1.2 - you can install that with ``apt``, but
+it is quite old now. To install a more recent version, uninstall the old version
+of ocrmypdf, and install the following dependencies:
 
 .. code-block:: bash
 
-    sudo apt-get update
-    sudo apt-get install \
-        ocrmypdf
-
-There are a few system dependency changes since ocrmypdf 6.1.2. Let's
-get these, too.
-
-.. code-block:: bash
-
-    sudo apt-get install \
+    sudo apt-get -y remove ocrmypdf
+    sudo apt-get -y update
+    sudo apt-get -y install \
+        ghostscript \
+        icc-profiles-free \
+        liblept5 \
         libxml2 \
-        pngquant
+        pngquant \
+        python3-cffi \
+        python3-distutils \
+        python3-pkg-resources \
+        python3-reportlab \
+        qpdf \
+        tesseract-ocr \
+        zlib1g
 
 We will need a newer version of ``pip`` then was available for Ubuntu 18.04:
 
@@ -214,7 +218,8 @@ of ``pip`` at ``/usr/local/bin/pip``.
 **Install OCRmyPDF**
 
 OCRmyPDF requires the locale to be set for UTF-8. **On some minimal
-Ubuntu installations systems**, it may be necessary to set the locale.
+Ubuntu installations**, such as the Ubuntu 16.04 Docker images it may be
+necessary to set the locale.
 
 .. code-block:: bash
 
@@ -228,7 +233,7 @@ environment variable contains ``$HOME/.local/bin``.
 .. code-block:: bash
 
     export PATH=$HOME/.local/bin:$PATH
-    pip3 install --user ocrmypdf
+    pip3.6 install --user ocrmypdf
 
 To add JBIG2 encoding, see :ref:`jbig2`.
 
@@ -294,23 +299,72 @@ compiled by hand.
 
 To add JBIG2 encoding, see :ref:`jbig2`.
 
-ArchLinux (AUR)
----------------
+Arch Linux (AUR)
+----------------
 
 .. image:: https://repology.org/badge/version-for-repo/aur/ocrmypdf.svg
     :alt: ArchLinux
     :target: https://repology.org/metapackage/ocrmypdf
 
-There is an `ArchLinux User Repository package for
-ocrmypdf <https://aur.archlinux.org/packages/ocrmypdf/>`__. You can use
-the following command.
+There is an `Arch User Repository (AUR) package for OCRmyPDF
+<https://aur.archlinux.org/packages/ocrmypdf/>`__.
+
+Installing AUR packages as root is not allowed, so you must first `setup a
+non-root user
+<https://wiki.archlinux.org/index.php/Users_and_groups#User_management>`__ and
+`configure sudo <https://wiki.archlinux.org/index.php/Sudo#Configuration>`__.
+The standard Docker image, ``archlinux/base:latest``, does **not** have a
+non-root user configured, so users of that image must follow these guides. If
+you are using a VM image, such as `the official Vagrant image
+<https://app.vagrantup.com/archlinux/boxes/archlinux>`__, this work may already
+be completed for you.
+
+Next you should install the `base-devel package group
+<https://www.archlinux.org/groups/x86_64/base-devel/>`__. This includes the
+standard tooling needed to build packages, such as a compiler and binary tools.
 
 .. code-block:: bash
 
-    yaourt -S ocrmypdf
+   sudo pacman -S base-devel
 
-If you have any difficulties with installation, check the repository
-package page.
+Now you are ready to install the OCRmyPDF package.
+
+.. code-block:: bash
+
+   curl -O https://aur.archlinux.org/cgit/aur.git/snapshot/ocrmypdf.tar.gz
+   tar xvzf ocrmypdf.tar.gz
+   cd ocrmypdf
+   makepkg -sri
+
+At this point you will have a working install of OCRmyPDF, but the Tesseract
+install won’t include any OCR language data. You can install `the
+tesseract-data package group
+<https://www.archlinux.org/groups/any/tesseract-data/>`__ to add all supported
+languages, or use that package listing to identify the appropriate package for
+your desired language.
+
+.. code-block:: bash
+
+   sudo pacman -S tesseract-data-eng
+
+As an alternative to this manual procedure, consider using an `AUR helper
+<https://wiki.archlinux.org/index.php/AUR_helpers>`__. Such a tool will
+automatically fetch, build and install the AUR package, resolve dependencies
+(including dependencies on AUR packages), and ease the upgrade procedure.
+
+If you have any difficulties with installation, check the repository package
+page.
+
+.. note::
+
+    The OCRmyPDF AUR package currently omits the JBIG2 encoder. OCRmyPDF works
+    fine without it but will produce larger output files. The encoder is
+    available from `the jbig2enc-git AUR package
+    <https://aur.archlinux.org/packages/jbig2enc-git/>`__ and may be installed
+    using the same series of steps as for the installation OCRmyPDF AUR
+    package. Alternatively, it may be built manually from source following the
+    instructions in `Installing the JBIG2 encoder <jbig2>`__.  If JBIG2 is
+    installed, OCRmyPDF 7.0.0 and later will automatically detect it.
 
 Alpine Linux
 ------------
@@ -375,7 +429,10 @@ languages you can optionally install them all:
 Manual installation on macOS
 ----------------------------
 
-These instructions probably work on all macOS supported by Homebrew.
+These instructions probably work on all macOS supported by Homebrew, and are
+for installing a more current version of OCRmyPDF than is available from
+Homebrew. Note that the Homebrew versions usually track the release versions
+fairly closely.
 
 If it's not already present, `install Homebrew <http://brew.sh/>`__.
 
@@ -386,14 +443,8 @@ Update Homebrew:
     brew update
 
 Install or upgrade the required Homebrew packages, if any are missing.
-To do this, download the ``Brewfile`` that lists all of the dependencies
-to the current directory, and run ``brew bundle`` to process them
-(installing or upgrading as needed). ``Brewfile`` is a plain text file.
-
-.. code-block:: bash
-
-    wget https://github.com/jbarlow83/OCRmyPDF/raw/master/.travis/Brewfile
-    brew bundle
+To do this, use ``brew edit ocrmypdf`` to obtain a recent list of Homebrew
+dependencies. You could also check the ``azure-pipelines.yml``.
 
 This will include the English, French, German and Spanish language
 packs. If you need other languages you can optionally install them all:
@@ -435,12 +486,12 @@ Installing on FreeBSD
     :alt: FreeBSD
     :target: https://repology.org/project/python:ocrmypdf/versions
 
-FreeBSD 11.2, 11.3, 12.0-RELEASE and 13.0-CURRENT are supported. Other
+FreeBSD 11.3, 12.0, 12.1-RELEASE and 13.0-CURRENT are supported. Other
 versions likely work but have not been tested.
 
 .. code-block:: bash
 
-    pkg install py36-ocrmypdf
+    pkg install py37-ocrmypdf
 
 To install a more recent version, you could attempt to first install the system
 version with ``pkg``, then use ``pip install --user ocrmypdf``.
@@ -507,7 +558,7 @@ Installing on Windows Subsystem for Linux
 
 .. code-block:: powershell
 
-    wsl sudo ln -s  /home/user/.local/bin/ocrmypdf /usr/local/bin/ocrmypdf
+    wsl sudo ln -s  /home/$USER/.local/bin/ocrmypdf /usr/local/bin/ocrmypdf
 
 Then confirm that the expected version from PyPI (|latest|) is installed:
 
@@ -524,6 +575,53 @@ Docker
 You can also :ref:`Install the Docker <docker-install>` container on Windows. Ensure that
 your command prompt can run the docker "hello world" container.
 
+Installing on Cygwin64 under Windows
+====================================
+
+First install the the following prerequisite Cygwin packages using ``setup-x86_64.exe``::
+
+    python36 (or later)
+    python3?-devel
+    python3?-pip
+    python3?-lxml
+    python3?-imaging
+
+       (where 3? means match the version of python3 you installed)
+
+    gcc-g++
+    ghostscript (<=9.50 or >=9.52-2 see note below)
+    libexempi3
+    libexempi-devel
+    libffi6
+    libffi-devel
+    pngquant
+    qpdf
+    libqpdf-devel
+    tesseract-ocr
+    tesseract-ocr-devel
+
+.. note::
+
+    The Cygwin package for Ghostscript in versions 9.52 and
+    9.52-1 contained a bug that caused an exception to occur when
+    ocrmypdf invoked gs.  Make sure you have either 9.50 (or earlier)
+    or 9.52-2 (or later).
+
+Then open a Cygwin terminal (i.e. ``mintty``), run the following commands. Note
+that if you are using the version of ``pip`` that was installed with the Cygwin
+Python package, the command name will be ``pip3``.  If you have since updated
+``pip`` (with, for instance ``pip3 install --upgrade pip``) the the command is
+likely just ``pip`` instead of ``pip3``:
+
+.. code-block:: bash
+
+    pip3 install wheel
+    pip3 install ocrmypdf
+
+The optional dependency "unpaper" that is currently not available under Cygwin.
+Without it, certain options such as ``--clean`` will produce an error message.
+However, the OCR-to-text-layer functionality is available.
+
 Installing with Python pip
 ==========================
 

docs/introduction.rst

@@ -10,7 +10,7 @@ About OCR
 `Optical character
 recognition <https://en.wikipedia.org/wiki/Optical_character_recognition>`__
 is technology that converts images of typed or handwritten text, such as
-in a scanned document, to computer text that can be searched and copied.
+in a scanned document, to computer text that can be selected, searched and copied.
 
 OCRmyPDF uses
 `Tesseract <https://github.com/tesseract-ocr/tesseract>`__, the best

docs/languages.rst

@@ -4,11 +4,20 @@
 Installing additional language packs
 ====================================
 
-OCRmyPDF uses Tesseract for OCR, and relies on its language packs for
-languages other than English.
+OCRmyPDF uses Tesseract for OCR, and relies on its language packs for all languages.
+On most platforms, English is installed with Tesseract by default, but not always.
 
 Tesseract supports `most
 languages <https://github.com/tesseract-ocr/tesseract/blob/master/doc/tesseract.1.asc#languages>`__.
+Languages are identified by standardized three-letter codes (called ISO 639-2 Alpha-3).
+Tesseract's documentation also lists the three-letter code for your language.
+Some are anglicized, e.g. Spanish is ``spa`` rather than ``esp``, while others
+are not, e.g. German is ``deu``.
+
+After you have installed a language pack, you can use it ``ocrmypdf -l <language>``,
+for example ``ocrmypdf -l spa``. For multilingual documents, you can specify
+all languages to be expected, e.g. ``ocrmypdf -l eng+fra`` for English and French.
+English is assumed by default unless other language(s) are specified.
 
 For Linux users, you can often find packages that provide language
 packs:

docs/performance.rst

@@ -0,0 +1,22 @@
+===========
+Performance
+===========
+
+Some users have noticed that current versions of OCRmyPDF do not run as quickly
+as some older versions (specifically 6.x and older). This is because OCRmyPDF
+added image optimization as a postprocessing step, and it is enabled by default.
+
+Speed
+=====
+
+If running OCRmyPDF quickly is your main goal, you can use settings such as:
+
+* ``--optimize 0`` to disable file size optimization
+* ``--output-type pdf`` to disable PDF/A generation
+* ``--fast-web-view 0`` to disable fast web view optimization
+* ``--skip-big`` to skip large images, if some pages have large images
+
+You can also avoid:
+
+* ``--force-ocr``
+* Image preprocessing

docs/release_notes.rst

@@ -13,6 +13,115 @@ Note that it is licensed under GPLv3, so scripts that
 ``import ocrmypdf`` and are released publicly should probably also be
 licensed under GPLv3.
 
+v9.8.2
+======
+
+-  Fixed an issue where OCRmyPDF would ignore text inside Form XObject when
+   making certain decisions about whether a document already had text.
+-  Fixed file size increase warning to take overhead of small files into account.
+-  Added instructions for installing on Cygwin.
+
+v9.8.1
+======
+
+-  Fixed an issue where unexpected files in the ``%PROGRAMFILES%\gs`` directory
+   (Windows) caused an exception.
+-  Mark pdfminer.six 20200517 as supported.
+-  If jbig2enc is missing and optimization is requested, a warning is issued
+   instead of an error, which was the intended behavior.
+-  Documentation updates.
+
+v9.8.0
+======
+
+-  Fixed issue where only the first PNG (FlateDecode) image in a file would be
+   considered for optimization. File sizes should be improved from here on.
+-  Fixed a startup crash when the chosen language was Japanese (#543).
+-  Added options to configure polling and log level to watcher.py.
+
+v9.7.2
+======
+
+-  Fixed an issue with ``ocrmypdf.ocr(...language=)`` not accepting a list of
+   languages as documented.
+-  Updated setup.py to confirm that pdfminer.six version 20200402 is supported.
+
+v9.7.1
+======
+
+-  Fixed version check failing when used with qpdf 10.0.0.
+-  Added some missing type annotations.
+-  Updated documentation to warn about need for "ifmain" guard and Windows.
+
+v9.7.0
+======
+
+-  Fixed an error in watcher.py if ``OCR_JSON_SETTINGS`` was not defined.
+-  Ghostscript 9.51 is now blacklisted, due to numerous problems with this version.
+-  Added a workaround for a problem with "txtwrite" in Ghostscript 9.52.
+-  Fixed an issue where the incorrect number of threads used was shown when
+   ``OMP_THREAD_LIMIT`` was manipulated.
+-  Removed a possible performance bottlenecks for files that use hundreds to
+   thousands of images on the same page.
+-  Documentation improvements.
+-  Optimization will now be applied to some monochrome images that have a color
+   profile defined instead of only black and white.
+-  ICC profiles are consulted when determining the simplified colorspace of an
+   image.
+
+v9.6.1
+======
+
+-  Documentation improvements - thanks to many users for their contributions!
+
+      - Fixed installation instructions for ArchLinux (@pigmonkey)
+      - Updated installation instructions for FreeBSD and other OSes (@knobix)
+      - Added instructions for using Docker Compose with watchdog (@ianalexander,
+        @deisi)
+      - Other miscellany (@mb720, @toy, @caiofacchinato)
+      - Some scripts provided in the documentation have been migrated out so that
+        they can be copied out as whole files, and to ensure syntax checking
+        is maintained.
+
+-  Fixed an error that caused bash completions to fail on macOS. (#502, #504;
+   @AlexanderWillner)
+-  Fixed a rare case where OCRmyPDF threw an exception while processing a PDF
+   with the wrong object type in its ``/Trailer /Info``. The error is now logged
+   and incorrect object is ignored. (#497)
+-  Removed potentially non-free file ``enron1.pdf`` and simplified the test that
+   used it.
+-  Removed potentially non-free file ``misc/media/logo.afdesign``.
+
+v9.6.0
+======
+
+-  Fixed a regression with transferring metadata from the input PDF to the output
+   PDF in certain situations.
+-  pdfminer.six is now supported up to version 2020-01-24.
+-  Messages are explaining page rotation decisions are now shown at the standard
+   verbosity level again when ``--rotate-pages``. In some previous version they
+   were set to debug level messages that only appeared with the parameter ``-v1``.
+-  Improvements to ``misc/watcher.py``. Thanks to @ianalexander and @svenihoney.
+-  Documentation improvements.
+
+v9.5.0
+======
+
+-  Added API functions to measure OCR quality.
+-  Modest improvements to handling PDFs with difficult/non compliant metadata.
+
+v9.4.0
+======
+
+-  Updated recommended dependency versions.
+-  Improvements to test coverage and changes to facilitate better measurement of
+   test coverage, such as when tests run in subprocesses.
+-  Improvements to error messages when Leptonica is not installed correctly.
+-  Fixed use of pytest "session scope" that may have caused some intermittent
+   CI failures.
+-  When the argument ``--keep-temporary-files`` or verbosity is set to ``-v1``,
+   a debug log file is generated in the working temporary folder.
+
 v9.3.0
 ======
 

misc/batch.py

@@ -0,0 +1,50 @@
+#!/usr/bin/env python3
+# Original version by DeliciousPickle@github; modified
+
+# This script must be edited to meet your needs.
+
+import logging
+import os
+import sys
+
+import ocrmypdf
+
+# pylint: disable=logging-format-interpolation
+# pylint: disable=logging-not-lazy
+
+script_dir = os.path.dirname(os.path.realpath(__file__))
+print(script_dir + '/batch.py: Start')
+
+if len(sys.argv) > 1:
+    start_dir = sys.argv[1]
+else:
+    start_dir = '.'
+
+if len(sys.argv) > 2:
+    log_file = sys.argv[2]
+else:
+    log_file = script_dir + '/ocr-tree.log'
+
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s %(message)s',
+    filename=log_file,
+    filemode='w',
+)
+
+ocrmypdf.configure_logging(ocrmypdf.Verbosity.default)
+
+for dir_name, subdirs, file_list in os.walk(start_dir):
+    logging.info(dir_name + '\n')
+    os.chdir(dir_name)
+    for filename in file_list:
+        file_ext = os.path.splitext(filename)[1]
+        if file_ext == '.pdf':
+            full_path = dir_name + '/' + filename
+            print(full_path)
+            result = ocrmypdf.ocr(filename, filename, deskew=True)
+            if result == ocrmypdf.ExitCode.already_done_ocr:
+                print("Skipped document because it already contained text")
+            elif result == ocrmypdf.ExitCode.ok:
+                print("OCR complete")
+            logging.info(result)

misc/completion/ocrmypdf.bash

@@ -5,7 +5,33 @@ set -o errexit
 _ocrmypdf()
 {
     local cur prev cword words split
-    _init_completion -s || return
+
+    # Homebrew on Macs have version 1.3 of bash-completion which doesn't include - see #502
+    if declare -F _init_completions >/dev/null 2>&1; then
+        _init_completion -s || return
+    else
+        COMPREPLY=()
+        _get_comp_words_by_ref cur prev words cword
+    fi
+
+    if [[ $cur == -* ]]; then
+        COMPREPLY=( $( compgen -W '--language --image-dpi --output-type
+            --sidecar --version --jobs --quiet --verbose --title --author
+            --subject --keywords --rotate-pages --remove-background --deskew
+            --clean --clean-final --unpaper-args --oversample --remove-vectors
+            --threshold --force-ocr --skip-text --redo-ocr
+            --skip-big --jpeg-quality --png-quality --jbig2-lossy
+            --max-image-mpixels --tesseract-config --tesseract-pagesegmode
+            --help --tesseract-oem --pdf-renderer --tesseract-timeout
+            --rotate-pages-threshold --pdfa-image-compression --user-words
+            --user-patterns --keep-temporary-files --output-type
+            --no-progress-bar --pages --fast-web-view' \
+            --  "$cur" ) )
+        return
+    else
+        _filedir
+        return
+    fi
 
     case $prev in
         --version|-h|--help)
@@ -65,25 +91,6 @@ _ocrmypdf()
     esac
 
     $split && return
-
-    if [[ $cur == -* ]]; then
-        COMPREPLY=( $( compgen -W '--language --image-dpi --output-type
-            --sidecar --version --jobs --quiet --verbose --title --author
-            --subject --keywords --rotate-pages --remove-background --deskew
-            --clean --clean-final --unpaper-args --oversample --remove-vectors
-            --threshold --force-ocr --skip-text --redo-ocr
-            --skip-big --jpeg-quality --png-quality --jbig2-lossy
-            --max-image-mpixels --tesseract-config --tesseract-pagesegmode
-            --help --tesseract-oem --pdf-renderer --tesseract-timeout
-            --rotate-pages-threshold --pdfa-image-compression --user-words
-            --user-patterns --keep-temporary-files --output-type
-            --no-progress-bar --pages --fast-web-view' \
-            --  "$cur" ) )
-        return
-    else
-        _filedir
-        return
-    fi
 } &&
 complete -F _ocrmypdf ocrmypdf
 

misc/docker-compose.example.yml

@@ -0,0 +1,15 @@
+---
+version: "3.3"
+services:
+  ocrmypdf:
+    restart: always
+    container_name: ocrmypdf
+    image: jbarlow83/ocrmypdf
+    volumes:
+      - "/media/scan:/input"
+      - "/mnt/scan:/output"
+    environment:
+      - OCR_OUTPUT_DIRECTORY_YEAR_MONT=0
+    user: "<SET TO YOUR USER ID>:<SET TO YOUR GROUP ID>"
+    entrypoint: python3
+    command: watcher.py

misc/synology.py

@@ -0,0 +1,72 @@
+#!/bin/env python3
+# Contributed by github.com/Enantiomerie
+
+# This script must be edited to meet your needs.
+
+import logging
+import os
+import shutil
+import subprocess
+import sys
+import time
+
+# pylint: disable=logging-format-interpolation
+# pylint: disable=logging-not-lazy
+
+script_dir = os.path.dirname(os.path.realpath(__file__))
+timestamp = time.strftime("%Y-%m-%d-%H%M_")
+log_file = script_dir + '/' + timestamp + 'ocrmypdf.log'
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s %(message)s',
+    filename=log_file,
+    filemode='w',
+)
+
+if len(sys.argv) > 1:
+    start_dir = sys.argv[1]
+else:
+    start_dir = '.'
+
+for dir_name, subdirs, file_list in os.walk(start_dir):
+    logging.info(dir_name)
+    os.chdir(dir_name)
+    for filename in file_list:
+        file_stem, file_ext = os.path.splitext(filename)
+        if file_ext != '.pdf':
+            continue
+        full_path = os.path.join(dir_name, filename)
+        timestamp_ocr = time.strftime("%Y-%m-%d-%H%M_OCR_")
+        filename_ocr = timestamp_ocr + file_stem + '.pdf'
+        # create string for pdf processing
+        # the script is processed as root user via chron
+        cmd = [
+            'docker',
+            'run',
+            '--rm',
+            '-i',
+            'jbarlow83/ocrmypdf',
+            '--deskew',
+            '-',
+            '-',
+        ]
+        logging.info(cmd)
+        full_path_ocr = os.path.join(dir_name, filename_ocr)
+        with open(filename, 'rb') as input_file, open(
+            full_path_ocr, 'wb'
+        ) as output_file:
+            proc = subprocess.run(
+                cmd,
+                stdin=input_file,
+                stdout=output_file,
+                stderr=subprocess.PIPE,
+                check=False,
+            )
+        logging.info(proc.stderr.read())
+        os.chmod(full_path_ocr, 0o664)
+        os.chmod(full_path, 0o664)
+        full_path_ocr_archive = sys.argv[2]
+        full_path_archive = sys.argv[2] + '/no_ocr'
+        shutil.move(full_path_ocr, full_path_ocr_archive)
+        shutil.move(full_path, full_path_archive)
+logging.info('Finished.\n')

misc/watcher.py

@@ -1,4 +1,5 @@
 # Copyright (C) 2019 Ian Alexander: https://github.com/ianalexander
+# Copyright (C) 2020 James R Barlow: https://github.com/jbarlow83
 #
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -13,53 +14,134 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
+import json
+import logging
 import os
+import sys
 import time
 from datetime import datetime
 from pathlib import Path
 
+import pikepdf
 from watchdog.events import PatternMatchingEventHandler
 from watchdog.observers import Observer
+from watchdog.observers.polling import PollingObserver
 
 import ocrmypdf
 
+# pylint: disable=logging-format-interpolation
+
 INPUT_DIRECTORY = os.getenv('OCR_INPUT_DIRECTORY', '/input')
 OUTPUT_DIRECTORY = os.getenv('OCR_OUTPUT_DIRECTORY', '/output')
 OUTPUT_DIRECTORY_YEAR_MONTH = bool(os.getenv('OCR_OUTPUT_DIRECTORY_YEAR_MONTH', False))
+ON_SUCCESS_DELETE = bool(os.getenv('OCR_ON_SUCCESS_DELETE', False))
+DESKEW = bool(os.getenv('OCR_DESKEW', False))
+OCR_JSON_SETTINGS = json.loads(os.getenv('OCR_JSON_SETTINGS', '{}'))
+POLL_NEW_FILE_SECONDS = os.getenv('OCR_POLL_NEW_FILE_SECONDS', 1)
+USE_POLLING = bool(os.getenv('OCR_USE_POLLING', False))
+LOGLEVEL = os.getenv('OCR_LOGLEVEL', 'INFO').upper()
 PATTERNS = ['*.pdf']
 
+log = logging.getLogger('ocrmypdf-watcher')
 
-def execute_ocrmypdf(file_path):
-    filename = Path(file_path).name
+
+def get_output_dir(root, basename):
     if OUTPUT_DIRECTORY_YEAR_MONTH:
         today = datetime.today()
-        output_directory_year_month = Path(
-            f'{OUTPUT_DIRECTORY}/{today.year}/{today.month}'
+        output_directory_year_month = (
+            Path(root) / str(today.year) / f'{today.month:02d}'
         )
         if not output_directory_year_month.exists():
             output_directory_year_month.mkdir(parents=True, exist_ok=True)
-        output_path = Path(output_directory_year_month) / filename
+        output_path = Path(output_directory_year_month) / basename
     else:
-        output_path = Path(OUTPUT_DIRECTORY) / filename
-    print(f'New file: {file_path}.\nAttempting to OCRmyPDF to: {output_path}')
-    ocrmypdf.ocr(file_path, output_path)
+        output_path = Path(OUTPUT_DIRECTORY) / basename
+    return output_path
+
+
+def wait_for_file_ready(file_path):
+    # This loop waits to make sure that the file is completely loaded on
+    # disk before attempting to read. Docker sometimes will publish the
+    # watchdog event before the file is actually fully on disk, causing
+    # pikepdf to fail.
+
+    retries = 5
+    while retries:
+        try:
+            pdf = pikepdf.open(file_path)
+        except (FileNotFoundError, pikepdf.PdfError) as e:
+            log.info(f"File {file_path} is not ready yet")
+            log.debug("Exception was", exc_info=e)
+            time.sleep(POLL_NEW_FILE_SECONDS)
+            retries -= 1
+        else:
+            pdf.close()
+            return True
+
+    return False
+
+
+def execute_ocrmypdf(file_path):
+    file_path = Path(file_path)
+    output_path = get_output_dir(OUTPUT_DIRECTORY, file_path.name)
+
+    log.info("-" * 20)
+    log.info(f'New file: {file_path}. Waiting until fully loaded...')
+    if not wait_for_file_ready(file_path):
+        log.info(f"Gave up waiting for {file_path} to become ready")
+        return
+    log.info(f'Attempting to OCRmyPDF to: {output_path}')
+    exit_code = ocrmypdf.ocr(
+        input_file=file_path,
+        output_file=output_path,
+        deskew=DESKEW,
+        **OCR_JSON_SETTINGS,
+    )
+    if exit_code == 0 and ON_SUCCESS_DELETE:
+        log.info(f'OCR is done. Deleting: {file_path}')
+        file_path.unlink()
+    else:
+        log.info('OCR is done')
 
 
 class HandleObserverEvent(PatternMatchingEventHandler):
     def on_any_event(self, event):
-        if event.event_type in ['created', 'modified']:
+        if event.event_type in ['created']:
             execute_ocrmypdf(event.src_path)
 
 
-if __name__ == "__main__":
-    print(
+def main():
+    ocrmypdf.configure_logging(
+        verbosity=ocrmypdf.Verbosity.default, manage_root_logger=True
+    )
+    log.setLevel(LOGLEVEL)
+    log.info(
         f"Starting OCRmyPDF watcher with config:\n"
         f"Input Directory: {INPUT_DIRECTORY}\n"
         f"Output Directory: {OUTPUT_DIRECTORY}\n"
         f"Output Directory Year & Month: {OUTPUT_DIRECTORY_YEAR_MONTH}"
     )
+    log.debug(
+        f"INPUT_DIRECTORY: {INPUT_DIRECTORY}\n"
+        f"OUTPUT_DIRECTORY: {OUTPUT_DIRECTORY}\n"
+        f"OUTPUT_DIRECTORY_YEAR_MONTH: {OUTPUT_DIRECTORY_YEAR_MONTH}\n"
+        f"ON_SUCCESS_DELETE: {ON_SUCCESS_DELETE}\n"
+        f"DESKEW: {DESKEW}\n"
+        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"
+    )
+
+    if 'input_file' in OCR_JSON_SETTINGS or 'output_file' in OCR_JSON_SETTINGS:
+        log.error('OCR_JSON_SETTINGS should not specify input file or output file')
+        sys.exit(1)
+
     handler = HandleObserverEvent(patterns=PATTERNS)
-    observer = Observer()
+    if USE_POLLING:
+        observer = PollingObserver()
+    else:
+        observer = Observer()
     observer.schedule(handler, INPUT_DIRECTORY, recursive=True)
     observer.start()
     try:
@@ -68,3 +150,7 @@ if __name__ == "__main__":
     except KeyboardInterrupt:
         observer.stop()
     observer.join()
+
+
+if __name__ == "__main__":
+    main()

requirements/dev.txt

@@ -1,2 +0,0 @@
-twine >= 1.8.1
-coverage >= 4.5

requirements/main.txt

@@ -1,10 +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
-cffi == 1.13.2
-img2pdf == 0.3.3
-pdfminer.six == 20191110
-pikepdf == 1.8.1
-Pillow >= 6.2.0
-reportlab == 3.5.32
-tqdm == 4.37.0
+cffi == 1.14.0
+img2pdf == 0.3.4
+pdfminer.six == 20200402
+pikepdf == 1.11.1
+Pillow == 7.1.1
+reportlab == 3.5.34
+tqdm == 4.45.0

requirements/test.txt

@@ -1,8 +1,7 @@
 pytest >= 5.0.0
 pytest-helpers-namespace >= 2019.1.8
-pytest-xdist >= 1.29.0  # For DumpError fix
-pytest-cov >= 2.6.1
+pytest-xdist >= 1.31.0
+pytest-cov >= 2.8.0
 python-xmp-toolkit == 2.0.1  # requires apt-get install libexempi3
                      # or brew install exempi
-PyPDF2 >= 1.26.0
 #PyMuPDF == 1.13.4    # optional

setup.cfg

@@ -23,7 +23,7 @@ force_grid_wrap=0
 use_parentheses=True
 line_length=88
 known_first_party = ocrmypdf
-known_third_party = PIL,PyPDF2,_cffi_backend,cffi,flask,gs,img2pdf,pdfminer,pikepdf,pkg_resources,pytest,reportlab,setuptools,sphinx_rtd_theme,tqdm,watchdog,werkzeug
+known_third_party = PIL,_cffi_backend,cffi,flask,gs,img2pdf,pdfminer,pikepdf,pkg_resources,pytest,reportlab,setuptools,sphinx_rtd_theme,tqdm,watchdog,werkzeug
 
 [metadata]
 license_file = LICENSE

setup.py

@@ -21,11 +21,12 @@ from __future__ import print_function, unicode_literals
 
 import sys
 
+from setuptools import find_packages, setup
+
 if sys.version_info < (3, 6):
     print("Python 3.6 or newer is required", file=sys.stderr)
     sys.exit(1)
 
-from setuptools import setup, find_packages
 
 # pylint: disable=w0613
 
@@ -61,7 +62,7 @@ setup(
     long_description_content_type='text/markdown',
     url='https://github.com/jbarlow83/OCRmyPDF',
     author='James R. Barlow',
-    author_email='jim@purplerock.ca',
+    author_email='james@purplerock.ca',
     packages=find_packages('src', exclude=["tests", "tests.*"]),
     package_dir={'': 'src'},
     keywords=['PDF', 'OCR', 'optical character recognition', 'PDF/A', 'scanning'],
@@ -97,7 +98,7 @@ 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, <= 20191110',
+        'pdfminer.six >= 20181108, <= 20200517',
         'pikepdf >= 1.8.1, < 2',
         'Pillow >= 6.2.0',
         'reportlab >= 3.3.0',  # oldest released version with sane image handling

src/ocrmypdf/__init__.py

@@ -23,6 +23,7 @@ from .exceptions import (
     DpiError,
     EncryptedPdfError,
     ExitCode,
+    ExitCodeException,
     InputFileError,
     MissingDependencyError,
     OutputFileAccessError,

src/ocrmypdf/_pipeline.py

@@ -19,6 +19,7 @@ import os
 import re
 import sys
 from datetime import datetime, timezone
+from pathlib import Path
 from shutil import copyfileobj
 
 import img2pdf
@@ -123,7 +124,7 @@ def _pdf_guess_version(input_file, search_window=1024):
     return ''
 
 
-def triage(input_file, output_file, options, log):
+def triage(original_filename, input_file, output_file, options, log):
     try:
         if _pdf_guess_version(input_file):
             if options.image_dpi:
@@ -135,8 +136,9 @@ def triage(input_file, output_file, options, log):
             safe_symlink(input_file, output_file)
             return output_file
     except EnvironmentError as e:
-        log.error(e)
-        raise InputFileError() from e
+        log.debug(f"Temporary file was at: {input_file}")
+        msg = str(e).replace(input_file, original_filename)
+        raise InputFileError(msg) from e
 
     triage_image_file(input_file, output_file, options, log)
     return output_file
@@ -181,7 +183,7 @@ def validate_pdfinfo_options(context):
             )
             raise InputFileError()
         else:
-            log.warn(
+            log.warning(
                 "This PDF has a fillable form. "
                 "Chances are it is a pure digital "
                 "document that does not need OCR."
@@ -327,6 +329,35 @@ def rasterize_preview(input_file, page_context):
     return output_file
 
 
+def describe_rotation(page_context, orient_conf, correction):
+    """
+    Describe the page rotation we are going to perform.
+    """
+    direction = {0: '⇧', 90: '⇨', 180: '⇩', 270: '⇦'}
+    turns = {0: ' ', 90: '⬏', 180: '↻', 270: '⬑'}
+
+    existing_rotation = page_context.pageinfo.rotation
+    action = ''
+    if orient_conf.confidence >= page_context.options.rotate_pages_threshold:
+        if correction != 0:
+            action = 'will rotate ' + turns[correction]
+        else:
+            action = 'rotation appears correct'
+    else:
+        if correction != 0:
+            action = 'confidence too low to rotate'
+        else:
+            action = 'no change'
+
+    facing = ''
+
+    if existing_rotation != 0:
+        facing = f"with existing rotation {direction.get(existing_rotation, '?')}, "
+    facing += f"page is facing {direction.get(orient_conf.angle, '?')}"
+
+    return f"{facing}, confidence {orient_conf.confidence:.2f} - {action}"
+
+
 def get_orientation_correction(preview, page_context):
     """
     Work out orientation correct for each page.
@@ -353,44 +384,14 @@ def get_orientation_correction(preview, page_context):
         tesseract_env=page_context.options.tesseract_env,
     )
 
-    direction = {0: '⇧', 90: '⇨', 180: '⇩', 270: '⇦'}
-
-    existing_rotation = page_context.pageinfo.rotation
-
     correction = orient_conf.angle % 360
-
-    apply_correction = False
-    action = ''
-    if orient_conf.confidence >= page_context.options.rotate_pages_threshold:
-        if correction != 0:
-            apply_correction = True
-            action = ' - will rotate'
-        else:
-            action = ' - rotation appears correct'
-    else:
-        if correction != 0:
-            action = ' - confidence too low to rotate'
-        else:
-            action = ' - no change'
-
-    facing = ''
-    if existing_rotation != 0:
-        facing = 'with existing rotation {}, '.format(
-            direction.get(existing_rotation, '?')
-        )
-    facing += 'page is facing {}'.format(direction.get(orient_conf.angle, '?'))
-
-    page_context.log.debug(
-        '{pagenum:4d}: {facing}, confidence {conf:.2f}{action}'.format(
-            pagenum=page_context.pageinfo.pageno,
-            facing=facing,
-            conf=orient_conf.confidence,
-            action=action,
-        )
-    )
-
-    if apply_correction:
+    page_context.log.info(describe_rotation(page_context, orient_conf, correction))
+    if (
+        orient_conf.confidence >= page_context.options.rotate_pages_threshold
+        and correction != 0
+    ):
         return correction
+
     return 0
 
 
@@ -693,11 +694,18 @@ def convert_to_pdfa(input_pdf, input_ps_stub, context):
     # stamping them out as soon as possible.
     modified = False
     with pikepdf.open(input_pdf) as pdf_file:
-        if pdf_file.docinfo:
-            for k, v in pdf_file.docinfo.items():
-                if b'\x00' in bytes(v):
-                    pdf_file.docinfo[k] = bytes(v).replace(b'\x00', b'')
-                    modified = True
+        try:
+            len(pdf_file.docinfo)
+        except TypeError:
+            context.log.error(
+                "File contains a malformed DocumentInfo block - continuing anyway"
+            )
+        else:
+            if pdf_file.docinfo:
+                for k, v in pdf_file.docinfo.items():
+                    if b'\x00' in bytes(v):
+                        pdf_file.docinfo[k] = bytes(v).replace(b'\x00', b'')
+                        modified = True
         if modified:
             pdf_file.save(fix_docinfo_file)
         else:
@@ -725,47 +733,53 @@ def should_linearize(working_file, context):
 def metadata_fixup(working_file, context):
     output_file = context.get_path('metafix.pdf')
     options = context.options
-    original = pikepdf.open(context.origin)
-    docinfo = get_docinfo(original, options)
-    pdf = pikepdf.open(working_file)
-    with pdf.open_metadata() as meta:
-        meta.load_from_docinfo(docinfo, delete_missing=False)
-        # If xmp:CreateDate is missing, set it to the modify date to
-        # match Ghostscript, for consistency
-        if 'xmp:CreateDate' not in meta:
-            meta['xmp:CreateDate'] = meta.get('xmp:ModifyDate', '')
 
-        meta_original = original.open_metadata()
-        not_copied = set(meta_original.keys()) - set(meta.keys())
-        if not_copied:
-            if options.output_type.startswith('pdfa'):
-                context.log.warning(
-                    "Some input metadata could not be copied because it is not "
-                    "permitted in PDF/A. You may wish to examine the output "
-                    "PDF's XMP metadata."
-                )
-                context.log.debug(
-                    "The following metadata fields were not copied: %r", not_copied
-                )
-            else:
-                context.log.error(
-                    "Some input metadata could not be copied."
-                    "You may wish to examine the output PDF's XMP metadata."
-                )
-                context.log.info(
-                    "The following metadata fields were not copied: %r", not_copied
-                )
-    pdf.save(
-        output_file,
-        compress_streams=True,
-        preserve_pdfa=True,
-        object_stream_mode=pikepdf.ObjectStreamMode.generate,
-        linearize=(  # Don't linearize if optimize() will be linearizing too
-            should_linearize(working_file, context) if options.optimize == 0 else False
-        ),
-    )
-    original.close()
-    pdf.close()
+    def report_on_metadata(missing):
+        if not missing:
+            return
+        if options.output_type.startswith('pdfa'):
+            context.log.warning(
+                "Some input metadata could not be copied because it is not "
+                "permitted in PDF/A. You may wish to examine the output "
+                "PDF's XMP metadata."
+            )
+            context.log.debug(
+                "The following metadata fields were not copied: %r", missing
+            )
+        else:
+            context.log.error(
+                "Some input metadata could not be copied."
+                "You may wish to examine the output PDF's XMP metadata."
+            )
+            context.log.info(
+                "The following metadata fields were not copied: %r", missing
+            )
+
+    with pikepdf.open(context.origin) as original, pikepdf.open(working_file) as pdf:
+        docinfo = get_docinfo(original, options)
+        with pdf.open_metadata() as meta:
+            meta.load_from_docinfo(docinfo, delete_missing=False, raise_failure=False)
+            # If xmp:CreateDate is missing, set it to the modify date to
+            # match Ghostscript, for consistency
+            if 'xmp:CreateDate' not in meta:
+                meta['xmp:CreateDate'] = meta.get('xmp:ModifyDate', '')
+
+            meta_original = original.open_metadata()
+            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,
+            linearize=(  # Don't linearize if optimize() will be linearizing too
+                should_linearize(working_file, context)
+                if options.optimize == 0
+                else False
+            ),
+        )
+
     return output_file
 
 

src/ocrmypdf/_sync.py

@@ -23,6 +23,7 @@ import signal
 import sys
 import threading
 from collections import namedtuple
+from pathlib import Path
 from tempfile import mkdtemp
 
 import PIL
@@ -230,7 +231,7 @@ def exec_concurrent(context):
     # Run exec_page_sync on every page context
     max_workers = min(len(context.pdfinfo), context.options.jobs)
     if max_workers > 1:
-        context.log.info("Start processing %d pages concurrent", max_workers)
+        context.log.info("Start processing %d pages concurrently", max_workers)
 
     # 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.
@@ -244,6 +245,10 @@ def exec_concurrent(context):
     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))
+    try:
+        tess_threads = int(context.options.tesseract_env['OMP_THREAD_LIMIT'])
+    except ValueError:  # OMP_THREAD_LIMIT initialized to non-numeric
+        context.log.error("Environment variable OMP_THREAD_LIMIT is not numeric")
     if tess_threads > 1:
         context.log.info("Using Tesseract OpenMP thread limit %d", tess_threads)
 
@@ -267,28 +272,43 @@ def exec_concurrent(context):
         unit='page',
         unit_scale=0.5,
         disable=not context.options.progress_bar,
-    ) as pbar, Pool(
-        processes=max_workers,
-        initializer=initializer,
-        initargs=(log_queue, PIL.Image.MAX_IMAGE_PIXELS),
-    ) as pool:
-        results = pool.imap_unordered(exec_page_sync, context.get_page_contexts())
-        while True:
-            try:
-                page_result = results.next()
-                sidecars[page_result.pageno] = page_result.text
-                pbar.update()
-                ocrgraft.graft_page(page_result)
-                pbar.update()
-            except StopIteration:
-                break
-            except (Exception, KeyboardInterrupt):
+    ) as pbar:
+        pool = Pool(
+            processes=max_workers,
+            initializer=initializer,
+            initargs=(log_queue, PIL.Image.MAX_IMAGE_PIXELS),
+        )
+        try:
+            results = pool.imap_unordered(exec_page_sync, context.get_page_contexts())
+            while True:
+                try:
+                    page_result = results.next()
+                    sidecars[page_result.pageno] = page_result.text
+                    pbar.update()
+                    ocrgraft.graft_page(page_result)
+                    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()
-                log_queue.put_nowait(None)  # Terminate log listener
-                # Don't try listener.join() here, will deadlock
-                raise
+            raise
+        finally:
+            # Terminate log listener
+            log_queue.put_nowait(None)
+            pool.close()
+            pool.join()
 
-    log_queue.put_nowait(None)
     listener.join()
 
     # Output sidecar text
@@ -320,6 +340,17 @@ def samefile(f1, f2):
         return os.path.samefile(f1, f2)
 
 
+def configure_debug_logging(log_filename, prefix=''):
+    log_file_handler = logging.FileHandler(log_filename, delay=True)
+    log_file_handler.setLevel(logging.DEBUG)
+    formatter = logging.Formatter(
+        '[%(asctime)s] - %(name)s - %(levelname)7s - %(message)s'
+    )
+    log_file_handler.setFormatter(formatter)
+    logging.getLogger(prefix).addHandler(log_file_handler)
+    return log_file_handler
+
+
 def run_pipeline(options, api=False):
     log = make_logger(options, __name__)
 
@@ -330,13 +361,23 @@ def run_pipeline(options, api=False):
         options.jobs = available_cpu_count()
 
     work_folder = mkdtemp(prefix="com.github.ocrmypdf.")
+    debug_log_handler = None
+    if (options.keep_temporary_files or options.verbose >= 1) and not os.environ.get(
+        'PYTEST_CURRENT_TEST', ''
+    ):
+        debug_log_handler = configure_debug_logging(Path(work_folder) / "debug.log")
+
     try:
         check_requested_output_file(options)
-        start_input_file = create_input_file(options, work_folder)
+        start_input_file, original_filename = create_input_file(options, work_folder)
 
         # Triage image or pdf
         origin_pdf = triage(
-            start_input_file, os.path.join(work_folder, 'origin.pdf'), options, log
+            original_filename,
+            start_input_file,
+            os.path.join(work_folder, 'origin.pdf'),
+            options,
+            log,
         )
 
         # Gather pdfinfo and create context
@@ -345,6 +386,7 @@ def run_pipeline(options, api=False):
             detailed_page_analysis=options.redo_ocr,
             progbar=options.progress_bar,
         )
+
         context = PDFContext(options, work_folder, origin_pdf, pdfinfo)
 
         # Validate options are okay for this pdf
@@ -391,6 +433,12 @@ def run_pipeline(options, api=False):
         log.exception("An exception occurred while executing the pipeline")
         return ExitCode.other_error
     finally:
+        if debug_log_handler:
+            try:
+                debug_log_handler.close()
+                log.removeHandler(debug_log_handler)
+            except EnvironmentError as e:
+                print(e, file=sys.stderr)
         cleanup_working_files(work_folder, options)
 
     return ExitCode.ok

src/ocrmypdf/_validation.py

@@ -24,6 +24,7 @@ import sys
 from pathlib import Path
 from shutil import copyfileobj
 
+import pikepdf
 import PIL
 
 from ._unicodefun import verify_python3_env
@@ -314,7 +315,7 @@ def check_options(options):
     check_dependency_versions(options)
 
 
-def check_closed_streams(options):
+def check_closed_streams(options):  # pragma: no cover
     """Work around Python issue with multiprocessing forking on closed streams
 
     https://bugs.python.org/issue28326
@@ -380,12 +381,12 @@ def create_input_file(options, work_folder):
         target = os.path.join(work_folder, 'stdin')
         with open(target, 'wb') as stream_buffer:
             copyfileobj(sys.stdin.buffer, stream_buffer)
-        return target
+        return target, "<stdin>"
     else:
         try:
             target = os.path.join(work_folder, 'origin')
             safe_symlink(options.input_file, target)
-            return target
+            return target, os.fspath(options.input_file)
         except FileNotFoundError:
             raise InputFileError(f"File not found - {options.input_file}")
 
@@ -410,8 +411,15 @@ def report_output_file_size(options, input_file, output_file):
         input_size = Path(input_file).stat().st_size
     except FileNotFoundError:
         return  # Outputting to stream or something
+    with pikepdf.open(output_file) as p:
+        # Overhead constants obtained by estimating amount of data added by OCR
+        # PDF/A conversion, and possible XMP metadata addition, with compression
+        FILE_OVERHEAD = 4000
+        OCR_PER_PAGE_OVERHEAD = 3000
+        reasonable_overhead = FILE_OVERHEAD + OCR_PER_PAGE_OVERHEAD * len(p.pages)
     ratio = output_size / input_size
-    if ratio < 1.35 or input_size < 25000:
+    reasonable_ratio = output_size / (input_size + reasonable_overhead)
+    if reasonable_ratio < 1.35 or input_size < 25000:
         return  # Seems fine
 
     reasons = []
@@ -466,11 +474,12 @@ def check_dependency_versions(options):
         version_checker=ghostscript.version,
         need_version='9.15',  # limited by Travis CI / Ubuntu 14.04 backports
     )
-    if ghostscript.version() == '9.24':
+    gs_version = ghostscript.version()
+    if gs_version in ('9.24', '9.51'):
         raise MissingDependencyError(
-            "Ghostscript 9.24 contains serious regressions and is not "
-            "supported. Please upgrade to Ghostscript 9.25 or use an older "
-            "version."
+            f"Ghostscript {gs_version} contains serious regressions and is not "
+            "supported. Please upgrade to a newer version, or downgrade to the "
+            "previous version."
         )
     check_external_program(
         program='qpdf',

src/ocrmypdf/api.py

@@ -18,10 +18,10 @@
 import logging
 import os
 import sys
-import warnings
+from contextlib import suppress
 from enum import IntEnum
 from pathlib import Path
-from typing import Dict, List, Optional
+from typing import Dict, Iterable
 
 from tqdm import tqdm
 
@@ -31,7 +31,15 @@ from .cli import parser
 
 
 class TqdmConsole:
-    """Wrapper to log messages in a way that is compatible with tqdm progress bar"""
+    """Wrapper to log messages in a way that is compatible with tqdm progress bar
+
+    This routes log messages through tqdm so that it can print them above the
+    progress bar, and then refresh the progress bar, rather than overwriting
+    it which looks messy.
+
+    For some reason Python 3.6 prints extra empty messages from time to time,
+    so we suppress those.
+    """
 
     def __init__(self, file):
         self.file = file
@@ -46,7 +54,7 @@ class TqdmConsole:
             tqdm.write(msg.rstrip(), end='\n', file=self.file)
 
     def flush(self):
-        if hasattr(self.file, "flush"):
+        with suppress(AttributeError):
             self.file.flush()
 
 
@@ -59,7 +67,11 @@ class Verbosity(IntEnum):
     debug_all = 2  #: More detailed debugging from ocrmypdf and dependent modules
 
 
-def configure_logging(verbosity, progress_bar_friendly=True, manage_root_logger=False):
+def configure_logging(
+    verbosity: Verbosity,
+    progress_bar_friendly: bool = True,
+    manage_root_logger: bool = False,
+):
     """Set up logging.
 
     Library users may wish to use this function if they want their log output to be
@@ -80,11 +92,14 @@ def configure_logging(verbosity, progress_bar_friendly=True, manage_root_logger=
             overwrite the progress bar
         manage_root_logger (bool): Configure the process's root logger, to ensure
             all log output is sent through
+
+    Returns:
+        The toplevel logger for ocrmypdf (or the root logger, if we are managing it).
     """
 
     prefix = '' if manage_root_logger else 'ocrmypdf'
     log = logging.getLogger(prefix)
-    log.setLevel(logging.INFO)
+    log.setLevel(logging.DEBUG)
 
     if progress_bar_friendly:
         console = logging.StreamHandler(stream=TqdmConsole(sys.stderr))
@@ -99,8 +114,6 @@ def configure_logging(verbosity, progress_bar_friendly=True, manage_root_logger=
         console.setLevel(logging.INFO)
 
     formatter = logging.Formatter('%(levelname)7s - %(message)s')
-    if verbosity >= 1:
-        log.setLevel(logging.DEBUG)
     if verbosity >= 2:
         formatter = logging.Formatter('%(name)s - %(levelname)7s - %(message)s')
 
@@ -116,8 +129,10 @@ def configure_logging(verbosity, progress_bar_friendly=True, manage_root_logger=
     if manage_root_logger:
         logging.captureWarnings(True)
 
+    return log
 
-def create_options(*, input_file, output_file, **kwargs):
+
+def create_options(*, input_file: os.PathLike, output_file: os.PathLike, **kwargs):
     cmdline = []
     deferred = []
 
@@ -139,6 +154,12 @@ def create_options(*, input_file, output_file, **kwargs):
                 cmdline.append(f"--{cmd_style_arg}")
             continue
 
+        if isinstance(val, Iterable) and not isinstance(val, str):
+            for elem in val:
+                cmdline.append(f"--{cmd_style_arg}")
+                cmdline.append(elem)
+            continue
+
         # We have a parameter
         cmdline.append(f"--{cmd_style_arg}")
         if isinstance(val, (int, float)):
@@ -169,7 +190,7 @@ def ocr(  # pylint: disable=unused-argument
     input_file: os.PathLike,
     output_file: os.PathLike,
     *,
-    language: List[str] = None,
+    language: Iterable[str] = None,
     image_dpi: int = None,
     output_type=None,
     sidecar: os.PathLike = None,
@@ -199,7 +220,7 @@ def ocr(  # pylint: disable=unused-argument
     jbig2_page_group_size: int = None,
     pages: str = None,
     max_image_mpixels: float = None,
-    tesseract_config: List[str] = None,
+    tesseract_config: Iterable[str] = None,
     tesseract_pagesegmode: int = None,
     tesseract_oem: int = None,
     pdf_renderer=None,

src/ocrmypdf/cli.py

@@ -20,6 +20,8 @@ import argparse
 from ._version import PROGRAM_NAME as _PROGRAM_NAME
 from ._version import __version__ as _VERSION
 
+__all__ = ['parser']
+
 
 def numeric(basetype, min_=None, max_=None):
     """Validator for numeric params"""

src/ocrmypdf/exec/__init__.py

@@ -23,13 +23,15 @@ import re
 import shutil
 import sys
 from collections.abc import Mapping
+from distutils.version import LooseVersion
 from functools import lru_cache
+from pathlib import Path
 from subprocess import PIPE, STDOUT, CalledProcessError
 from subprocess import run as subprocess_run
 
 from ..exceptions import ExitCode, MissingDependencyError
 
-log = logging.Logger(__name__)
+log = logging.getLogger(__name__)
 
 
 def _get_program(args, env=None):
@@ -77,12 +79,21 @@ def run(args, *, env=None, **kwargs):
             if new_args0:
                 args[0] = new_args0
 
-    log.debug(args)
+    process_log = log.getChild(os.path.basename(program))
+    process_log.debug("Running: %s", args)
     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
-    return subprocess_run(args, env=env, **kwargs)
+    proc = subprocess_run(args, env=env, **kwargs)
+    if process_log.isEnabledFor(logging.DEBUG):
+        try:
+            stderr = proc.stderr.decode('utf-8', 'replace')
+        except AttributeError:
+            stderr = proc.stderr
+        if stderr:
+            process_log.debug("stderr = %s", stderr)
+    return proc
 
 
 def get_version(program, *, version_arg='--version', regex=r'(\d+(\.\d+)*)', env=None):
@@ -128,24 +139,25 @@ def shim_paths_with_program_files(env=None):
     program_files = env.get('PROGRAMFILES', '')
     if not program_files:
         return env.get('PATH', '')
-    paths = []
-    try:
-        for dirname in os.listdir(program_files):
-            if dirname.lower() == 'tesseract-ocr':
-                paths.append(os.path.join(program_files, dirname))
-            elif dirname.lower() == 'gs':
-                try:
-                    latest_gs = max(
-                        os.listdir(os.path.join(program_files, dirname)),
-                        key=lambda d: float(d[2:]),
-                    )
-                except (FileNotFoundError, NotADirectoryError):
-                    continue
-                paths.append(os.path.join(program_files, dirname, latest_gs, 'bin'))
-    except EnvironmentError:
-        pass
-    paths.extend(path for path in os.get_exec_path(env) if path not in set(paths))
-    return os.pathsep.join(paths)
+
+    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 = '''
@@ -223,10 +235,10 @@ def _error_trailer(program, package, **kwargs):
 
 
 def _error_missing_program(program, package, required_for, recommended):
-    if required_for:
+    if recommended:
+        log.warning(missing_recommend_program.format(**locals()))
+    elif required_for:
         log.error(missing_optional_program.format(**locals()))
-    elif recommended:
-        log.info(missing_recommend_program.format(**locals()))
     else:
         log.error(missing_program.format(**locals()))
     _error_trailer(**locals())
@@ -261,7 +273,15 @@ def check_external_program(
             raise MissingDependencyError()
         return
 
-    if found_version < need_version:
+    def remove_leading_v(s):
+        if s.startswith('v'):
+            return s[1:]
+        return s
+
+    found_version = remove_leading_v(found_version)
+    need_version = remove_leading_v(need_version)
+
+    if found_version and LooseVersion(found_version) < LooseVersion(need_version):
         _error_old_version(program, package, need_version, found_version, required_for)
         if not recommended:
             raise MissingDependencyError()

src/ocrmypdf/exec/ghostscript.py

@@ -195,8 +195,6 @@ def rasterize_pdf(
     try:
         p = run(args_gs, stdout=PIPE, stderr=PIPE, check=True)
     except CalledProcessError as e:
-        with suppress(OSError):
-            Path(output_file).unlink()  # no unfinished files
         log.error(e.stderr.decode(errors='replace'))
         raise SubprocessOutputError('Ghostscript rasterizing failed')
     else:
@@ -313,15 +311,12 @@ def generate_pdfa(
         ]
     )
     args_gs.extend(fspath(s) for s in pdf_pages)  # Stringify Path objs
-    log.debug(args_gs)
     try:
         with Path(output_file).open('wb') as output:
             p = run(args_gs, stdout=output, stderr=PIPE, check=True)
     except CalledProcessError as e:
         # Ghostscript does not change return code when it fails to create
         # PDF/A - check PDF/A status elsewhere
-        with suppress(OSError):
-            Path(output_file).unlink()
         log.error(e.stderr.decode(errors='replace'))
         raise SubprocessOutputError('Ghostscript PDF/A rendering failed')
     else:
@@ -346,5 +341,3 @@ def generate_pdfa(
                 "Ghostscript had to remove PDF 'overprinting' from the "
                 "input file to complete PDF/A conversion. "
             )
-        else:
-            log.debug(stderr)

src/ocrmypdf/exec/tesseract.py

@@ -77,20 +77,14 @@ def has_textonly_pdf(tesseract_env=None, langs=None):
     args_tess = tess_base_args(langs, engine_mode=None) + ['--print-parameters', 'pdf']
     params = ''
     try:
-        proc = run(
-            args_tess,
-            check=True,
-            universal_newlines=True,
-            stdout=PIPE,
-            stderr=STDOUT,
-            env=tesseract_env,
-        )
+        # print-parameters can return non-UTF8 if the parameters are so initialized
+        proc = run(args_tess, check=True, stdout=PIPE, stderr=STDOUT, env=tesseract_env)
         params = proc.stdout
     except CalledProcessError as e:
         raise MissingDependencyError(
             "Could not --print-parameters from tesseract"
         ) from e
-    if 'textonly_pdf' in params:
+    if b'textonly_pdf' in params:
         return True
     return False
 
@@ -260,8 +254,8 @@ def generate_hocr(
     log,
 ):
 
-    output_hocr = next(o for o in output_files if o.endswith('.hocr'))
-    output_sidecar = next(o for o in output_files if o.endswith('.txt'))
+    output_hocr = next(o for o in output_files if fspath(o).endswith('.hocr'))
+    output_sidecar = next(o for o in output_files if fspath(o).endswith('.txt'))
     prefix = os.path.splitext(output_hocr)[0]
 
     args_tesseract = tess_base_args(language, engine_mode)
@@ -275,14 +269,10 @@ 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)
     try:
-        log.debug(args_tesseract)
         p = run(
             args_tesseract,
             stdout=PIPE,
@@ -381,7 +371,6 @@ def generate_pdf(
 
     args_tesseract.extend([input_image, prefix, 'pdf', 'txt'] + tessconfig)
     try:
-        log.debug(args_tesseract)
         p = run(
             args_tesseract,
             stdout=PIPE,

src/ocrmypdf/helpers.py

@@ -28,7 +28,7 @@ from pathlib import Path
 log = logging.getLogger(__name__)
 
 
-def safe_symlink(input_file, soft_link_name, *args, **kwargs):
+def safe_symlink(input_file: os.PathLike, soft_link_name: os.PathLike, *args, **kwargs):
     """
     Helper function: relinks soft symbolic link if necessary
     """
@@ -76,12 +76,12 @@ def is_iterable_notstr(thing):
     return isinstance(thing, Iterable) and not isinstance(thing, str)
 
 
-def monotonic(L):
+def monotonic(L: Iterable):
     """Does list increase monotonically?"""
     return all(b > a for a, b in zip(L, L[1:]))
 
 
-def page_number(input_file):
+def page_number(input_file: os.PathLike):
     """Get one-based page number implied by filename (000002.pdf -> 2)"""
     return int(os.path.basename(os.fspath(input_file))[0:6])
 
@@ -97,35 +97,43 @@ def available_cpu_count():
     return 1
 
 
-def is_file_writable(test_file):
+def is_file_writable(test_file: os.PathLike):
     """Intentionally racy test if target is writable.
 
     We intend to write to the output file if and only if we succeed and
     can replace it atomically. Before doing the OCR work, make sure
     the location is writable.
     """
-    p = Path(test_file)
-
-    if p.is_symlink():
-        p = p.resolve(strict=False)
-
-    # p.is_file() throws an exception in some cases
-    if p.exists() and p.is_file():
-        return os.access(
-            os.fspath(p),
-            os.W_OK,
-            effective_ids=(os.access in os.supports_effective_ids),
-        )
-    else:
-        try:
-            fp = p.open('wb')
-        except OSError:
-            return False
+    try:
+        if not isinstance(test_file, Path):
+            p = Path(test_file)
         else:
-            fp.close()
-            with suppress(OSError):
-                p.unlink()
-        return True
+            p = test_file
+
+        if p.is_symlink():
+            p = p.resolve(strict=False)
+
+        # p.is_file() throws an exception in some cases
+        if p.exists() and p.is_file():
+            return os.access(
+                os.fspath(p),
+                os.W_OK,
+                effective_ids=(os.access in os.supports_effective_ids),
+            )
+        else:
+            try:
+                fp = p.open('wb')
+            except OSError:
+                return False
+            else:
+                fp.close()
+                with suppress(OSError):
+                    p.unlink()
+            return True
+    except (EnvironmentError, RuntimeError) as e:
+        log.debug(e)
+        log.error(str(e))
+        return False
 
 
 def deprecated(func):

src/ocrmypdf/hocrtransform.py

@@ -88,7 +88,7 @@ class HocrTransform:
         if self.width is None or self.height is None:
             raise HocrTransformError("hocr file is missing page dimensions")
 
-    def __str__(self):
+    def __str__(self):  # pragma: no cover
         """
         Return the textual content of the HTML body
         """
@@ -190,7 +190,7 @@ class HocrTransform:
             pt = self.pt_from_pixel(pxl_coords)
 
             # draw the bbox border
-            if showBoundingboxes:
+            if showBoundingboxes:  # pragma: no cover
                 pdf.rect(
                     pt.x1, self.height - pt.y2, pt.x2 - pt.x1, pt.y2 - pt.y1, fill=1
                 )
@@ -231,7 +231,7 @@ class HocrTransform:
         pdf.save()
 
     @classmethod
-    def polyval(cls, poly, x):
+    def polyval(cls, poly, x):  # pragma: no cover
         return x * poly[0] + poly[1]
 
     def _do_line(
@@ -269,7 +269,7 @@ class HocrTransform:
         # of the line box
         baseline_y2 = self.height - (line_box.y2 + intercept)
 
-        if showBoundingboxes:
+        if showBoundingboxes:  # pragma: no cover
             # draw the baseline in magenta, dashed
             pdf.setDash()
             pdf.setStrokeColorRGB(0.95, 0.65, 0.95)
@@ -318,7 +318,7 @@ class HocrTransform:
             font_width = pdf.stringWidth(elemtxt, fontname, fontsize)
 
             # draw the bbox border
-            if showBoundingboxes:
+            if showBoundingboxes:  # pragma: no cover
                 pdf.rect(
                     box.x1, self.height - line_box.y2, box_width, line_height, fill=0
                 )

src/ocrmypdf/leptonica.py

@@ -47,21 +47,32 @@ if os.name == 'nt':
 else:
     libname = 'lept'
 _libpath = find_library(libname)
-if not _libpath and os.name == 'nt':
+if not _libpath:
     raise MissingDependencyError(
         """
         ---------------------------------------------------------------------
-        This error normally occurs when ocrmypdf can't find a file named
-        liblept-5.dll (Leptonica).  Please ensure Tesseract-OCR is installed
-        and its location is added to the system PATH environment variable.
+        This error normally occurs when ocrmypdf can't find the Leptonica
+        library, which is usually installed with Tesseract OCR. It could be that
+        Tesseract is not installed properly, we can't find the installation
+        on your system PATH environment variable.
 
-        For details see:
+        The library we are looking for is usually called:
+            liblept-5.dll   (Windows)
+            liblept*.dylib  (macOS)
+            liblept*.so     (Linux/BSD)
+
+        Please review our installation procedures to find a solution:
             https://ocrmypdf.readthedocs.io/en/latest/installation.html
         ---------------------------------------------------------------------
         """
     )
-lept = ffi.dlopen(_libpath)
-lept.setMsgSeverity(lept.L_SEVERITY_WARNING)
+try:
+    lept = ffi.dlopen(_libpath)
+    lept.setMsgSeverity(lept.L_SEVERITY_WARNING)
+except ffi.error as e:
+    raise MissingDependencyError(
+        f"Leptonica library found at {_libpath}, but we could not access it"
+    ) from e
 
 
 class _LeptonicaErrorTrap:

src/ocrmypdf/lib/compile_leptonica.py

@@ -443,6 +443,12 @@ pixReadBarcodes(PIXA     *pixa,
                 SARRAY  **psaw,
                 l_int32 debugflag);
 
+PIX *
+pixGenHalftoneMask(PIX      *pixs,
+                        PIX     **ppixtext,
+                        l_int32  *phtfound,
+                        PIXA     *pixadb);
+
 l_int32
 l_generateCIDataForPdf(const char *fname,
                        PIX *pix,

src/ocrmypdf/optimize.py

@@ -155,6 +155,17 @@ def extract_image_generic(*, pike, root, log, image, xref, options):
         # generating a PNG from compressed data
         pim.as_pil_image().save(png_name(root, xref))
         return xref, '.png'
+    elif (
+        not pim.indexed
+        and pim.colorspace == Name.ICCBased
+        and pim.bits_per_component == 1
+        and not options.jbig2_lossy
+    ):
+        # We can losslessly optimize 1-bit images to CCITT or JBIG2 without
+        # paying any attention to the ICC profile, provided we're not doing
+        # lossy JBIG2
+        pim.as_pil_image().save(png_name(root, xref))
+        return xref, '.png'
 
     return None
 
@@ -410,9 +421,9 @@ def transcode_pngs(pike, images, image_name_fn, root, log, options):
             )
             continue
         if compdata.type == leptonica.lept.L_FLATE_ENCODE:
-            return rewrite_png(pike, im_obj, compdata, log)
+            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)
+            rewrite_png_as_g4(pike, im_obj, compdata, log)
 
 
 def rewrite_png_as_g4(pike, im_obj, compdata, log):

src/ocrmypdf/pdfinfo/info.py

@@ -18,11 +18,11 @@
 
 import logging
 import re
-from collections import namedtuple
+from collections import defaultdict, namedtuple
 from decimal import Decimal
 from enum import Enum
 from math import hypot, isclose
-from os import fspath
+from os import PathLike, fspath
 from pathlib import Path
 from warnings import warn
 
@@ -30,17 +30,17 @@ import pikepdf
 from pikepdf import PdfMatrix
 from tqdm import tqdm
 
-from ocrmypdf.exceptions import EncryptedPdfError, MissingDependencyError
-
-from . import ghosttext
-from .layout import get_page_analysis, get_text_boxes
+from ocrmypdf.exceptions import EncryptedPdfError
+from ocrmypdf.exec import ghostscript
+from ocrmypdf.pdfinfo import ghosttext
+from ocrmypdf.pdfinfo.layout import get_page_analysis, get_text_boxes
 
 logger = logging.getLogger()
 
 Colorspace = Enum('Colorspace', 'gray rgb cmyk lab icc index sep devn pattern jpeg2000')
 
 Encoding = Enum(
-    'Encoding', 'ccitt jpeg jpeg2000 jbig2 asciihex ascii85 lzw flate ' + 'runlength'
+    'Encoding', 'ccitt jpeg jpeg2000 jbig2 asciihex ascii85 lzw flate runlength'
 )
 
 FRIENDLY_COLORSPACE = {
@@ -98,7 +98,7 @@ XobjectSettings = namedtuple('XobjectSettings', ['name', 'shorthand', 'stack_dep
 InlineSettings = namedtuple('InlineSettings', ['iimage', 'shorthand', 'stack_depth'])
 
 ContentsInfo = namedtuple(
-    'ContentsInfo', ['xobject_settings', 'inline_images', 'found_vector']
+    'ContentsInfo', ['xobject_settings', 'inline_images', 'found_vector', 'name_index']
 )
 
 TextboxInfo = namedtuple('TextboxInfo', ['bbox', 'is_visible', 'is_corrupt'])
@@ -151,6 +151,7 @@ def _interpret_contents(contentstream, initial_shorthand=UNIT_SQUARE):
     ctm = PdfMatrix(initial_shorthand)
     xobject_settings = []
     inline_images = []
+    name_index = defaultdict(lambda: [])
     found_vector = False
     vector_ops = set('S s f F f* B B* b b*'.split())
     image_ops = set('BI ID EI q Q Do cm'.split())
@@ -185,6 +186,7 @@ def _interpret_contents(contentstream, initial_shorthand=UNIT_SQUARE):
                 name=image_name, shorthand=ctm.shorthand, stack_depth=len(stack)
             )
             xobject_settings.append(settings)
+            name_index[image_name].append(settings)
         elif operator == 'INLINE IMAGE':  # BI/ID/EI are grouped into this
             iimage = operands[0]
             inline = InlineSettings(
@@ -198,6 +200,7 @@ def _interpret_contents(contentstream, initial_shorthand=UNIT_SQUARE):
         xobject_settings=xobject_settings,
         inline_images=inline_images,
         found_vector=found_vector,
+        name_index=name_index,
     )
 
 
@@ -303,14 +306,22 @@ class ImageInfo:
         if self._enc == Encoding.jpeg2000:
             self._color = Colorspace.jpeg2000
 
-        self._comp = FRIENDLY_COMP.get(self._color, '?')
+        if self._color == Colorspace.icc:
+            # Check the ICC profile to determine actual colorspace
+            pim_icc = pim.icc
+            if pim_icc.profile.xcolor_space == 'GRAY':
+                self._comp = 1
+            elif pim_icc.profile.xcolor_space == 'CMYK':
+                self._comp = 4
+            else:
+                self._comp = 3
+        else:
+            self._comp = FRIENDLY_COMP.get(self._color, '?')
 
-        # Bit of a hack... infer grayscale if component count is uncertain
-        # but encoding must be monochrome. This happens if a monochrome image
-        # has an ICC profile attached. Better solution would be to examine
-        # the ICC profile.
-        if self._comp == '?' and self._enc in (Encoding.ccitt, Encoding.jbig2):
-            self._comp = FRIENDLY_COMP[Colorspace.gray]
+            # Bit of a hack... infer grayscale if component count is uncertain
+            # but encoding only supports monochrome.
+            if self._comp == '?' and self._enc in (Encoding.ccitt, Encoding.jbig2):
+                self._comp = FRIENDLY_COMP[Colorspace.gray]
 
     @property
     def name(self):
@@ -411,13 +422,9 @@ def _find_regular_images(container, contentsinfo):
     """
 
     for pdfimage, xobj in _image_xobjects(container):
-
-        # For each image that is drawn on this, check if we drawing the
-        # current image - yes this is O(n^2), but n == 1 almost always
-        for draw in contentsinfo.xobject_settings:
-            if draw.name != xobj:
-                continue
-
+        if xobj not in contentsinfo.name_index:
+            continue
+        for draw in contentsinfo.name_index[xobj]:
             if draw.stack_depth == 0 and _is_unit_square(draw.shorthand):
                 # At least one PDF in the wild (and test suite) draws an image
                 # when the graphics stack depth is 0, meaning that the image
@@ -551,7 +558,7 @@ def simplify_textboxes(miner, textbox_getter):
         yield TextboxInfo(box.bbox, visible, corrupt)
 
 
-def _pdf_get_pageinfo(pdf, pageno: int, infile, xmltext):
+def _pdf_get_pageinfo(pdf, pageno: int, infile: PathLike, xmltext: str):
     pageinfo = {}
     pageinfo['pageno'] = pageno
     pageinfo['images'] = []
@@ -611,25 +618,28 @@ def _pdf_get_pageinfo(pdf, pageno: int, infile, xmltext):
 
 def _pdf_get_all_pageinfo(infile, detailed_analysis=False, log=None, progbar=False):
     pdf = pikepdf.open(infile)  # Do not close in this function
-    if pdf.is_encrypted:
-        pdf.close()
-        raise EncryptedPdfError()  # Triggered by encryption with empty passwd
-    if detailed_analysis:
-        pages_xml = None
-    else:
-        pages_xml = ghosttext.extract_text_xml(infile, pdf, pageno=None, log=log)
+    try:
+        if pdf.is_encrypted:
+            raise EncryptedPdfError()  # Triggered by encryption with empty passwd
+        if detailed_analysis:
+            pages_xml = None
+        else:
+            pages_xml = ghosttext.extract_text_xml(infile, pdf, pageno=None, log=log)
 
-    pages = []
-    for n, _ in tqdm(
-        enumerate(pdf.pages),
-        total=len(pdf.pages),
-        desc="Scan",
-        unit='page',
-        disable=not progbar,
-    ):
-        page_xml = pages_xml[n] if pages_xml else None
-        page = PageInfo(pdf, n, infile, page_xml, detailed_analysis)
-        pages.append(page)
+        pages = []
+        for n, _ in tqdm(
+            enumerate(pdf.pages),
+            total=len(pdf.pages),
+            desc="Scan",
+            unit='page',
+            disable=not progbar,
+        ):
+            page_xml = pages_xml[n] if pages_xml else None
+            page = PageInfo(pdf, n, infile, page_xml, detailed_analysis)
+            pages.append(page)
+    except Exception:
+        pdf.close()
+        raise
 
     return pages, pdf
 
@@ -750,6 +760,8 @@ class PdfInfo:
 
     def __init__(self, infile, detailed_page_analysis=False, log=logger, progbar=False):
         self._infile = infile
+        if ghostscript.version() in ('9.52',):
+            detailed_page_analysis = True  # txtwrite doesn't work in these versions
         self._pages, pdf = _pdf_get_all_pageinfo(
             infile, detailed_page_analysis, log=log, progbar=progbar
         )
@@ -805,10 +817,14 @@ def main():
     parser = argparse.ArgumentParser()
     parser.add_argument('infile')
     args = parser.parse_args()
-    info = _pdf_get_all_pageinfo(args.infile)
+    pagesinfo, pdfinfo = _pdf_get_all_pageinfo(args.infile)
     from pprint import pprint
 
-    pprint(info)
+    pprint(pdfinfo)
+    for page in pagesinfo:
+        pprint(page)
+        for im in page.images:
+            pprint(im)
 
 
 if __name__ == '__main__':

src/ocrmypdf/pdfinfo/layout.py

@@ -269,7 +269,9 @@ class TextPositionTracker(PDFLayoutAnalyzer):
 
 def get_page_analysis(infile, pageno, pscript5_mode):
     rman = pdfminer.pdfinterp.PDFResourceManager(caching=True)
-    dev = TextPositionTracker(rman, laparams=LAParams())
+    dev = TextPositionTracker(
+        rman, laparams=LAParams(all_texts=True, detect_vertical=True)
+    )
     interp = pdfminer.pdfinterp.PDFPageInterpreter(rman, dev)
 
     if pscript5_mode:

src/ocrmypdf/quality.py

@@ -0,0 +1,60 @@
+# © 2020 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 re
+from typing import Iterable
+
+"""Utilities to measure OCR quality"""
+
+
+class OcrQualityDictionary:
+    """Manages a dictionary for simple OCR quality checks."""
+
+    def __init__(self, *, wordlist: Iterable[str] = []):
+        """Construct a dictionary from a list of words.
+
+        Words for which capitalization is important should be capitalized in the
+        dictionary. Words that contain spaces or other punctuation will never match.
+        """
+        self.dictionary = set()
+        self.dictionary.update(w for w in wordlist)
+
+    def measure_words_matched(self, ocr_text: str) -> float:
+        """Check how many unique words in the OCR text match a dictionary.
+
+        Words with mixed capitalized are only considered a match if the test word
+        matches that capitalization.
+
+        Returns:
+            number of words that match / number
+        """
+        text = re.sub(r"[0-9_]+", ' ', ocr_text)
+        text = re.sub(r'\W+', ' ', text)
+        text_words_list = re.split(r'\s+', text)
+        text_words = {w for w in text_words_list if len(w) >= 3}
+
+        matches = 0
+        for w in text_words:
+            if w in self.dictionary or (
+                w != w.lower() and w.lower() in self.dictionary
+            ):
+                matches += 1
+        if matches > 0:
+            hit_ratio = matches / len(text_words)
+        else:
+            hit_ratio = 0.0
+        return hit_ratio

tests/conftest.py

@@ -28,12 +28,6 @@ from ocrmypdf import api, cli
 
 pytest_plugins = ['helpers_namespace']
 
-try:
-    from pytest_cov.embed import cleanup_on_sigterm
-except ImportError:
-    pass
-else:
-    cleanup_on_sigterm()
 
 # pylint: disable=E1101
 # pytest.helpers is dynamic so it confuses pylint
@@ -103,10 +97,48 @@ assert ast.parse(WINDOWS_SHIM_TEMPLATE.format(spoofer=repr(r"C:\\Temp\\file.py")
 def spoof(tmp_path_factory, **kwargs):
     """Modify PATH to override subprocess executables
 
-    spoof(program1='replacement', ...)
+    spoof(tmp_path_factory, program1='replacement', ...)
 
-    Creates temporary directory with symlinks to targets.
+    For the test suite we need a way override executables, so that we can
+    substitute desired results such as errors or just speed up OCR.
 
+    On POSIXish platforms we create a temporary folder with overrides that
+    are symlinks to the executables we want to run. We do not actually override
+    PATH. We also set an environment variable _OCRMYPDF_TEST_PATH, which
+    OCRmyPDF's subprocess wrapper will check before they use regular PATH. The
+    output is a folder full of executables we are overriding. We can override
+    multiple executables. The end result is a folder we can use in a PATH-style
+    lookup to override some executables:
+
+        /tmp/abcxyz/tesseract -> ocrmypdf/tests/resources/spoof/tesseract_crash.py
+        /tmp/abcxyz/gs -> ocrmypdf/tests/resources/spoof/gs_backflip.py
+
+    Windows needs extra help from us because usually, only the Administrator
+    can create symlinks. Instead we create small Python scripts that call
+    the programs we want, implementing the effect of a symlink. This is cleaner
+    than creating Windows executables or trying to use non-Python scripts.
+    The temporary folder generated for Windows could like:
+
+        %TEMP%\abcxyz\tesseract.py:
+            (script that runs ocrmypdf/tests/resources/spoof/tesseract_crash.py)
+        %TEMP%\abcxyz\gswin32c.py:
+            (script that runs ocrmypdf/tests/resources/spoof/gs_backflip.py)
+        %TEMP%\abcxyz\gswin64c.py:
+            (script that runs ocrmypdf/tests/resources/spoof/gs_backflip.py)
+
+    We also address one quirk here, that Ghostscript may be known as gswin32c
+    or gswin64c, depending on what the user installed (regardless of Windows
+    itself). On POSIX, Ghostscript is just 'gs'.  We handle the special case here
+    too.
+
+    All of this is intimately dependent on the machinery in ocrmypdf.exec.run().
+    In particular, for Windows, that code has to know that if there is a .py
+    file, it needs to run it with Python, since Windows does not like being
+    asked to execute files.
+
+    We don't overload PATH directly because we have some tests where we call
+    ocrmypdf as a subprocess (to exercise the command line interface) and some
+    tests where we call it as an API.
     """
     env = os.environ.copy()
     slug = '-'.join(v.replace('.py', '') for v in sorted(kwargs.values()))
@@ -137,12 +169,12 @@ def spoof(tmp_path_factory, **kwargs):
     return env
 
 
-@pytest.fixture(scope='session')
+@pytest.fixture
 def spoof_tesseract_noop(tmp_path_factory):
     return spoof(tmp_path_factory, tesseract='tesseract_noop.py')
 
 
-@pytest.fixture(scope='session')
+@pytest.fixture
 def spoof_tesseract_cache(tmp_path_factory):
     if running_in_docker():
         return os.environ.copy()
@@ -201,7 +233,10 @@ def check_ocrmypdf(input_file, output_file, *args, env=None):
 
 @pytest.helpers.register
 def run_ocrmypdf_api(input_file, output_file, *args, env=None):
-    "Run ocrmypdf and let caller deal with results"
+    """Run ocrmypdf via API and let caller deal with results
+
+    Does not currently have a way to manipulate the PATH except for Tesseract.
+    """
 
     options = cli.parser.parse_args(
         [str(input_file), str(output_file)]
@@ -211,6 +246,10 @@ def run_ocrmypdf_api(input_file, output_file, *args, env=None):
     if env:
         options.tesseract_env = env.copy()
         options.tesseract_env['_OCRMYPDF_TEST_INFILE'] = os.fspath(input_file)
+        first_path = env.get('_OCRMYPDF_TEST_PATH', '').split(os.pathsep)[0]
+        if 'spoof' in first_path:
+            assert 'gs' not in first_path, "use run_ocrmypdf() for gs"
+            assert 'tesseract' in first_path
     if options.tesseract_env:
         assert all(isinstance(v, (str, bytes)) for v in options.tesseract_env.values())
 
@@ -222,13 +261,21 @@ def run_ocrmypdf(input_file, output_file, *args, env=None, universal_newlines=Tr
     "Run ocrmypdf and let caller deal with results"
 
     if env is None:
-        env = os.environ
+        env = os.environ.copy()
 
     p_args = (
         OCRMYPDF
         + [str(arg) for arg in args if arg is not None]
         + [str(input_file), str(output_file)]
     )
+
+    # Tell subprocess where to find coverage.py configuration
+    # This has no effect except when coverage is running
+    # Details: https://coverage.readthedocs.io/en/coverage-5.0/subprocess.html
+    coverage_rc = Path(__file__).parent.parent / '.coveragerc'
+    assert coverage_rc.exists()
+    env['COVERAGE_PROCESS_START'] = os.fspath(coverage_rc)
+
     p = run(
         p_args, stdout=PIPE, stderr=PIPE, universal_newlines=universal_newlines, env=env
     )

tests/spoof/gs_raster_failure.py

@@ -35,13 +35,13 @@ def main():
         print('SPOOFED: ' + os.path.basename(__file__))
         sys.exit(0)
 
-    # For any rendering calls (device == pdfwrite) call real ghostscript
-    if '-sDEVICE=pdfwrite' in sys.argv:
+    # For non-image rastering calls, use real ghostscript
+    if '-sDEVICE=pdfwrite' in sys.argv or '-sDEVICE=txtwrite' in sys.argv:
         real_ghostscript(sys.argv)
         return
 
     # Fail
-    print("ERROR: Ghost story archive not found")
+    print("ERROR: Ghost story archive not found", file=sys.stderr)
     sys.exit(1)
 
 

tests/spoof/gs_render_failure.py

@@ -40,7 +40,7 @@ def main():
         return
 
     # Fail
-    print("ERROR: Casper is not a friendly ghost")
+    print("ERROR: Casper is not a friendly ghost", file=sys.stderr)
     sys.exit(1)
 
 

tests/spoof/tesseract_badutf8.py

@@ -1,80 +0,0 @@
-#!/usr/bin/env python3
-# © 2017 James R. Barlow: github.com/jbarlow83
-#
-# Permission is hereby granted, free of charge, to any person obtaining a
-# copy of this software and associated documentation files (the
-# "Software"), to deal in the Software without restriction, including
-# without limitation the rights to use, copy, modify, merge, publish,
-# distribute, sublicense, and/or sell copies of the Software, and to
-# permit persons to whom the Software is furnished to do so, subject to
-# the following conditions:
-#
-# The above copyright notice and this permission notice shall be included
-# in all copies or substantial portions of the Software.
-#
-# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
-# OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
-# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
-# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
-# CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
-# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
-# SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
-
-import sys
-
-"""Tesseract bad utf8 spoof
-
-In 'hocr' mode or 'pdf' mode, return error code 1 and some non-Unicode
-text because tesseract seems to do that in some cases related to
-language pack version mismatches
-
-"""
-
-
-VERSION_STRING = '''tesseract 4.0.0
- leptonica-1.77.0
-  libjpeg 9c : libpng 1.6.35 : libtiff 4.0.10 : zlib 1.2.11 : libopenjp2 2.3.0
- Found AVX2
- Found AVX
- Found SSE
-SPOOFED
-'''
-
-# Japanese "Invalid UTF-8" encoded in Shift JIS
-BAD_UTF8 = b'\x96\xb3\x8c\xf8\x82\xc8UTF-8\x0a'
-
-
-def main():
-    if sys.argv[1] == '--version':
-        print(VERSION_STRING, file=sys.stderr)
-        sys.exit(0)
-    elif sys.argv[1] == '--list-langs':
-        print('List of available languages (1):\neng', file=sys.stderr)
-        sys.exit(0)
-    elif sys.argv[-2] == '--print-parameters':
-        print("Some parameters", file=sys.stderr)
-        print("textonly_pdf\t1\tSome help text")
-        sys.exit(0)
-    elif sys.argv[-2] in ('hocr', 'pdf'):
-        sys.stdout.buffer.write(BAD_UTF8)
-        sys.exit(1)
-    elif sys.argv[-1] == 'stdout':
-        # input file is at sys.argv[-2] but we don't look at it
-        print(
-            """Orientation: 0
-Orientation in degrees: 0
-Orientation confidence: 100.00
-Script: 1
-Script confidence: 100.00""",
-            file=sys.stderr,
-        )
-    else:
-        print("Spoof doesn't understand arguments", file=sys.stderr)
-        print(sys.argv, file=sys.stderr)
-        sys.exit(1)
-
-    sys.exit(0)
-
-
-if __name__ == '__main__':
-    main()

tests/spoof/tesseract_noop.py

@@ -32,9 +32,10 @@ In orientation check mode, report the orientation is upright.
 """
 
 import sys
+from pathlib import Path
 
 import img2pdf
-import PyPDF2 as pypdf
+import pikepdf
 from PIL import Image
 
 VERSION_STRING = '''tesseract 4.0.0
@@ -99,12 +100,10 @@ def main():
                 pagesize = im.size[0] / dpi[0], im.size[1] / dpi[1]
                 ptsize = pagesize[0] * 72, pagesize[1] * 72
 
-            pdf_out = pypdf.PdfFileWriter()
-            pdf_out.addBlankPage(ptsize[0], ptsize[1])
-            with open(output + '.pdf', 'wb') as f:
-                pdf_out.write(f)
-            with open(output + '.txt', 'w') as f:
-                f.write('')
+            pdf_out = pikepdf.new()
+            pdf_out.add_blank_page(page_size=ptsize)
+            pdf_out.save(Path(output).with_suffix('.pdf'), static_id=True)
+            Path(output).with_suffix('.txt').write_text('')
         else:
             inputf = sys.argv[-4]
             output = sys.argv[-3]

tests/test_acroform.py

@@ -15,6 +15,8 @@
 # You should have received a copy of the GNU General Public License
 # along with OCRmyPDF.  If not, see <http://www.gnu.org/licenses/>.
 
+import logging
+
 import pytest
 
 import ocrmypdf
@@ -30,4 +32,11 @@ def acroform(resources):
 def test_acroform_and_redo(acroform, caplog, no_outpdf):
     with pytest.raises(ocrmypdf.exceptions.InputFileError):
         check_ocrmypdf(acroform, no_outpdf, '--redo-ocr')
-        assert '--redo-ocr is not currently possible' in caplog.text
+    assert '--redo-ocr is not currently possible' in caplog.text
+
+
+def test_acroform_message(acroform, caplog, spoof_tesseract_noop, outpdf):
+    caplog.set_level(logging.INFO)
+    check_ocrmypdf(acroform, outpdf, env=spoof_tesseract_noop)
+    assert 'fillable form' in caplog.text
+    assert '--force-ocr' in caplog.text

tests/test_api.py

@@ -0,0 +1,68 @@
+# © 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 logging
+from io import StringIO
+
+import pytest
+from tqdm import tqdm
+
+import ocrmypdf
+
+
+def test_raw_console():
+    bio = StringIO()
+    tqconsole = ocrmypdf.api.TqdmConsole(file=bio)
+    tqconsole.write("Test")
+    tqconsole.flush()
+    assert "Test" in bio.getvalue()
+
+
+def test_tqdm_console():
+    log = logging.getLogger()
+    log.setLevel(logging.INFO)
+
+    formatter = logging.Formatter('%(message)s')
+
+    bio = StringIO()
+    console = logging.StreamHandler(ocrmypdf.api.TqdmConsole(file=bio))
+    console.setFormatter(formatter)
+
+    log.addHandler(console)
+
+    def before_pbar(message):
+        # Ensure that log messages appear before the progress bar, even when
+        # printed after the progress bar updates.
+        v = bio.getvalue()
+        pbar_start_marker = '|#'
+        return v.index(message) < v.index(pbar_start_marker)
+
+    with tqdm(total=2, file=bio, disable=False) as pbar:
+        pbar.update()
+        msg = "1/2 above progress bar"
+        log.info(msg)
+        assert before_pbar(msg)
+
+    log.info("done")
+    assert not before_pbar("done")
+
+
+def test_language_list():
+    with pytest.raises(
+        (ocrmypdf.exceptions.InputFileError, ocrmypdf.exceptions.MissingDependencyError)
+    ):
+        ocrmypdf.ocr('doesnotexist.pdf', '_.pdf', language=['eng', 'deu'])

tests/test_ghostscript.py

@@ -31,28 +31,28 @@ run_ocrmypdf_api = pytest.helpers.run_ocrmypdf_api
 spoof = pytest.helpers.spoof
 
 
-@pytest.fixture(scope='session')
+@pytest.fixture
 def spoof_no_tess_gs_render_fail(tmp_path_factory):
     return spoof(
         tmp_path_factory, tesseract='tesseract_noop.py', gs='gs_render_failure.py'
     )
 
 
-@pytest.fixture(scope='session')
+@pytest.fixture
 def spoof_no_tess_gs_raster_fail(tmp_path_factory):
     return spoof(
         tmp_path_factory, tesseract='tesseract_noop.py', gs='gs_raster_failure.py'
     )
 
 
-@pytest.fixture(scope='session')
+@pytest.fixture
 def spoof_no_tess_no_pdfa(tmp_path_factory):
     return spoof(
         tmp_path_factory, tesseract='tesseract_noop.py', gs='gs_pdfa_failure.py'
     )
 
 
-@pytest.fixture(scope='session')
+@pytest.fixture
 def spoof_no_tess_pdfa_warning(tmp_path_factory):
     return spoof(
         tmp_path_factory, tesseract='tesseract_noop.py', gs='gs_feature_elision.py'
@@ -60,18 +60,18 @@ def spoof_no_tess_pdfa_warning(tmp_path_factory):
 
 
 @pytest.fixture
-def linn(resources):
-    path = resources / 'linn.pdf'
+def francais(resources):
+    path = resources / 'francais.pdf'
     return path, pikepdf.open(path)
 
 
-def test_rasterize_size(linn, outdir, caplog):
-    path, pdf = linn
+def test_rasterize_size(francais, outdir, caplog):
+    path, pdf = francais
     page_size_pts = (pdf.pages[0].MediaBox[2], pdf.pages[0].MediaBox[3])
     assert pdf.pages[0].MediaBox[0] == pdf.pages[0].MediaBox[1] == 0
     page_size = (page_size_pts[0] / Decimal(72), page_size_pts[1] / Decimal(72))
-    target_size = Decimal('200.0'), Decimal('150.0')
-    target_dpi = 42.0, 4242.0
+    target_size = Decimal('50.0'), Decimal('30.0')
+    forced_dpi = 42.0, 4242.0
 
     log = logging.getLogger()
     rasterize_pdf(
@@ -81,21 +81,21 @@ def test_rasterize_size(linn, outdir, caplog):
         target_size[1] / page_size[1],
         raster_device='pngmono',
         log=log,
-        page_dpi=target_dpi,
+        page_dpi=forced_dpi,
     )
 
     with Image.open(outdir / 'out.png') as im:
         assert im.size == target_size
-        assert im.info['dpi'] == target_dpi
+        assert im.info['dpi'] == forced_dpi
 
 
-def test_rasterize_rotated(linn, outdir, caplog):
-    path, pdf = linn
+def test_rasterize_rotated(francais, outdir, caplog):
+    path, pdf = francais
     page_size_pts = (pdf.pages[0].MediaBox[2], pdf.pages[0].MediaBox[3])
     assert pdf.pages[0].MediaBox[0] == pdf.pages[0].MediaBox[1] == 0
     page_size = (page_size_pts[0] / Decimal(72), page_size_pts[1] / Decimal(72))
-    target_size = Decimal('200.0'), Decimal('150.0')
-    target_dpi = 42.0, 4242.0
+    target_size = Decimal('50.0'), Decimal('30.0')
+    forced_dpi = 42.0, 4242.0
 
     log = logging.getLogger()
     caplog.set_level(logging.DEBUG)
@@ -106,34 +106,34 @@ def test_rasterize_rotated(linn, outdir, caplog):
         target_size[1] / page_size[1],
         raster_device='pngmono',
         log=log,
-        page_dpi=target_dpi,
+        page_dpi=forced_dpi,
         rotation=90,
     )
 
     with Image.open(outdir / 'out.png') as im:
         assert im.size == (target_size[1], target_size[0])
-        assert im.info['dpi'] == (target_dpi[1], target_dpi[0])
+        assert im.info['dpi'] == (forced_dpi[1], forced_dpi[0])
 
 
 def test_gs_render_failure(spoof_no_tess_gs_render_fail, resources, outpdf):
     p, out, err = run_ocrmypdf(
         resources / 'blank.pdf', outpdf, env=spoof_no_tess_gs_render_fail
     )
-    print(err)
+    assert 'Casper is not a friendly ghost' in err
     assert p.returncode == ExitCode.child_process_error
 
 
 def test_gs_raster_failure(spoof_no_tess_gs_raster_fail, resources, outpdf):
     p, out, err = run_ocrmypdf(
-        resources / 'ccitt.pdf', outpdf, env=spoof_no_tess_gs_raster_fail
+        resources / 'francais.pdf', outpdf, env=spoof_no_tess_gs_raster_fail
     )
-    print(err)
+    assert 'Ghost story archive not found' in err
     assert p.returncode == ExitCode.child_process_error
 
 
 def test_ghostscript_pdfa_failure(spoof_no_tess_no_pdfa, resources, outpdf):
     p, out, err = run_ocrmypdf(
-        resources / 'ccitt.pdf', outpdf, env=spoof_no_tess_no_pdfa
+        resources / 'francais.pdf', outpdf, env=spoof_no_tess_no_pdfa
     )
     assert (
         p.returncode == ExitCode.pdfa_conversion_failed
@@ -141,4 +141,4 @@ def test_ghostscript_pdfa_failure(spoof_no_tess_no_pdfa, resources, outpdf):
 
 
 def test_ghostscript_feature_elision(spoof_no_tess_pdfa_warning, resources, outpdf):
-    check_ocrmypdf(resources / 'ccitt.pdf', outpdf, env=spoof_no_tess_pdfa_warning)
+    check_ocrmypdf(resources / 'francais.pdf', outpdf, env=spoof_no_tess_pdfa_warning)

tests/test_graft.py

@@ -25,25 +25,30 @@ import ocrmypdf
 
 
 def test_no_glyphless_graft(resources, outdir):
-    pdf = pikepdf.open(resources / 'francais.pdf')
-    pdf_aspect = pikepdf.open(resources / 'aspect.pdf')
-    pdf_cmyk = pikepdf.open(resources / 'cmyk.pdf')
-    pdf.pages.extend(pdf_aspect.pages)
-    pdf.pages.extend(pdf_cmyk.pages)
-    pdf.save(outdir / 'test.pdf')
+    with pikepdf.open(resources / 'francais.pdf') as pdf, pikepdf.open(
+        resources / 'aspect.pdf'
+    ) as pdf_aspect, pikepdf.open(resources / 'cmyk.pdf') as pdf_cmyk:
+        pdf.pages.extend(pdf_aspect.pages)
+        pdf.pages.extend(pdf_cmyk.pages)
+        pdf.save(outdir / 'test.pdf')
 
     with patch('ocrmypdf._graft.MAX_REPLACE_PAGES', 2):
         ocrmypdf.ocr(
-            outdir / 'test.pdf', outdir / 'out.pdf', deskew=True, tesseract_timeout=0
+            outdir / 'test.pdf',
+            outdir / 'out.pdf',
+            deskew=True,
+            tesseract_timeout=0,
+            force_ocr=True,
         )
+    # This test needs asserts
 
 
 def test_links(resources, outpdf):
     ocrmypdf.ocr(
         resources / 'link.pdf', outpdf, redo_ocr=True, oversample=200, output_type='pdf'
     )
-    pdf = pikepdf.open(outpdf)
-    p1 = pdf.pages[0]
-    p2 = pdf.pages[1]
-    assert p1.Annots[0].A.D[0].objgen == p2.objgen
-    assert p2.Annots[0].A.D[0].objgen == p1.objgen
+    with pikepdf.open(outpdf) as pdf:
+        p1 = pdf.pages[0]
+        p2 = pdf.pages[1]
+        assert p1.Annots[0].A.D[0].objgen == p2.objgen
+        assert p2.Annots[0].A.D[0].objgen == p1.objgen

tests/test_helpers.py

@@ -0,0 +1,116 @@
+# © 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 logging
+import multiprocessing
+import os
+from pathlib import Path
+from unittest.mock import MagicMock
+
+import pytest
+
+import ocrmypdf.helpers as helpers
+
+
+class TestSafeSymlink:
+    def test_safe_symlink_link_self(self, tmp_path, caplog):
+        helpers.safe_symlink(tmp_path / 'self', tmp_path / 'self')
+        assert caplog.record_tuples[0][1] == logging.WARNING
+
+    def test_safe_symlink_overwrite(self, tmp_path):
+        (tmp_path / 'regular_file').touch()
+        with pytest.raises(FileExistsError):
+            helpers.safe_symlink(tmp_path / 'input', tmp_path / 'regular_file')
+
+    def test_safe_symlink_relink(self, tmp_path):
+        (tmp_path / 'regular_file_a').touch()
+        (tmp_path / 'regular_file_b').write_bytes(b'ABC')
+        (tmp_path / 'link').symlink_to(tmp_path / 'regular_file_a')
+        helpers.safe_symlink(tmp_path / 'regular_file_b', tmp_path / 'link')
+        assert (tmp_path / 'link').samefile(tmp_path / 'regular_file_b') or (
+            tmp_path / 'link'
+        ).read_bytes() == b'ABC'
+
+
+def test_no_cpu_count(monkeypatch):
+    def cpu_count_raises():
+        raise NotImplementedError()
+
+    monkeypatch.setattr(multiprocessing, 'cpu_count', cpu_count_raises)
+    with pytest.warns(expected_warning=UserWarning):
+        assert helpers.available_cpu_count() == 1
+
+
+def test_deprecated():
+    @helpers.deprecated
+    def old_function():
+        return 42
+
+    with pytest.deprecated_call():
+        assert old_function() == 42
+
+
+class TestFileIsWritable:
+    @pytest.fixture
+    def non_existent(self, tmp_path):
+        return tmp_path / 'nofile'
+
+    @pytest.fixture
+    def basic_file(self, tmp_path):
+        basic = tmp_path / 'basic'
+        basic.touch()
+        return basic
+
+    def test_plain(self, non_existent):
+        assert helpers.is_file_writable(non_existent)
+
+    def test_symlink_loop(self, tmp_path):
+        loop = tmp_path / 'loop'
+        loop.symlink_to(loop)
+        assert not helpers.is_file_writable(loop)
+
+    def test_chmod(self, basic_file):
+        assert helpers.is_file_writable(basic_file)
+        basic_file.chmod(0o400)
+        assert not helpers.is_file_writable(basic_file)
+        basic_file.chmod(0o000)
+        assert not helpers.is_file_writable(basic_file)
+
+    def test_permission_error(self, basic_file):
+        pathmock = MagicMock(spec_set=basic_file)
+        pathmock.is_symlink.return_value = False
+        pathmock.exists.return_value = True
+        pathmock.is_file.side_effect = PermissionError
+        assert not helpers.is_file_writable(pathmock)
+
+
+def test_shim_paths(tmp_path):
+    progfiles = tmp_path / 'Program Files'
+    progfiles.mkdir()
+    (progfiles / 'tesseract-ocr').mkdir()
+    (progfiles / 'gs' / '9.51' / 'bin').mkdir(parents=True)
+    (progfiles / 'gs' / '9.52' / 'bin').mkdir(parents=True)
+    syspath = tmp_path / 'bin'
+    env = {'PROGRAMFILES': str(progfiles), 'PATH': str(syspath)}
+    from ocrmypdf.exec import shim_paths_with_program_files
+
+    result_str = shim_paths_with_program_files(env=env)
+    results = result_str.split(os.pathsep)
+    assert results[0].endswith('tesseract-ocr')
+    assert results[1].endswith(os.path.join('gs', '9.52', 'bin'))
+    assert results[2].endswith(os.path.join('gs', '9.51', 'bin'))
+    assert results[3] == str(syspath)

tests/test_main.py

@@ -45,12 +45,12 @@ spoof = pytest.helpers.spoof
 RENDERERS = ['hocr', 'sandwich']
 
 
-@pytest.fixture(scope='session')
+@pytest.fixture
 def spoof_tesseract_crash(tmp_path_factory):
     return spoof(tmp_path_factory, tesseract='tesseract_crash.py')
 
 
-@pytest.fixture(scope='session')
+@pytest.fixture
 def spoof_tesseract_big_image_error(tmp_path_factory):
     return spoof(tmp_path_factory, tesseract='tesseract_big_image_error.py')
 
@@ -270,6 +270,16 @@ def test_input_file_not_found(caplog, no_outpdf):
     assert input_file in caplog.text
 
 
+@pytest.mark.skipif(os.name == 'nt', reason="chmod")
+def test_input_file_not_readable(caplog, resources, outdir, no_outpdf):
+    input_file = outdir / 'trivial.pdf'
+    shutil.copy(resources / 'trivial.pdf', input_file)
+    input_file.chmod(0o000)
+    result = run_ocrmypdf_api(input_file, no_outpdf)
+    assert result == ExitCode.input_file
+    assert str(input_file) in caplog.text
+
+
 def test_input_file_not_a_pdf(caplog, no_outpdf):
     input_file = __file__  # Try to OCR this file
     result = run_ocrmypdf_api(input_file, no_outpdf)

tests/test_metadata.py

@@ -31,8 +31,11 @@ import pytest
 from pikepdf.models.metadata import decode_pdf_date
 
 from ocrmypdf._jobcontext import PDFContext
+from ocrmypdf._pipeline import convert_to_pdfa
+from ocrmypdf.cli import parser
 from ocrmypdf.exceptions import ExitCode
 from ocrmypdf.pdfa import SRGB_ICC_PROFILE, file_claims_pdfa, generate_pdfa_ps
+from ocrmypdf.pdfinfo import PdfInfo
 
 try:
     import fitz
@@ -313,19 +316,21 @@ def test_metadata_fixup_warning(resources, outdir, caplog):
 
 
 def test_prevent_gs_invalid_xml(resources, outdir):
-    from ocrmypdf.__main__ import parser
-    from ocrmypdf._pipeline import convert_to_pdfa
-    from ocrmypdf.pdfa import generate_pdfa_ps
-    from ocrmypdf.pdfinfo import PdfInfo
-
     generate_pdfa_ps(outdir / 'pdfa.ps')
-    copyfile(resources / 'enron1.pdf', outdir / 'layers.rendered.pdf')
+    copyfile(resources / 'trivial.pdf', outdir / 'layers.rendered.pdf')
+
+    # Inject a string with a trailing nul character into the DocumentInfo
+    # dictionary of this PDF, as often occurs in practice.
+    with pikepdf.open(outdir / 'layers.rendered.pdf') as pike:
+        pike.Root.DocumentInfo = pikepdf.Dictionary(
+            Title=b'String with trailing nul\x00'
+        )
 
     options = parser.parse_args(
         args=['-j', '1', '--output-type', 'pdfa-2', 'a.pdf', 'b.pdf']
     )
-    pdfinfo = PdfInfo(resources / 'enron1.pdf')
-    context = PDFContext(options, outdir, resources / 'enron1.pdf', pdfinfo)
+    pdfinfo = PdfInfo(outdir / 'layers.rendered.pdf')
+    context = PDFContext(options, outdir, outdir / 'layers.rendered.pdf', pdfinfo)
 
     convert_to_pdfa(
         str(outdir / 'layers.rendered.pdf'), str(outdir / 'pdfa.ps'), context
@@ -339,5 +344,30 @@ def test_prevent_gs_invalid_xml(resources, outdir):
             xmp_start = mm.find(XMP_MAGIC)
             xmp_end = mm.rfind(b'<?xpacket end', xmp_start)
             assert 0 < xmp_start < xmp_end
+            # Ensure we did not carry the nul forward.
             assert mm.find(b'&#0;', xmp_start, xmp_end) == -1, "found escaped nul"
             assert mm.find(b'\x00', xmp_start, xmp_end) == -1
+
+
+def test_malformed_docinfo(caplog, resources, outdir):
+    generate_pdfa_ps(outdir / 'pdfa.ps')
+    # copyfile(resources / 'trivial.pdf', outdir / 'layers.rendered.pdf')
+
+    with pikepdf.open(resources / 'trivial.pdf') as pike:
+        pike.trailer.Info = pikepdf.Stream(pike, b"<xml></xml>")
+        pike.save(outdir / 'layers.rendered.pdf', fix_metadata_version=False)
+
+    options = parser.parse_args(
+        args=['-j', '1', '--output-type', 'pdfa-2', 'a.pdf', 'b.pdf']
+    )
+    pdfinfo = PdfInfo(outdir / 'layers.rendered.pdf')
+    context = PDFContext(options, outdir, outdir / 'layers.rendered.pdf', pdfinfo)
+
+    convert_to_pdfa(
+        str(outdir / 'layers.rendered.pdf'), str(outdir / 'pdfa.ps'), context
+    )
+
+    print(caplog.records)
+    assert any(
+        'malformed DocumentInfo block' in record.message for record in caplog.records
+    )

tests/test_optimize.py

@@ -18,10 +18,12 @@
 import logging
 from os import fspath
 from pathlib import Path
+from unittest.mock import patch
 
+import img2pdf
 import pikepdf
 import pytest
-from PIL import Image
+from PIL import Image, ImageDraw
 
 from ocrmypdf import optimize as opt
 from ocrmypdf.exec import jbig2enc, pngquant
@@ -130,3 +132,41 @@ def test_flate_to_jbig2(resources, outdir, spoof_tesseract_noop):
     pdf = pikepdf.open(outdir / 'out.pdf')
     pim = pikepdf.PdfImage(next(iter(pdf.pages[0].images.values())))
     assert pim.filters[0] == '/JBIG2Decode'
+
+
+def test_multiple_pngs(resources, outdir, spoof_tesseract_noop):
+    with Path.open(outdir / 'in.pdf', 'wb') as inpdf:
+        img2pdf.convert(
+            fspath(resources / 'baiona_colormapped.png'),
+            fspath(resources / 'baiona_gray.png'),
+            with_pdfrw=False,
+            outputstream=inpdf,
+        )
+
+    def mockquant(input_file, output_file, _quality_min, _quality_max):
+        with Image.open(input_file) as im:
+            draw = ImageDraw.Draw(im)
+            draw.rectangle((0, 0, im.width, im.height), fill=128)
+            im.save(output_file)
+
+    with patch('ocrmypdf.optimize.pngquant.quantize', new=mockquant):
+        check_ocrmypdf(
+            outdir / 'in.pdf',
+            outdir / 'out.pdf',
+            '--optimize',
+            '3',
+            '--jobs',
+            '1',
+            '--use-threads',
+            '--output-type',
+            'pdf',
+            env=spoof_tesseract_noop,
+        )
+
+    with pikepdf.open(outdir / 'in.pdf') as inpdf, pikepdf.open(
+        outdir / 'out.pdf'
+    ) as outpdf:
+        for n in range(len(inpdf.pages)):
+            inim = next(iter(inpdf.pages[n].images.values()))
+            outim = next(iter(outpdf.pages[n].images.values()))
+            assert len(outim.read_raw_bytes()) < len(inim.read_raw_bytes()), n

tests/test_pdfinfo.py

@@ -26,6 +26,7 @@ from PIL import Image
 from reportlab.pdfgen.canvas import Canvas
 
 from ocrmypdf import pdfinfo
+from ocrmypdf.exec import ghostscript
 from ocrmypdf.pdfinfo import Colorspace, Encoding
 
 # pylint: disable=protected-access
@@ -183,6 +184,9 @@ def test_ocr_detection(resources):
 @pytest.mark.parametrize(
     'testfile', ('truetype_font_nomapping.pdf', 'type3_font_nomapping.pdf')
 )
+@pytest.mark.xfail(
+    ghostscript.version() in ('9.52',), reason="gs 9.52 txtwrite doesn't work"
+)
 def test_corrupt_font_detection(resources, testfile):
     filename = resources / testfile
     with pytest.raises(NotImplementedError):

tests/test_quality.py

@@ -0,0 +1,35 @@
+# © 2020 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 pytest
+
+import ocrmypdf.quality as qual
+
+
+def test_quality_measurement():
+    oqd = qual.OcrQualityDictionary(
+        wordlist=["words", "words", "quick", "brown", "fox", "dog", "lazy"]
+    )
+    assert len(oqd.dictionary) == 6  # 6 unique
+
+    assert (
+        oqd.measure_words_matched("The quick brown fox jumps quickly over the lazy dog")
+        == 0.5
+    )
+    assert oqd.measure_words_matched("12345 10% _f  7fox -brown   | words") == 1.0
+
+    assert oqd.measure_words_matched("quick quick quick") == 1.0

tests/test_stdio.py

@@ -33,11 +33,6 @@ run_ocrmypdf_api = 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)
@@ -56,6 +51,9 @@ def test_stdin(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf):
 
 
 def test_stdout(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf):
+    if 'COV_CORE_DATAFILE' in spoof_tesseract_noop:
+        pytest.skip(msg="Coverage uses stdout")
+
     input_file = str(resources / 'francais.pdf')
     output_file = str(outpdf)
 
@@ -121,6 +119,9 @@ def test_bad_locale():
     reason="Windows does not like this; not sure how to fix",
 )
 def test_dev_null(spoof_tesseract_noop, resources):
+    if 'COV_CORE_DATAFILE' in spoof_tesseract_noop:
+        pytest.skip(msg="Coverage uses stdout")
+
     p, out, err = run_ocrmypdf(
         resources / 'trivial.pdf', os.devnull, '--force-ocr', env=spoof_tesseract_noop
     )

tests/test_tess4.py

@@ -15,7 +15,9 @@
 # You should have received a copy of the GNU General Public License
 # along with OCRmyPDF.  If not, see <http://www.gnu.org/licenses/>.
 
+import logging
 import os
+import subprocess
 from contextlib import contextmanager
 from os import fspath
 from pathlib import Path
@@ -81,3 +83,95 @@ def test_no_languages(tmp_path):
 
     with pytest.raises(MissingDependencyError):
         tesseract.languages(tesseract_env=env)
+
+
+def test_image_too_large_hocr(monkeypatch, resources, outdir):
+    log = logging.getLogger('test_image_too_large_hocr')
+
+    def dummy_run(args, *, env=None, **kwargs):
+        raise subprocess.CalledProcessError(1, 'tesseract', output=b'Image too large')
+
+    monkeypatch.setattr(tesseract, 'run', dummy_run)
+    tesseract.generate_hocr(
+        input_file=resources / 'crom.png',
+        output_files=[outdir / 'out.hocr', outdir / 'out.txt'],
+        language=['eng'],
+        engine_mode=None,
+        tessconfig=[],
+        timeout=180.0,
+        pagesegmode=None,
+        log=log,
+        user_words=None,
+        user_patterns=None,
+        tesseract_env=None,
+    )
+    assert "name='ocr-capabilities'" in Path(outdir / 'out.hocr').read_text()
+
+
+def test_image_too_large_pdf(monkeypatch, resources, outdir):
+    log = logging.getLogger('test_image_too_large_pdf')
+
+    def dummy_run(args, *, env=None, **kwargs):
+        raise subprocess.CalledProcessError(1, 'tesseract', output=b'Image too large')
+
+    monkeypatch.setattr(tesseract, 'run', dummy_run)
+    tesseract.generate_pdf(
+        input_image=resources / 'crom.png',
+        skip_pdf=resources / 'blank.pdf',
+        output_pdf=outdir / 'pdf.pdf',
+        output_text=outdir / 'txt.txt',
+        language=['eng'],
+        engine_mode=None,
+        text_only=False,
+        tessconfig=[],
+        timeout=180.0,
+        pagesegmode=None,
+        log=log,
+        user_words=None,
+        user_patterns=None,
+        tesseract_env=None,
+    )
+    assert Path(outdir / 'txt.txt').read_text() == '[skipped page]'
+    if os.name != 'nt':  # different semantics
+        assert Path(outdir / 'pdf.pdf').samefile(resources / 'blank.pdf')
+
+
+def test_timeout(caplog):
+    log = logging.getLogger('test_timeout')
+    tesseract.page_timedout(log, '123456.png', 5)
+    assert "123456" in caplog.text
+    assert "took too long" in caplog.text
+
+
+@pytest.mark.parametrize(
+    'in_, logged',
+    [
+        (b'Tesseract Open Source', ''),
+        (b'lots of diacritics blah blah', 'diacritics'),
+        (b'Warning in pixReadMem', ''),
+        (b'OSD: Weak margin', 'unsure about page orientation'),
+        (b'Error in pixScanForForeground', ''),
+        (b'Error in boxClipToRectangle', ''),
+        (b'an unexpected error', 'an unexpected error'),
+        (b'a dire warning', 'a dire warning'),
+        (b'read_params_file something', 'read_params_file'),
+        (b'an innocent message', 'innocent'),
+        (b'\x7f\x7f\x80innocent unicode failure', 'innocent'),
+    ],
+)
+def test_tesseract_log_output(caplog, in_, logged):
+    log = logging.getLogger('tesseract_log_output')
+    log.setLevel(logging.INFO)
+
+    tesseract.tesseract_log_output(log, in_, 'dummy')
+    if logged == '':
+        assert caplog.text == ''
+    else:
+        assert logged in caplog.text
+
+
+def test_tesseract_log_output_raises(caplog):
+    log = logging.getLogger('tesseract_log_output')
+    with pytest.raises(tesseract.TesseractConfigError):
+        tesseract.tesseract_log_output(log, b'parameter not found: moo', 'dummy')
+    assert 'not found' in caplog.text

tests/test_unpaper.py

@@ -43,7 +43,7 @@ def have_unpaper():
         return True
 
 
-@pytest.fixture(scope="session")
+@pytest.fixture
 def spoof_unpaper_oldversion(tmp_path_factory):
     return spoof(tmp_path_factory, unpaper="unpaper_oldversion.py")
 

tests/test_validation.py

@@ -19,6 +19,7 @@ import logging
 import os
 from unittest.mock import patch
 
+import pikepdf
 import pytest
 
 import ocrmypdf._validation as vd
@@ -111,15 +112,20 @@ def test_output_tty():
 def test_report_file_size(tmp_path, caplog):
     in_ = tmp_path / 'a.pdf'
     out = tmp_path / 'b.pdf'
-    in_.write_bytes(b'123')
-    out.write_bytes(b'')
+    pdf = pikepdf.new()
+    pdf.save(in_)
+    pdf.save(out)
     opts = make_opts()
     vd.report_output_file_size(opts, in_, out)
     assert caplog.text == ''
     caplog.clear()
 
-    os.truncate(in_, 25001)
-    os.truncate(out, 50000)
+    waste_of_space = b'Dummy' * 5000
+    pdf.root.Dummy = waste_of_space
+    pdf.save(in_)
+    pdf.root.Dummy2 = waste_of_space + waste_of_space
+    pdf.save(out)
+
     with patch('ocrmypdf._validation.jbig2enc.available', return_value=True), patch(
         'ocrmypdf._validation.pngquant.available', return_value=True
     ):
@@ -176,3 +182,58 @@ def test_language_warning(caplog):
         vd.check_options_languages(opts)
         assert opts.language == ['eng']
         assert 'assuming --language' in caplog.text
+
+
+def test_version_comparison():
+    vd.check_external_program(
+        program="dummy_basic",
+        package="dummy",
+        version_checker=lambda: '9.0',
+        need_version='8.0.2',
+    )
+    vd.check_external_program(
+        program="dummy_doubledigit",
+        package="dummy",
+        version_checker=lambda: '10.0',
+        need_version='8.0.2',
+    )
+    vd.check_external_program(
+        program="tesseract",
+        package="tesseract",
+        version_checker=lambda: '4.0.0-beta.1',
+        need_version='4.0.0',
+    )
+    vd.check_external_program(
+        program="tesseract",
+        package="tesseract",
+        version_checker=lambda: 'v5.0.0-alpha.20200201',
+        need_version='4.0.0',
+    )
+    with pytest.raises(MissingDependencyError):
+        vd.check_external_program(
+            program="dummy_fails",
+            package="dummy",
+            version_checker=lambda: '1.0',
+            need_version='2.0',
+        )
+
+
+def test_optional_program_recommended(caplog):
+    caplog.clear()
+
+    def raiser():
+        raise FileNotFoundError('jbig2')
+
+    with caplog.at_level(logging.WARNING):
+        vd.check_external_program(
+            program="jbig2",
+            package="jbig2enc",
+            version_checker=raiser,
+            need_version='42',
+            required_for='this test case',
+            recommended=True,
+        )
+        assert any(
+            (loglevel == logging.WARNING and "recommended" in msg)
+            for _logger_name, loglevel, msg in caplog.record_tuples
+        )