Fixed setup.py syntax error

Better get the last one out

.coveragerc

@@ -0,0 +1,16 @@
+# Coverage isn't really compatible with subprocesses so results are unreliable
+
+[run]
+branch = True
+#concurrency = multiprocessing
+source = ocrmypdf/
+
+[report]
+exclude_lines =
+	pragma: no cover
+	def __repr__
+	raise AssertionError
+	raise NotImplementedError
+	if 0:
+	if False:
+	if __name__ == .__main__.:

.docker/Dockerfile

@@ -1,8 +1,6 @@
 # OCRmyPDF
 #
-# VERSION               4.4.2
-FROM      ubuntu:16.10
-MAINTAINER James R. Barlow <jim@purplerock.ca>
+FROM      ubuntu:17.10
 
 RUN apt-get update && apt-get install -y --no-install-recommends \
   software-properties-common python-software-properties \
@@ -20,6 +18,8 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
   tesseract-ocr-spa \
   tesseract-ocr-deu
 
+ENV LANG=C.UTF-8
+
 RUN python3 -m venv --system-site-packages /appenv
 
 # This installs the latest binary wheel instead of the code in the current
@@ -51,4 +51,4 @@ WORKDIR /home/docker
 
 # Must use array form of ENTRYPOINT
 # Non-array form does not append other arguments, because that is "intuitive"
-ENTRYPOINT ["/application/docker-wrapper.sh"]
+ENTRYPOINT ["/application/.docker/docker-wrapper.sh"]

.docker/Dockerfile.polyglot

@@ -1,8 +1,6 @@
 # OCRmyPDF polyglot
 #
-# VERSION               4.4.2
 FROM      jbarlow83/ocrmypdf:latest
-MAINTAINER James R. Barlow <jim@purplerock.ca>
 
 USER root
 
@@ -16,4 +14,4 @@ USER docker
 
 # Must use array form of ENTRYPOINT
 # Non-array form does not append other arguments, because that is "intuitive"
-ENTRYPOINT ["/application/docker-wrapper.sh"]
+ENTRYPOINT ["/application/.docker/docker-wrapper.sh"]

.docker/Dockerfile.tess4

@@ -1,8 +1,6 @@
 # OCRmyPDF
 #
-# VERSION               4.4.2
-FROM 	ubuntu:16.10
-MAINTAINER James R. Barlow <jim@purplerock.ca>
+FROM 	ubuntu:17.10
 
 RUN apt-get update && apt-get install -y --no-install-recommends \
   software-properties-common python-software-properties \
@@ -30,6 +28,8 @@ RUN apt-get update \
 		tesseract-ocr-rus \
 		tesseract-ocr-chi-sim 
 
+ENV LANG=C.UTF-8
+
 RUN python3 -m venv --system-site-packages /appenv
 
 # This installs the latest binary wheel instead of the code in the current
@@ -61,4 +61,4 @@ WORKDIR /home/docker
 
 # Must use array form of ENTRYPOINT
 # Non-array form does not append other arguments, because that is "intuitive"
-ENTRYPOINT ["/application/docker-wrapper.sh"]
+ENTRYPOINT ["/application/.docker/docker-wrapper.sh"]

.dockerignore

@@ -3,18 +3,9 @@
 *.pyc
 *.rst
 *.sublime*
-*/*.pyc
-*/*/*.pyc
-*/*/*/*.pyc
-*/*/*/*/*.pyc
-*/*/*/*/*/*.pyc
-*/*/*/*/*/*/*.pyc
-*/*/*/*/*/*/*/*.pyc
-.cache/
-.eggs/
-.github/
-.git/
-.ipynb_checkpoints/
+**/*.pyc
+.*/
+!.docker/
 .ruffus_history.sqlite
 bin/
 build/

.gitattributes

@@ -8,6 +8,6 @@
 *.PDF	binary
 *.png	binary
 *.jpg	binary
-
+*.bin   binary
 
 .git_archival.txt  export-subst

.gitignore

@@ -7,6 +7,7 @@ tasks.py
 .bash_history
 .ruffus_history.sqlite
 .idea/
+.pytest_cache/
 
 # Package building
 *.egg-info/
@@ -34,8 +35,10 @@ log/
 /*.qdf
 *.ipynb
 .ipynb_checkpoints/
-tests/cache/
 tests/output/
 tests/resources/private/
 tmp/
-pdfbox-app*.jar
+pdfbox-app*.jar
+.vscode/
+IDEAS
+_Dockerfile.local

.travis.yml

@@ -4,13 +4,10 @@ cache:
   pip: true
   directories:
   - packages
-  - tests/cache
   - $HOME/Library/Caches/Homebrew
 
 env:
   global:
-    - secure: "oyX5xesoHD7qcDXKIxMyGZPi+H/WxcvfFkaviEmq84K1DDyHk48+9e92IKgrw8/lcTADnEo/AgVKfnhCPflFimk1xTkgaK4sUg1WLI2YjmaHcwl5SlBHa2rN3uGBwy1hyP92qyv/mMc9R59NtRJ8u76lbn6eN9wi7lkFWdE6BTw="  # DOCKERHUB_OCRMYPDF_TOKEN
-    - secure: "WlyII8YLsiUUyLtEA563GvEZmbneDb/T8q/P1uNbyQ2ps1U82tH0nSUV2CspSMxOFtZzPHCrRvnAmuTYKshBj+GNnBb1J9FKQmFwF+4NPeqsFdUkQ1NeeCmfIRShuNC3Otg2GGwj4Zssdg+QnVy43t2L11qizzfY+lY+MVzAYcM="  # DOCKERHUB_OCRMYPDF_TESS4_TOKEN
     - secure: "hsf6MT+n2x3OiDM2fQyJZdV0/PWYmv81LdVqC6cfnHBE/8N3DloJRqQ7WfO14TxhiK9PEC7MpyCj0lSabUHEO7gSH6Vks6I1asoSkt8S9/bSMlhT4hei+pwVpeGEiU5xHVATNjY+D919VC3IFvc3XmjT74h/2SLhaZ+jhEmDggM="  # HOMEBREW_OCRMYPDF_TOKEN
 
 matrix:
@@ -19,13 +16,25 @@ matrix:
       sudo: required
       language: python
       python: 3.5
+      env: EXTRAS=
     - os: linux
       sudo: required
       language: python
       python: 3.6
+      env: EXTRAS=
+    - os: linux
+      sudo: required
+      language: python
+      python: 3.6
+      env: EXTRAS=[fitz]
     - os: osx
       osx_image: xcode8
       language: generic
+      env: EXTRAS=
+    - os: osx
+      osx_image: xcode8
+      language: generic
+      env: EXTRAS=[fitz]
 
 before_cache:
 - rm -f $HOME/.cache/pip/log/debug.log
@@ -34,21 +43,23 @@ before_install: |
   if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then
     bash .travis/linux_before_install.sh
   elif [[ "$TRAVIS_OS_NAME" == "osx" ]]; then
-    bash .travis/osx_before_install.sh
+    brew update && brew bundle --file=.travis/Brewfile
+    pip3 install --upgrade pip
+    pip3 install wheel
   fi
 
 install:
-- pip3 install .
+- pip3 install ".$EXTRAS"
 - pip3 install -r requirements.txt -r test_requirements.txt
 
 script:
-- mv ocrmypdf dont_import_this_ocrmypdf
-- pytest -n auto
-- mv dont_import_this_ocrmypdf ocrmypdf
+- tesseract --version
+- qpdf --version
+- pytest
 
 deploy:
   # release for main pypi
-  # 3.6 is considered the build leader and does the deploy, otherwise there is
+  # 3.6 w/o fitz 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
@@ -60,12 +71,12 @@ deploy:
   on:
     branch: master
     tags: true
-    condition: $TRAVIS_PYTHON_VERSION == "3.6" && $TRAVIS_OS_NAME == "linux"
+    condition: $TRAVIS_PYTHON_VERSION == "3.6" && $TRAVIS_OS_NAME == "linux" && $EXTRAS == ""
   skip_upload_docs: true
 
   # test pypi
 - provider: pypi
-  server: https://testpypi.python.org/pypi
+  server: https://testpypi.pypi.org/legacy/
   user: ocrmypdf-travis
   password:
     secure: "DTFOmmNL6olA0+yXvp4u9jXZlZeqrJsJ0526jzqf4a3gZ6jnGTq5UI6WzRsslSyoMMfXKtHQebqHM6ogSgCZinyZ3ufHJo8fn9brxbEc2gsiWkbj5o3bGwdWMT1vNNE7XW0VCpw87rZ1EEwjl4FJHFudMlPR1yfU5+uq0k0PACo="
@@ -80,16 +91,13 @@ deploy:
   # we really just want to run after_deploy *after* pypi upload is done, but
   # after_deploy on runs if a given box deployed
 - provider: script
-  script: .travis/null_deploy.sh
+  script: /usr/bin/true
   on:
     branch: master
-    tags: false
+    tags: true
     condition: $TRAVIS_OS_NAME == "osx"
 
 after_deploy: |
   if [[ "$TRAVIS_OS_NAME" == "osx" ]]; then
     bash .travis/osx_brew.sh
-  elif [[ "$TRAVIS_PYTHON_VERSION" == "3.6" && "$TRAVIS_OS_NAME" == "linux" ]]; then
-    curl -H "Content-Type: application/json" --data '{"build": true}' -X POST https://registry.hub.docker.com/u/jbarlow83/ocrmypdf/trigger/$DOCKERHUB_OCRMYPDF_TOKEN/
-    curl -H "Content-Type: application/json" --data '{"build": true}' -X POST https://registry.hub.docker.com/u/jbarlow83/ocrmypdf-tess4/trigger/$DOCKERHUB_OCRMYPDF_TESS4_TOKEN/
-  fi
+  fi

.travis/Brewfile

@@ -0,0 +1,13 @@
+brew 'python'
+
+brew 'ghostscript'
+brew 'jbig2dec'
+brew 'leptonica'
+brew 'libffi'
+brew 'libtiff'
+brew 'libxml2'
+brew 'mupdf-tools'
+brew 'openjpeg'
+brew 'qpdf'
+brew 'tesseract'
+brew 'unpaper'

.travis/autobrew.py

@@ -1,3 +1,6 @@
+#!/usr/bin/env python3
+# © 2017-18 James R. Barlow: github.com/jbarlow83
+
 from string import Template
 from subprocess import run, PIPE
 import re
@@ -11,35 +14,51 @@ class Ocrmypdf < Formula
   ${ocrmypdf_url}
   ${ocrmypdf_sha256}
 
-  depends_on :python3
   depends_on "pkg-config" => :build
-  depends_on "libffi"
-  depends_on "tesseract"
-  depends_on "ghostscript"
-  depends_on "unpaper"
-  depends_on "qpdf"
-
-  # For Pillow source install
-  depends_on "openjpeg"
+  depends_on "mupdf-tools" => :build  # statically links libmupdf.a
   depends_on "freetype"
+  depends_on "ghostscript"
+  depends_on "jpeg"
   depends_on "libpng"
+  depends_on "python"
+  depends_on "qpdf"
+  depends_on "tesseract"
+  depends_on "unpaper"
 
 ${resources}
-
   def install
-    ENV.append ["SETUPTOOLS_SCM_PRETEND_VERSION"], "v${ocrmypdf_version}"
-    ENV.each do |key, value|
-      puts "#{key}:#{value}"
+    venv = virtualenv_create(libexec, "python3")
+
+    resource("Pillow").stage do
+      inreplace "setup.py" do |s|
+        sdkprefix = MacOS::CLT.installed? ? "" : MacOS.sdk_path
+        s.gsub! "openjpeg.h", "probably_not_a_header_called_this_eh.h"
+        s.gsub! "ZLIB_ROOT = None", "ZLIB_ROOT = ('#{sdkprefix}/usr/lib', '#{sdkprefix}/usr/include')"
+        s.gsub! "JPEG_ROOT = None", "JPEG_ROOT = ('#{Formula["jpeg"].opt_prefix}/lib', '#{Formula["jpeg"].opt_prefix}/include')"
+        s.gsub! "FREETYPE_ROOT = None", "FREETYPE_ROOT = ('#{Formula["freetype"].opt_prefix}/lib', '#{Formula["freetype"].opt_prefix}/include')"
+      end
+
+      # avoid triggering "helpful" distutils code that doesn't recognize Xcode 7 .tbd stubs
+      ENV.append "CFLAGS", "-I#{MacOS.sdk_path}/System/Library/Frameworks/Tk.framework/Versions/8.5/Headers" unless MacOS::CLT.installed?
+      venv.pip_install Pathname.pwd
     end
-    virtualenv_install_with_resources
+
+    res = resources.map(&:name).to_set - ["Pillow"]
+
+    res.each do |r|
+      venv.pip_install resource(r)
+    end
+
+    venv.pip_install_and_link buildpath
   end
 
   test do
-    # `test do` will create, run in and delete a temporary directory.
-    #
-    # The installed folder is not in the path, so use the entire path to any
-    # executables being tested: `system "#{bin}/program", "do", "something"`.
-    system "#{bin}/ocrmypdf", "--version"
+    # Since we use Python 3, we require a UTF-8 locale
+    ENV["LC_ALL"] = "en_US.UTF-8"
+
+    system "#{bin}/ocrmypdf", "-f", "-q", "--deskew",
+                              test_fixtures("test.pdf"), "ocr.pdf"
+    assert_predicate testpath/"ocr.pdf", :exist?
   end
 end
 """)

.travis/linux_before_install.sh

@@ -1,4 +1,5 @@
 #!/bin/bash
+# © 2017 James R. Barlow: github.com/jbarlow83
 set -euo pipefail
 set -x
 
@@ -26,5 +27,5 @@ sudo apt-get install -y --no-install-recommends \
 
 pip install --upgrade pip
 mkdir -p packages
-[ -f packages/unpaper_6.1-1.deb ] || wget -q https://dl.dropboxusercontent.com/u/28971240/unpaper_6.1-1.deb -O packages/unpaper_6.1-1.deb
+[ -f packages/unpaper_6.1-1.deb ] || wget -q 'https://www.dropbox.com/s/vaq0kbwi6e6au80/unpaper_6.1-1.deb?raw=1' -O packages/unpaper_6.1-1.deb
 sudo dpkg -i packages/unpaper_6.1-1.deb

.travis/null_deploy.sh

@@ -1,3 +0,0 @@
-#!/usr/bin/env bash
-
-exit 0

.travis/osx_before_install.sh

@@ -1,16 +0,0 @@
-#!/bin/bash
-set -euo pipefail
-set -x
-
-brew update
-
-brew install openjpeg jbig2dec libtiff     # image libraries
-brew install qpdf
-brew install ghostscript
-brew install python3
-brew install libxml2 libffi leptonica
-brew install unpaper   # optional
-brew install tesseract
-
-pip3 install --upgrade pip
-pip3 install wheel

.travis/osx_brew.sh

@@ -1,11 +1,14 @@
 #!/bin/bash
-set -euo pipefail
+# © 2017 James R. Barlow: github.com/jbarlow83
+set -uo pipefail
 set -x
 
 pip3 install homebrew-pypi-poet
 python3 .travis/autobrew.py
 cat ocrmypdf.rb
-brew audit ocrmypdf.rb
+
+# brew audit crashes Travis
+#brew audit ocrmypdf.rb
 
 # Important: disable debug output so token is hidden
 set +x

LICENSE

@@ -0,0 +1,674 @@
+                    GNU GENERAL PUBLIC LICENSE
+                       Version 3, 29 June 2007
+
+ Copyright (C) 2007 Free Software Foundation, Inc. <https://fsf.org/>
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+                            Preamble
+
+  The GNU General Public License is a free, copyleft license for
+software and other kinds of works.
+
+  The licenses for most software and other practical works are designed
+to take away your freedom to share and change the works.  By contrast,
+the GNU General Public License is intended to guarantee your freedom to
+share and change all versions of a program--to make sure it remains free
+software for all its users.  We, the Free Software Foundation, use the
+GNU General Public License for most of our software; it applies also to
+any other work released this way by its authors.  You can apply it to
+your programs, too.
+
+  When we speak of free software, we are referring to freedom, not
+price.  Our General Public Licenses are designed to make sure that you
+have the freedom to distribute copies of free software (and charge for
+them if you wish), that you receive source code or can get it if you
+want it, that you can change the software or use pieces of it in new
+free programs, and that you know you can do these things.
+
+  To protect your rights, we need to prevent others from denying you
+these rights or asking you to surrender the rights.  Therefore, you have
+certain responsibilities if you distribute copies of the software, or if
+you modify it: responsibilities to respect the freedom of others.
+
+  For example, if you distribute copies of such a program, whether
+gratis or for a fee, you must pass on to the recipients the same
+freedoms that you received.  You must make sure that they, too, receive
+or can get the source code.  And you must show them these terms so they
+know their rights.
+
+  Developers that use the GNU GPL protect your rights with two steps:
+(1) assert copyright on the software, and (2) offer you this License
+giving you legal permission to copy, distribute and/or modify it.
+
+  For the developers' and authors' protection, the GPL clearly explains
+that there is no warranty for this free software.  For both users' and
+authors' sake, the GPL requires that modified versions be marked as
+changed, so that their problems will not be attributed erroneously to
+authors of previous versions.
+
+  Some devices are designed to deny users access to install or run
+modified versions of the software inside them, although the manufacturer
+can do so.  This is fundamentally incompatible with the aim of
+protecting users' freedom to change the software.  The systematic
+pattern of such abuse occurs in the area of products for individuals to
+use, which is precisely where it is most unacceptable.  Therefore, we
+have designed this version of the GPL to prohibit the practice for those
+products.  If such problems arise substantially in other domains, we
+stand ready to extend this provision to those domains in future versions
+of the GPL, as needed to protect the freedom of users.
+
+  Finally, every program is threatened constantly by software patents.
+States should not allow patents to restrict development and use of
+software on general-purpose computers, but in those that do, we wish to
+avoid the special danger that patents applied to a free program could
+make it effectively proprietary.  To prevent this, the GPL assures that
+patents cannot be used to render the program non-free.
+
+  The precise terms and conditions for copying, distribution and
+modification follow.
+
+                       TERMS AND CONDITIONS
+
+  0. Definitions.
+
+  "This License" refers to version 3 of the GNU General Public License.
+
+  "Copyright" also means copyright-like laws that apply to other kinds of
+works, such as semiconductor masks.
+
+  "The Program" refers to any copyrightable work licensed under this
+License.  Each licensee is addressed as "you".  "Licensees" and
+"recipients" may be individuals or organizations.
+
+  To "modify" a work means to copy from or adapt all or part of the work
+in a fashion requiring copyright permission, other than the making of an
+exact copy.  The resulting work is called a "modified version" of the
+earlier work or a work "based on" the earlier work.
+
+  A "covered work" means either the unmodified Program or a work based
+on the Program.
+
+  To "propagate" a work means to do anything with it that, without
+permission, would make you directly or secondarily liable for
+infringement under applicable copyright law, except executing it on a
+computer or modifying a private copy.  Propagation includes copying,
+distribution (with or without modification), making available to the
+public, and in some countries other activities as well.
+
+  To "convey" a work means any kind of propagation that enables other
+parties to make or receive copies.  Mere interaction with a user through
+a computer network, with no transfer of a copy, is not conveying.
+
+  An interactive user interface displays "Appropriate Legal Notices"
+to the extent that it includes a convenient and prominently visible
+feature that (1) displays an appropriate copyright notice, and (2)
+tells the user that there is no warranty for the work (except to the
+extent that warranties are provided), that licensees may convey the
+work under this License, and how to view a copy of this License.  If
+the interface presents a list of user commands or options, such as a
+menu, a prominent item in the list meets this criterion.
+
+  1. Source Code.
+
+  The "source code" for a work means the preferred form of the work
+for making modifications to it.  "Object code" means any non-source
+form of a work.
+
+  A "Standard Interface" means an interface that either is an official
+standard defined by a recognized standards body, or, in the case of
+interfaces specified for a particular programming language, one that
+is widely used among developers working in that language.
+
+  The "System Libraries" of an executable work include anything, other
+than the work as a whole, that (a) is included in the normal form of
+packaging a Major Component, but which is not part of that Major
+Component, and (b) serves only to enable use of the work with that
+Major Component, or to implement a Standard Interface for which an
+implementation is available to the public in source code form.  A
+"Major Component", in this context, means a major essential component
+(kernel, window system, and so on) of the specific operating system
+(if any) on which the executable work runs, or a compiler used to
+produce the work, or an object code interpreter used to run it.
+
+  The "Corresponding Source" for a work in object code form means all
+the source code needed to generate, install, and (for an executable
+work) run the object code and to modify the work, including scripts to
+control those activities.  However, it does not include the work's
+System Libraries, or general-purpose tools or generally available free
+programs which are used unmodified in performing those activities but
+which are not part of the work.  For example, Corresponding Source
+includes interface definition files associated with source files for
+the work, and the source code for shared libraries and dynamically
+linked subprograms that the work is specifically designed to require,
+such as by intimate data communication or control flow between those
+subprograms and other parts of the work.
+
+  The Corresponding Source need not include anything that users
+can regenerate automatically from other parts of the Corresponding
+Source.
+
+  The Corresponding Source for a work in source code form is that
+same work.
+
+  2. Basic Permissions.
+
+  All rights granted under this License are granted for the term of
+copyright on the Program, and are irrevocable provided the stated
+conditions are met.  This License explicitly affirms your unlimited
+permission to run the unmodified Program.  The output from running a
+covered work is covered by this License only if the output, given its
+content, constitutes a covered work.  This License acknowledges your
+rights of fair use or other equivalent, as provided by copyright law.
+
+  You may make, run and propagate covered works that you do not
+convey, without conditions so long as your license otherwise remains
+in force.  You may convey covered works to others for the sole purpose
+of having them make modifications exclusively for you, or provide you
+with facilities for running those works, provided that you comply with
+the terms of this License in conveying all material for which you do
+not control copyright.  Those thus making or running the covered works
+for you must do so exclusively on your behalf, under your direction
+and control, on terms that prohibit them from making any copies of
+your copyrighted material outside their relationship with you.
+
+  Conveying under any other circumstances is permitted solely under
+the conditions stated below.  Sublicensing is not allowed; section 10
+makes it unnecessary.
+
+  3. Protecting Users' Legal Rights From Anti-Circumvention Law.
+
+  No covered work shall be deemed part of an effective technological
+measure under any applicable law fulfilling obligations under article
+11 of the WIPO copyright treaty adopted on 20 December 1996, or
+similar laws prohibiting or restricting circumvention of such
+measures.
+
+  When you convey a covered work, you waive any legal power to forbid
+circumvention of technological measures to the extent such circumvention
+is effected by exercising rights under this License with respect to
+the covered work, and you disclaim any intention to limit operation or
+modification of the work as a means of enforcing, against the work's
+users, your or third parties' legal rights to forbid circumvention of
+technological measures.
+
+  4. Conveying Verbatim Copies.
+
+  You may convey verbatim copies of the Program's source code as you
+receive it, in any medium, provided that you conspicuously and
+appropriately publish on each copy an appropriate copyright notice;
+keep intact all notices stating that this License and any
+non-permissive terms added in accord with section 7 apply to the code;
+keep intact all notices of the absence of any warranty; and give all
+recipients a copy of this License along with the Program.
+
+  You may charge any price or no price for each copy that you convey,
+and you may offer support or warranty protection for a fee.
+
+  5. Conveying Modified Source Versions.
+
+  You may convey a work based on the Program, or the modifications to
+produce it from the Program, in the form of source code under the
+terms of section 4, provided that you also meet all of these conditions:
+
+    a) The work must carry prominent notices stating that you modified
+    it, and giving a relevant date.
+
+    b) The work must carry prominent notices stating that it is
+    released under this License and any conditions added under section
+    7.  This requirement modifies the requirement in section 4 to
+    "keep intact all notices".
+
+    c) You must license the entire work, as a whole, under this
+    License to anyone who comes into possession of a copy.  This
+    License will therefore apply, along with any applicable section 7
+    additional terms, to the whole of the work, and all its parts,
+    regardless of how they are packaged.  This License gives no
+    permission to license the work in any other way, but it does not
+    invalidate such permission if you have separately received it.
+
+    d) If the work has interactive user interfaces, each must display
+    Appropriate Legal Notices; however, if the Program has interactive
+    interfaces that do not display Appropriate Legal Notices, your
+    work need not make them do so.
+
+  A compilation of a covered work with other separate and independent
+works, which are not by their nature extensions of the covered work,
+and which are not combined with it such as to form a larger program,
+in or on a volume of a storage or distribution medium, is called an
+"aggregate" if the compilation and its resulting copyright are not
+used to limit the access or legal rights of the compilation's users
+beyond what the individual works permit.  Inclusion of a covered work
+in an aggregate does not cause this License to apply to the other
+parts of the aggregate.
+
+  6. Conveying Non-Source Forms.
+
+  You may convey a covered work in object code form under the terms
+of sections 4 and 5, provided that you also convey the
+machine-readable Corresponding Source under the terms of this License,
+in one of these ways:
+
+    a) Convey the object code in, or embodied in, a physical product
+    (including a physical distribution medium), accompanied by the
+    Corresponding Source fixed on a durable physical medium
+    customarily used for software interchange.
+
+    b) Convey the object code in, or embodied in, a physical product
+    (including a physical distribution medium), accompanied by a
+    written offer, valid for at least three years and valid for as
+    long as you offer spare parts or customer support for that product
+    model, to give anyone who possesses the object code either (1) a
+    copy of the Corresponding Source for all the software in the
+    product that is covered by this License, on a durable physical
+    medium customarily used for software interchange, for a price no
+    more than your reasonable cost of physically performing this
+    conveying of source, or (2) access to copy the
+    Corresponding Source from a network server at no charge.
+
+    c) Convey individual copies of the object code with a copy of the
+    written offer to provide the Corresponding Source.  This
+    alternative is allowed only occasionally and noncommercially, and
+    only if you received the object code with such an offer, in accord
+    with subsection 6b.
+
+    d) Convey the object code by offering access from a designated
+    place (gratis or for a charge), and offer equivalent access to the
+    Corresponding Source in the same way through the same place at no
+    further charge.  You need not require recipients to copy the
+    Corresponding Source along with the object code.  If the place to
+    copy the object code is a network server, the Corresponding Source
+    may be on a different server (operated by you or a third party)
+    that supports equivalent copying facilities, provided you maintain
+    clear directions next to the object code saying where to find the
+    Corresponding Source.  Regardless of what server hosts the
+    Corresponding Source, you remain obligated to ensure that it is
+    available for as long as needed to satisfy these requirements.
+
+    e) Convey the object code using peer-to-peer transmission, provided
+    you inform other peers where the object code and Corresponding
+    Source of the work are being offered to the general public at no
+    charge under subsection 6d.
+
+  A separable portion of the object code, whose source code is excluded
+from the Corresponding Source as a System Library, need not be
+included in conveying the object code work.
+
+  A "User Product" is either (1) a "consumer product", which means any
+tangible personal property which is normally used for personal, family,
+or household purposes, or (2) anything designed or sold for incorporation
+into a dwelling.  In determining whether a product is a consumer product,
+doubtful cases shall be resolved in favor of coverage.  For a particular
+product received by a particular user, "normally used" refers to a
+typical or common use of that class of product, regardless of the status
+of the particular user or of the way in which the particular user
+actually uses, or expects or is expected to use, the product.  A product
+is a consumer product regardless of whether the product has substantial
+commercial, industrial or non-consumer uses, unless such uses represent
+the only significant mode of use of the product.
+
+  "Installation Information" for a User Product means any methods,
+procedures, authorization keys, or other information required to install
+and execute modified versions of a covered work in that User Product from
+a modified version of its Corresponding Source.  The information must
+suffice to ensure that the continued functioning of the modified object
+code is in no case prevented or interfered with solely because
+modification has been made.
+
+  If you convey an object code work under this section in, or with, or
+specifically for use in, a User Product, and the conveying occurs as
+part of a transaction in which the right of possession and use of the
+User Product is transferred to the recipient in perpetuity or for a
+fixed term (regardless of how the transaction is characterized), the
+Corresponding Source conveyed under this section must be accompanied
+by the Installation Information.  But this requirement does not apply
+if neither you nor any third party retains the ability to install
+modified object code on the User Product (for example, the work has
+been installed in ROM).
+
+  The requirement to provide Installation Information does not include a
+requirement to continue to provide support service, warranty, or updates
+for a work that has been modified or installed by the recipient, or for
+the User Product in which it has been modified or installed.  Access to a
+network may be denied when the modification itself materially and
+adversely affects the operation of the network or violates the rules and
+protocols for communication across the network.
+
+  Corresponding Source conveyed, and Installation Information provided,
+in accord with this section must be in a format that is publicly
+documented (and with an implementation available to the public in
+source code form), and must require no special password or key for
+unpacking, reading or copying.
+
+  7. Additional Terms.
+
+  "Additional permissions" are terms that supplement the terms of this
+License by making exceptions from one or more of its conditions.
+Additional permissions that are applicable to the entire Program shall
+be treated as though they were included in this License, to the extent
+that they are valid under applicable law.  If additional permissions
+apply only to part of the Program, that part may be used separately
+under those permissions, but the entire Program remains governed by
+this License without regard to the additional permissions.
+
+  When you convey a copy of a covered work, you may at your option
+remove any additional permissions from that copy, or from any part of
+it.  (Additional permissions may be written to require their own
+removal in certain cases when you modify the work.)  You may place
+additional permissions on material, added by you to a covered work,
+for which you have or can give appropriate copyright permission.
+
+  Notwithstanding any other provision of this License, for material you
+add to a covered work, you may (if authorized by the copyright holders of
+that material) supplement the terms of this License with terms:
+
+    a) Disclaiming warranty or limiting liability differently from the
+    terms of sections 15 and 16 of this License; or
+
+    b) Requiring preservation of specified reasonable legal notices or
+    author attributions in that material or in the Appropriate Legal
+    Notices displayed by works containing it; or
+
+    c) Prohibiting misrepresentation of the origin of that material, or
+    requiring that modified versions of such material be marked in
+    reasonable ways as different from the original version; or
+
+    d) Limiting the use for publicity purposes of names of licensors or
+    authors of the material; or
+
+    e) Declining to grant rights under trademark law for use of some
+    trade names, trademarks, or service marks; or
+
+    f) Requiring indemnification of licensors and authors of that
+    material by anyone who conveys the material (or modified versions of
+    it) with contractual assumptions of liability to the recipient, for
+    any liability that these contractual assumptions directly impose on
+    those licensors and authors.
+
+  All other non-permissive additional terms are considered "further
+restrictions" within the meaning of section 10.  If the Program as you
+received it, or any part of it, contains a notice stating that it is
+governed by this License along with a term that is a further
+restriction, you may remove that term.  If a license document contains
+a further restriction but permits relicensing or conveying under this
+License, you may add to a covered work material governed by the terms
+of that license document, provided that the further restriction does
+not survive such relicensing or conveying.
+
+  If you add terms to a covered work in accord with this section, you
+must place, in the relevant source files, a statement of the
+additional terms that apply to those files, or a notice indicating
+where to find the applicable terms.
+
+  Additional terms, permissive or non-permissive, may be stated in the
+form of a separately written license, or stated as exceptions;
+the above requirements apply either way.
+
+  8. Termination.
+
+  You may not propagate or modify a covered work except as expressly
+provided under this License.  Any attempt otherwise to propagate or
+modify it is void, and will automatically terminate your rights under
+this License (including any patent licenses granted under the third
+paragraph of section 11).
+
+  However, if you cease all violation of this License, then your
+license from a particular copyright holder is reinstated (a)
+provisionally, unless and until the copyright holder explicitly and
+finally terminates your license, and (b) permanently, if the copyright
+holder fails to notify you of the violation by some reasonable means
+prior to 60 days after the cessation.
+
+  Moreover, your license from a particular copyright holder is
+reinstated permanently if the copyright holder notifies you of the
+violation by some reasonable means, this is the first time you have
+received notice of violation of this License (for any work) from that
+copyright holder, and you cure the violation prior to 30 days after
+your receipt of the notice.
+
+  Termination of your rights under this section does not terminate the
+licenses of parties who have received copies or rights from you under
+this License.  If your rights have been terminated and not permanently
+reinstated, you do not qualify to receive new licenses for the same
+material under section 10.
+
+  9. Acceptance Not Required for Having Copies.
+
+  You are not required to accept this License in order to receive or
+run a copy of the Program.  Ancillary propagation of a covered work
+occurring solely as a consequence of using peer-to-peer transmission
+to receive a copy likewise does not require acceptance.  However,
+nothing other than this License grants you permission to propagate or
+modify any covered work.  These actions infringe copyright if you do
+not accept this License.  Therefore, by modifying or propagating a
+covered work, you indicate your acceptance of this License to do so.
+
+  10. Automatic Licensing of Downstream Recipients.
+
+  Each time you convey a covered work, the recipient automatically
+receives a license from the original licensors, to run, modify and
+propagate that work, subject to this License.  You are not responsible
+for enforcing compliance by third parties with this License.
+
+  An "entity transaction" is a transaction transferring control of an
+organization, or substantially all assets of one, or subdividing an
+organization, or merging organizations.  If propagation of a covered
+work results from an entity transaction, each party to that
+transaction who receives a copy of the work also receives whatever
+licenses to the work the party's predecessor in interest had or could
+give under the previous paragraph, plus a right to possession of the
+Corresponding Source of the work from the predecessor in interest, if
+the predecessor has it or can get it with reasonable efforts.
+
+  You may not impose any further restrictions on the exercise of the
+rights granted or affirmed under this License.  For example, you may
+not impose a license fee, royalty, or other charge for exercise of
+rights granted under this License, and you may not initiate litigation
+(including a cross-claim or counterclaim in a lawsuit) alleging that
+any patent claim is infringed by making, using, selling, offering for
+sale, or importing the Program or any portion of it.
+
+  11. Patents.
+
+  A "contributor" is a copyright holder who authorizes use under this
+License of the Program or a work on which the Program is based.  The
+work thus licensed is called the contributor's "contributor version".
+
+  A contributor's "essential patent claims" are all patent claims
+owned or controlled by the contributor, whether already acquired or
+hereafter acquired, that would be infringed by some manner, permitted
+by this License, of making, using, or selling its contributor version,
+but do not include claims that would be infringed only as a
+consequence of further modification of the contributor version.  For
+purposes of this definition, "control" includes the right to grant
+patent sublicenses in a manner consistent with the requirements of
+this License.
+
+  Each contributor grants you a non-exclusive, worldwide, royalty-free
+patent license under the contributor's essential patent claims, to
+make, use, sell, offer for sale, import and otherwise run, modify and
+propagate the contents of its contributor version.
+
+  In the following three paragraphs, a "patent license" is any express
+agreement or commitment, however denominated, not to enforce a patent
+(such as an express permission to practice a patent or covenant not to
+sue for patent infringement).  To "grant" such a patent license to a
+party means to make such an agreement or commitment not to enforce a
+patent against the party.
+
+  If you convey a covered work, knowingly relying on a patent license,
+and the Corresponding Source of the work is not available for anyone
+to copy, free of charge and under the terms of this License, through a
+publicly available network server or other readily accessible means,
+then you must either (1) cause the Corresponding Source to be so
+available, or (2) arrange to deprive yourself of the benefit of the
+patent license for this particular work, or (3) arrange, in a manner
+consistent with the requirements of this License, to extend the patent
+license to downstream recipients.  "Knowingly relying" means you have
+actual knowledge that, but for the patent license, your conveying the
+covered work in a country, or your recipient's use of the covered work
+in a country, would infringe one or more identifiable patents in that
+country that you have reason to believe are valid.
+
+  If, pursuant to or in connection with a single transaction or
+arrangement, you convey, or propagate by procuring conveyance of, a
+covered work, and grant a patent license to some of the parties
+receiving the covered work authorizing them to use, propagate, modify
+or convey a specific copy of the covered work, then the patent license
+you grant is automatically extended to all recipients of the covered
+work and works based on it.
+
+  A patent license is "discriminatory" if it does not include within
+the scope of its coverage, prohibits the exercise of, or is
+conditioned on the non-exercise of one or more of the rights that are
+specifically granted under this License.  You may not convey a covered
+work if you are a party to an arrangement with a third party that is
+in the business of distributing software, under which you make payment
+to the third party based on the extent of your activity of conveying
+the work, and under which the third party grants, to any of the
+parties who would receive the covered work from you, a discriminatory
+patent license (a) in connection with copies of the covered work
+conveyed by you (or copies made from those copies), or (b) primarily
+for and in connection with specific products or compilations that
+contain the covered work, unless you entered into that arrangement,
+or that patent license was granted, prior to 28 March 2007.
+
+  Nothing in this License shall be construed as excluding or limiting
+any implied license or other defenses to infringement that may
+otherwise be available to you under applicable patent law.
+
+  12. No Surrender of Others' Freedom.
+
+  If conditions are imposed on you (whether by court order, agreement or
+otherwise) that contradict the conditions of this License, they do not
+excuse you from the conditions of this License.  If you cannot convey a
+covered work so as to satisfy simultaneously your obligations under this
+License and any other pertinent obligations, then as a consequence you may
+not convey it at all.  For example, if you agree to terms that obligate you
+to collect a royalty for further conveying from those to whom you convey
+the Program, the only way you could satisfy both those terms and this
+License would be to refrain entirely from conveying the Program.
+
+  13. Use with the GNU Affero General Public License.
+
+  Notwithstanding any other provision of this License, you have
+permission to link or combine any covered work with a work licensed
+under version 3 of the GNU Affero General Public License into a single
+combined work, and to convey the resulting work.  The terms of this
+License will continue to apply to the part which is the covered work,
+but the special requirements of the GNU Affero General Public License,
+section 13, concerning interaction through a network will apply to the
+combination as such.
+
+  14. Revised Versions of this License.
+
+  The Free Software Foundation may publish revised and/or new versions of
+the GNU General Public License from time to time.  Such new versions will
+be similar in spirit to the present version, but may differ in detail to
+address new problems or concerns.
+
+  Each version is given a distinguishing version number.  If the
+Program specifies that a certain numbered version of the GNU General
+Public License "or any later version" applies to it, you have the
+option of following the terms and conditions either of that numbered
+version or of any later version published by the Free Software
+Foundation.  If the Program does not specify a version number of the
+GNU General Public License, you may choose any version ever published
+by the Free Software Foundation.
+
+  If the Program specifies that a proxy can decide which future
+versions of the GNU General Public License can be used, that proxy's
+public statement of acceptance of a version permanently authorizes you
+to choose that version for the Program.
+
+  Later license versions may give you additional or different
+permissions.  However, no additional obligations are imposed on any
+author or copyright holder as a result of your choosing to follow a
+later version.
+
+  15. Disclaimer of Warranty.
+
+  THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY
+APPLICABLE LAW.  EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT
+HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY
+OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO,
+THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+PURPOSE.  THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM
+IS WITH YOU.  SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF
+ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
+
+  16. Limitation of Liability.
+
+  IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
+WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS
+THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY
+GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE
+USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF
+DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD
+PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS),
+EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF
+SUCH DAMAGES.
+
+  17. Interpretation of Sections 15 and 16.
+
+  If the disclaimer of warranty and limitation of liability provided
+above cannot be given local legal effect according to their terms,
+reviewing courts shall apply local law that most closely approximates
+an absolute waiver of all civil liability in connection with the
+Program, unless a warranty or assumption of liability accompanies a
+copy of the Program in return for a fee.
+
+                     END OF TERMS AND CONDITIONS
+
+            How to Apply These Terms to Your New Programs
+
+  If you develop a new program, and you want it to be of the greatest
+possible use to the public, the best way to achieve this is to make it
+free software which everyone can redistribute and change under these terms.
+
+  To do so, attach the following notices to the program.  It is safest
+to attach them to the start of each source file to most effectively
+state the exclusion of warranty; and each file should have at least
+the "copyright" line and a pointer to where the full notice is found.
+
+    <one line to give the program's name and a brief idea of what it does.>
+    Copyright (C) <year>  <name of author>
+
+    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
+    the Free Software Foundation, either version 3 of the License, or
+    (at your option) any later version.
+
+    This program 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 this program.  If not, see <https://www.gnu.org/licenses/>.
+
+Also add information on how to contact you by electronic and paper mail.
+
+  If the program does terminal interaction, make it output a short
+notice like this when it starts in an interactive mode:
+
+    <program>  Copyright (C) <year>  <name of author>
+    This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
+    This is free software, and you are welcome to redistribute it
+    under certain conditions; type `show c' for details.
+
+The hypothetical commands `show w' and `show c' should show the appropriate
+parts of the General Public License.  Of course, your program's commands
+might be different; for a GUI interface, you would use an "about box".
+
+  You should also get your employer (if you work as a programmer) or school,
+if any, to sign a "copyright disclaimer" for the program, if necessary.
+For more information on this, and how to apply and follow the GNU GPL, see
+<https://www.gnu.org/licenses/>.
+
+  The GNU General Public License does not permit incorporating your program
+into proprietary programs.  If your program is a subroutine library, you
+may consider it more useful to permit linking proprietary applications with
+the library.  If this is what you want to do, use the GNU Lesser General
+Public License instead of this License.  But first, please read
+<https://www.gnu.org/licenses/why-not-lgpl.html>.

LICENSE.rst

@@ -1,20 +0,0 @@
-Copyright (c) 2013-2016, The OCRmyPDF Authors
-
-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.

MANIFEST.in

@@ -7,26 +7,25 @@ include dev_requirements.txt
 include .git_archival.txt
 
 # docker
-include Dockerfile
-include Dockerfile.polyglot
 include .dockerignore
+recursive-include .docker *
 
 # tests
-include pytest.ini
+include .coveragerc
 recursive-include tests *.jpg
 recursive-include tests *.png
 recursive-include tests *.pdf
 recursive-include tests *.py
 recursive-include tests *.rst
 recursive-include tests *.txt
+recursive-include tests/cache *
 recursive-exclude tests/output *
 recursive-exclude tests/output_pageinfo *
 recursive-exclude tests/resources/private *
-recursive-exclude tests/cache *
 
 # documentation
+include LICENSE
 include *.rst
-include pipeline.svg
 recursive-exclude .github * 
 recursive-include docs *.py
 recursive-include docs *.rst
@@ -35,14 +34,16 @@ recursive-exclude docs/_build *
 
 
 # support files
-recursive-include ocrmypdf/data *
+recursive-include src/ocrmypdf/data *
 include *.py
 exclude tasks.py
+recursive-exclude .travis *
+exclude .travis*
+
 
 # code
-recursive-include ocrmypdf *.py
+recursive-include src/ocrmypdf *.py
 exclude ocrmypdf/lib/_leptonica.py
-include OCRmyPDF.sh
-include docker-wrapper.sh
+
 
 

README.rst

@@ -1,6 +1,17 @@
 OCRmyPDF
 ========
 
+.. image:: https://travis-ci.org/jbarlow83/OCRmyPDF.svg?branch=master
+    :target: https://travis-ci.org/jbarlow83/OCRmyPDF
+
+.. image:: https://img.shields.io/pypi/v/ocrmypdf.svg   
+    :target: https://pypi.org/project/ocrmypdf/
+
+.. image:: https://img.shields.io/homebrew/v/ocrmypdf.svg   
+    :alt: homebrew
+    :target: http://brewformulas.org/Ocrmypdf
+
+
 OCRmyPDF adds an OCR text layer to scanned PDF files, allowing them to
 be searched or copy-pasted.
 
@@ -65,11 +76,10 @@ Users of Debian 9 or later or Ubuntu 16.10 or later may simply
 
    apt-get install ocrmypdf
 
-and macOS users may simply
+and macOS users with Homebrew may simply
 
 .. code-block:: bash
 
-   brew tap jbarlow83/ocrmypdf
    brew install ocrmypdf
 
 For everyone else, `see our documentation <https://ocrmypdf.readthedocs.io/en/latest/installation.html>`_ for installation steps.
@@ -113,6 +123,13 @@ If you detect an issue, please:
 -  If possible provide your input PDF file as well as the content of the
    temporary folder (using a file sharing service like Dropbox)
 
+Requirements
+------------
+
+Runs on CPython 3.6, and requires external program installations of Ghostscript, Tesseract OCR, QPDF, and Leptonica. ocrmypdf is pure Python, but uses CFFI to portably generate library bindings.
+
+Python 3.5 is also supported.
+
 Press & Media
 -------------
 
@@ -122,6 +139,15 @@ Press & Media
 -  `heise Open Source, 09/2014: Texterkennung mit
    OCRmyPDF <http://heise.de/-2356670>`_
 
+License
+-------
+
+The OCRmyPDF software is licensed under the GNU GPLv3. Certain files are covered by other licenses, as noted in their source files. 
+
+The license for each test file varies, and is noted in tests/resources/README.rst. The documentation is licensed under Creative Commons Attribution-ShareAlike 4.0 (CC-BY-SA 4.0).
+
+OCRmyPDF versions prior to 6.0 were licensed under the MIT License.
+
 Disclaimer
 ----------
 

dev_requirements.txt

@@ -1,10 +1,4 @@
 check-manifest >= 0.35
 twine >= 1.8.1
 coverage >= 4.4
-pytest-xdist >= 1.16.0
 GitPython == 2.1.3
-
-# Known good versions: 1.11.1
-# Known broken versions: 1.15.0
-setuptools-scm == 1.11.1
-setuptools-scm-git-archive == 1.0

docs/advanced.rst

@@ -33,20 +33,23 @@ If you want to adjust the amount of time spent on OCR, change ``--tesseract-time
 Overriding default tesseract
 """"""""""""""""""""""""""""
 
-OCRmyPDF checks the environment variable ``OCRMYPDF_TESSERACT`` for the full path *to the tesseract executable* first. 
+OCRmyPDF checks the system ``PATH`` for the ``tesseract`` binary.
 
-For example, if you are testing tesseract 4.00 and don't wish to disturb your tesseract 3.04 installation, you can launch OCRmyPDF as follows:
+.. envvar:: TESSDATA_PREFIX
+
+	A Tesseract environment variable that overrides the path to Tesseract's data files.
+
+For example, if you are testing tesseract 4.00 and don't wish to use an existing tesseract 3.04 installation, you can launch OCRmyPDF as follows:
 
 .. code-block:: bash
 
 	env \
-		OCRMYPDF_TESSERACT=/home/user/src/tesseract4/api/tesseract \
+		PATH=/home/user/src/tesseract4/api:$PATH \
 		TESSDATA_PREFIX=/home/user/src/tesseract4 \
-		ocrmypdf --pdf-renderer tess4 --tesseract-oem 2 input.pdf output.pdf
+		ocrmypdf --tesseract-oem 2 input.pdf output.pdf
+
+In this example ``TESSDATA_PREFIX`` directs Tesseract 4.0 to use LSTM training data. ``--tesseract-oem 1`` requests tesseract 4.0's new LSTM engine. (Tesseract 4.0 only.)
 
-* ``TESSDATA_PREFIX`` directs tesseract 4.0 to use LSTM training data. This is a tesseract environment variable.
-* ``--pdf-renderer tess4`` takes advantage of new tesseract 4.0 PDF renderer in OCRmyPDF. (Tesseract 4.0 only.)
-* ``--tesseract-oem 1`` requests tesseract 4.0's new LSTM engine. (Tesseract 4.0 only.)
 
 Overriding other support programs
 """""""""""""""""""""""""""""""""
@@ -57,7 +60,8 @@ In addition to tesseract, OCRmyPDF uses the following external binaries:
 * ``unpaper``
 * ``qpdf``
 
-In each case OCRmyPDF will check the environment variable ``OCRMYPDF_{program}`` before asking the system to find ``{program}`` on the PATH. For example, you could redirect OCRmyPDF to ``OCRMYPDF_GS`` to override Ghostscript.
+In each case OCRmyPDF will search the ``PATH`` environment variable to locate the binaries.
+
 
 Changing tesseract configuration variables
 """"""""""""""""""""""""""""""""""""""""""
@@ -95,33 +99,29 @@ rendering
   Creating a new PDF from other data (such as an existing PDF).
 
 
-OCRmyPDF has three PDF renderers: ``hocr``, ``tesseract`` and ``tess4``. The renderer may be selected using ``--pdf-renderer``. The default is ``auto`` which lets OCRmyPDF select the renderer to use. Currently, ``auto`` always selects ``hocr``. 
+OCRmyPDF has three PDF renderers: ``sandwich``, ``hocr``, ``tesseract``. The renderer may be selected using ``--pdf-renderer``. The default is ``auto`` which lets OCRmyPDF select the renderer to use. Currently, ``auto`` selects ``sandwich`` for Tesseract 3.05.01 or newer, or ``hocr`` for older versions of Tesseract.
+
+The ``sandwich`` renderer
+"""""""""""""""""""""""""
+
+The ``sandwich`` renderer uses Tesseract's new text-only PDF feature, which produces a PDF page that lays out the OCR in invisible text. This page is then "sandwiched" onto the original PDF page, allowing lossless application of OCR even to PDF pages that contain other vector objects.
+
+When image preprocessing features like ``--deskew`` are used, the original PDF will be rendered as a full page and the OCR layer will be placed on top.
+
+This renderer requires Tesseract 3.05.01 or newer.
 
 The ``hocr`` renderer
 """""""""""""""""""""
 
-The ``hocr`` renderer is the default because it works in most cases. In this mode the whole PDF is rasterized, the raster image is run through OCR to generate a .hocr file, which is an HTML-like file that specifies the location of all identified words.
-
-The .hocr file is then rendered as a PDF and merged with the image layer.
-
-The image layer is copied from the original PDF page if possible, avoiding potentially lossy transcoding or loss of other PDF information. If preprocessing is specified, then the image layer is a new PDF.
-
-This is the only option for tesseract 3.02 and older.
+The ``hocr`` renderer works with older versions of Tesseract. The image layer is copied from the original PDF page if possible, avoiding potentially lossy transcoding or loss of other PDF information. If preprocessing is specified, then the image layer is a new PDF.
 
+This works in all versions of Tesseract.
 
 The ``tesseract`` renderer
 """"""""""""""""""""""""""
 
-The tesseract renderer uses tesseract's capability to produce a PDF directly. In version 3, tesseract automatically combined the image layer and text, meaning that this mode *always* transcodes and loses potentially loses image quality and other PDF information.
+The ``tesseract`` renderer creates a PDF with the image and text layers precomposed, meaning that it always transcodes, loses image quality and rasterizes any vector objects. It does a better job on non-Latin text and document structure than ``hocr``.
 
-It does a much better job on non-Latin text.
+If a PDF created with this renderer using Tesseract versions older than 3.05.00 is then passed through Ghostscript's pdfwrite feature, the OCR text *may* be corrupted. The ``--output-type=pdfa`` argument will produce a warning in this situation.
 
-In a future release this will become the "tess3" renderer and ultimately will be dropped.
-
-
-The ``tess4`` renderer
-""""""""""""""""""""""
-
-The tess4 renderer uses tesseract 4.00 alpha's text-only PDF feature added in January 2017. This combines the advantages of the tesseract and hocr renderers, transcoding the image layer only if required by preprocessing options.
-
-Ghostscript PDF/A still sometimes inserts spaces between words when the tess4 renderer is used, affecting search quality.  ``--output-pdf pdf`` may be used to avoid this issue.
+*This renderer is deprecated and will be removed whenever support for older versions of Tesseract is dropped.*

docs/batch.rst

@@ -18,6 +18,8 @@ The ``--tag`` argument tells parallel to print the filename as a prefix whenever
 
 	parallel --tag -j 2 ocrmypdf '{}' 'output/{}' ::: *.pdf
 
+OCRmyPDF automaticaly repairs PDFs before parsing and gathering information from them.  If you are already repairing PDFs with ``qpdf`` prior to attempting OCR, or you can use ``--skip-repair`` to skip this step.  It may improve performance for large files, since repairing PDFs is single-threaded.
+
 Directory trees
 ---------------
 
@@ -90,13 +92,83 @@ This user contributed script also provides an example of batch processing.
 API
 """
 
-OCRmyPDF is currently supported as a command line interface. Due to limitations in one of the libraries OCRmyPDF depends on, it is not yet usable as an API.
+OCRmyPDF is currently supported as a command line interface. This means that even if you are using OCRmyPDF in a Python script, you should run it in a subprocess rather importing the ocrmypdf package.
+
+The reason for this limitation is that the `ruffus <https://github.com/bunbun/ruffus/>`_ library that OCRmyPDF depends on is unfortunately not reentrant. OCRmyPDF works by defining each operation it does as a ruffus task that takes one or more files as input and generates one or more files as output. As such ruffus is fairly fundamental.
+
+(If you find individual functions implemented in OCRmyPDF useful (such as ``ocrmypdf.pdfinfo``), you can use these if you wish to.)
+
+
+Synology DiskStations
+"""""""""""""""""""""
+
+Synology DiskStations (Network Attached Storage devices) can run the Docker image of OCRmyPDF if the Synology `Docker package <https://www.synology.com/en-global/dsm/packages/Docker>`_ is installed. Attached is a script to address particular quirks of using OCRmyPDF on one of these devices.
+
+This is only possible for x86-based Synology products. Some Synology 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.Popen(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')
+
 
 
 Huge batch jobs
 """""""""""""""
 
-If you have thousands of files to work with, contact the author.
+If you have thousands of files to work with, contact the author. Consulting work related to OCRmyPDF helps fund this open source project and all inquiries are appreciated.
 
 
 Hot (watched) folders

docs/conf.py

@@ -52,7 +52,7 @@ master_doc = 'index'
 
 # General information about the project.
 project = 'ocrmypdf'
-copyright = '2017, James R. Barlow'
+copyright = '2018, James R. Barlow. Licensed under Creative Commons Attribution-ShareAlike 4.0.'
 author = 'James R. Barlow'
 
 # The version info for the project you're documenting, acts as replacement for
@@ -60,7 +60,7 @@ author = 'James R. Barlow'
 # built documents.
 #
 # The short X.Y version.
-from ocrmypdf import VERSION as OCRMYPDF_VERSION
+from ocrmypdf import __version__ as OCRMYPDF_VERSION
 
 _version_parts = OCRMYPDF_VERSION.split('.')
 

docs/cookbook.rst

@@ -11,7 +11,7 @@ ocrmypdf has built-in help.
 
 .. code-block:: bash
 
-	ocrmypdf --help
+    ocrmypdf --help
 
 
 Add an OCR layer and convert to PDF/A
@@ -19,21 +19,21 @@ Add an OCR layer and convert to PDF/A
 
 .. code-block:: bash
 
-	ocrmypdf input.pdf output.pdf
+    ocrmypdf input.pdf output.pdf
 
 Add an OCR layer and output a standard PDF
 """"""""""""""""""""""""""""""""""""""""""
 
 .. code-block:: bash
 
-	ocrmypdf --output-type pdf input.pdf output.pdf
+    ocrmypdf --output-type pdf input.pdf output.pdf
 
 Create a PDF/A with all color and grayscale images converted to JPEG
 """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
 
 .. code-block:: bash
 
-	ocrmypdf --output-type pdfa --pdfa-image-compression jpeg input.pdf output.pdf
+    ocrmypdf --output-type pdfa --pdfa-image-compression jpeg input.pdf output.pdf
 
 Modify a file in place
 """"""""""""""""""""""
@@ -42,7 +42,7 @@ The file will only be overwritten if OCRmyPDF is successful.
 
 .. code-block:: bash
 
-	ocrmypdf myfile.pdf myfile.pdf
+    ocrmypdf myfile.pdf myfile.pdf
 
 Correct page rotation
 """""""""""""""""""""
@@ -51,7 +51,7 @@ OCR will attempt to automatic correct the rotation of each page. This can help f
 
 .. code-block:: bash
 
-	ocrmypdf --rotate-pages myfile.pdf myfile.pdf
+    ocrmypdf --rotate-pages myfile.pdf myfile.pdf
 
 You can increase (decrease) the parameter ``--rotate-pages-threshold`` to make page rotation more (less) aggressive.
 
@@ -63,8 +63,8 @@ By default OCRmyPDF assumes the document is English.
 
 .. code-block:: bash
 
-	ocrmypdf -l fre LeParisien.pdf LeParisien.pdf
-	ocrmypdf -l eng+fre Bilingual-English-French.pdf Bilingual-English-French.pdf
+    ocrmypdf -l fre LeParisien.pdf LeParisien.pdf
+    ocrmypdf -l eng+fre Bilingual-English-French.pdf Bilingual-English-French.pdf
 
 Language packs must be installed for all languages specified. See :ref:`Installing additional language packs <lang-packs>`.
 
@@ -72,19 +72,11 @@ Language packs must be installed for all languages specified. See :ref:`Installi
 Produce PDF and text file containing OCR text
 """""""""""""""""""""""""""""""""""""""""""""
 
-This produces a file named "output.pdf" and a companion text file named "output.txt". The ``pdftotext`` program from 
-`Poppler <https://poppler.freedesktop.org/>`_ is used to extract text from the finished PDF.
-
+This produces a file named "output.pdf" and a companion text file named "output.txt".
 
 .. code-block:: bash
 
-	ocrmypdf input.pdf - | tee output.pdf | pdftotext - output.txt
-
-.. note::
-
-	To get pdftotext, Debian/Ubuntu users may ``apt-get install poppler-utils`` 
-	and macOS users may ``brew install poppler`` respectively.
-
+    ocrmypdf --sidecar output.txt input.pdf output.pdf
 
 OCR images, not PDFs
 --------------------
@@ -93,21 +85,27 @@ Use a program like `img2pdf <https://gitlab.mister-muffin.de/josch/img2pdf>`_ to
 
 .. code-block:: bash
 
-	img2pdf my-images*.jpg | ocrmypdf - myfile.pdf
+    img2pdf my-images*.jpg | ocrmypdf - myfile.pdf
 
-If given a single image as input, OCRmyPDF will try converting it to a PDF on its own.  If the DPI specified in the image is incorrect, it can be overridden with ``--image-dpi``:
+``img2pdf`` also has features to control the position of images on a page, if desired.
+
+For convenience, OCRmyPDF can convert single images to PDFs on its own. If the resolution (dots per inch, DPI) of an image is not set or is incorrect, it can be overridden with ``--image-dpi``. (As 1 inch is 2.54 cm, 1 dpi = 0.39 dpcm).
 
 .. code-block:: bash
 
-	ocrmypdf --image-dpi 300 image.png myfile.pdf
+    ocrmypdf --image-dpi 300 image.png myfile.pdf
 
-This feature may be removed at some point, because OCRmyPDF does not specialize in converting images to PDFs.
+If you have multiple images, you must use ``img2pdf`` to convert the images to PDF.
+
+.. note::
+
+    ImageMagick ``convert`` can also convert a group of images to PDF, but in the author's experience it takes a long time, transcodes unnecessarily and gives poor results.
 
 You can also use Tesseract 3.04+ directly to convert single page images or multi-page TIFFs to PDF:
 
 .. code-block:: bash
 
-	tesseract my-image.jpg output-prefix pdf 
+    tesseract my-image.jpg output-prefix pdf
 
 Image processing
 ----------------
@@ -126,11 +124,11 @@ OCRmyPDF perform some image processing on each page of a PDF, if desired.  The s
 
 .. note::
 
-	In many cases image processing will rasterize PDF pages as images, potentially losing quality.
+    In many cases image processing will rasterize PDF pages as images, potentially losing quality.
 
 .. warning::
 
-	``--clean-final`` and ``-remove-background`` may leave undesirable visual artifacts in some images where their algorithms have shortcomings. Files should be visually reviewed after using these options.
+    ``--clean-final`` and ``-remove-background`` may leave undesirable visual artifacts in some images where their algorithms have shortcomings. Files should be visually reviewed after using these options.
 
 
 OCR and correct document skew (crooked scan)
@@ -140,12 +138,54 @@ Deskew:
 
 .. code-block:: bash
 
-	ocrmypdf --deskew input.pdf output.pdf
+    ocrmypdf --deskew input.pdf output.pdf
 
 Image processing commands can be combined. The order in which options are given does not matter. OCRmyPDF always applies the steps of the image processing pipeline in the same order (rotate, remove background, deskew, clean).
 
 .. code-block:: bash
 
-	ocrmypdf --deskew --clean --rotate-pages input.pdf output.pdf
+    ocrmypdf --deskew --clean --rotate-pages input.pdf output.pdf
 
 
+Don't actually OCR my PDF
+"""""""""""""""""""""""""
+
+If you set ``--tesseract-timeout 0`` OCRmyPDF will apply its image processing without performing OCR, if all you want to is to apply image processing or PDF/A conversion.
+
+.. code-block:: bash
+
+    ocrmypdf --tesseract-timeout=0 --remove-background input.pdf output.pdf
+
+
+Redo OCR
+""""""""
+
+To redo OCR on a file OCRed with other OCR software or a previous version of OCRmyPDF and/or Tesseract, you may use the ``--force-ocr`` argument. Normally, OCRmyPDF does not modify files that already appear to contain OCR text.
+
+.. code-block:: bash
+
+    ocrmypdf --force-ocr input.pdf output.pdf
+
+Note that the method above will force rasterization of all pages, potentially reducing quality or losing vector content. 
+
+To ensure quality is preserved, one could extract all of the images and rebuild the PDF for a lossless transformation. This recipe does not work when PDFs contain multiple images per page, as many do in practice. It will also lose any page rotation information.
+
+.. code-block:: bash
+
+    pdfimages -all old-ocr.pdf prefix  # extract all images
+    img2pdf -o temp.pdf prefix*        # construct new PDF from the images
+    # review the new PDF to ensure it visually matches the old one
+    ocrmypdf --output-type pdf temp.pdf new-ocr.pdf
+
+``--output-type pdf`` is used here to avoid using Ghostscript which will also rasterize images.
+
+
+Improving OCR quality
+---------------------
+
+The `Image processing`_ features can improve OCR quality.
+
+Rotating pages and deskewing helps to ensure that the page orientation is correct before OCR begins. Removing the background and/or cleaning the page can also improve results. The ``--oversample DPI`` argument can be specified to resample images to higher resolution before attempting OCR; this can improve results as well.
+
+OCR quality will suffer if the resolution of input images is not correct (since the range of pixel sizes that will be checked for possible fonts will also be incorrect).
+

docs/index.rst

@@ -9,7 +9,7 @@ OCRmyPDF documentation
 OCRmyPDF adds an OCR text layer to scanned PDF files, allowing them to
 be searched.
 
-PDFs are the best format for scanned documents.  Unfortunately, PDFs can be difficult to work with. OCRmyPDF makes it easy to apply image processing and OCR to existing PDFs.
+PDF is the best format for storing and exchanging scanned documents.  Unfortunately, PDFs can be difficult to modify. OCRmyPDF makes it easy to apply image processing and OCR to existing PDFs.
 
 .. toctree::
    :maxdepth: 1

docs/installation.rst

@@ -15,15 +15,21 @@ Users of Debian 9 ("stretch") or later or Ubuntu 16.10 or later may simply
 Installing on macOS
 -------------------
 
+OCRmyPDF is now a standard `Homebrew <https://brew.sh>`_ formula. To install on macOS:
+
 .. code-block:: bash
-    brew tap jbarlow83/ocrmypdf
+
     brew install ocrmypdf
 
-.. warning::
+.. note::
 
     Users who previously installed OCRmyPDF on macOS using ``pip install ocrmypdf`` should remove the pip version (``pip3 uninstall ocrmypdf``) before switching to the Homebrew version.
 
-.. _Docker:
+.. note::
+
+    Users who previously installed OCRmyPDF from the private tap should switch to the mainline version (``brew untap jbarlow83/ocrmypdf``) and install from there.
+
+.. _Docker-install:
 
 Installing the Docker image
 ---------------------------
@@ -37,6 +43,7 @@ Follow the Docker installation instructions for your platform.  If you can run t
 successfully, your system is ready to download and execute the image:
 
 .. code-block:: bash
+
     docker run hello-world
    
 OCRmyPDF will use all available CPU cores.  By default, the VirtualBox machine instance on Windows and macOS has only a single CPU core enabled. Use the VirtualBox Manager to determine the name of your Docker engine host, and then follow these optional steps to enable multiple CPUs:
@@ -100,6 +107,12 @@ In this worked example, the current working directory contains an input file cal
 
 Note that ``ocrmypdf`` has its own separate ``-v VERBOSITYLEVEL`` argument to control debug verbosity. All Docker arguments should before the ``ocrmypdf`` image name and all arguments to ``ocrmypdf`` should be listed after.
 
+In some environments the permissions associated with Docker can be complex to configure. The process that executes Docker may end up not having the permissions to write the specified file system. In that case one can stream the file into and out of the Docker process and avoid all permission hassles, using ``-`` as the input and output filename:
+
+.. code-block:: bash
+
+    docker run --rm -i   ocrmypdf <other arguments to ocrmypdf> - - <input.pdf >output.pdf
+
 For convenience, a shell alias can hide the docker command:
 
 .. code-block:: bash
@@ -114,6 +127,9 @@ Or in the wonderful `fish shell <https://fishshell.com/>`_:
     alias ocrmypdf 'docker run --rm -v (pwd):/home/docker ocrmypdf'
     funcsave ocrmypdf
 
+.. note::
+
+    The ocrmypdf Docker containers are designed to be used for a single OCR job. The ``docker run --rm`` argument tells Docker to delete temporary storage associated with container when it is done executing.
 
 Manual installation on macOS
 ----------------------------
@@ -160,7 +176,13 @@ Update the homebrew pip and install Pillow:
     pip3 install --upgrade pip
     pip3 install --upgrade pillow
 
-You can then install OCRmyPDF from PyPI:
+You can then install OCRmyPDF from PyPI, for the current user:
+
+.. code-block:: bash
+
+    pip3 install --user ocrmypdf
+
+or system-wide:
 
 .. code-block:: bash
 
@@ -189,8 +211,13 @@ No package is currently available for Ubuntu 16.04, but you can install the depe
         python3-pip \
         python3-cffi
 
-If you wish install OCRmyPDF to the system Python, then install as follows (note this installs new packages
-into your system Python, which could interfere with other programs):
+If you wish install OCRmyPDF for the current user:
+
+.. code-block:: bash
+
+    pip3 install --user ocrmypdf
+
+Alternately, system-wide. Note that this may modify the system Python environment:
 
 .. code-block:: bash
 
@@ -209,15 +236,8 @@ follow these steps.
 Installing on Ubuntu 14.04 LTS
 ------------------------------
 
-Installing on Ubuntu 14.04 LTS (trusty) is more difficult than some other options, because of bugs in Python package installation and because OCRmyPDF depends on some packages newer than are available in the main distribution.
-
-Add new "apt" repositories needed for backports of Ghostscript 9.16, libav-11 (for unpaper 6.1) and Tesseract 4.00 (alpha). This will replace Ghostscript and Tesseract 3.x on your system. If you prefer to not modify your system in this matter, consider using a Docker container.
-
-.. code-block:: bash
-
-    sudo add-apt-repository ppa:vshn/ghostscript -y
-    sudo add-apt-repository ppa:heyarje/libav-11 -y
-    sudo add-apt-repository ppa:alex-p/tesseract-ocr
+Installing on Ubuntu 14.04 LTS (trusty) is more difficult than some other options,
+because it is older and does not provide ``pip``.
 
 Update apt-get:
 
@@ -234,45 +254,46 @@ Install system dependencies:
         zlib1g-dev \
         libjpeg-dev \
         libffi-dev \
-        libavformat56 libavcodec56 libavutil54 \
+        qpdf
+
+We will need backports of Ghostscript 9.16, libav-11 (for unpaper 6.1),
+Tesseract 4.00 (alpha), and Python 3.6. This will replace Ghostscript and
+Tesseract 3.x on your system. Python 3.6 will be installed alongside the system
+Python 3.
+
+If you prefer to not modify your system in this matter, consider using a Docker container.
+
+.. code-block:: bash
+
+    sudo add-apt-repository ppa:vshn/ghostscript -y
+    sudo add-apt-repository ppa:heyarje/libav-11 -y
+    sudo add-apt-repository ppa:alex-p/tesseract-ocr -y
+    sudo add-apt-repository ppa:jonathonf/python-3.6 -y
+
+    sudo apt-get update
+
+    sudo apt-get install \
+        python3.6 \
         ghostscript \
-        qpdf \
-        python3-pip \
-        python3-pil \
-        python3-pytest \
-        python3-reportlab \
-        python3-wheel \
-        python3-venv \
         tesseract-ocr \
-        tesseract-ocr-eng
+        tesseract-ocr-eng \
+        libavformat56 libavcodec56 libavutil54 \
+        wget
 
-If you wish install OCRmyPDF to the system Python, then install as follows (note this installs new packages
-into your system Python, which could interfere with other programs):
+Now we need to install ``pip`` and let it install ocrmypdf:
 
 .. code-block:: bash
 
-    sudo pip3 install ocrmypdf
+    wget -O - -o /dev/null https://bootstrap.pypa.io/get-pip.py | python3.6
+    pip3.6 install ocrmypdf
 
-If you wish to install OCRmyPDF to a virtual environment to isolate the system Python, you can
-follow these steps.  This includes a workaround `for a known, unresolved issue in Ubuntu 14.04's ensurepip
-package <http://www.thefourtheye.in/2014/12/Python-venv-problem-with-ensurepip-in-Ubuntu.html>`_:
-
-.. code-block:: bash
-
-    sudo apt-get install python3-venv
-    python3 -m venv venv-ocrmypdf --without-pip
-    source venv-ocrmypdf/bin/activate
-    wget -O - -o /dev/null https://bootstrap.pypa.io/get-pip.py | python
-    deactivate
-    python3 -m venv --system-site-packages venv-ocrmypdf
-    source venv-ocrmypdf/bin/activate
-    pip install ocrmypdf
+The ``wget`` command will download a program and run it.
 
 These installation instructions omit the optional dependency ``unpaper``, which is only available at version 0.4.2 in Ubuntu 14.04. The author could not find a backport of ``unpaper``, and created a .deb package to do the job of installing unpaper 6.1 (for x86 64-bit only):
 
 .. code-block:: bash
 
-    wget -q https://dl.dropboxusercontent.com/u/28971240/unpaper_6.1-1.deb -O unpaper_6.1-1.deb
+    wget -q 'https://www.dropbox.com/s/vaq0kbwi6e6au80/unpaper_6.1-1.deb?raw=1' -O unpaper_6.1-1.deb
     sudo dpkg -i unpaper_6.1-1.deb
 
 
@@ -281,7 +302,7 @@ Installing on ArchLinux
 
 The author is aware of an `ArchLinux package for ocrmypdf <https://aur.archlinux.org/packages/ocrmypdf/>`_. It seems like the following command might work.
 
-.. code-block::
+.. code-block:: bash
 
     pacman -S ocrmypdf
 
@@ -302,6 +323,8 @@ The command line syntax to run ocrmypdf from a command prompt will resemble:
 
 where /c/Users/sampleuser is a Unix representation of the Windows path C:\\Users\\sampleuser, assuming a user named "sampleuser" is running ocrmypdf on a file in their home directory, and the files "test.pdf" and "output.pdf" are in the sampleuser folder. The Windows user must have read and write permissions.
 
+`Bash on Ubuntu on Windows <https://github.com/Microsoft/BashOnWindows>`_ should also be a viable route for running the OCRmyPDF Docker container.
+
 Installing HEAD revision from sources
 -------------------------------------
 

docs/introduction.rst

@@ -16,7 +16,7 @@ OCRmyPDF uses `Tesseract <https://github.com/tesseract-ocr/tesseract>`_, the bes
 About PDFs
 ----------
 
-PDFs are page description files that attempts to preserve a layout exactly. They can contain `vector graphic files <http://vector-conversions.com/vectorizing/raster_vs_vector.html>`_ that can contain raster objects such as scanned images. Because PDFs can contain multiple pages (unlike many image formats) and can contain fonts and text, it is a good formats for exchanging scanned documents.
+PDFs are page description files that attempts to preserve a layout exactly. They  contain `vector graphics <http://vector-conversions.com/vectorizing/raster_vs_vector.html>`_ that can contain raster objects such as scanned images. Because PDFs can contain multiple pages (unlike many image formats) and can contain fonts and text, it is a good formats for exchanging scanned documents.
 
 .. image:: bitmap_vs_svg.svg
 
@@ -42,23 +42,23 @@ PDF/A has a few drawbacks.  Some PDF viewers include an alert that the file is a
 What OCRmyPDF does
 ------------------
 
-OCRmyPDF analyzes each page of a PDF to determine the colorspace and resolution (DPI) needed to capture all of the information on that page without losing content.  It uses `Ghostscript <http://ghostscript.com/>`_ to rasterize the page, and then performs on OCR on the rasterized image.  It is not enough to simply extract the images from each page and run OCR on them individually.  Of course one could use Ghostscript or another PDF rasterizer and then pass the image to Tesseract.  OCRmyPDF automates this process and produces a minimally changed output file that contains the same information, colorspace and resolution.
+OCRmyPDF analyzes each page of a PDF to determine the colorspace and resolution (DPI) needed to capture all of the information on that page without losing content.  It uses `Ghostscript <http://ghostscript.com/>`_ to rasterize the page, and then performs on OCR on the rasterized image to create an OCR "layer". The layer is then grafted back onto the original PDF.
 
-The Tesseract OCR engine can output 'hOCR' files, which are XML files that contain a description of the text it found on the page.  OCRmyPDF will render a new PDF that contains only the hidden text layer, and merge this with the original page.
+While one can use a program like Ghostscript or ImageMagick to get an image and put the image through Tesseract, that actually creates a new PDF and many details may be lost. OCRmyPDF can produce a minimally changed PDF as output.
 
-Alternately, OCRmyPDF can use the Tesseract OCR engine to directly output PDFs for each page, then merge them.
+OCRmyPDF also some image processing options like deskew which improve the appearance of files and quality of OCR. When these are used, the OCR layer is grafted onto the processed image instead.
 
-By default, OCRmyPDF will convert the file to a PDF/A.  This behavior can be disabled with the ``--output-type pdf`` argument.
-
-Depending on the settings selected, OCRmyPDF may "graft" the OCR layer into the existing PDF, or reconstruct a visually equivalent new PDF.
+By default, OCRmyPDF produces archival PDFs – PDF/A, which are a stricter subset of PDF features designed for long term archives. If regular PDFs are desired, this can be disabled with ``--output-type pdf``.
 
 
 Why you shouldn't do this manually
 ----------------------------------
 
-There are two routes to manually applying OCR to an existing PDF, both of which destroy information in the original PDF.
+A PDF is similar to an HTML file, in that it contains document structure along with images.  Sometimes a PDF does nothing more than present a full page image, but often there is additional content that would be lost.
 
-1. Rasterize each page as an image, OCR the images, and combine the output into a PDF. This preserves the appearance of each page, but resamples all images (possibly losing quality, increasing file size, introducing compression artifacts, etc.)
+A manual process could work like either of these:
+
+1. Rasterize each page as an image, OCR the images, and combine the output into a PDF. This preserves the layout of each page, but resamples all images (possibly losing quality, increasing file size, introducing compression artifacts, etc.).
 
 2. Extract each image, OCR, and combine the output into a PDF. This loses the context in which images are used in the PDF, meaning that cropping, rotation and scaling of pages may be lost. Some scanned PDFs use multiple images segmented into black and white, grayscale and color regions, with stencil masks to prevent overlap, as this can enhance the appearance of a file while reducing file size. Clearly, reassembling these images will be easy. This also loses and text or vector art on any pages in a PDF with both scanned and pure digital content.
 
@@ -83,6 +83,7 @@ OCRmyPDF is limited by the Tesseract OCR engine.  As such it experiences these l
 OCRmyPDF is also limited by the PDF specification:
 
 * PDF encodes the position of text glyphs but does not encode document structure.  There is no markup that divides a document in sections, paragraphs, sentences, or even words (since blank spaces are not represented). As such all elements of document structure including the spaces between words must be derived heuristically.  Some PDF viewers do a better job of this than others.
+* Because some popular open source PDF viewers have a particularly hard time with spaces betweem words, OCRmyPDF appends a space to each text element as a workaround. While this mixes document structure with graphical information that ideally should be left to the PDF viewer to interpret, it improves compatibility with some viewers and does not cause problems for better ones.
 
 Ghostscript also imposes some limitations:
 
@@ -101,4 +102,13 @@ To the author's knowledge, OCRmyPDF is the most feature-rich and thoroughly test
 * pdf2pdfocr
 * pdfsandwich
 * pypdfocr
-* pdfbeads
+* pdfbeads
+
+Web front-ends
+--------------
+
+* `Nextcloud OCR <https://github.com/janis91/ocr>`_ is a free software plugin for the Nextcloud private cloud software
+* `OCRmyPDF-web <https://github.com/sseemayer/OCRmyPDF-web>`_, a micro web-frontend for OCRmyPDF (third-party, not actively maintained)
+
+Bear in mind that OCRmyPDF is not designed to be secure against malware-bearing PDFs (see `Using OCRmyPDF online`_).
+

docs/languages.rst

@@ -33,6 +33,34 @@ Docker users
 
 Users of the Docker image may use the alternative :ref:`"polyglot" container <docker-polyglot>` which includes all languages.
 
+Adding individual language packs to a Docker image
+""""""""""""""""""""""""""""""""""""""""""""""""""
+
+If you wish to add a single language pack, you could do the following:
+
+* Download the desired ``.trainedata`` file from the `tessdata <https://github.com/tesseract-ocr/tessdata>`_ repository. Let's use Hebrew in this example (``heb.traineddata``)
+
+* Copy the file to ``/home/user/downloads/heb.traineddata``.
+
+* Create a new container based on the ocrmypdf-tess4 image and jump into it with a terminal:
+
+.. code-block:: bash
+
+	host$ docker run  -v /home/user/downloads:/home/docker -it --entrypoint /bin/bash ocrmypdf-tess4
+
+* Put the file where Tesseract expects it:
+
+.. code-block:: bash
+
+	docker$ cp /home/docker/heb.traineddata /usr/share/tesseract-ocr/tessdata
+
+* Note the container id, and save it as a new image (in this example, ``ocrmypdf-tess4-heb``)
+
+.. code-block:: bash
+
+    host$ docker commit <container_id> ocrmypdf-tess4-heb
+
+
 Known limitations
 -----------------
 

docs/release_notes.rst

@@ -1,7 +1,207 @@
 Release notes
 =============
 
-OCRmyPDF uses `semantic versioning <http://semver.org/>`_.
+OCRmyPDF uses `semantic versioning <http://semver.org/>`_ for its command line interface.
+
+The OCRmyPDF package itself does not contain a public API, although it is fairly stable and breaking changes are usually timed with a major release. A future release will clearly define the stable public API.
+
+v6.1.2
+------
+
+-   Upgrade to PyMuPDF v1.12.5 which includes a more complete fix to #239.
+
+-   Add ``defusedxml`` dependency.
+
+
+v6.1.1
+------
+
+-   Fix text being reported as found on all pages if PyMuPDF is not installed.
+
+
+v6.1.0
+------
+
+-   PyMuPDF is now an optional but recommended dependency, to alleviate installation difficulties on platforms that have less access to PyMuPDF than the author anticipated.  Install OCRmyPDF with ``pip install ocrmypdf[fitz]`` to use it to its full potential.
+
+-   Fix ``FileExistsError`` that could occur if OCR timed out while it was generating the output file. (#218)
+
+-   Fix table of contents/bookmarks all being redirected to page 1 when generating a PDF/A (with PyMuPDF).  (Without PyMuPDF the table of contents is removed in PDF/A mode.)
+
+-   Fix "RuntimeError: invalid key in dict" when table of contents/bookmarks titles contained the character ``)``. (#239)
+
+-   Added a new argument ``--skip-repair`` to skip the initial PDF repair step if the PDF is already well-formed (because another program repaired it).
+
+
+v6.0.0
+------
+
+-   The software license has been changed to GPLv3. Test resource files and some individual sources may have other licenses.
+
+-   OCRmyPDF now depends on `PyMuPDF <https://pymupdf.readthedocs.io/en/latest/installation/>`_. Including PyMuPDF is the primary reason for the change to GPLv3.
+
+-   Other backward incompatible changes
+
+    + The ``OCRMYPDF_TESSERACT``, ``OCRMYPDF_QPDF``, ``OCRMYPDF_GS`` and ``OCRMYPDF_UNPAPER`` environment variables are no longer used. Change ``PATH`` if you need to override the external programs OCRmyPDF uses.
+
+    + The ``ocrmypdf`` package has been moved to ``src/ocrmypdf`` to avoid issues with accidental import.
+
+    + The function ``ocrmypdf.exec.get_program`` was removed.
+
+    + The deprecated module ``ocrmypdf.pageinfo`` was removed.
+
+    + The ``--pdf-renderer tess4`` alias for ``sandwich`` was removed.
+
+-   Fixed an issue where OCRmyPDF failed to detect existing text on pages, depending on how the text and fonts were encoded within the PDF. (#233, #232)
+
+-   Fixed an issue that caused dramatic inflation of file sizes when ``--skip-text --output-type pdf`` was used. OCRmyPDF now removes duplicate resources such as fonts, images and other objects that it generates. (#237)
+
+-   Improved performance of the initial page splitting step. Originally this step was not believed to be expensive and ran in a process. Large file testing revealed it to be a bottleneck, so it is now parallelized. On a 700 page file with quad core machine, this change saves about 2 minutes. (#234)
+
+-   The test suite now includes a cache that can be used to speed up test runs across platforms. This also does not require computing checksums, so it's faster. (#217)
+
+
+v5.7.0
+------
+
+-   Fixed an issue that caused poor CPU utilization on machines more than 4 cores when running Tesseract 4. (Related to issue #217.)
+
+-   The 'hocr' renderer has been improved. The 'sandwich' and 'tesseract' renderers are still better for most use cases, but 'hocr' may be useful for people who work with the PDF.js renderer in English/ASCII languages. (#225)
+
+    + It now formats text in a matter that is easier for certain PDF viewers to select and extract copy and paste text. This should help macOS Preview and PDF.js in particular.
+    + The appearance of selected text and behavior of selecting text is improved.
+    + The PDF content stream now uses relative moves, making it more compact and easier for viewers to determine when two words on the same line.
+    + It can now deal with text on a skewed baseline.
+    + Thanks to @cforcey for the pull request, @jbreiden for many helpful suggestions, @ctbarbour for another round of improvements, and @acaloiaro for an independent review.
+
+v5.6.3
+------
+
+-   Suppress two debug messages that were too verbose
+
+
+v5.6.2
+------
+
+-   Development branch accidentally tagged as release. Do not use.
+
+
+v5.6.1
+------
+
+-   Fix issue #219: change how the final output file is created to avoid triggering permission errors when the output is a special file such as ``/dev/null``
+-   Fix test suite failures due to a qpdf 8.0.0 regression and Python 3.5's handling of symlink
+-   The "encrypted PDF" error message was different depending on the type of PDF encryption. Now a single clear message appears for all types of PDF encryption.
+-   ocrmypdf is now in Homebrew. Homebrew users are advised to the version of ocrmypdf in the official homebrew-core formulas rather than the private tap.
+-   Some linting
+
+
+v5.6.0
+------
+
+-   Fix issue #216: preserve "text as curves" PDFs without rasterizing file
+-   Related to the above, messages about rasterizing are more consistent
+-   For consistency versions minor releases will now get the trailing .0 they always should have had.
+
+
+v5.5
+----
+
+-   Add new argument ``--max-image-mpixels``. Pillow 5.0 now raises an exception when images may be decompression bombs. This argument can be used to override the limit Pillow sets.
+-   Fix output page cropped when using the sandwich renderer and OCR is skipped on a rotated and image-processed page
+-   A warning is now issued when old versions of Ghostscript are used in cases known to cause issues with non-Latin characters
+-   Fix a few parameter validation checks for ``-output-type pdfa-1`` and ``pdfa-2`` 
+
+
+v5.4.4
+------
+
+-   Fix issue #181: fix final merge failure for PDFs with more pages than the system file handle limit (``ulimit -n``)
+-   Fix issue #200: an uncommon syntax for formatting decimal numbers in a PDF would cause qpdf to issue a warning, which ocrmypdf treated as an error. Now this the warning is relayed.
+-   Fix an issue where intermediate PDFs would be created at version 1.3 instead of the version of the original file. It's possible but unlikely this had side effects.
+-   A warning is now issued when older versions of qpdf are used since issues like #200 cause qpdf to infinite-loop
+-   Address issue #140: if Tesseract outputs invalid UTF-8, escape it and print its message instead of aborting with a Unicode error 
+-   Adding previously unlisted setup requirement, pytest-runner
+-   Update documentation: fix an error in the example script for Synology with Docker images, improved security guidance, advised ``pip install --user``
+
+
+v5.4.3
+------
+
+-   If a subprocess fails to report its version when queried, exit cleanly with an error instead of throwing an exception
+-   Added test to confirm that the system locale is Unicode-aware and fail early if it's not
+-   Clarified some copyright information
+-   Updated pinned requirements.txt so the homebrew formula captures more recent versions
+
+
+v5.4.2
+------
+
+-   Fixed a regression from v5.4.1 that caused sidecar files to be created as empty files
+
+
+v5.4.1
+------
+
+-   Add workaround for Tesseract v4.00alpha crash when trying to obtain orientation and the latest language packs are installed
+
+
+v5.4
+----
+
+-   Change wording of a deprecation warning to improve clarity
+-   Added option to generate PDF/A-1b output if desired (``--output-type pdfa-1``); default remains PDF/A-2b generation
+-   Update documentation
+
+
+v5.3.3
+------
+
+-   Fixed missing error message that should occur when trying to force ``--pdf-renderer sandwich`` on old versions of Tesseract
+-   Update copyright information in test files
+-   Set system ``LANG`` to UTF-8 in Dockerfiles to avoid UTF-8 encoding errors
+
+
+v5.3.2
+------
+
+-   Fixed a broken test case related to language packs
+
+
+v5.3.1
+------
+
+-   Fixed wrong return code given for missing Tesseract language packs
+-   Fixed "brew audit" crashing on Travis when trying to auto-brew
+
+
+v5.3
+----
+
+-   Added ``--user-words`` and ``--user-patterns`` arguments which are forwarded to Tesseract OCR as words and regular expressions respective to use to guide OCR. Supplying a list of subject-domain words should assist Tesseract with resolving words. (#165)
+-   Using a non Latin-1 language with the "hocr" renderer now warns about possible OCR quality and recommends workarounds (#176)
+-   Output file path added to error message when that location is not writable (#175)
+-   Otherwise valid PDFs with leading whitespace at the beginning of the file are now accepted
+
+
+v5.2
+----
+
+-   When using Tesseract 3.05.01 or newer, OCRmyPDF will select the "sandwich" PDF renderer by default, unless another PDF renderer is specified with the ``--pdf-renderer`` argument. The previous behavior was to select ``--pdf-renderer=hocr``.
+-   The "tesseract" PDF renderer is now deprecated, since it can cause problems with Ghostscript on Tesseract 3.05.00
+-   The "tess4" PDF renderer has been renamed to "sandwich". "tess4" is now a deprecated alias for "sandwich".
+
+
+v5.1
+----
+
+-   Files with pages larger than 200" (5080 mm) in either dimension are now supported with ``--output-type=pdf`` with the page size preserved (in the PDF specification this feature is called UserUnit scaling). Due to Ghostscript limitations this is not available in conjunction with PDF/A output.
+
+
+v5.0.1
+------
+
+-   Fixed issue #169, exception due to failure to create sidecar text files on some versions of Tesseract 3.04, including the jbarlow83/ocrmypdf Docker image
 
 
 v5.0
@@ -9,9 +209,9 @@ v5.0
 
 -   Backward incompatible changes
 
-    + Support for Python 3.4 dropped. Python 3.5 is now required.
-    + Support for Tesseract 3.02 and 3.03 dropped. Tesseract 3.04 or newer is required. Tesseract 4.00 (alpha) is supported.
-    + The OCRmyPDF.sh script was removed.
+     + Support for Python 3.4 dropped. Python 3.5 is now required.
+     + Support for Tesseract 3.02 and 3.03 dropped. Tesseract 3.04 or newer is required. Tesseract 4.00 (alpha) is supported.
+     + The OCRmyPDF.sh script was removed.
 
 -   Add a new feature, ``--sidecar``, which allows creating "sidecar" text files which contain the OCR results in plain text. These OCR text is more reliable than extracting text from PDFs. Closes #126.
 -   New feature: ``--pdfa-image-compression``, which allows overriding Ghostscript's lossy-or-lossless image encoding heuristic and making all images JPEG encoded or lossless encoded as desired. Fixes #163.
@@ -45,242 +245,242 @@ v4.5.4
 v4.5.3
 ------
 
--  Added a workaround for Ghostscript 9.21 and probably earlier versions would fail with the error message "VMerror -25", due to a Ghostscript bug in XMP metadata handling
--  High Unicode characters (U+10000 and up) are no longer accepted for setting metadata on the command line, as Ghostscript may not handle them correctly.
--  Fixed an issue where the ``tess4`` renderer would duplicate content onto output pages if tesseract failed or timed out
--  Fixed ``tess4`` renderer not recognized when lossless reconstruction is possible
+-   Added a workaround for Ghostscript 9.21 and probably earlier versions would fail with the error message "VMerror -25", due to a Ghostscript bug in XMP metadata handling
+-   High Unicode characters (U+10000 and up) are no longer accepted for setting metadata on the command line, as Ghostscript may not handle them correctly.
+-   Fixed an issue where the ``tess4`` renderer would duplicate content onto output pages if tesseract failed or timed out
+-   Fixed ``tess4`` renderer not recognized when lossless reconstruction is possible
 
 
 v4.5.2
 ------
 
--  Fix issue #147. ``--pdf-renderer tess4 --clean`` will produce an oversized page containing the original image in the bottom left corner, due to loss DPI information.
--  Make "using Tesseract 4.0" warning less ominous
--  Set up machinery for homebrew OCRmyPDF tap
+-   Fix issue #147. ``--pdf-renderer tess4 --clean`` will produce an oversized page containing the original image in the bottom left corner, due to loss DPI information.
+-   Make "using Tesseract 4.0" warning less ominous
+-   Set up machinery for homebrew OCRmyPDF tap
 
 
 v4.5.1
 ------
 
--  Fix issue #137, proportions of images with a non-square pixel aspect ratio would be distorted in output for ``--force-ocr`` and some other combinations of flags
+-   Fix issue #137, proportions of images with a non-square pixel aspect ratio would be distorted in output for ``--force-ocr`` and some other combinations of flags
 
 
 v4.5
 ----
 
--  Exotic PDFs containing "Form XObjects" are now supported (issue #134; PDF reference manual 8.10), and images they contain are taken into account when determining the resolution for rasterizing
--  The Tesseract 4 Docker image no longer includes all languages, because it took so long to build something would tend to fail
--  OCRmyPDF now warns about using ``--pdf-renderer tesseract`` with Tesseract 3.04 or lower due to issues with Ghostscript corrupting the OCR text in these cases
+-   Exotic PDFs containing "Form XObjects" are now supported (issue #134; PDF reference manual 8.10), and images they contain are taken into account when determining the resolution for rasterizing
+-   The Tesseract 4 Docker image no longer includes all languages, because it took so long to build something would tend to fail
+-   OCRmyPDF now warns about using ``--pdf-renderer tesseract`` with Tesseract 3.04 or lower due to issues with Ghostscript corrupting the OCR text in these cases
 
 
 v4.4.2
 ------
 
--  The Docker images (ocrmypdf, ocrmypdf-polyglot, ocrmypdf-tess4) are now based on Ubuntu 16.10 instead of Debian stretch
+-   The Docker images (ocrmypdf, ocrmypdf-polyglot, ocrmypdf-tess4) are now based on Ubuntu 16.10 instead of Debian stretch
 
-   + This makes supporting the Tesseract 4 image easier
-   + This could be a disruptive change for any Docker users who built customized these images with their own changes, and made those changes in a way that depends on Debian and not Ubuntu
+    + This makes supporting the Tesseract 4 image easier
+    + This could be a disruptive change for any Docker users who built customized these images with their own changes, and made those changes in a way that depends on Debian and not Ubuntu
 
--  OCRmyPDF now prevents running the Tesseract 4 renderer with Tesseract 3.04, which was permitted in v4.4 and v4.4.1 but will not work
+-   OCRmyPDF now prevents running the Tesseract 4 renderer with Tesseract 3.04, which was permitted in v4.4 and v4.4.1 but will not work
 
 
 v4.4.1
 ------
 
--  To prevent a `TIFF output error <https://github.com/python-pillow/Pillow/issues/2206>`_ caused by img2pdf >= 0.2.1 and Pillow <= 3.4.2, dependencies have been tightened
--  The Tesseract 4.00 simultaneous process limit was increased from 1 to 2, since it was observed that 1 lowers performance
--  Documentation improvements to describe the ``--tesseract-config`` feature 
--  Added test cases and fixed error handling for ``--tesseract-config``
--  Tweaks to setup.py to deal with issues in the v4.4 release
+-   To prevent a `TIFF output error <https://github.com/python-pillow/Pillow/issues/2206>`_ caused by img2pdf >= 0.2.1 and Pillow <= 3.4.2, dependencies have been tightened
+-   The Tesseract 4.00 simultaneous process limit was increased from 1 to 2, since it was observed that 1 lowers performance
+-   Documentation improvements to describe the ``--tesseract-config`` feature 
+-   Added test cases and fixed error handling for ``--tesseract-config``
+-   Tweaks to setup.py to deal with issues in the v4.4 release
 
 v4.4
 ----
 
--  Tesseract 4.00 is now supported on an experimental basis.
+-   Tesseract 4.00 is now supported on an experimental basis.
 
-   +  A new rendering option ``--pdf-renderer tess4`` exploits Tesseract 4's new text-only output PDF mode. See the documentation on PDF Renderers for details.
-   +  The ``--tesseract-oem`` argument allows control over the Tesseract 4 OCR engine mode (tesseract's ``--oem``). Use ``--tesseract-oem 2`` to enforce the new LSTM mode.
-   +  Fixed poor performance with Tesseract 4.00 on Linux
+    +  A new rendering option ``--pdf-renderer tess4`` exploits Tesseract 4's new text-only output PDF mode. See the documentation on PDF Renderers for details.
+    +  The ``--tesseract-oem`` argument allows control over the Tesseract 4 OCR engine mode (tesseract's ``--oem``). Use ``--tesseract-oem 2`` to enforce the new LSTM mode.
+    +  Fixed poor performance with Tesseract 4.00 on Linux
 
--  Fixed an issue that caused corruption of output to stdout in some cases
--  Removed test for Pillow JPEG and PNG support, as the minimum supported version of Pillow now enforces this
--  OCRmyPDF now tests that the intended destination file is writable before proceeding
--  The test suite now requires ``pytest-helpers-namespace`` to run (but not install)
--  Significant code reorganization to make OCRmyPDF re-entrant and improve performance. All changes should be backward compatible for the v4.x series.
+-   Fixed an issue that caused corruption of output to stdout in some cases
+-   Removed test for Pillow JPEG and PNG support, as the minimum supported version of Pillow now enforces this
+-   OCRmyPDF now tests that the intended destination file is writable before proceeding
+-   The test suite now requires ``pytest-helpers-namespace`` to run (but not install)
+-   Significant code reorganization to make OCRmyPDF re-entrant and improve performance. All changes should be backward compatible for the v4.x series.
 
-   + However, OCRmyPDF's dependency "ruffus" is not re-entrant, so no Python API is available. Scripts should continue to use the command line interface.
+    + However, OCRmyPDF's dependency "ruffus" is not re-entrant, so no Python API is available. Scripts should continue to use the command line interface.
 
 
 v4.3.5
 ------
 
--  Update documentation to confirm Python 3.6.0 compatibility. No code changes were needed, so many earlier versions are likely supported.
+-   Update documentation to confirm Python 3.6.0 compatibility. No code changes were needed, so many earlier versions are likely supported.
 
 
 v4.3.4
 ------
 
--  Fixed "decimal.InvalidOperation: quantize result has too many digits" for high DPI images
+-   Fixed "decimal.InvalidOperation: quantize result has too many digits" for high DPI images
 
 
 v4.3.3
 ------
 
--  Fixed PDF/A creation with Ghostscript 9.20 properly
--  Fixed an exception on inline stencil masks with a missing optional parameter
+-   Fixed PDF/A creation with Ghostscript 9.20 properly
+-   Fixed an exception on inline stencil masks with a missing optional parameter
 
 
 v4.3.2
 ------
 
--  Fixed a PDF/A creation issue with Ghostscript 9.20 (note: this fix did not actually work)
+-   Fixed a PDF/A creation issue with Ghostscript 9.20 (note: this fix did not actually work)
 
 
 v4.3.1
 ------
 
--  Fixed an issue where pages produced by the "hocr" renderer after a Tesseract timeout would be rotated incorrectly if the input page was rotated with a /Rotate marker
--  Fixed a file handle leak in LeptonicaErrorTrap that would cause a "too many open files" error for files around hundred pages of pages long when ``--deskew`` or ``--remove-background`` or other Leptonica based image processing features were in use, depending on the system value of ``ulimit -n``
--  Ability to specify multiple languages for multilingual documents is now advertised in documentation
--  Reduced the file sizes of some test resources
--  Cleaned up debug output
--  Tesseract caching in test cases is now more cautious about false cache hits and reproducing exact output, not that any problems were observed
+-   Fixed an issue where pages produced by the "hocr" renderer after a Tesseract timeout would be rotated incorrectly if the input page was rotated with a /Rotate marker
+-   Fixed a file handle leak in LeptonicaErrorTrap that would cause a "too many open files" error for files around hundred pages of pages long when ``--deskew`` or ``--remove-background`` or other Leptonica based image processing features were in use, depending on the system value of ``ulimit -n``
+-   Ability to specify multiple languages for multilingual documents is now advertised in documentation
+-   Reduced the file sizes of some test resources
+-   Cleaned up debug output
+-   Tesseract caching in test cases is now more cautious about false cache hits and reproducing exact output, not that any problems were observed
 
 
 v4.3
 ----
 
--  New feature ``--remove-background`` to detect and erase the background of color and grayscale images
--  Better documentation
--  Fixed an issue with PDFs that draw images when the raster stack depth is zero 
--  ocrmypdf can now redirect its output to stdout for use in a shell pipeline
+-   New feature ``--remove-background`` to detect and erase the background of color and grayscale images
+-   Better documentation
+-   Fixed an issue with PDFs that draw images when the raster stack depth is zero 
+-   ocrmypdf can now redirect its output to stdout for use in a shell pipeline
 
-   +  This does not improve performance since temporary files are still used for buffering
-   +  Some output validation is disabled in this mode
+    +  This does not improve performance since temporary files are still used for buffering
+    +  Some output validation is disabled in this mode
 
 v4.2.5
 ------
 
--  Fixed an issue (#100) with PDFs that omit the optional /BitsPerComponent parameter on images
--  Removed non-free file milk.pdf
+-   Fixed an issue (#100) with PDFs that omit the optional /BitsPerComponent parameter on images
+-   Removed non-free file milk.pdf
 
 
 v4.2.4
 ------
 
--  Fixed an error (#90) caused by PDFs that use stencil masks properly
--  Fixed handling of PDFs that try to draw images or stencil masks without properly setting up the graphics state (such images are now ignored for the purposes of calculating DPI)
+-   Fixed an error (#90) caused by PDFs that use stencil masks properly
+-   Fixed handling of PDFs that try to draw images or stencil masks without properly setting up the graphics state (such images are now ignored for the purposes of calculating DPI)
 
 v4.2.3
 ------
 
--  Fixed an issue with PDFs that store page rotation (/Rotate) in an indirect object
--  Integrated a few fixes to simplify downstream packaging (Debian)
+-   Fixed an issue with PDFs that store page rotation (/Rotate) in an indirect object
+-   Integrated a few fixes to simplify downstream packaging (Debian)
 
-   +  The test suite no longer assumes it is installed
-   +  If running Linux, skip a test that passes Unicode on the command line
+    +  The test suite no longer assumes it is installed
+    +  If running Linux, skip a test that passes Unicode on the command line
 
--  Added a test case to check explicit masks and stencil masks
--  Added a test case for indirect objects and linearized PDFs
--  Deprecated the OCRmyPDF.sh shell script
+-   Added a test case to check explicit masks and stencil masks
+-   Added a test case for indirect objects and linearized PDFs
+-   Deprecated the OCRmyPDF.sh shell script
 
 
 v4.2.2
 ------
 
--  Improvements to documentation
+-   Improvements to documentation
 
 
 v4.2.1
 ------
 
--  Fixed an issue where PDF pages that contained stencil masks would report an incorrect DPI and cause Ghostscript to abort
--  Implemented stdin streaming
+-   Fixed an issue where PDF pages that contained stencil masks would report an incorrect DPI and cause Ghostscript to abort
+-   Implemented stdin streaming
 
 
 v4.2
 ----
 
--  ocrmypdf will now try to convert single image files to PDFs if they are provided as input (#15)
+-   ocrmypdf will now try to convert single image files to PDFs if they are provided as input (#15)
 
-   +  This is a basic convenience feature. It only supports a single image and always makes the image fill the whole page.
-   +  For better control over image to PDF conversion, use ``img2pdf`` (one of ocrmypdf's dependencies)
+    +  This is a basic convenience feature. It only supports a single image and always makes the image fill the whole page.
+    +  For better control over image to PDF conversion, use ``img2pdf`` (one of ocrmypdf's dependencies)
 
--  New argument ``--output-type {pdf|pdfa}`` allows disabling Ghostscript PDF/A generation
+-   New argument ``--output-type {pdf|pdfa}`` allows disabling Ghostscript PDF/A generation
 
-   +  ``pdfa`` is the default, consistent with past behavior
-   +  ``pdf`` provides a workaround for users concerned about the increase in file size from Ghostscript forcing JBIG2 images to CCITT and transcoding JPEGs
-   +  ``pdf`` preserves as much as it can about the original file, including problems that PDF/A conversion fixes
+    +  ``pdfa`` is the default, consistent with past behavior
+    +  ``pdf`` provides a workaround for users concerned about the increase in file size from Ghostscript forcing JBIG2 images to CCITT and transcoding JPEGs
+    +  ``pdf`` preserves as much as it can about the original file, including problems that PDF/A conversion fixes
 
--  PDFs containing images with "non-square" pixel aspect ratios, such as 200x100 DPI, are now handled and converted properly (fixing a bug that caused to be cropped)
--  ``--force-ocr`` rasterizes pages even if they contain no images
+-   PDFs containing images with "non-square" pixel aspect ratios, such as 200x100 DPI, are now handled and converted properly (fixing a bug that caused to be cropped)
+-   ``--force-ocr`` rasterizes pages even if they contain no images
 
-   +  supports users who want to use OCRmyPDF to reconstruct text information in PDFs with damaged Unicode maps (copy and paste text does not match displayed text)
-   +  supports reinterpreting PDFs where text was rendered as curves for printing, and text needs to be recovered
-   +  fixes issue #82
+    +  supports users who want to use OCRmyPDF to reconstruct text information in PDFs with damaged Unicode maps (copy and paste text does not match displayed text)
+    +  supports reinterpreting PDFs where text was rendered as curves for printing, and text needs to be recovered
+    +  fixes issue #82
 
--  Fixes an issue where, with certain settings, monochrome images in PDFs would be converted to 8-bit grayscale, increasing file size (#79)
--  Support for Ubuntu 12.04 LTS "precise" has been dropped in favor of (roughly) Ubuntu 14.04 LTS "trusty" 
+-   Fixes an issue where, with certain settings, monochrome images in PDFs would be converted to 8-bit grayscale, increasing file size (#79)
+-   Support for Ubuntu 12.04 LTS "precise" has been dropped in favor of (roughly) Ubuntu 14.04 LTS "trusty" 
 
-   +  Some Ubuntu "PPAs" (backports) are needed to make it work
-      
--  Support for some older dependencies dropped
+    +  Some Ubuntu "PPAs" (backports) are needed to make it work
 
-   +  Ghostscript 9.15 or later is now required (available in Ubuntu trusty with backports)
-   +  Tesseract 3.03 or later is now required (available in Ubuntu trusty)
+-   Support for some older dependencies dropped
 
--  Ghostscript now runs in "safer" mode where possible
+    +  Ghostscript 9.15 or later is now required (available in Ubuntu trusty with backports)
+    +  Tesseract 3.03 or later is now required (available in Ubuntu trusty)
+
+-   Ghostscript now runs in "safer" mode where possible
 
 v4.1.4
 ------
 
--  Bug fix: monochrome images with an ICC profile attached were incorrectly converted to full color images if lossless reconstruction was not possible due to other settings; consequence was increased file size for these images
+-   Bug fix: monochrome images with an ICC profile attached were incorrectly converted to full color images if lossless reconstruction was not possible due to other settings; consequence was increased file size for these images
 
 
 v4.1.3
 ------
 
--  More helpful error message for PDFs with version 4 security handler
--  Update usage instructions for Windows/Docker users
--  Fix order of operations for matrix multiplication (no effect on most users)
--  Add a few leptonica wrapper functions (no effect on most users)
+-   More helpful error message for PDFs with version 4 security handler
+-   Update usage instructions for Windows/Docker users
+-   Fix order of operations for matrix multiplication (no effect on most users)
+-   Add a few leptonica wrapper functions (no effect on most users)
 
 
 v4.1.2
 ------
 
--  Replace IEC sRGB ICC profile with Debian's sRGB (from icc-profiles-free) which is more compatible with the MIT license
--  More helpful error message for an error related to certain types of malformed PDFs
+-   Replace IEC sRGB ICC profile with Debian's sRGB (from icc-profiles-free) which is more compatible with the MIT license
+-   More helpful error message for an error related to certain types of malformed PDFs
 
 
 v4.1
 ----
 
--  ``--rotate-pages`` now only rotates pages when reasonably confidence in the orientation. This behavior can be adjusted with the new argument ``--rotate-pages-threshold``
--  Fixed problems in error checking if ``unpaper`` is uninstalled or missing at run-time
--  Fixed problems with "RethrownJobError" errors during error handling that suppressed the useful error messages
+-   ``--rotate-pages`` now only rotates pages when reasonably confidence in the orientation. This behavior can be adjusted with the new argument ``--rotate-pages-threshold``
+-   Fixed problems in error checking if ``unpaper`` is uninstalled or missing at run-time
+-   Fixed problems with "RethrownJobError" errors during error handling that suppressed the useful error messages
 
 
 v4.0.7
 ------
 
--  Minor correction to Ghostscript output settings
+-   Minor correction to Ghostscript output settings
 
 
 v4.0.6
 ------
 
--  Update install instructions
--  Provide a sRGB profile instead of using Ghostscript's
+-   Update install instructions
+-   Provide a sRGB profile instead of using Ghostscript's
 
 
 v4.0.5
 ------
 
--  Remove some verbose debug messages from v4.0.4
--  Fixed temporary that wasn't being deleted
--  DPI is now calculated correctly for cropped images, along with other image transformations
--  Inline images are now checked during DPI calculation instead of rejecting the image
+-   Remove some verbose debug messages from v4.0.4
+-   Fixed temporary that wasn't being deleted
+-   DPI is now calculated correctly for cropped images, along with other image transformations
+-   Inline images are now checked during DPI calculation instead of rejecting the image
 
 v4.0.4
 ------
@@ -294,16 +494,16 @@ v4.0.3
 New features
 ^^^^^^^^^^^^
 
--  Page orientations detected are now reported in a summary comment
+-   Page orientations detected are now reported in a summary comment
 
 
 Fixes
 ^^^^^
 
--  Show stack trace if unexpected errors occur
--  Treat "too few characters" error message from Tesseract as a reason to skip that page rather than
-   abort the file
--  Docker: fix blank JPEG2000 issue by insisting on Ghostscript versions that have this fixed
+-   Show stack trace if unexpected errors occur
+-   Treat "too few characters" error message from Tesseract as a reason to skip that page rather than
+    abort the file
+-   Docker: fix blank JPEG2000 issue by insisting on Ghostscript versions that have this fixed
 
 
 v4.0.2
@@ -312,10 +512,10 @@ v4.0.2
 Fixes
 ^^^^^
 
--  Fixed compatibility with Tesseract 3.04.01 release, particularly its different way of outputting
-   orientation information
--  Improved handling of Tesseract errors and crashes
--  Fixed use of chmod on Docker that broke most test cases
+-   Fixed compatibility with Tesseract 3.04.01 release, particularly its different way of outputting
+    orientation information
+-   Improved handling of Tesseract errors and crashes
+-   Fixed use of chmod on Docker that broke most test cases
 
 
 v4.0.1
@@ -324,7 +524,7 @@ v4.0.1
 Fixes
 ^^^^^
 
--  Fixed a KeyError if tesseract fails to find page orientation information
+-   Fixed a KeyError if tesseract fails to find page orientation information
 
 
 v4.0
@@ -333,33 +533,33 @@ v4.0
 New features
 ^^^^^^^^^^^^
 
--  Automatic page rotation (``-r``) is now available. It uses ignores any prior rotation information
-   on PDFs and sets rotation based on the dominant orientation of detectable text. This feature is
-   fairly reliable but some false positives occur especially if there is not much text to work with. (#4) 
--  Deskewing is now performed using Leptonica instead of unpaper. Leptonica is faster and more reliable
-   at image deskewing than unpaper.
+-   Automatic page rotation (``-r``) is now available. It uses ignores any prior rotation information
+    on PDFs and sets rotation based on the dominant orientation of detectable text. This feature is
+    fairly reliable but some false positives occur especially if there is not much text to work with. (#4) 
+-   Deskewing is now performed using Leptonica instead of unpaper. Leptonica is faster and more reliable
+    at image deskewing than unpaper.
 
 
 Fixes
 ^^^^^
 
--  Fixed an issue where lossless reconstruction could cause some pages to be appear incorrectly
-   if the page was rotated by the user in Acrobat after being scanned (specifically if it a /Rotate tag)
--  Fixed an issue where lossless reconstruction could misalign the graphics layer with respect to
-   text layer if the page had been cropped such that its origin is not (0, 0) (#49)
+-   Fixed an issue where lossless reconstruction could cause some pages to be appear incorrectly
+    if the page was rotated by the user in Acrobat after being scanned (specifically if it a /Rotate tag)
+-   Fixed an issue where lossless reconstruction could misalign the graphics layer with respect to
+    text layer if the page had been cropped such that its origin is not (0, 0) (#49)
 
 
 Changes
 ^^^^^^^
 
--  Logging output is now much easier to read
--  ``--deskew`` is now performed by Leptonica instead of unpaper (#25)
--  libffi is now required
--  Some changes were made to the Docker and Travis build environments to support libffi
--  ``--pdf-renderer=tesseract`` now displays a warning if the Tesseract version is less than 3.04.01,
-   the planned release that will include fixes to an important OCR text rendering bug in Tesseract 3.04.00.
-   You can also manually install ./share/sharp2.ttf on top of pdf.ttf in your Tesseract tessdata folder
-   to correct the problem.
+-   Logging output is now much easier to read
+-   ``--deskew`` is now performed by Leptonica instead of unpaper (#25)
+-   libffi is now required
+-   Some changes were made to the Docker and Travis build environments to support libffi
+-   ``--pdf-renderer=tesseract`` now displays a warning if the Tesseract version is less than 3.04.01,
+    the planned release that will include fixes to an important OCR text rendering bug in Tesseract 3.04.00.
+    You can also manually install ./share/sharp2.ttf on top of pdf.ttf in your Tesseract tessdata folder
+    to correct the problem.
 
 
 v3.2.1
@@ -368,8 +568,8 @@ v3.2.1
 Changes
 ^^^^^^^
 
--  Fixed issue #47 "convert() got and unexpected keyword argument 'dpi'" by upgrading to img2pdf 0.2
--  Tweaked the Dockerfiles
+-   Fixed issue #47 "convert() got and unexpected keyword argument 'dpi'" by upgrading to img2pdf 0.2
+-   Tweaked the Dockerfiles
 
 
 v3.2
@@ -378,19 +578,19 @@ v3.2
 New features
 ^^^^^^^^^^^^
 
--  Lossless reconstruction: when possible, OCRmyPDF will inject text layers without 
-   otherwise manipulating the content and layout of a PDF page. For example, a PDF containing a mix
-   of vector and raster content would see the vector content preserved. Images may still be transcoded
-   during PDF/A conversion.  (``--deskew`` and ``--clean-final`` disable this mode, necessarily.)
--  New argument ``--tesseract-pagesegmode`` allows you to pass page segmentation arguments to Tesseract OCR.
-   This helps for two column text and other situations that confuse Tesseract.
--  Added a new "polyglot" version of the Docker image, that generates Tesseract with all languages packs installed,
-   for the polyglots among us. It is much larger.
+-   Lossless reconstruction: when possible, OCRmyPDF will inject text layers without 
+    otherwise manipulating the content and layout of a PDF page. For example, a PDF containing a mix
+    of vector and raster content would see the vector content preserved. Images may still be transcoded
+    during PDF/A conversion.  (``--deskew`` and ``--clean-final`` disable this mode, necessarily.)
+-   New argument ``--tesseract-pagesegmode`` allows you to pass page segmentation arguments to Tesseract OCR.
+    This helps for two column text and other situations that confuse Tesseract.
+-   Added a new "polyglot" version of the Docker image, that generates Tesseract with all languages packs installed,
+    for the polyglots among us. It is much larger.
 
 Changes
 ^^^^^^^
 
--  JPEG transcoding quality is now 95 instead of the default 75. Bigger file sizes for less degradation.
+-   JPEG transcoding quality is now 95 instead of the default 75. Bigger file sizes for less degradation.
 
 
 
@@ -400,7 +600,7 @@ v3.1.1
 Changes
 ^^^^^^^
 
--  Fixed bug that caused incorrect page size and DPI calculations on documents with mixed page sizes
+-   Fixed bug that caused incorrect page size and DPI calculations on documents with mixed page sizes
 
 v3.1
 ----
@@ -408,17 +608,17 @@ v3.1
 Changes
 ^^^^^^^
 
--  Default output format is now PDF/A-2b instead of PDF/A-1b
--  Python 3.5 and macOS El Capitan are now supported platforms - no changes were
-   needed to implement support
--  Improved some error messages related to missing input files
--  Fixed issue #20 - uppercase .PDF extension not accepted
--  Fixed an issue where OCRmyPDF failed to text that certain pages contained previously OCR'ed text, 
-   such as OCR text produced by Tesseract 3.04
--  Inserts /Creator tag into PDFs so that errors can be traced back to this project
--  Added new option ``--pdf-renderer=auto``, to let OCRmyPDF pick the best PDF renderer. 
-   Currently it always chooses the 'hocrtransform' renderer but that behavior may change.
--  Set up Travis CI automatic integration testing
+-   Default output format is now PDF/A-2b instead of PDF/A-1b
+-   Python 3.5 and macOS El Capitan are now supported platforms - no changes were
+    needed to implement support
+-   Improved some error messages related to missing input files
+-   Fixed issue #20 - uppercase .PDF extension not accepted
+-   Fixed an issue where OCRmyPDF failed to text that certain pages contained previously OCR'ed text, 
+    such as OCR text produced by Tesseract 3.04
+-   Inserts /Creator tag into PDFs so that errors can be traced back to this project
+-   Added new option ``--pdf-renderer=auto``, to let OCRmyPDF pick the best PDF renderer. 
+    Currently it always chooses the 'hocrtransform' renderer but that behavior may change.
+-   Set up Travis CI automatic integration testing
 
 v3.0
 ----
@@ -426,55 +626,55 @@ v3.0
 New features
 ^^^^^^^^^^^^
 
--  Easier installation with a Docker container or Python's ``pip`` package manager 
--  Eliminated many external dependencies, so it's easier to setup
--  Now installs ``ocrmypdf`` to ``/usr/local/bin`` or equivalent for system-wide
-   access and easier typing
--  Improved command line syntax and usage help (``--help``)
--  Tesseract 3.03+ PDF page rendering can be used instead for better positioning
-   of recognized text (``--pdf-renderer tesseract``)
--  PDF metadata (title, author, keywords) are now transferred to the 
-   output PDF
--  PDF metadata can also be set from the command line (``--title``, etc.)
--  Automatic repairs malformed input PDFs if possible
--  Added test cases to confirm everything is working
--  Added option to skip extremely large pages that take too long to OCR and are 
-   often not OCRable (e.g. large scanned maps or diagrams); other pages are still
-   processed (``--skip-big``)
--  Added option to kill Tesseract OCR process if it seems to be taking too long on
-   a page, while still processing other pages (``--tesseract-timeout``)
--  Less common colorspaces (CMYK, palette) are now supported by conversion to RGB
--  Multiple images on the same PDF page are now supported
+-   Easier installation with a Docker container or Python's ``pip`` package manager 
+-   Eliminated many external dependencies, so it's easier to setup
+-   Now installs ``ocrmypdf`` to ``/usr/local/bin`` or equivalent for system-wide
+    access and easier typing
+-   Improved command line syntax and usage help (``--help``)
+-   Tesseract 3.03+ PDF page rendering can be used instead for better positioning
+    of recognized text (``--pdf-renderer tesseract``)
+-   PDF metadata (title, author, keywords) are now transferred to the 
+    output PDF
+-   PDF metadata can also be set from the command line (``--title``, etc.)
+-   Automatic repairs malformed input PDFs if possible
+-   Added test cases to confirm everything is working
+-   Added option to skip extremely large pages that take too long to OCR and are 
+    often not OCRable (e.g. large scanned maps or diagrams); other pages are still
+    processed (``--skip-big``)
+-   Added option to kill Tesseract OCR process if it seems to be taking too long on
+    a page, while still processing other pages (``--tesseract-timeout``)
+-   Less common colorspaces (CMYK, palette) are now supported by conversion to RGB
+-   Multiple images on the same PDF page are now supported
 
 Changes
 ^^^^^^^
 
--  New, robust rewrite in Python 3.4+ with ruffus_ pipelines
--  Now uses Ghostscript 9.14's improved color conversion model to preserve PDF colors
--  OCR text is now rendered in the PDF as invisible text. Previous versions of OCRmyPDF
-   incorrectly rendered visible text with an image on top.
--  All "tasks" in the pipeline can be executed in parallel on any
-   available CPUs, increasing performance
--  The ``-o DPI`` argument has been phased out, in favor of ``--oversample DPI``, in
-   case we need ``-o OUTPUTFILE`` in the future
--  Removed several dependencies, so it's easier to install.  We no 
-   longer use:
-   
-   - GNU parallel_
-   - ImageMagick_
-   - Python 2.7
-   - Poppler
-   - MuPDF_ tools
-   - shell scripts
-   - Java and JHOVE_
-   - libxml2
+-   New, robust rewrite in Python 3.4+ with ruffus_ pipelines
+-   Now uses Ghostscript 9.14's improved color conversion model to preserve PDF colors
+-   OCR text is now rendered in the PDF as invisible text. Previous versions of OCRmyPDF
+    incorrectly rendered visible text with an image on top.
+-   All "tasks" in the pipeline can be executed in parallel on any
+    available CPUs, increasing performance
+-   The ``-o DPI`` argument has been phased out, in favor of ``--oversample DPI``, in
+    case we need ``-o OUTPUTFILE`` in the future
+-   Removed several dependencies, so it's easier to install.  We no 
+    longer use:
+    
+    - GNU parallel_
+    - ImageMagick_
+    - Python 2.7
+    - Poppler
+    - MuPDF_ tools
+    - shell scripts
+    - Java and JHOVE_
+    - libxml2
 
--  Some new external dependencies are required or optional, compared to v2.x:
+-   Some new external dependencies are required or optional, compared to v2.x:
 
-   - Ghostscript 9.14+
-   - qpdf_ 5.0.0+
-   - Unpaper_ 6.1 (optional)
-   - some automatically managed Python packages
+    - Ghostscript 9.14+
+    - qpdf_ 5.0.0+
+    - Unpaper_ 6.1 (optional)
+    - some automatically managed Python packages
   
 .. _ruffus: http://www.ruffus.org.uk/index.html
 .. _parallel: https://www.gnu.org/software/parallel/
@@ -487,87 +687,87 @@ Changes
 Release candidates
 ^^^^^^^^^^^^^^^^^^
 
--  rc9:
+-   rc9:
 
-   - fix issue #118: report error if ghostscript iccprofiles are missing
-   - fixed another issue related to #111: PDF rasterized to palette file
-   - add support image files with a palette
-   - don't try to validate PDF file after an exception occurs
+    - fix issue #118: report error if ghostscript iccprofiles are missing
+    - fixed another issue related to #111: PDF rasterized to palette file
+    - add support image files with a palette
+    - don't try to validate PDF file after an exception occurs
 
--  rc8:
+-   rc8:
 
-   - fix issue #111: exception thrown if PDF is missing DocumentInfo dictionary
+    - fix issue #111: exception thrown if PDF is missing DocumentInfo dictionary
 
--  rc7:
+-   rc7:
 
-   - fix error when installing direct from pip, "no such file 'requirements.txt'"
+    - fix error when installing direct from pip, "no such file 'requirements.txt'"
 
--  rc6:
+-   rc6:
 
-   - dropped libxml2 (Python lxml) since Python 3's internal XML parser is sufficient
-   - set up Docker container
-   - fix Unicode errors if recognized text contains Unicode characters and system locale is not UTF-8
+    - dropped libxml2 (Python lxml) since Python 3's internal XML parser is sufficient
+    - set up Docker container
+    - fix Unicode errors if recognized text contains Unicode characters and system locale is not UTF-8
 
--  rc5:
+-   rc5:
 
-   - dropped Java and JHOVE in favour of qpdf
-   - improved command line error output
-   - additional tests and bug fixes
-   - tested on Ubuntu 14.04 LTS
+    - dropped Java and JHOVE in favour of qpdf
+    - improved command line error output
+    - additional tests and bug fixes
+    - tested on Ubuntu 14.04 LTS
 
--  rc4:
+-   rc4:
 
-   - dropped MuPDF in favour of qpdf
-   - fixed some installer issues and errors in installation instructions
-   - improve performance: run Ghostscript with multithreaded rendering
-   - improve performance: use multiple cores by default
-   - bug fix: checking for wrong exception on process timeout 
+    - dropped MuPDF in favour of qpdf
+    - fixed some installer issues and errors in installation instructions
+    - improve performance: run Ghostscript with multithreaded rendering
+    - improve performance: use multiple cores by default
+    - bug fix: checking for wrong exception on process timeout 
 
--  rc3: skipping version number intentionally to avoid confusion with Tesseract
--  rc2: first release for public testing to test-PyPI, Github
--  rc1: testing release process
+-   rc3: skipping version number intentionally to avoid confusion with Tesseract
+-   rc2: first release for public testing to test-PyPI, Github
+-   rc1: testing release process
 
 Compatibility notes
 -------------------
 
--  ``./OCRmyPDF.sh`` script is still available for now
--  Stacking the verbosity option like ``-vvv`` is no longer supported
+-   ``./OCRmyPDF.sh`` script is still available for now
+-   Stacking the verbosity option like ``-vvv`` is no longer supported
 
--  The configuration file ``config.sh`` has been removed.  Instead, you can
-   feed a file to the arguments for common settings:
+-   The configuration file ``config.sh`` has been removed.  Instead, you can
+    feed a file to the arguments for common settings:
 
 ::
 
-   ocrmypdf input.pdf output.pdf @settings.txt
+    ocrmypdf input.pdf output.pdf @settings.txt
 
 where ``settings.txt`` contains *one argument per line*, for example:
 
 ::
 
-   -l 
-   deu 
-   --author 
-   A. Merkel 
-   --pdf-renderer 
-   tesseract
+    -l 
+    deu 
+    --author 
+    A. Merkel 
+    --pdf-renderer 
+    tesseract
 
 
 Fixes
 ^^^^^
 
--  Handling of filenames containing spaces: fixed
+-   Handling of filenames containing spaces: fixed
 
 Notes and known issues
 ^^^^^^^^^^^^^^^^^^^^^^
 
--  Some dependencies may work with lower versions than tested, so try
-   overriding dependencies if they are "in the way" to see if they work.
+-   Some dependencies may work with lower versions than tested, so try
+    overriding dependencies if they are "in the way" to see if they work.
 
--  ``--pdf-renderer tesseract`` will output files with an incorrect page size in Tesseract 3.03,
-   due to a bug in Tesseract.
+-   ``--pdf-renderer tesseract`` will output files with an incorrect page size in Tesseract 3.03,
+    due to a bug in Tesseract.
 
--  PDF files containing "inline images" are not supported and won't be for the 3.0 release. Scanned
-   images almost never contain inline images.
+-   PDF files containing "inline images" are not supported and won't be for the 3.0 release. Scanned
+    images almost never contain inline images.
 
 
 v2.2-stable (2014-09-29)

docs/security.rst

@@ -10,7 +10,7 @@ The disclaimer applies: this software has no warranties of any kind.
 PDFs may contain malware
 ------------------------
 
-PDF is a rich, complex file format. The official PDF 1.7 specification, ISO 32000:2008, is hundreds of packages long and references several annexes each of which are similar in length. PDFs can contain video, audio, JavaScript and other programming, and forms. In some cases, they can open internet connections to pre-selected URLs. All of these possible attack vectors.
+PDF is a rich, complex file format. The official PDF 1.7 specification, ISO 32000:2008, is hundreds of pages long and references several annexes each of which are similar in length. PDFs can contain video, audio, XML, JavaScript and other programming, and forms. In some cases, they can open internet connections to pre-selected URLs. All of these possible attack vectors.
 
 In short, PDFs `may contain viruses <https://security.stackexchange.com/questions/64052/can-a-pdf-file-contain-a-virus>`_.
 
@@ -21,23 +21,58 @@ How OCRmyPDF processes PDFs
 
 OCRmyPDF must open and interpret your PDF in order to insert an OCR layer. First, it runs all PDFs through `qpdf <https://github.com/qpdf/qpdf>`_, a program that repairs PDFs with syntax errors. This is done because, in the author's experience, a significant number of PDFs in the wild especially those created by scanners are not well-formed files. qpdf makes it more likely that OCRmyPDF will succeed, but offers no security guarantees. qpdf is also used to split the PDF into single page PDFs.
 
-After qpdf, OCRmyPDF examines each page using `PyPDF2 <https://github.com/mstamy2/PyPDF2>`_. This library also has no warranties or guarantees.
+After qpdf, OCRmyPDF examines each page using `PyPDF2 <https://github.com/mstamy2/PyPDF2>`_. This library also has no warranties or guarantees. OCRmyPDF works with qpdf 5.0 and up, but version 7.0 is recommended because of known security vulnerabilities in early versions.
 
-Finally, OCRmyPDF rasterizes each page of the PDF using `Ghostscript <http://ghostscript.com/>`_ in ``-dSAFER`` mode. 
+Finally, OCRmyPDF rasterizes each page of the PDF using `Ghostscript <http://ghostscript.com/>`_ in ``-dSAFER`` mode.
 
 Depending on the options specified, OCRmyPDF may graft the OCR layer into the existing PDF or it may essentially reconstruct ("re-fry") a visually identical PDF that may be quite different at the binary level. That said, OCRmyPDF is not a tool designed for sanitizing PDFs.
 
-Using OCRmyPDF online
----------------------
+Using OCRmyPDF online or as a service
+-------------------------------------
 
-OCRmyPDF is not designed to be deployed "as a service", in a setting where a user/attacker could upload a file for OCR processing online. It is not designed to be secure in this case.
+OCRmyPDF should not be deployed as a public-facing service, such as a website where a potential attacker could upload a PDF of their choice for OCR. OCRmyPDF is not designed to be secure against PDF malware. Another concern is PDFs specifically designed to be a denial of service attack: PDFs can contain recursive data structures that sometimes send parsers into infinite loops, and issue complex graphics drawing commands.
+
+Setting aside these concerns, a side effect of OCRmyPDF is it may incidentally sanitize PDFs that contain malware. It runs ``qpdf`` to repair the PDF, which could correct malformed PDF structures that are part of an attack. When PDF/A output is selected (the default), the input PDF is partially reconstructed by Ghostscript. When ``--force-ocr`` is used, all pages are rasterized and reconverted to PDF, which could remove malware in embedded images. No guarantees.
+
+OCRmyPDF should be relatively safe to use in a trusted intranet, with some considerations:
+
+Limiting CPU usage
+^^^^^^^^^^^^^^^^^^
+
+OCRmyPDF will attempt to use all available CPUs and storage, so executing ``nice ocrmypdf`` or limiting the number of jobs with the ``-j`` argument may ensure the server remains available. Another option would be run OCRmyPDF jobs inside a Docker container, a virtual machine, or a cloud instance, which can impose its own limits on CPU usage and be terminated "from orbit" if it fails to complete.
+
+Temporary storage requirements
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+OCRmyPDF will use a large amount of temporary storage for its work, proportional to the total number of pixels needed to rasterize the PDF. The raster image of a 8.5×11" color page at 300 DPI takes 25 MB uncompressed; OCRmyPDF saves its intermediates as PNG, but that still means it requires about 9 MB per intermediate based on average compression ratios. Multiple intermediates per page are also required, depending on the command line given. A rule of thumb would be to allow 100 MB of temporary storage per page in a file – meaning that a small cloud servers or small VM partitions should be provisioned with plenty of extra space, if say, a 500 page file might be sent.
+
+To check temporary storage usage on actual files, run ``ocrmypdf -k ...`` which will preserve and print the path to temporary storage when the job is done.
+
+To change where temporary files are stored, change the ``TMPDIR`` environment variable for ocrmypdf's environment. (Python's ``tempfile.gettempdir()`` returns the root directory in which temporary files will be stored.) For example, one could redirect ``TMPDIR`` to a large RAM disk to avoid wear on HDD/SSD and potentially improve performance. On Amazon Web Services, ``TMPDIR`` can be set to `empheral storage <https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/InstanceStorage.html>`_.
+
+Timeouts
+^^^^^^^^
+
+To prevent excessively long OCR jobs consider setting ``--tesseract-timeout`` and/or ``--skip-big`` arguments. ``--skip-big`` is particularly helpful if your PDFs include documents such as reports on standard page sizes with large images attached - often large images are not worth OCR'ing anyway.
+
+Commercial alternatives
+^^^^^^^^^^^^^^^^^^^^^^^
+
+The author also provides professional services that include OCR and building databases around PDFs, and is happy to provide consultation.
+
+Abbyy Cloud OCR is a viable commercial alternative with a web services API. 
 
-Abbyy Cloud OCR is a viable commercial alternative with a web services API. The author also provides professional services that include OCR and building databases around PDFs, and is happy to provide consultation.
 
 Password protection, digital signatures and certification
 ---------------------------------------------------------
 
-OCRmyPDF cannot remove password protection from a PDF. ``qpdf``, one of its dependencies, has this capability. After OCR is applied, password protection is not permitted on PDF/A documents but the file can be converted to regular PDF.
+Password protected PDFs usually have two passwords, and owner and user password. When the user password is set to empty, PDF readers will open the file automatically and marked it as "(SECURED)". While not as reliable as a digital signature, this indicates that whoever set the password approved of the file at that time. When the user password is set, the document cannot be viewed without the password. 
+
+Either way, OCRmyPDF does not remove passwords from PDFs and exits with an error on encountering them.
+
+``qpdf``, one of OCRmyPDF's dependencies, can remove passwords. If the owner and user password are set, a password is required for ``qpdf``. If only the owner password is set, then the password can be stripped, even if one does not have the owner password.
+
+After OCR is applied, password protection is not permitted on PDF/A documents but the file can be converted to regular PDF.
 
 Many programs exist which are capable of inserting an image of someone's signature. On its own, this offers no security guarantees. It is trivial to remove the signature image and apply it to other files. This practice offers no real security.
 

ocrmypdf/__init__.py

@@ -1,7 +0,0 @@
-import pkg_resources
-
-PROGRAM_NAME = 'ocrmypdf'
-
-VERSION = pkg_resources.get_distribution('ocrmypdf').version
-
-

ocrmypdf/exec/__init__.py

@@ -1,11 +0,0 @@
-#!/usr/bin/env python3
-# © 2016 James R. Barlow: github.com/jbarlow83
-
-"""Wrappers to manage subprocess calls"""
-
-import os
-
-
-def get_program(name):
-    envvar = 'OCRMYPDF_' + name.upper()
-    return os.environ.get(envvar, name)

ocrmypdf/exec/qpdf.py

@@ -1,132 +0,0 @@
-#!/usr/bin/env python3
-# © 2015 James R. Barlow: github.com/jbarlow83
-
-from subprocess import CalledProcessError, STDOUT, PIPE, run, check_output
-from functools import lru_cache
-import sys
-import os
-import re
-
-from ..exceptions import InputFileError, SubprocessOutputError, \
-    MissingDependencyError, EncryptedPdfError
-from . import get_program
-
-
-@lru_cache(maxsize=1)
-def version():
-    args_qpdf = [
-        get_program('qpdf'),
-        '--version'
-    ]
-    try:
-        p = run(args_qpdf, universal_newlines=True, stderr=STDOUT,
-                stdout=PIPE)
-    except CalledProcessError as e:
-        print("Could not find qpdf executable on system PATH.",
-              file=sys.stderr)
-        raise MissingDependencyError() from e
-
-    qpdf_version = re.match(r'qpdf version (.+)', p.stdout).group(1)
-    return qpdf_version
-
-
-def check(input_file, log=None):
-    args_qpdf = [
-        get_program('qpdf'),
-        '--check',
-        input_file
-    ]
-
-    if log is None:
-        import logging as log
-
-    try:
-        check_output(args_qpdf, stderr=STDOUT, universal_newlines=True)
-    except CalledProcessError as e:
-        if e.returncode == 2:
-            log.error("{0}: not a valid PDF, and could not repair it.".format(
-                input_file))
-            log.error("Details:")
-            log.error(e.output)
-        elif e.returncode == 3:
-            log.info("qpdf --check returned warnings:")
-            log.info(e.output)
-        else:
-            log.warning(e.output)
-        return False
-    return True
-
-
-def _probably_encrypted(e):
-    """qpdf can report a false positive "file is encrypted" message for damaged
-    files - suppress this"""
-    return e.returncode == 2 and \
-        'invalid password' in e.output and \
-        'file is damaged' not in e.output
-
-
-def repair(input_file, output_file, log):
-    args_qpdf = [
-        get_program('qpdf'), input_file, output_file
-    ]
-    try:
-        check_output(args_qpdf, stderr=STDOUT, universal_newlines=True)
-    except CalledProcessError as e:
-        if e.returncode == 3 and e.output.find("operation succeeded"):
-            log.debug('qpdf found and fixed errors: ' + e.output)
-            log.debug(e.output)
-            return
-
-        if _probably_encrypted(e):
-            log.error("{0}: this PDF is password-protected - password must "
-                      "be removed for OCR".format(input_file))
-            raise EncryptedPdfError() from e
-        elif e.returncode == 2:
-            log.error("{0}: not a valid PDF, and could not repair it.".format(
-                      input_file))
-            log.error("Details: " + e.output)
-            raise InputFileError() from e
-        else:
-            log.error("{0}: unknown error".format(
-                      input_file))
-            log.error(e.output)
-            raise SubprocessOutputError() from e
-
-
-def get_npages(input_file, log):
-    try:
-        pages = check_output(
-            [get_program('qpdf'), '--show-npages', input_file],
-             universal_newlines=True, close_fds=True)
-    except CalledProcessError as e:
-        if e.returncode == 2 and e.output.find('No such file'):
-            log.error(e.output)
-            raise InputFileError() from e
-    return int(pages)
-
-
-def split_pages(input_file, work_folder, npages):
-    """Split multipage PDF into individual pages.
-
-    Incredibly enough, this multiple process approach is about 70 times
-    faster than using Ghostscript.
-    """
-    for n in range(int(npages)):
-        args_qpdf = [
-            get_program('qpdf'), input_file,
-            '--pages', input_file, '{0}'.format(n + 1), '--',
-            os.path.join(work_folder, '{0:06d}.page.pdf'.format(n + 1))
-        ]
-        run(args_qpdf, check=True)
-
-
-def merge(input_files, output_file):
-    """Merge the list of input files (all filenames) into the output file.
-
-    The input files may contain one or more pages.
-    """
-    args_qpdf = [
-        get_program('qpdf'), input_files[0], '--pages'
-    ] + input_files + ['--', output_file]
-    run(args_qpdf, check=True)
-

ocrmypdf/helpers.py

@@ -1,79 +0,0 @@
-#!/usr/bin/env python3
-# © 2016 James R. Barlow: github.com/jbarlow83
-
-from functools import partial
-from collections.abc import Iterable
-from contextlib import suppress
-import sys
-import os
-
-
-def re_symlink(input_file, soft_link_name, log=None):
-    """
-    Helper function: relinks soft symbolic link if necessary
-    """
-
-    if log is None:
-        prdebug = partial(print, file=sys.stderr)
-    else:
-        prdebug = log.debug
-
-    # Guard against soft linking to oneself
-    if input_file == soft_link_name:
-        prdebug("Warning: No symbolic link made. You are using " +
-                "the original data directory as the working directory.")
-        return
-
-    # Soft link already exists: delete for relink?
-    if os.path.lexists(soft_link_name):
-        # do not delete or overwrite real (non-soft link) file
-        if not os.path.islink(soft_link_name):
-            raise FileExistsError(
-                "%s exists and is not a link" % soft_link_name)
-        try:
-            os.unlink(soft_link_name)
-        except:
-            prdebug("Can't unlink %s" % (soft_link_name))
-
-    if not os.path.exists(input_file):
-        raise FileNotFoundError(
-            "trying to create a broken symlink to %s" % input_file)
-
-    prdebug("os.symlink(%s, %s)" % (input_file, soft_link_name))
-
-    # Create symbolic link using absolute path
-    os.symlink(
-        os.path.abspath(input_file),
-        soft_link_name
-    )
-
-
-def is_iterable_notstr(thing):
-    return isinstance(thing, Iterable) and not isinstance(thing, str)
-
-
-def page_number(input_file):
-    return int(os.path.basename(input_file)[0:6])
-
-
-def is_file_writable(test_file):
-    """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.
-    """
-    if os.path.exists(test_file):
-        return os.access(
-            test_file, os.W_OK,
-            effective_ids=(os.access in os.supports_effective_ids))
-    else:
-        try:
-            fp = open(test_file, 'wb')
-        except OSError as e:
-            return False
-        else:
-            fp.close()
-            with suppress(OSError):
-                os.unlink(test_file)
-        return True

ocrmypdf/hocrtransform.py

@@ -1,230 +0,0 @@
-#!/usr/bin/env python3
-##############################################################################
-# Copyright (c) 2013-14: fritz-hh from Github
-#   (https://github.com/fritz-hh)
-#
-# Copyright (c) 2010: Jonathan Brinley from Github
-#   (https://github.com/jbrinley/HocrConverter)
-#   Initial version by Jonathan Brinley, jonathanbrinley@gmail.com
-##############################################################################
-from reportlab.pdfgen.canvas import Canvas
-from reportlab.lib.units import inch
-from xml.etree import ElementTree
-from PIL import Image
-from collections import namedtuple
-import re
-import argparse
-
-
-Rect = namedtuple('Rect', ['x1', 'y1', 'x2', 'y2'])
-
-
-class HocrTransformError(Exception):
-    pass
-
-
-class HocrTransform():
-
-    """
-    A class for converting documents from the hOCR format.
-    For details of the hOCR format, see:
-    http://docs.google.com/View?docid=dfxcv4vc_67g844kf
-    """
-
-    def __init__(self, hocrFileName, dpi):
-        self.dpi = dpi
-        self.boxPattern = re.compile(r'bbox((\s+\d+){4})')
-
-        self.hocr = ElementTree.parse(hocrFileName)
-
-        # if the hOCR file has a namespace, ElementTree requires its use to
-        # find elements
-        matches = re.match(r'({.*})html', self.hocr.getroot().tag)
-        self.xmlns = ''
-        if matches:
-            self.xmlns = matches.group(1)
-
-        # get dimension in pt (not pixel!!!!) of the OCRed image
-        self.width, self.height = None, None
-        for div in self.hocr.findall(
-                ".//%sdiv[@class='ocr_page']" % (self.xmlns)):
-            coords = self.element_coordinates(div)
-            pt_coords = self.pt_from_pixel(coords)
-            self.width = pt_coords.x2 - pt_coords.x1
-            self.height = pt_coords.y2 - pt_coords.y1
-            # there shouldn't be more than one, and if there is, we don't want
-            # it
-            break
-        if self.width is None or self.height is None:
-            raise HocrTransformError("hocr file is missing page dimensions")
-
-    def __str__(self):
-        """
-        Return the textual content of the HTML body
-        """
-        if self.hocr is None:
-            return ''
-        body = self.hocr.find(".//%sbody" % (self.xmlns))
-        if body:
-            return self._get_element_text(body)
-        else:
-            return ''
-
-    def _get_element_text(self, element):
-        """
-        Return the textual content of the element and its children
-        """
-        text = ''
-        if element.text is not None:
-            text += element.text
-        for child in element.getchildren():
-            text += self._get_element_text(child)
-        if element.tail is not None:
-            text += element.tail
-        return text
-
-    def element_coordinates(self, element):
-        """
-        Returns a tuple containing the coordinates of the bounding box around
-        an element
-        """
-        out = (0, 0, 0, 0)
-        if 'title' in element.attrib:
-            matches = self.boxPattern.search(element.attrib['title'])
-            if matches:
-                coords = matches.group(1).split()
-                out = Rect._make(int(coords[n]) for n in range(4))
-        return out
-
-    def pt_from_pixel(self, pxl):
-        """
-        Returns the quantity in PDF units (pt) given quantity in pixels
-        """
-        return Rect._make(
-            (c / self.dpi * inch) for c in pxl)
-
-    def replace_unsupported_chars(self, s):
-        """
-        Given an input string, returns the corresponding string that:
-        - is available in the helvetica facetype
-        - does not contain any ligature (to allow easy search in the PDF file)
-        """
-        # The 'u' before the character to replace indicates that it is a
-        # unicode character
-        s = s.replace(u"fl", "fl")
-        s = s.replace(u"fi", "fi")
-        return s
-
-    def to_pdf(self, outFileName, imageFileName=None, showBoundingboxes=False,
-               fontname="Helvetica", invisibleText=False):
-        """
-        Creates a PDF file with an image superimposed on top of the text.
-        Text is positioned according to the bounding box of the lines in
-        the hOCR file.
-        The image need not be identical to the image used to create the hOCR
-        file.
-        It can have a lower resolution, different color mode, etc.
-        """
-        # create the PDF file
-        # page size in points (1/72 in.)
-        pdf = Canvas(
-            outFileName, pagesize=(self.width, self.height), pageCompression=1)
-
-        # draw bounding box for each paragraph
-        # light blue for bounding box of paragraph
-        pdf.setStrokeColorRGB(0, 1, 1)
-        # light blue for bounding box of paragraph
-        pdf.setFillColorRGB(0, 1, 1)
-        pdf.setLineWidth(0)		# no line for bounding box
-        for elem in self.hocr.findall(
-                ".//%sp[@class='%s']" % (self.xmlns, "ocr_par")):
-
-            elemtxt = self._get_element_text(elem).rstrip()
-            if len(elemtxt) == 0:
-                continue
-
-            pxl_coords = self.element_coordinates(elem)
-            pt = self.pt_from_pixel(pxl_coords)
-
-            # draw the bbox border
-            if showBoundingboxes:
-                pdf.rect(
-                    pt.x1, self.height - pt.y2, pt.x2 - pt.x1, pt.y2 - pt.y1,
-                    fill=1)
-
-        # check if element with class 'ocrx_word' are available
-        # otherwise use 'ocr_line' as fallback
-        elemclass = "ocr_line"
-        if self.hocr.find(
-                ".//%sspan[@class='ocrx_word']" % (self.xmlns)) is not None:
-            elemclass = "ocrx_word"
-
-        # itterate all text elements
-        # light green for bounding box of word/line
-        pdf.setStrokeColorRGB(1, 0, 0)
-        pdf.setLineWidth(0.5)		# bounding box line width
-        pdf.setDash(6, 3)		# bounding box is dashed
-        pdf.setFillColorRGB(0, 0, 0)  # text in black
-        for elem in self.hocr.findall(
-                ".//%sspan[@class='%s']" % (self.xmlns, elemclass)):
-
-            elemtxt = self._get_element_text(elem).rstrip()
-
-            elemtxt = self.replace_unsupported_chars(elemtxt)
-
-            if len(elemtxt) == 0:
-                continue
-
-            pxl_coords = self.element_coordinates(elem)
-            pt = self.pt_from_pixel(pxl_coords)
-
-            # draw the bbox border
-            if showBoundingboxes:
-                pdf.rect(
-                    pt.x1, self.height - pt.y2, pt.x2 - pt.x1, pt.y2 - pt.y1,
-                    fill=0)
-
-            text = pdf.beginText()
-            fontsize = pt.y2 - pt.y1
-            text.setFont(fontname, fontsize)
-            if invisibleText:
-                text.setTextRenderMode(3)  # Invisible (indicates OCR text)
-
-            # set cursor to bottom left corner of bbox (adjust for dpi)
-            text.setTextOrigin(pt.x1, self.height - pt.y2)
-
-            # scale the width of the text to fill the width of the bbox
-            text.setHorizScale(
-                100 * (pt.x2 - pt.x1) / pdf.stringWidth(
-                    elemtxt, fontname, fontsize))
-
-            # write the text to the page
-            text.textLine(elemtxt)
-            pdf.drawText(text)
-
-        # put the image on the page, scaled to fill the page
-        if imageFileName is not None:
-            pdf.drawImage(imageFileName, 0, 0,
-                          width=self.width, height=self.height)
-
-        # finish up the page and save it
-        pdf.showPage()
-        pdf.save()
-
-
-if __name__ == "__main__":
-    parser = argparse.ArgumentParser(description='Convert hocr file to PDF')
-    parser.add_argument('-b', '--boundingboxes', action="store_true",
-                        default=False, help='Show bounding boxes borders')
-    parser.add_argument('-r', '--resolution', type=int,
-                        default=300,
-                        help='Resolution of the image that was OCRed')
-    parser.add_argument('-i', '--image', default=None,
-                        help='Path to the image to be placed above the text')
-    parser.add_argument('hocrfile', help='Path to the hocr file to be parsed')
-    parser.add_argument(
-        'outputfile', help='Path to the PDF file to be generated')
-    args = parser.parse_args()
-
-    hocr = HocrTransform(args.hocrfile, args.resolution)
-    hocr.to_pdf(args.outputfile, args.image, args.boundingboxes)

ocrmypdf/lib/__init__.py

@@ -1 +0,0 @@
-"""Bindings to external libraries"""

ocrmypdf/pageinfo.py

@@ -1,534 +0,0 @@
-#!/usr/bin/env python3
-# © 2015 James R. Barlow: github.com/jbarlow83
-
-from subprocess import Popen, PIPE
-from decimal import Decimal
-from math import hypot, isclose
-import re
-import sys
-import PyPDF2 as pypdf
-from collections import namedtuple
-
-
-matrix_mult = pypdf.pdf.utils.matrixMultiply
-
-FRIENDLY_COLORSPACE = {
-    '/DeviceGray': 'gray',
-    '/CalGray': 'gray',
-    '/DeviceRGB': 'rgb',
-    '/CalRGB': 'rgb',
-    '/DeviceCMYK': 'cmyk',
-    '/Lab': 'lab',
-    '/ICCBased': 'icc',
-    '/Indexed': 'index',
-    '/Separation': 'sep',
-    '/DeviceN': 'devn',
-    '/Pattern': '-',
-    '/G': 'gray',  # Abbreviations permitted in inline images
-    '/RGB': 'rgb',
-    '/CMYK': 'cmyk',
-    '/I': 'index',
-}
-
-FRIENDLY_ENCODING = {
-    '/CCITTFaxDecode': 'ccitt',
-    '/DCTDecode': 'jpeg',
-    '/JPXDecode': 'jpx',
-    '/JBIG2Decode': 'jbig2',
-    '/CCF': 'ccitt',  # Abbreviations permitted in inline images
-    '/DCT': 'jpeg',
-    '/AHx': 'asciihex',
-    '/A85': 'ascii85',
-    '/LZW': 'lzw',
-    '/Fl': 'flate',
-    '/RL': 'runlength'
-}
-
-FRIENDLY_COMP = {
-    'gray': 1,
-    'rgb': 3,
-    'cmyk': 4,
-    'lab': 3,
-    'index': 1
-}
-
-
-UNIT_SQUARE = (1.0, 0.0, 0.0, 1.0, 0.0, 0.0)
-
-
-def _matrix_from_shorthand(shorthand):
-    """Convert from PDF matrix shorthand to full matrix
-
-    PDF 1.7 spec defines a shorthand for describing the entries of a matrix
-    since the last column is always (0, 0, 1).
-    """
-
-    a, b, c, d, e, f = map(float, shorthand)
-    return ((a, b, 0),
-            (c, d, 0),
-            (e, f, 1))
-
-
-def _shorthand_from_matrix(matrix):
-    """Convert from transformation matrix to PDF shorthand."""
-    a, b = matrix[0][0], matrix[0][1]
-    c, d = matrix[1][0], matrix[1][1]
-    e, f = matrix[2][0], matrix[2][1]
-    return tuple(map(float, (a, b, c, d, e, f)))
-
-
-def _is_unit_square(shorthand):
-    values = map(float, shorthand)
-    pairwise = zip(values, UNIT_SQUARE)
-    return all([isclose(a, b, rel_tol=1e-3) for a, b in pairwise])
-
-XobjectSettings = namedtuple('XobjectSettings',
-    ['name', 'shorthand', 'stack_depth'])
-
-InlineSettings = namedtuple('InlineSettings',
-    ['settings', 'shorthand', 'stack_depth'])
-
-ContentsInfo = namedtuple('ContentsInfo', ['xobject_settings', 'inline_images'])
-
-
-def _normalize_stack(operations):
-    """Fix runs of qQ's in the stack
-
-    For some reason PyPDF2 converts runs of qqq, QQ, QQQq, etc. into single
-    operations.  Break this silliness up and issue each stack operation
-    individually so we don't lose count.
-
-    """
-    for operands, command in operations:
-        if re.match(br'Q*q+$', command):   # Zero or more Q, one or more q
-            for char in command:           # Split into individual bytes
-                yield ([], bytes([char]))  # Yield individual bytes
-        else:
-            yield (operands, command)
-
-
-def _interpret_contents(contentstream, initial_shorthand=UNIT_SQUARE):
-    """Interpret the PDF content stream
-
-    The stack represents the state of the PDF graphics stack.  We are only
-    interested in the current transformation matrix (CTM) so we only track
-    this object; a full implementation would need to track many other items.
-
-    The CTM is initialized to the mapping from user space to device space.
-    PDF units are 1/72".  In a PDF viewer or printer this matrix is initialized
-    to the transformation to device space.  For example if set to
-    (1/72, 0, 0, 1/72, 0, 0) then all units would be calculated in inches.
-
-    Images are always considered to be (0, 0) -> (1, 1).  Before drawing an
-    image there should be a 'cm' that sets up an image coordinate system
-    where drawing from (0, 0) -> (1, 1) will draw on the desired area of the
-    page.
-
-    PDF units suit our needs so we initialize ctm to the identity matrix.
-
-    PyPDF2 replaces inline images with a fake "INLINE IMAGE" operator.
-
-    """
-
-    operations = contentstream.operations
-    stack = []
-    ctm = _matrix_from_shorthand(initial_shorthand)
-    xobject_settings = []
-    inline_images = []
-
-    for n, op in enumerate(_normalize_stack(operations)):
-        operands, command = op
-        if command == b'q':
-            stack.append(ctm)
-            if len(stack) > 32:
-                raise RuntimeError(
-                    "PDF graphics stack overflow, command %i" % n)
-        elif command == b'Q':
-            try:
-                ctm = stack.pop()
-            except IndexError:
-                raise RuntimeError(
-                    "PDF graphics stack underflow, command %i" % n)
-        elif command == b'cm':
-            ctm = matrix_mult(
-                _matrix_from_shorthand(operands), ctm)
-        elif command == b'Do':
-            image_name = operands[0]
-            settings = XobjectSettings(
-                name=image_name, shorthand=_shorthand_from_matrix(ctm),
-                stack_depth=len(stack))
-            xobject_settings.append(settings)
-        elif command == b'INLINE IMAGE':
-            settings = operands['settings']
-            inline = InlineSettings(
-                settings=settings, shorthand=_shorthand_from_matrix(ctm),
-                stack_depth=len(stack))
-            inline_images.append(inline)
-
-    return ContentsInfo(
-        xobject_settings=xobject_settings,
-        inline_images=inline_images)
-
-
-def _get_dpi(ctm_shorthand, image_size):
-    """Given the transformation matrix and image size, find the image DPI.
-
-    PDFs do not include image resolution information within image data.
-    Instead, the PDF page content stream describes the location where the
-    image will be rasterized, and the effective resolution is the ratio of the
-    pixel size to raster target size.
-
-    Normally a scanned PDF has the paper size set appropriately but this is
-    not guaranteed. The most common case is a cropped image will change the
-    page size (/CropBox) without altering the page content stream. That means
-    it is not sufficient to assume that the image fills the page, even though
-    that is the most common case.
-
-    A PDF image may be scaled (always), cropped, translated, rotated in place
-    to an arbitrary angle (rarely) and skewed. Only equal area mappings can
-    be expressed, that is, it is not necessary to consider distortions where
-    the effective DPI varies with position.
-
-    To determine the image scale, transform an offset axis vector v0 (0, 0),
-    width-axis vector v0 (1, 0), height-axis vector vh (0, 1) with the matrix,
-    which gives the dimensions of the image in PDF units. From there we can
-    compare to actual image dimensions. PDF uses
-    row vector * matrix_tranposed unlike the traditional
-    matrix * column vector.
-
-    The offset, width and height vectors can be combined in a matrix and
-    multiplied by the transform matrix. Then we want to calculated
-        magnitude(width_vector - offset_vector)
-    and
-        magnitude(height_vector - offset_vector)
-
-    When the above is worked out algebraically, the effect of translation
-    cancels out, and the vector magnitudes become functions of the nonzero
-    transformation matrix indices. The results of the derivation are used
-    in this code.
-
-    pdfimages -list does calculate the DPI in some way that is not completely
-    naive, but it does not get the DPI of rotated images right, so cannot be
-    used anymore to validate this. Photoshop works, or using Acrobat to
-    rotate the image back to normal.
-
-    It does not matter if the image is partially cropped, or even out of the
-    /MediaBox.
-
-    """
-
-    a, b, c, d, _, _ = ctm_shorthand
-
-    # Calculate the width and height of the image in PDF units
-    image_drawn_width = hypot(a, b)
-    image_drawn_height = hypot(c, d)
-
-    # The scale of the image is pixels per PDF unit (1/72")
-    scale_w = image_size[0] / image_drawn_width
-    scale_h = image_size[1] / image_drawn_height
-
-    # DPI = scale * 72
-    dpi_w = scale_w * 72.0
-    dpi_h = scale_h * 72.0
-
-    return (dpi_w, dpi_h)
-
-
-def _find_inline_images(contentsinfo):
-    "Find inline images in the contentstream"
-
-    for n, inline in enumerate(contentsinfo.inline_images):
-        image = {}
-        image['name'] = str('inline-%02d' % n)
-        image['width'] = inline.settings['/W']
-        image['height'] = inline.settings['/H']
-        if '/BPC' in inline.settings:
-            image['bpc'] = inline.settings['/BPC']
-        else:
-            image['bpc'] = 8
-        if '/CS' in inline.settings:
-            image['color'] = FRIENDLY_COLORSPACE.get(inline.settings['/CS'], '-')
-        else:
-            image['color'] = '-'
-        image['comp'] = FRIENDLY_COMP.get(image['color'], '?')
-        if '/F' in inline.settings:
-            filter_ = inline.settings['/F']
-            if isinstance(filter_, pypdf.generic.ArrayObject):
-                filter_ = filter_[0]
-            image['enc'] = FRIENDLY_ENCODING.get(filter_, 'image')
-        else:
-            image['enc'] = 'image'
-
-        dpi_w, dpi_h = _get_dpi(
-            inline.shorthand, (image['width'], image['height']))
-        image['dpi_w'], image['dpi_h'] = Decimal(dpi_w), Decimal(dpi_h)
-        yield image
-
-
-def _image_xobjects(container):
-    """Search for all XObject-based images in the container
-
-    Usually the container is a page, but it could also be a Form XObject
-    that contains images. Filter out the Form XObjects which are dealt with
-    elsewhere.
-
-    Generate a sequence of tuples (image, xobj container), where container,
-    where xobj is the name of the object and image is the object itself,
-    since the object does not know its own name.
-
-    """
-
-    if '/Resources' not in container:
-        return
-    resources = container['/Resources']
-    if '/XObject' not in resources:
-        return
-    for xobj in resources['/XObject']:
-        candidate = resources['/XObject'][xobj]
-        if candidate['/Subtype'] == '/Image':
-            image = candidate
-            yield (image, xobj)
-
-
-def _find_regular_images(container, contentsinfo):
-    """Find images stored in the container's /Resources /XObject
-
-    Usually the container is a page, but it could also be a Form XObject
-    that contains images.
-
-    Generates images with their DPI at time of drawing.
-
-    """
-
-    for pdfimage, xobj in _image_xobjects(container):
-        image = {}
-        image['name'] = xobj
-        image['width'] = pdfimage['/Width']
-        image['height'] = pdfimage['/Height']
-        if '/BitsPerComponent' in pdfimage:
-            image['bpc'] = pdfimage['/BitsPerComponent']
-        else:
-            image['bpc'] = 8
-
-        # Fixme: this is incorrectly treats explicit masks as stencil masks,
-        # but good enough for now. Explicit masks have /ImageMask true but are
-        # never called for in content stream, instead are drawn as a /Mask on
-        # other images. For our purposes finding out the details of /Mask
-        # will seldom matter.
-        if '/ImageMask' in pdfimage:
-            image['type'] = 'stencil' if pdfimage['/ImageMask'].value \
-                            else 'image'
-        else:
-            image['type'] = 'image'
-        if '/Filter' in pdfimage:
-            filter_ = pdfimage['/Filter']
-            if isinstance(filter_, pypdf.generic.ArrayObject):
-                filter_ = filter_[0]
-            image['enc'] = FRIENDLY_ENCODING.get(filter_, 'image')
-        else:
-            image['enc'] = 'image'
-        if '/ColorSpace' in pdfimage:
-            cs = pdfimage['/ColorSpace']
-            if isinstance(cs, pypdf.generic.ArrayObject):
-                cs = cs[0]
-            image['color'] = FRIENDLY_COLORSPACE.get(cs, '-')
-        else:
-            image['color'] = 'jpx' if image['enc'] == 'jpx' else '?'
-
-        image['comp'] = FRIENDLY_COMP.get(image['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 image['comp'] == '?' and image['enc'] in ('ccitt', 'jbig2'):
-            image['comp'] = FRIENDLY_COMP['gray']
-
-        image['dpi_w'] = image['dpi_h'] = 0
-
-        for xobj in contentsinfo.xobject_settings:
-            # Loop in case the same image is display multiple times on a page
-            if xobj.name != image['name']:
-                continue
-
-            if xobj.stack_depth == 0 and _is_unit_square(xobj.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
-                # gets drawn into a square of 1x1 PDF units (or 1/72",
-                # or 0.35 mm).  The equivalent DPI will be >100,000.  Exclude
-                # these from our DPI calculation for the page.
-                continue
-
-            dpi_w, dpi_h = _get_dpi(
-                xobj.shorthand, (image['width'], image['height']))
-
-            # When image is used multiple times take the highest DPI it is
-            # rendered at
-            image['dpi_w'] = max(dpi_w, image.get('dpi_w', 0))
-            image['dpi_h'] = max(dpi_h, image.get('dpi_h', 0))
-
-        DPI_PREC = Decimal('1.000')
-        dpi = Decimal(image['dpi_w'] * image['dpi_h']).sqrt()
-        image['dpi_w'] = Decimal(image['dpi_w']).quantize(DPI_PREC)
-        image['dpi_h'] = Decimal(image['dpi_h']).quantize(DPI_PREC)
-        image['dpi'] = dpi.quantize(DPI_PREC)
-        yield image
-
-
-def _find_form_xobject_images(pdf, container, contentsinfo):
-    """Find any images that are in Form XObjects in the container
-
-    The container may be a page, or a parent Form XObject.
-
-    """
-    if '/Resources' not in container:
-        return
-    resources = container['/Resources']
-    if '/XObject' not in resources:
-        return
-    for xobj in resources['/XObject']:
-        candidate = resources['/XObject'][xobj]
-        if candidate['/Subtype'] != '/Form':
-            continue
-
-        form_xobject = candidate
-        for settings in contentsinfo.xobject_settings:
-            if settings.name != xobj:
-                continue
-
-            # Find images once for each time this Form XObject is drawn.
-            # This could be optimized to cache the multiple drawing events
-            # but in practice both Form XObjects and multiple drawing of the
-            # same object are both very rare.
-            ctm_shorthand = settings.shorthand
-            yield from _find_images(pdf, form_xobject, ctm_shorthand)
-
-
-def _find_images(pdf, container, shorthand=None):
-    """Find all individual instances of images drawn in the container
-
-    Usually the container is a page, but it may also be a Form XObject.
-
-    On a typical page images are stored inline or as regular images
-    in an XObject.
-
-    Form XObjects may include inline images, XObject images,
-    and recursively, other Form XObjects; and also vector drawing commands.
-
-    Every instance of an image being drawn somewhere is flattened and
-    treated as a unique image, since if the same image is drawn multiple times
-    on one page it may be drawn at differing resolutions, and our objective
-    is to find the resolution at which the page can be rastered without
-    downsampling.
-
-    """
-
-    if container.get('/Type') == '/Page' and '/Contents' in container:
-        # For a /Page the content stream is attached to the page's /Contents
-        page = container
-        contentstream = pypdf.pdf.ContentStream(page.getContents(), pdf)
-        initial_shorthand = shorthand or UNIT_SQUARE
-    elif container.get('/Type') == '/XObject' and \
-            container['/Subtype'] == '/Form':
-        # For a Form XObject that content stream is attached to the XObject
-        contentstream = pypdf.pdf.ContentStream(container, pdf)
-
-        # Set the CTM to the state it was when the "Do" operator was
-        # encountered that is drawing this instance of the Form XObject
-        ctm = _matrix_from_shorthand(shorthand or UNIT_SQUARE)
-
-        # A Form XObject may provide its own matrix to map form space into
-        # user space. Get this if one exists
-        form_matrix = _matrix_from_shorthand(
-                container.get('/Matrix', UNIT_SQUARE))
-
-        # Concatenate form matrix with CTM to ensure CTM is correct for
-        # drawing this instance of the XObject
-        ctm = matrix_mult(form_matrix, ctm)
-        initial_shorthand = _shorthand_from_matrix(ctm)
-    else:
-        return
-
-    contentsinfo = _interpret_contents(contentstream, initial_shorthand)
-
-    yield from _find_inline_images(contentsinfo)
-    yield from _find_regular_images(container, contentsinfo)
-    yield from _find_form_xobject_images(pdf, container, contentsinfo)
-
-
-def _page_has_text(pdf, page):
-    if not '/Contents' in page:
-        return False
-
-    # Simple test
-    text = page.extractText()
-    if text.strip() != '':
-        return True
-
-    # More nuanced test to deal with quirks of Tesseract PDF generation
-    # Check if there's a Glyphless font
-    try:
-        font = page['/Resources']['/Font']
-    except KeyError:
-        pass
-    else:
-        font_objects = list(font.keys())
-        for font_object in font_objects:
-            basefont = font[font_object]['/BaseFont']
-            if basefont.endswith('GlyphLessFont'):
-                return True
-
-    return False
-
-
-def _pdf_get_pageinfo(infile, pageno: int):
-    pageinfo = {}
-    pageinfo['pageno'] = pageno
-    pageinfo['images'] = []
-
-    pdf = pypdf.PdfFileReader(infile)
-    page = pdf.pages[pageno]
-
-    pageinfo['has_text'] = _page_has_text(pdf, page)
-
-    width_pt = page.mediaBox.getWidth()
-    height_pt = page.mediaBox.getHeight()
-    pageinfo['width_inches'] = width_pt / Decimal(72.0)
-    pageinfo['height_inches'] = height_pt / Decimal(72.0)
-
-    try:
-        pageinfo['rotate'] = int(page['/Rotate'])
-    except KeyError:
-        pageinfo['rotate'] = 0
-
-    pageinfo['images'] = [im for im in
-                          _find_images(pdf, page)]
-    if pageinfo['images']:
-        xres = max(image['dpi_w'] for image in pageinfo['images'])
-        yres = max(image['dpi_h'] for image in pageinfo['images'])
-        pageinfo['xres'], pageinfo['yres'] = xres, yres
-        pageinfo['width_pixels'] = \
-            int(round(xres * pageinfo['width_inches']))
-        pageinfo['height_pixels'] = \
-            int(round(yres * pageinfo['height_inches']))
-
-    return pageinfo
-
-
-def pdf_get_all_pageinfo(infile):
-    pdf = pypdf.PdfFileReader(infile)
-    return [_pdf_get_pageinfo(infile, n) for n in range(pdf.numPages)]
-
-
-def main():
-    import argparse
-    parser = argparse.ArgumentParser()
-    parser.add_argument('infile')
-    args = parser.parse_args()
-    info = pdf_get_all_pageinfo(args.infile)
-    from pprint import pprint
-    pprint(info)
-
-
-if __name__ == '__main__':
-    main()

requirements.txt

@@ -2,8 +2,9 @@
 # setup.py lists a separate set of requirements that are looser to simplify
 # installation
 ruffus == 2.6.3
-Pillow == 4.1.1
+Pillow == 5.0.0
 reportlab == 3.4.0
 PyPDF2 == 1.26.0
-img2pdf == 0.2.3
-cffi == 1.10.0
+img2pdf == 0.2.4
+cffi == 1.11.5
+PyMuPDF == 1.12.4

setup.cfg

@@ -11,4 +11,7 @@ ignore =
 [tool:pytest]
 norecursedirs = lib .pc .git output cache resources
 testpaths = tests
-addopts = -n auto
+addopts = -n auto
+
+[metadata]
+license_file = LICENSE

setup.py

@@ -1,6 +1,21 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 # © 2015 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/>.
 
 from __future__ import print_function, unicode_literals
 
@@ -185,14 +200,20 @@ if 'upload' in sys.argv[1:]:
 
 tests_require = open('test_requirements.txt').read().splitlines()
 
+
+def readme():
+    with open('README.rst') as f:
+        return f.read()
+
 setup(
     name='ocrmypdf',
     description='OCRmyPDF adds an OCR text layer to scanned PDF files, allowing them to be searched',
+    long_description=readme(),
     url='https://github.com/jbarlow83/OCRmyPDF',
     author='James R. Barlow',
     author_email='jim@purplerock.ca',
-    license='MIT',
-    packages=find_packages(exclude=["tests", "tests.*"]),
+    packages=find_packages('src', exclude=["tests", "tests.*"]),
+    package_dir={'': 'src'},
     keywords=['PDF', 'OCR', 'optical character recognition', 'PDF/A', 'scanning'],
     classifiers=[
         "Programming Language :: Python :: 3.5",
@@ -202,7 +223,7 @@ setup(
         "Intended Audience :: End Users/Desktop",
         "Intended Audience :: Science/Research",
         "Intended Audience :: System Administrators",
-        "License :: OSI Approved :: MIT License",
+        "License :: OSI Approved :: GNU General Public License v3 (GPLv3)",
         "Operating System :: MacOS :: MacOS X",
         "Operating System :: POSIX",
         "Operating System :: POSIX :: BSD",
@@ -211,22 +232,28 @@ setup(
         "Topic :: Text Processing :: Indexing",
         "Topic :: Text Processing :: Linguistic",
         ],
+    python_requires=' >= 3.5',
     setup_requires=[
+        'cffi >= 1.9.1',        # to build the leptonica module
+        'pytest-runner',        # to enable python setup.py test
         'setuptools_scm',       # so that version will work
-        'cffi>=1.9.1'           # to build the leptonica module
     ],
     use_scm_version={'version_scheme': 'post-release'},
     cffi_modules=[
-        'ocrmypdf/lib/compile_leptonica.py:ffibuilder'
+        'src/ocrmypdf/lib/compile_leptonica.py:ffibuilder'
     ],
     install_requires=[
-        'ruffus==2.6.3',        # pinned - ocrmypdf implements a 2.6.3 workaround
-        'Pillow>=4.0.0',        # Pillow < 4 has BytesIO/TIFF bug w/img2pdf 0.2.3
-        'reportlab>=3.3.0',     # oldest released version with sane image handling
-        'PyPDF2>=1.26',         # pure Python, so track HEAD closely
-        'img2pdf>=0.2.3',       # pure Python, so track HEAD closely
-        'cffi>=1.9.1'           # must be a setup and install requirement
+        'cffi >= 1.9.1',          # must be a setup and install requirement
+        'defusedxml >= 0.5.0',    # pure Python, so track HEAD closely
+        'img2pdf >= 0.2.4',       # pure Python, so track HEAD closely
+        'Pillow >= 4.0.0',        # Pillow < 4 has BytesIO/TIFF bug w/img2pdf 0.2.3
+        'PyPDF2 >= 1.26',         # pure Python, so track HEAD closely
+        'reportlab >= 3.3.0',     # oldest released version with sane image handling
+        'ruffus == 2.6.3',        # pinned - ocrmypdf implements a 2.6.3 workaround
     ],
+    extras_require={
+        'fitz': ['PyMuPDF >= 1.12.5']     # for table of contents bug
+    },
     tests_require=tests_require,
     entry_points={
         'console_scripts': [

src/ocrmypdf/__init__.py

@@ -0,0 +1,25 @@
+# © 2017 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 pkg_resources
+
+PROGRAM_NAME = 'ocrmypdf'
+
+# Official PEP 396
+__version__ = pkg_resources.get_distribution('ocrmypdf').version
+
+VERSION = __version__

src/ocrmypdf/__main__.py

@@ -1,37 +1,54 @@
 #!/usr/bin/env python3
-# © 2015-16 James R. Barlow: github.com/jbarlow83
+# © 2015-17 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/>.
 
-from contextlib import suppress
 from tempfile import mkdtemp
 from collections.abc import Sequence
+from pathlib import Path
 import sys
 import os
 import re
-import shutil
 import warnings
 import multiprocessing
 import atexit
 import textwrap
-import img2pdf
 import logging
 import argparse
 
 import PyPDF2 as pypdf
-from PIL import Image
+import PIL
 
 import ruffus.ruffus_exceptions as ruffus_exceptions
 import ruffus.cmdline as cmdline
 import ruffus.proxy_logger as proxy_logger
 
-from .pipeline import JobContext, JobContextManager, re_symlink, \
+from .pipeline import JobContext, JobContextManager, \
     cleanup_working_files, build_pipeline
 from .pdfa import file_claims_pdfa
-from .helpers import is_iterable_notstr, re_symlink, is_file_writable
-from .exec import tesseract, qpdf
+from .helpers import is_iterable_notstr, re_symlink, is_file_writable, \
+    available_cpu_count
+from .exec import tesseract, qpdf, ghostscript
+from .lib import fitz
 from . import PROGRAM_NAME, VERSION
 
-from .exceptions import *
+from .exceptions import ExitCode, ExitCodeException, MissingDependencyError, \
+    InputFileError, BadArgsError, OutputFileAccessError
 from . import exceptions as ocrmypdf_exceptions
+from ._unicodefun import verify_python3_env
 
 warnings.simplefilter('ignore', pypdf.utils.PdfReadWarning)
 
@@ -41,6 +58,9 @@ warnings.simplefilter('ignore', pypdf.utils.PdfReadWarning)
 
 MINIMUM_TESS_VERSION = '3.04'
 
+HOCR_OK_LANGS = frozenset([
+    'eng', 'deu', 'spa', 'ita', 'por'
+])
 
 def complain(message):
     print(*textwrap.wrap(message), file=sys.stderr)
@@ -50,6 +70,11 @@ def complain(message):
 if 'IDE_PROJECT_ROOTS' in os.environ:
     os.environ['PATH'] = '/usr/local/bin:' + os.environ['PATH']
 
+# -------- 
+# Critical environment tests
+
+verify_python3_env()
+
 if tesseract.version() < MINIMUM_TESS_VERSION:
     complain(
         "Please install tesseract {0} or newer "
@@ -131,12 +156,15 @@ parser.add_argument(
     '--image-dpi', metavar='DPI', type=int,
     help="For input image instead of PDF, use this DPI instead of file's.")
 parser.add_argument(
-    '--output-type', choices=['pdfa', 'pdf'], default='pdfa',
+    '--output-type', choices=['pdfa', 'pdf', 'pdfa-1', 'pdfa-2'], 
+    default='pdfa',
     help="Choose output type. 'pdfa' creates a PDF/A-2b compliant file for "
          "long term archiving (default, recommended) but may not suitable "
          "for users who want their file altered as little as possible. 'pdfa' "
          "also has problems with full Unicode text. 'pdf' attempts to "
-         "preserve file contents as much as possible.")
+         "preserve file contents as much as possible. 'pdf-a1' creates a "
+         "PDF/A1-b file. 'pdf-a2' is equivalent to 'pdfa'."
+         )
 
 # Use null string '\0' as sentinel to indicate the user supplied no argument,
 # since that is the only invalid character for filepaths on all platforms
@@ -233,6 +261,11 @@ ocrsettings.add_argument(
 advanced = parser.add_argument_group(
     "Advanced",
     "Advanced options to control Tesseract's OCR behavior")
+advanced.add_argument(
+    '--max-image-mpixels', action='store', type=float, metavar='MPixels',
+    help="Set maximum number of pixels to unpack before treating an image as a "
+         "decompression bomb",
+    default=128.0)
 advanced.add_argument(
     '--tesseract-config', action='append', metavar='CFG', default=[],
     help="Additional Tesseract configuration files -- see documentation")
@@ -250,15 +283,17 @@ advanced.add_argument(
          "3 - default.")
     )
 advanced.add_argument(
-    '--pdf-renderer', choices=['auto', 'tesseract', 'hocr', 'tess4'], default='auto',
+    '--pdf-renderer',
+    choices=['auto', 'tesseract', 'hocr', 'sandwich'], default='auto',
     help="Choose OCR PDF renderer - the default option is to let OCRmyPDF "
-         "choose.  The 'tesseract' PDF renderer is more accurate and does a "
-         "better job and document structure such as recognizing columns. It "
-         "also does a better job on non-Latin languages. However, it does "
-         "not work as well when older versions of Tesseract or Ghostscript "
-         "are installed, and some combinations of arguments to do not work "
-         "with --pdf-renderer tesseract.  The 'tess4' PDF renderer is similar "
-         "to 'tesseract', requires tesseract 4, and gives superior results.")
+         "choose."
+         "auto - let OCRmyPDF choose; "
+         "sandwich - default renderer for Tesseract 3.05.01 and newer; "
+         "hocr - default renderer for older versions of Tesseract; "
+         "tesseract - gives better results for non-Latin languages and "
+         "Tesseract older than 3.05.01 but has problems with some versions "
+         " of Ghostscript; deprecated"
+    )
 advanced.add_argument(
     '--tesseract-timeout', default=180.0, type=float, metavar='SECONDS',
     help='Give up on OCR after the timeout, but copy the preprocessed page '
@@ -278,6 +313,23 @@ advanced.add_argument(
          "are applied to all pages, including those for which OCR was "
          "skipped.  Not supported for --output-type=pdf ; that setting "
          "preserves the original compression of all images.")
+advanced.add_argument(
+    '--user-words', metavar='FILE',
+    help="Specify the location of the Tesseract user words file. This is a "
+         "list of words Tesseract should consider while performing OCR in "
+         "addition to its standard language dictionaries. This can improve "
+         "OCR quality especially for specialized and technical documents.")
+advanced.add_argument(
+    '--user-patterns', metavar='FILE',
+    help="Specify the location of the Tesseract user patterns file.")
+advanced.add_argument(
+    '--skip-repair', action='store_true',
+    help="Normally OCRmyPDF automatically repairs PDFs using qpdf before "
+         "processing.  If you have already run qpdf or a similar program "
+         "that repairs PDF errors, you can tell OCRmyPDF to skip repair with "
+         "this option.  This may be helpful in batch processing where all "
+         "files are repaired prior to OCR occurs, since repair is single "
+         "threaded and time consuming for large files.")
 
 debugging = parser.add_argument_group(
     "Debugging",
@@ -301,39 +353,49 @@ def check_options_languages(options, _log):
     if '+' in options.language[0]:
         options.language = options.language[0].split('+')
 
-    if not set(options.language).issubset(tesseract.languages()):
+    languages = set(options.language)
+    if not languages.issubset(tesseract.languages()):
         msg = (
             "The installed version of tesseract does not have language "
             "data for the following requested languages: \n")
-        for lang in (set(options.language) - tesseract.languages()):
+        for lang in (languages - tesseract.languages()):
             msg += lang + '\n'
-        raise argparse.ArgumentError(None, msg)
+        raise MissingDependencyError(msg)
 
 
 def check_options_output(options, log):
     if options.pdf_renderer == 'auto':
-        options.pdf_renderer = 'hocr'
+        if tesseract.has_textonly_pdf():
+            options.pdf_renderer = 'sandwich'
+        else:
+            options.pdf_renderer = 'hocr'
 
-    if options.pdf_renderer in ('tesseract', 'tess4'):
-        if tesseract.version() < '3.05':
-            log.warning(
-                "The setting --pdf-renderer=tesseract is not recommend for "
-                " use with tesseract versions less than 3.05, because it "
-                " produces OCR text that is incompatible with Ghostscript and "
-                " some other software.")
-        elif tesseract.version() == '4.00.00alpha':
-            log.warning(
-                "The setting --pdf-renderer={tesseract,tess4} is not"
-                " recommended for builds of tesseract 4.00.00alpha older than"
-                " February 2017. Make sure you are using a recent build.")
+    if options.pdf_renderer == 'sandwich' and not tesseract.has_textonly_pdf():
+        raise MissingDependencyError(
+            "The 'sandwich' renderer requires Tesseract 3.05.01 or newer; "
+            "or Tesseract 4.00 alpha newer than February 2017.")
 
-    if options.debug_rendering and options.pdf_renderer == 'tesseract':
+    if options.pdf_renderer == 'tesseract':
+        if tesseract.version() < '3.05' and \
+                options.output_type.startswith('pdfa'):
+            log.warning(
+                "For best results use --pdf-renderer=tesseract "
+                "--output-type=pdf to disable PDF/A generation via "
+                "Ghostscript, which is known to corrupt the OCR text of "
+                "some PDFs produced your version of Tesseract.")
+        elif tesseract.has_textonly_pdf():
+            log.warning(
+                "The argument --pdf-renderer=tesseract provides support for "
+                "versions of tesseract older than your version. For best "
+                "results omit this argument and let OCRmyPDF choose the "
+                "best available renderer.")
+
+    if options.debug_rendering and options.pdf_renderer != 'hocr':
         log.info(
-            "Ignoring --debug-rendering because it is not supported with"
-            "--pdf-renderer=tesseract.")
+            "Ignoring --debug-rendering because it requires --pdf-renderer=hocr")
 
     lossless_reconstruction = False
-    if options.pdf_renderer in ('hocr', 'tess4'):
+    if options.pdf_renderer in ('hocr', 'sandwich'):
         if not any((options.deskew, options.clean_final, options.force_ocr,
                    options.remove_background)):
             lossless_reconstruction = True
@@ -379,39 +441,53 @@ def check_options_ocr_behavior(options, log):
     # if options.redo_ocr and (options.skip_text or options.force_ocr):
     #     raise argparse.ArgumentError(
     #         "Error: --redo-ocr and other OCR options are incompatible.")
+    languages = set(options.language)
+    if options.pdf_renderer == 'hocr' and \
+            not languages.issubset(HOCR_OK_LANGS):
+        msg = (
+            "The 'hocr' PDF renderer is known to cause problems with one "
+            "or more of the languages in your document. ")
 
-    if set(options.language) & {'chi_sim', 'chi_tra'} and \
-            (options.pdf_renderer == 'hocr' or options.output_type == 'pdfa'):
-        log.warning(
-            "Your settings are known to cause problems with OCR of Chinese text. "
-            "Try adding these arguments: "
-            "    ocrmypdf --pdf-renderer tesseract --output-type pdf")
+        if tesseract.has_textonly_pdf():
+            msg += (
+                "Use --pdf-renderer auto (the default) to avoid this issue.")
+        else:
+            msg += (
+                "Use --pdf-renderer tesseract --output-type pdf to avoid "
+                "this issue")
+        log.warning(msg)
+    elif ghostscript.version() < '9.20' and \
+            not languages.issubset(HOCR_OK_LANGS) \
+            and options.output_type != 'pdf':
+        msg = (
+            "The installed version of Ghostscript does not work correctly "
+            "with the OCR languages you specified. Use --output-type pdf or "
+            "upgrade to Ghostscript 9.20 or later to avoid this issue.")
+        msg += "Found Ghostscript {}".format(ghostscript.version())
+        log.warning(msg)
 
 
 def check_options_advanced(options, log):
-    if tesseract.v4():
-        log.info(
-            "Tesseract v4.x.alpha found. OCRmyPDF support is experimental.")
     if options.tesseract_oem and not tesseract.v4():
         log.warning(
             "--tesseract-oem requires Tesseract 4.x -- argument ignored")
-    if options.pdf_renderer == 'tess4' and not tesseract.has_textonly_pdf():
+    if options.pdf_renderer == 'sandwich' and not tesseract.has_textonly_pdf():
         raise MissingDependencyError(
-            "--pdf-renderer tess4 requires Tesseract 4.x "
+            "--pdf-renderer sandwich requires Tesseract 4.x "
             "commit 3d9fb3b or later")
     if options.pdfa_image_compression != 'auto' and \
-            options.output_type != 'pdfa':
+            options.output_type.startswith('pdfa'):
         log.warning(
             "--pdfa-image-compression argument has no effect when "
-            "--output-type is not 'pdfa'"
+            "--output-type is not 'pdfa', 'pdfa-1', or 'pdfa-2'"
         )
 
 
 def check_options_metadata(options, log):
     import unicodedata
-    metadata = [options.title, options.author, options.keywords,
-                options.subject]
-    for s in (m for m in metadata if m):
+    docinfo = [options.title, options.author, options.keywords,
+               options.subject]
+    for s in (m for m in docinfo if m):
         for c in s:
             if unicodedata.category(c) == 'Co' or ord(c) >= 0x10000:
                 raise ValueError(
@@ -421,6 +497,12 @@ def check_options_metadata(options, log):
                 ))
 
 
+def check_options_pillow(options, log):
+    PIL.Image.MAX_IMAGE_PIXELS = int(options.max_image_mpixels * 1000000)
+    if PIL.Image.MAX_IMAGE_PIXELS == 0:
+        PIL.Image.MAX_IMAGE_PIXELS = None
+
+
 def check_options(options, log):
     try:
         check_options_languages(options, log)
@@ -430,6 +512,7 @@ def check_options(options, log):
         check_options_preprocessing(options, log)
         check_options_ocr_behavior(options, log)
         check_options_advanced(options, log)
+        check_options_pillow(options, log)
     except ValueError as e:
         log.error(e)
         sys.exit(ExitCode.bad_args)
@@ -465,24 +548,6 @@ def logging_factory(logger_name, logger_args):
     return root_logger
 
 
-def available_cpu_count():
-    try:
-        return multiprocessing.cpu_count()
-    except NotImplementedError:
-        pass
-
-    try:
-        import psutil
-        return psutil.cpu_count()
-    except (ImportError, AttributeError):
-        pass
-
-    complain(
-        "Could not get CPU count.  Assuming one (1) CPU."
-        "Use -j N to set manually.")
-    return 1
-
-
 def cleanup_ruffus_error_message(msg):
     msg = re.sub(r'\s+', r' ', msg)
     msg = re.sub(r"\((.+?)\)", r'\1', msg)
@@ -493,31 +558,47 @@ def cleanup_ruffus_error_message(msg):
 def do_ruffus_exception(ruffus_five_tuple, options, log):
     """Replace the elaborate ruffus stack trace with a user friendly
     description of the error message that occurred."""
+    exit_code = None
 
     task_name, job_name, exc_name, exc_value, exc_stack = ruffus_five_tuple
+    job_name = job_name  # unused
     if exc_name == 'builtins.SystemExit':
         match = re.search(r"\.(.+?)\)", exc_value)
         exit_code_name = match.groups()[0]
-        exit_code = getattr(ExitCode, exit_code_name, 'other_error')
-        return exit_code
+        exit_code = getattr(ExitCode, exit_code_name, 'other_error')        
     elif exc_name == 'ruffus.ruffus_exceptions.MissingInputFileError':
         log.error(cleanup_ruffus_error_message(exc_value))
-        return ExitCode.input_file
+        exit_code = ExitCode.input_file
     elif exc_name == 'builtins.TypeError':
         # Even though repair_pdf will fail, ruffus will still try
         # to call split_pages with no input files, likely due to a bug
         if task_name == 'split_pages':
             log.error("Input file '{0}' is not a valid PDF".format(
                 options.input_file))
-            return ExitCode.input_file
+            exit_code = ExitCode.input_file
     elif exc_name == 'builtins.KeyboardInterrupt':
         log.error("Interrupted by user")
-        return ExitCode.ctrl_c
+        exit_code = ExitCode.ctrl_c
     elif exc_name == 'subprocess.CalledProcessError':
         # It's up to the subprocess handler to report something useful
         msg = "Error occurred while running this command:"
         log.error(msg + '\n' + exc_value)
-        return ExitCode.child_process_error
+        exit_code = ExitCode.child_process_error
+    elif (exc_name == 'PyPDF2.utils.PdfReadError' and \
+            'not been decrypted' in exc_value) or \
+            (exc_name == 'ocrmypdf.exceptions.EncryptedPdfError'):
+        log.error(textwrap.dedent("""\
+            Input PDF is encrypted. The encryption must be removed to
+            perform OCR. 
+
+            For information about this PDF's security use
+                qpdf --show-encryption infilename
+
+            You can remove the encryption using
+                qpdf --decrypt [--password=[password]] infilename
+
+            """))
+        exit_code = ExitCode.encrypted_pdf        
     elif exc_name == 'ocrmypdf.exceptions.PdfMergeFailedError':
         log.error(textwrap.dedent("""\
             Failed to merge PDF image layer with OCR layer
@@ -528,24 +609,20 @@ def do_ruffus_exception(ruffus_five_tuple, options, log):
             Try using
                 ocrmypdf --pdf-renderer tesseract  [..other args..]
             """))
-        return ExitCode.input_file
+        exit_code = ExitCode.input_file
     elif exc_name.startswith('ocrmypdf.exceptions.'):
         base_exc_name = exc_name.replace('ocrmypdf.exceptions.', '')
         exc_class = getattr(ocrmypdf_exceptions, base_exc_name)
-        return exc_class.exit_code
-    elif exc_name == 'PyPDF2.utils.PdfReadError' and \
-            'not been decrypted' in exc_value:
-        log.error(textwrap.dedent("""\
-            Input PDF uses either an encryption algorithm or a PDF security
-            handler that is not supported by ocrmypdf.
+        exit_code = exc_class.exit_code
+    elif exc_name == 'PIL.Image.DecompressionBombError':
+        msg = cleanup_ruffus_error_message(exc_value)
+        msg += ("\nUse the --max-image-mpixels argument to set increase the "
+                "maximum number of megapixels to accept.")
+        log.error(msg)
+        exit_code = ExitCode.input_file
 
-            For information about this PDF's security use
-                qpdf --show-encryption [...input PDF...]
-
-            (Only algorithms "R = 1" and "R = 2" are supported.)
-
-            """))
-        return ExitCode.encrypted_pdf
+    if exit_code is not None:
+        return exit_code
 
     if not options.verbose:
         log.error(exc_stack)
@@ -555,6 +632,21 @@ def do_ruffus_exception(ruffus_five_tuple, options, log):
 def traverse_ruffus_exception(e_args, options, log):
     """Walk through a RethrownJobError and find the first exception.
 
+    Ruffus flattens exception to 5 element tuples. Because of a bug
+    in <= 2.6.3 it may present either the single:
+      (task, job, exc, value, stack)
+    or something like:
+      [[(task, job, exc, value, stack)]]
+    
+    Generally cross-process exception marshalling doesn't work well
+    and ruffus doesn't support because BaseException has its own
+    implementation of __reduce__ that attempts to reconstruct the
+    exception based on e.__init__(e.args).
+    
+    Attempting to log the exception directly marshalls it to the logger
+    which is probably in another process, so it's better to log only
+    data from the exception at this point.
+
     The exit code will be based on this, even if multiple exceptions occurred
     at the same time."""
 
@@ -586,6 +678,9 @@ def check_closed_streams(options):
 
     """
 
+    if sys.version_info[0:3] >= (3, 6, 4):
+        return True  # Issued fixed in Python 3.6.4+
+
     if sys.stderr is None:
         sys.stderr = open(os.devnull, 'w')
 
@@ -611,6 +706,103 @@ def check_closed_streams(options):
     return True
 
 
+def log_page_orientations(pdfinfo, _log):
+    direction = {0: 'n', 90: 'e',
+                180: 's', 270: 'w'}
+    orientations = []
+    for n, page in enumerate(pdfinfo):
+        angle = page.rotation or 0
+        if angle != 0:
+            orientations.append('{0}{1}'.format(
+                n + 1,
+                direction.get(angle, '')))
+    if orientations:
+        _log.info('Page orientations detected: ' + ' '.join(orientations))
+
+
+def preamble(_log):
+    _log.debug('ocrmypdf ' + VERSION)
+    _log.debug('tesseract ' + tesseract.version())
+    _log.debug('qpdf ' + qpdf.version())
+    if fitz:
+        _log.debug('PyMuPDF ' + fitz.version[0])
+        _log.debug('libmupdf ' + fitz.version[1])
+    else:
+        _log.debug('PyMuPDF not installed')
+
+
+def check_environ(options, _log):
+    old_envvars = (
+        'OCRMYPDF_TESSERACT',
+        'OCRMYPDF_QPDF',
+        'OCRMYPDF_GS',
+        'OCRMYPDF_UNPAPER')
+    for k in old_envvars:
+        if k in os.environ:
+            _log.warning(textwrap.dedent("""\
+                OCRmyPDF no longer uses the environment variable {}. 
+                Change PATH to select alternate programs.""".format(k)))
+
+
+def check_input_file(options, _log, start_input_file):
+    if options.input_file == '-':
+        # stdin
+        _log.info('reading file from standard input')
+        with open(start_input_file, 'wb') as stream_buffer:
+            from shutil import copyfileobj
+            copyfileobj(sys.stdin.buffer, stream_buffer)
+    else:
+        try:
+            re_symlink(options.input_file, start_input_file, _log)
+        except FileNotFoundError:
+            _log.error("File not found - " + options.input_file)
+            raise InputFileError()
+
+
+def check_requested_output_file(options, _log):
+    if options.output_file == '-':
+        if sys.stdout.isatty():
+            _log.error(textwrap.dedent("""\
+                Output was set to stdout '-' but it looks like stdout
+                is connected to a terminal.  Please redirect stdout to a
+                file."""))
+            raise BadArgsError()
+    elif not is_file_writable(options.output_file):
+        _log.error(
+            "Output file location (" + options.output_file + ") " +
+            "is not a writable file.")
+        raise OutputFileAccessError()
+
+
+def report_output_file_size(options, _log, input_file, output_file):
+    try:
+        output_size = Path(output_file).stat().st_size
+        input_size = Path(input_file).stat().st_size
+    except FileNotFoundError:
+        return  # Outputting to stream or something
+    ratio = output_size / input_size
+    if ratio < 1.35 or input_size < 25000:
+        return  # Seems fine
+    
+    reasons = []
+    if not fitz:
+        reasons.append("The optional dependency PyMuPDF is not installed.")
+    if options.force_ocr:
+        reasons.append("The argument --force-ocr was issued.")
+
+    if reasons:
+        explanation = (
+            "Possible reasons for this include:\n" + '\n'.join(reasons) + "\n")
+    else:
+        explanation = (
+            "No reason for this increase is known.  Please report this issue.")
+
+    _log.warning(textwrap.dedent("""\
+        The output file size is {:.2f}× larger than the input file.
+        {}
+        """.format(ratio, explanation)))
+
+
 def run_pipeline():
     options = parser.parse_args()
     options.verbose_abbreviated_path = 1
@@ -622,16 +814,33 @@ def run_pipeline():
 
     _log, _log_mutex = proxy_logger.make_shared_logger_and_proxy(
         logging_factory, __name__, logger_args)
-    _log.debug('ocrmypdf ' + VERSION)
-    _log.debug('tesseract ' + tesseract.version())
-
+    preamble(_log)
     check_options(options, _log)
 
+    # Complain about qpdf version < 7.0.0
+    # Suppress the warning if in the test suite, since there are no PPAs
+    # for qpdf 7.0.0 for Ubuntu trusty (i.e. Travis)
+    if qpdf.version() < '7.0.0' and not os.environ.get('PYTEST_CURRENT_TEST'):
+        complain(
+            "You are using qpdf version {0} which has known issues including "
+            "security vulnerabilities with certain malformed PDFs. Consider "
+            "upgrading to version 7.0.0 or newer.".format(qpdf.version()))
+
     # Any changes to options will not take effect for options that are already
     # bound to function parameters in the pipeline. (For example
     # options.input_file, options.pdf_renderer are already bound.)
     if not options.jobs:
         options.jobs = available_cpu_count()
+
+    # Performance is improved by setting Tesseract to single threaded. In tests
+    # this gives better throughput than letting a smaller number of Tesseract
+    # jobs run multithreaded.
+    if tesseract.v4():
+        os.environ.setdefault('OMP_THREAD_LIMIT', '1')
+    check_environ(options, _log)
+    if os.environ.get('PYTEST_CURRENT_TEST'):
+        os.environ['_OCRMYPDF_TEST_INFILE'] = options.input_file
+
     try:
         work_folder = mkdtemp(prefix="com.github.ocrmypdf.")
         options.history_file = os.path.join(
@@ -639,36 +848,14 @@ def run_pipeline():
         start_input_file = os.path.join(
             work_folder, 'origin')
 
-        if options.input_file == '-':
-            # stdin
-            _log.info('reading file from standard input')
-            with open(start_input_file, 'wb') as stream_buffer:
-                from shutil import copyfileobj
-                copyfileobj(sys.stdin.buffer, stream_buffer)
-        else:
-            try:
-                re_symlink(options.input_file, start_input_file, _log)
-            except FileNotFoundError:
-                _log.error("File not found - " + options.input_file)
-                return ExitCode.input_file
-
-        if options.output_file == '-':
-            if sys.stdout.isatty():
-                _log.error(textwrap.dedent("""\
-                    Output was set to stdout '-' but it looks like stdout
-                    is connected to a terminal.  Please redirect stdout to a
-                    file."""))
-                return ExitCode.bad_args
-        elif not is_file_writable(options.output_file):
-            _log.error(textwrap.dedent("""\
-                Output file location is not writable."""))
-            return ExitCode.file_access_error
+        check_input_file(options, _log, start_input_file)
+        check_requested_output_file(options, _log)
 
         manager = JobContextManager()
-        manager.register('JobContext', JobContext)
+        manager.register('JobContext', JobContext)  # pylint: disable=no-member
         manager.start()
 
-        context = manager.JobContext()
+        context = manager.JobContext()  # pylint: disable=no-member
         context.set_options(options)
         context.set_work_folder(work_folder)
 
@@ -678,29 +865,12 @@ def run_pipeline():
     except ruffus_exceptions.RethrownJobError as e:
         if options.verbose:
             _log.debug(str(e))  # stringify exception so logger doesn't have to
-
-        # Ruffus flattens exception to 5 element tuples. Because of a bug
-        # in <= 2.6.3 it may present either the single:
-        #   (task, job, exc, value, stack)
-        # or something like:
-        #   [[(task, job, exc, value, stack)]]
-        #
-        # Generally cross-process exception marshalling doesn't work well
-        # and ruffus doesn't support because BaseException has its own
-        # implementation of __reduce__ that attempts to reconstruct the
-        # exception based on e.__init__(e.args).
-        #
-        # Attempting to log the exception directly marshalls it to the logger
-        # which is probably in another process, so it's better to log only
-        # data from the exception at this point.
-
         exitcode = traverse_ruffus_exception(e.args, options, _log)
         if exitcode is None:
             _log.error("Unexpected ruffus exception: " + str(e))
             _log.error(repr(e))
             return ExitCode.other_error
-        else:
-            return exitcode
+        return exitcode
     except ExitCodeException as e:
         return e.exit_code
     except Exception as e:
@@ -709,8 +879,12 @@ def run_pipeline():
 
     if options.flowchart:
         _log.info("Flowchart saved to {}".format(options.flowchart))
-    elif options.output_file != '-':
-        if options.output_type == 'pdfa':
+    elif options.output_file == '-':
+        _log.info("Output sent to stdout")
+    elif os.path.samefile(options.output_file, os.devnull):
+        pass  # Say nothing when sending to dev null
+    else:
+        if options.output_type.startswith('pdfa'):
             pdfa_info = file_claims_pdfa(options.output_file)
             if pdfa_info['pass']:
                 msg = 'Output file is a {} (as expected)'
@@ -718,29 +892,20 @@ def run_pipeline():
             else:
                 msg = 'Output file is okay but is not PDF/A (seems to be {})'
                 _log.warning(msg.format(pdfa_info['conformance']))
-
                 return ExitCode.invalid_output_pdf
         if not qpdf.check(options.output_file, _log):
             _log.warning('Output file: The generated PDF is INVALID')
             return ExitCode.invalid_output_pdf
-    else:
-        _log.info("Output sent to stdout")
+
+        report_output_file_size(options, _log, start_input_file, 
+                                options.output_file)
 
     pdfinfo = context.get_pdfinfo()
     if options.verbose:
         from pprint import pformat
         _log.debug(pformat(pdfinfo))
-    direction = {0: 'n', 90: 'e',
-                 180: 's', 270: 'w'}
-    orientations = []
-    for n, page in enumerate(pdfinfo):
-        angle = pdfinfo[n].get('rotated', 0)
-        if angle != 0:
-            orientations.append('{0}{1}'.format(
-                n + 1,
-                direction.get(angle, '')))
-    if orientations:
-        _log.info('Page orientations detected: ' + ' '.join(orientations))
+
+    log_page_orientations(pdfinfo, _log)
 
     return ExitCode.ok
 

src/ocrmypdf/_unicodefun.py

@@ -0,0 +1,108 @@
+# Copyright (c) 2014, Armin Ronacher
+#
+# Copyright (c) 2017, James R Barlow
+#
+# Some rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions are
+# met:
+#
+#     * Redistributions of source code must retain the above copyright
+#       notice, this list of conditions and the following disclaimer.
+#
+#     * Redistributions in binary form must reproduce the above
+#       copyright notice, this list of conditions and the following
+#       disclaimer in the documentation and/or other materials provided
+#       with the distribution.
+#
+#     * The names of the contributors may not be used to endorse or
+#       promote products derived from this software without specific
+#       prior written permission.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+import os
+import sys
+import codecs
+
+
+def verify_python3_env():
+    """Ensures that the environment is good for unicode on Python 3."""
+    try:
+        import locale
+        fs_enc = codecs.lookup(locale.getpreferredencoding()).name
+    except Exception:
+        fs_enc = 'ascii'
+    if fs_enc != 'ascii':
+        return
+
+    extra = ''
+    if os.name == 'posix':
+        import subprocess
+        rv = subprocess.Popen(['locale', '-a'], stdout=subprocess.PIPE,
+                              stderr=subprocess.PIPE).communicate()[0]
+        good_locales = set()
+        has_c_utf8 = False
+
+        # Make sure we're operating on text here.
+        if isinstance(rv, bytes):
+            rv = rv.decode('ascii', 'replace')
+
+        for line in rv.splitlines():
+            locale = line.strip()
+            if locale.lower().endswith(('.utf-8', '.utf8')):
+                good_locales.add(locale)
+                if locale.lower() in ('c.utf8', 'c.utf-8'):
+                    has_c_utf8 = True
+
+        extra += '\n\n'
+        if not good_locales:
+            extra += (
+                'Additional information: on this system no suitable UTF-8\n'
+                'locales were discovered.  This most likely requires resolving\n'
+                'by reconfiguring the locale system.'
+            )
+        elif has_c_utf8:
+            extra += (
+                'This system supports the C.UTF-8 locale which is recommended.\n'
+                'You might be able to resolve your issue by exporting the\n'
+                'following environment variables:\n\n'
+                '    export LC_ALL=C.UTF-8\n'
+                '    export LANG=C.UTF-8'
+            )
+        else:
+            extra += (
+                'This system lists a couple of UTF-8 supporting locales that\n'
+                'you can pick from.  The following suitable locales were\n'
+                'discovered: %s'
+            ) % ', '.join(sorted(good_locales))
+
+        bad_locale = None
+        for locale in os.environ.get('LC_ALL'), os.environ.get('LANG'):
+            if locale and locale.lower().endswith(('.utf-8', '.utf8')):
+                bad_locale = locale
+            if locale is not None:
+                break
+        if bad_locale is not None:
+            extra += (
+                '\nocrmypdf discovered that you exported a UTF-8 locale\n'
+                'but the locale system could not pick up from it because\n'
+                'it does not exist.  The exported locale is "%s" but it\n'
+                'is not supported'
+            ) % bad_locale
+
+    raise RuntimeError('ocrmypdf will abort further execution because Python 3 '
+                       'was configured to use ASCII as encoding for the '
+                       'environment.' + extra)

src/ocrmypdf/exceptions.py

@@ -1,5 +1,19 @@
-#!/usr/bin/env python3
 # © 2016 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/>.
 
 
 from enum import IntEnum
@@ -20,7 +34,11 @@ class ExitCode(IntEnum):
 
 
 class ExitCodeException(Exception):
-    pass
+    exit_code = ExitCode.other_error
+
+
+class BadArgsError(ExitCodeException):
+    exit_code = ExitCode.bad_args
 
 
 class PdfMergeFailedError(ExitCodeException):
@@ -39,6 +57,10 @@ class DpiError(ExitCodeException):
     exit_code = ExitCode.input_file
 
 
+class OutputFileAccessError(ExitCodeException):
+    exit_code = ExitCode.file_access_error
+
+
 class PriorOcrFoundError(ExitCodeException):
     exit_code = ExitCode.already_done_ocr
 

src/ocrmypdf/exec/__init__.py

@@ -0,0 +1,51 @@
+# © 2016 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/>.
+
+"""Wrappers to manage subprocess calls"""
+
+import os
+import re
+import sys
+from subprocess import run, STDOUT, PIPE, CalledProcessError
+from ..exceptions import MissingDependencyError
+
+
+def get_version(program, *, 
+        version_arg='--version', regex=r'(\d+(\.\d+)*)'):
+    "Get the version of the specified program"
+    args_prog = [
+        program,
+        version_arg
+    ]
+    try:
+        proc = run(
+            args_prog, close_fds=True, universal_newlines=True,
+            stdout=PIPE, stderr=STDOUT, check=True)
+        output = proc.stdout
+    except CalledProcessError as e:
+        raise MissingDependencyError(
+            "Could not find program '{}' on the PATH".format(
+                program)) from e
+    try:
+        version = re.match(regex, output.strip()).group(1)
+    except AttributeError as e:
+        raise MissingDependencyError(
+            ("The program '{}' did not report its version. "
+            "Message was:\n{}").format(program, output)
+        )
+
+    return version

src/ocrmypdf/exec/ghostscript.py

@@ -1,31 +1,35 @@
-#!/usr/bin/env python3
-# © 2015 James R. Barlow: github.com/jbarlow83
+# © 2017 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/>.
 
 from tempfile import NamedTemporaryFile
 from subprocess import run, PIPE, STDOUT, CalledProcessError
 from shutil import copy
 from functools import lru_cache
 import re
-from . import get_program
-from ..exceptions import SubprocessOutputError
+import sys
+from PIL import Image
+from . import get_version
+from ..exceptions import SubprocessOutputError, MissingDependencyError
+from ..helpers import fspath
 
 
 @lru_cache(maxsize=1)
 def version():
-    args_gs = [
-        get_program('gs'),
-        '--version'
-    ]
-    try:
-        version = check_output(
-                args_gs, close_fds=True, universal_newlines=True,
-                stderr=STDOUT)
-    except CalledProcessError as e:
-        print("Could not find Ghostscript executable on system PATH.",
-              file=sys.stderr)
-        raise MissingDependencyError from e
-
-    return version.strip()
+    return get_version('gs')
 
 
 def _gs_error_reported(stream):
@@ -33,10 +37,31 @@ def _gs_error_reported(stream):
 
 
 def rasterize_pdf(input_file, output_file, xres, yres, raster_device, log,
-                  pageno=1):
+                  pageno=1, page_dpi=None):
+    """
+    Rasterize one page of a PDF at resolution (xres, yres) in canvas units.
+    
+    The image is sized to match the integer pixels dimensions implied by 
+    (xres, yres) even if those numbers are noninteger. The image's DPI will
+     be overridden with the values in page_dpi.
+    
+    :param input_file: pathlike
+    :param output_file: pathlike
+    :param xres: resolution at which to rasterize page
+    :param yres: 
+    :param raster_device: 
+    :param log: 
+    :param pageno: page number to rasterize (beginning at page 1)
+    :param page_dpi: resolution tuple (x, y) overriding output image DPI 
+    :return: 
+    """
+    res = xres, yres
+    int_res = round(xres), round(yres)
+    if not page_dpi:
+        page_dpi = res
     with NamedTemporaryFile(delete=True) as tmp:
         args_gs = [
-            get_program('gs'),
+            'gs',
             '-dQUIET',
             '-dSAFER',
             '-dBATCH',
@@ -44,11 +69,11 @@ def rasterize_pdf(input_file, output_file, xres, yres, raster_device, log,
             '-sDEVICE=%s' % raster_device,
             '-dFirstPage=%i' % pageno,
             '-dLastPage=%i' % pageno,
+            '-r{0}x{1}'.format(str(int_res[0]), str(int_res[1])),
             '-o', tmp.name,
-            '-r{0}x{1}'.format(str(round(xres)), str(round(yres))),
-            input_file
+            fspath(input_file)
         ]
-
+        
         p = run(args_gs, stdout=PIPE, stderr=STDOUT,
                 universal_newlines=True)
         if _gs_error_reported(p.stdout):
@@ -56,14 +81,31 @@ def rasterize_pdf(input_file, output_file, xres, yres, raster_device, log,
         else:
             log.debug(p.stdout)
 
-        if p.returncode == 0:
-            copy(tmp.name, output_file)
-        else:
+        if p.returncode != 0:
             log.error('Ghostscript rasterizing failed')
             raise SubprocessOutputError()
 
+        # Ghostscript only accepts integers for output resolution
+        # if the resolution happens to be fractional, then the discrepancy
+        # would change the size of the output page, especially if the DPI
+        # is quite low. Resize the image to the expected size
 
-def generate_pdfa(pdf_pages, output_file, compression, log, threads=1):
+        tmp.seek(0)
+        with Image.open(tmp) as im:
+            expected_size = round(im.size[0] / int_res[0] * res[0]), \
+                            round(im.size[1] / int_res[1] * res[1])
+            if expected_size != im.size or page_dpi != (xres, yres):
+                log.debug(
+                    "Ghostscript: resize output image {} -> {}".format(
+                        im.size, expected_size))
+                im.resize(expected_size).save(
+                    fspath(output_file), dpi=page_dpi)
+            else:
+                copy(tmp.name, fspath(output_file))
+
+
+def generate_pdfa(pdf_pages, output_file, compression, log,
+                  threads=1, pdf_version='1.5', pdfa_part='2'):
     compression_args = []
     if compression == 'jpeg':
         compression_args = [
@@ -87,18 +129,19 @@ def generate_pdfa(pdf_pages, output_file, compression, log, threads=1):
 
     with NamedTemporaryFile(delete=True) as gs_pdf:
         args_gs = [
-            get_program("gs"),
+            "gs",
             "-dQUIET",
             "-dBATCH",
             "-dNOPAUSE",
-            '-dNumRenderingThreads=' + str(threads),
+            "-dCompatibilityLevel=" + str(pdf_version),
+            "-dNumRenderingThreads=" + str(threads),
             "-sDEVICE=pdfwrite",
             "-dAutoRotatePages=/None",
             "-sColorConversionStrategy=/RGB",
             "-sProcessColorModel=DeviceRGB"
         ] + compression_args + [
             "-dJPEGQ=95",
-            "-dPDFA=2",
+            "-dPDFA=" + pdfa_part,
             "-dPDFACompatibilityPolicy=1",
             "-sOutputFile=" + gs_pdf.name,
         ]
@@ -125,4 +168,4 @@ def generate_pdfa(pdf_pages, output_file, compression, log, threads=1):
             copy(gs_pdf.name, output_file)
         else:
             log.error('Ghostscript PDF/A rendering failed')
-            raise SubprocessOutputError()
+            raise SubprocessOutputError()

src/ocrmypdf/exec/qpdf.py

@@ -0,0 +1,227 @@
+# © 2017 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/>.
+
+from subprocess import CalledProcessError, STDOUT, PIPE, run
+from functools import lru_cache
+import sys
+import os
+import re
+import resource
+
+from ..exceptions import InputFileError, SubprocessOutputError, \
+    MissingDependencyError, EncryptedPdfError
+from . import  get_version
+from ..helpers import re_symlink
+
+
+@lru_cache(maxsize=1)
+def version():
+    return get_version('qpdf', regex=r'qpdf version (.+)')    
+
+
+def check(input_file, log=None):
+    args_qpdf = [
+        'qpdf',
+        '--check',
+        input_file
+    ]
+
+    if log is None:
+        import logging as log
+
+    try:
+        run(args_qpdf, stderr=STDOUT, stdout=PIPE, universal_newlines=True, 
+            check=True)
+    except CalledProcessError as e:
+        if e.returncode == 2:
+            log.error("{0}: not a valid PDF, and could not repair it.".format(
+                input_file))
+            log.error("Details:")
+            log.error(e.output)
+        elif e.returncode == 3:
+            log.info("qpdf --check returned warnings:")
+            log.info(e.output)
+        else:
+            log.warning(e.output)
+        return False
+    return True
+
+
+def _probably_encrypted(e):
+    """qpdf can report a false positive "file is encrypted" message for damaged
+    files - suppress this"""
+    return e.returncode == 2 and \
+        'invalid password' in e.output and \
+        'file is damaged' not in e.output
+
+
+def repair(input_file, output_file, log):
+    args_qpdf = [
+        'qpdf', input_file, output_file
+    ]
+    try:
+        run(args_qpdf, stderr=STDOUT, stdout=PIPE, universal_newlines=True, 
+            check=True)
+    except CalledProcessError as e:
+        if e.returncode == 3 and e.output.find("operation succeeded"):
+            log.debug('qpdf found and fixed errors: ' + e.output)
+            return
+
+        if _probably_encrypted(e):
+            raise EncryptedPdfError() from e
+        elif e.returncode == 2:
+            log.error("{0}: not a valid PDF, and could not repair it.".format(
+                      input_file))
+            log.error("Details: " + e.output)
+            raise InputFileError() from e
+        else:
+            log.error("{0}: unknown error".format(
+                      input_file))
+            log.error(e.output)
+            raise SubprocessOutputError() from e
+
+
+def get_npages(input_file, log):
+    try:
+        pages = run(
+            ['qpdf', '--show-npages', input_file],
+            universal_newlines=True, check=True, stdout=PIPE, stderr=STDOUT)
+    except CalledProcessError as e:
+        if e.returncode == 2 and e.output.find('No such file'):
+            log.error(e.output)
+            raise InputFileError() from e
+    return int(pages)
+
+
+def split_pages(input_file, work_folder, npages):
+    """Split multipage PDF into individual pages.
+
+    Incredibly enough, this multiple process approach is about 70 times
+    faster than using Ghostscript.
+    """
+    for n in range(int(npages)):
+        args_qpdf = [
+            'qpdf', input_file,
+            '--pages', input_file, '{0}'.format(n + 1), '--',
+            os.path.join(work_folder, '{0:06d}.page.pdf'.format(n + 1))
+        ]
+        run(args_qpdf, check=True)
+
+
+def extract_page(input_file, output_file, pageno):
+    args_qpdf = [
+        'qpdf', input_file,
+        '--pages', input_file, '{0}'.format(pageno + 1), '--',
+        output_file
+    ]
+    run(args_qpdf, check=True)
+    
+
+def _merge_inner(input_files, output_file, min_version=None, log=None):
+    """Merge the list of input files (all filenames) into the output file.
+
+    The input files may contain one or more pages.
+    """
+
+    # Single page 'merges' should still be attempted to that the same error
+    # checking is applied to single page case
+
+    version_arg = ['--min-version={}'.format(min_version)] \
+                  if min_version else []
+
+    if log is None:
+        import logging as log
+
+    args_qpdf = [
+        'qpdf'
+    ] + version_arg + [
+        input_files[0], '--pages'
+    ] + input_files + ['--', output_file]
+
+    try:
+        run(args_qpdf, check=True, stderr=PIPE, universal_newlines=True)
+    except CalledProcessError as e:
+        if e.returncode == 3 and \
+                e.stderr.find("unknown token while reading object") and \
+                e.stderr.find("operation succeeded"):
+            # Only whitelist the 'unknown token' problem (decimal/string issue)
+            # qpdf issue #165
+            log.warning('qpdf found and fixed errors: ' + e.stderr)
+            return
+        raise e from e
+
+
+def merge(input_files, output_file, min_version=None, log=None, max_files=None):
+    """Merge the list of input files (all filenames) into the output file.
+
+    The input files may contain one or more pages.
+
+    """
+    # qpdf requires that every file that contributes to the output has a file 
+    # descriptor that remains open. That means, given our approach of one 
+    # intermediate PDF per, we can practically hit the number of file 
+    # descriptors. 
+
+    if max_files is None or max_files < 2:
+        # Find out how many open file descriptors we can get away with
+        ulimits = resource.getrlimit(resource.RLIMIT_NOFILE)
+        max_open_files = ulimits[0]
+        max_files = max_open_files // 2  # Conservative guess
+
+    # We'll write things alongside the output file
+    output_dir = os.path.dirname(output_file)
+
+    import random
+    import string    
+
+    def randstr():
+        return ''.join(random.sample(string.ascii_lowercase, 6))
+
+    # How many files to grab at once, merging all their contents
+    step_size = max_files
+
+    workqueue = input_files.copy()
+    counter = 1
+    next_workqueue = []
+    while len(workqueue) > 1 or len(next_workqueue) > 0:
+        # Take n files out of the queue
+        n = min(step_size, len(workqueue))
+        job = workqueue[0:n]
+        del workqueue[0:n]
+        log.debug('merging ' + repr(job))
+
+        # Merge them into 1 file, which will contain n^depth pages
+        merge_file = os.path.join(
+            output_dir, "merge-{:06d}-{}.pdf".format(counter, randstr()))
+        counter += 1
+        _merge_inner(job, merge_file, min_version=min_version, log=log)
+
+        # On the next 
+        next_workqueue.append(merge_file)
+        log.debug('next_workqueue ' + repr(next_workqueue))
+
+        # If we're out of things to do in this queue, move on to the next
+        # queue. On the counter-th pass of the workqueue we can chew through
+        # (step_size)**N pages, so on most systems the second pass finishes
+        # the job.
+        if len(workqueue) == 0:
+            workqueue = next_workqueue
+            next_workqueue = []
+
+    re_symlink(workqueue.pop(), output_file, log)
+
+

src/ocrmypdf/exec/tesseract.py

@@ -1,93 +1,91 @@
-#!/usr/bin/env python3
-# © 2015 James R. Barlow: github.com/jbarlow83
+# © 2017 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 sys
 import os
 import re
 import shutil
 from functools import lru_cache
-from ..exceptions import MissingDependencyError, TesseractConfigError
-from ..helpers import page_number
-from . import get_program
 from collections import namedtuple
 from textwrap import dedent
+from subprocess import PIPE, CalledProcessError, \
+    TimeoutExpired, check_output, STDOUT
+from contextlib import suppress
+
 import PyPDF2 as pypdf
 
-from subprocess import Popen, PIPE, CalledProcessError, \
-    TimeoutExpired, check_output, STDOUT, DEVNULL
-
+from ..exceptions import MissingDependencyError, TesseractConfigError
+from ..helpers import page_number
+from . import get_version
 
 OrientationConfidence = namedtuple(
     'OrientationConfidence',
     ('angle', 'confidence'))
 
-HOCR_TEMPLATE = '''<?xml version="1.0" encoding="UTF-8"?>
+HOCR_TEMPLATE = """<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
     "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
 <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
  <head>
   <title></title>
-  <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
-  <meta name='ocr-system' content='tesseract 3.02.02' />
+<meta http-equiv="Content-Type" content="text/html;charset=utf-8" />
+  <meta name='ocr-system' content='tesseract 3.05.00' />
   <meta name='ocr-capabilities' content='ocr_page ocr_carea ocr_par ocr_line ocrx_word'/>
- </head>
- <body>
-  <div class='ocr_page' id='page_1' title='image "x.tif"; bbox 0 0 {0} {1}; ppageno 0'>
-   <div class='ocr_carea' id='block_1_1' title="bbox 0 1 {0} {1}">
-    <p class='ocr_par' dir='ltr' id='par_1' title="bbox 0 1 {0} {1}">
-     <span class='ocr_line' id='line_1' title="bbox 0 1 {0} {1}"><span class='ocrx_word' id='word_1' title="bbox 0 1 {0} {1}"> </span>
-     </span>
-    </p>
-   </div>
+</head>
+<body>
+  <div class='ocr_page' id='page_1' title='image "_blank.png"; bbox 0 0 {0} {1}; ppageno 0'>
   </div>
  </body>
-</html>'''
+</html>
+"""
 
 
 @lru_cache(maxsize=1)
 def version():
-    args_tess = [
-        get_program('tesseract'),
-        '--version'
-    ]
-    try:
-        versions = check_output(
-                args_tess, close_fds=True, universal_newlines=True,
-                stderr=STDOUT)
-    except CalledProcessError as e:
-        print("Could not find Tesseract executable on system PATH.",
-              file=sys.stderr)
-        raise MissingDependencyError from e
-
-    tesseract_version = re.match(r'tesseract\s(.+)', versions).group(1)
-    return tesseract_version
+    return get_version('tesseract', regex=r'tesseract\s(.+)')
 
 
 def v4():
     "Is this Tesseract v4.0?"
-    return (version() >= '4')
+    return version() >= '4'
 
 
+@lru_cache(maxsize=1)
 def has_textonly_pdf():
-    if version() == '4.00.00alpha':
-        # textonly_pdf added during the 4.00.00alpha cycle, so we must test
-        # more carefully to see if it is present
-        args_tess = [
-            get_program('tesseract'),
-            '--print-parameters'
-        ]
-        try:
-            params = check_output(
-                    args_tess, close_fds=True, universal_newlines=True,
-                    stderr=STDOUT)
-        except CalledProcessError as e:
-            print("Could not --print-parameters from tesseract",
-                  file=sys.stderr)
-            raise MissingDependencyError from e
-        if 'textonly_pdf' in params:
-            return True
-    else:
-        return v4()
+    """Does Tesseract have textonly_pdf capability?
+    
+    Available in 3.05.01, and v4.00.00alpha since January 2017. Best to 
+    parse the parameter list
+    """
+    args_tess = [
+        'tesseract',
+        '--print-parameters'
+    ]
+    params = ''
+    try:
+        params = check_output(
+                args_tess, universal_newlines=True, stderr=STDOUT)
+    except CalledProcessError as e:
+        print("Could not --print-parameters from tesseract",
+              file=sys.stderr)
+        raise MissingDependencyError from e
+    if 'textonly_pdf' in params:
+        return True
+    return False
 
 
 def psm():
@@ -98,13 +96,12 @@ def psm():
 @lru_cache(maxsize=1)
 def languages():
     args_tess = [
-        get_program('tesseract'),
+        'tesseract',
         '--list-langs'
     ]
     try:
         langs = check_output(
-                args_tess, close_fds=True, universal_newlines=True,
-                stderr=STDOUT)
+                args_tess, universal_newlines=True, stderr=STDOUT)
     except CalledProcessError as e:
         msg = dedent("""Tesseract failed to report available languages.
         Output from Tesseract:
@@ -116,12 +113,12 @@ def languages():
     return set(lang.strip() for lang in langs.splitlines()[1:])
 
 
-def tess_base_args(languages, engine_mode):
+def tess_base_args(langs, engine_mode):
     args = [
-        get_program('tesseract'),
+        'tesseract',
     ]
-    if languages:
-        args.extend(['-l', '+'.join(languages)])
+    if langs:
+        args.extend(['-l', '+'.join(langs)])
     if engine_mode is not None and v4():
         args.extend(['--oem', str(engine_mode)])
     return args
@@ -129,7 +126,7 @@ def tess_base_args(languages, engine_mode):
 
 def get_orientation(input_file, language: list, engine_mode, timeout: float,
                     log):
-    args_tesseract = tess_base_args(language, engine_mode) + [
+    args_tesseract = tess_base_args(['osd'], engine_mode) + [
         psm(), '0',
         input_file,
         'stdout'
@@ -137,19 +134,18 @@ def get_orientation(input_file, language: list, engine_mode, timeout: float,
 
     try:
         stdout = check_output(
-            args_tesseract, close_fds=True, stderr=STDOUT,
-            universal_newlines=True, timeout=timeout)
+            args_tesseract, stderr=STDOUT, timeout=timeout)
     except TimeoutExpired:
         return OrientationConfidence(angle=0, confidence=0.0)
     except CalledProcessError as e:
         tesseract_log_output(log, e.output, input_file)
-        if ('Too few characters. Skipping this page' in e.output or
-                'Image too large' in e.output):
+        if (b'Too few characters. Skipping this page' in e.output or
+                b'Image too large' in e.output):
             return OrientationConfidence(0, 0)
         raise e from e
     else:
         osd = {}
-        for line in stdout.splitlines():
+        for line in stdout.decode().splitlines():
             line = line.strip()
             parts = line.split(':', maxsplit=2)
             if len(parts) == 2:
@@ -174,8 +170,17 @@ def get_orientation(input_file, language: list, engine_mode, timeout: float,
 
 
 def tesseract_log_output(log, stdout, input_file):
-    lines = stdout.splitlines()
     prefix = "{0:4d}: [tesseract] ".format(page_number(input_file))
+
+    try:
+        text = stdout.decode()
+    except UnicodeDecodeError:
+        log.error(prefix + "command line output was not utf-8. " +
+            "This usually means Tesseract's language packs do not match "
+            "the installed version of Tesseract.")
+        text = stdout.decode('utf-8', 'backslashreplace')
+
+    lines = text.splitlines()
     for line in lines:
         if line.startswith("Tesseract Open Source"):
             continue
@@ -216,7 +221,8 @@ def _generate_null_hocr(output_hocr, output_sidecar, image):
 
 def generate_hocr(input_file, output_files, language: list, engine_mode,
                   tessconfig: list,
-                  timeout: float, pagesegmode: int, log):
+                  timeout: float, pagesegmode: int, user_words, user_patterns,
+                  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'))
@@ -227,19 +233,28 @@ def generate_hocr(input_file, output_files, language: list, engine_mode,
     if pagesegmode is not None:
         args_tesseract.extend([psm(), str(pagesegmode)])
 
+    if user_words:
+        args_tesseract.extend(['--user-words', user_words])
+
+    if user_patterns:
+        args_tesseract.extend(['--user-patterns', user_patterns])
+
     # Reminder: test suite tesseract spoofers will break after any changes
     # to the number of order parameters here
+    # Tesseract 3.04 requires the order here to be "hocr txt" and will fail
+    # on "txt hocr"
+
     args_tesseract.extend([
         input_file,
         prefix,
-        'txt',
-        'hocr'
+        'hocr',
+        'txt'
     ] + tessconfig)
     try:
         log.debug(args_tesseract)
         stdout = check_output(
-            args_tesseract, close_fds=True, stderr=STDOUT,
-            universal_newlines=True, timeout=timeout)
+            args_tesseract, stderr=STDOUT,
+            timeout=timeout)
     except TimeoutExpired:
         # Generate a HOCR file with no recognized text if tesseract times out
         # Temporary workaround to hocrTransform not being able to function if
@@ -248,9 +263,9 @@ def generate_hocr(input_file, output_files, language: list, engine_mode,
         _generate_null_hocr(output_hocr, output_sidecar, input_file)
     except CalledProcessError as e:
         tesseract_log_output(log, e.output, input_file)
-        if 'read_params_file: parameter not found' in e.output:
+        if b'read_params_file: parameter not found' in e.output:
             raise TesseractConfigError() from e
-        if 'Image too large' in e.output:
+        if b'Image too large' in e.output:
             _generate_null_hocr(output_hocr, output_sidecar, input_file)
             return
 
@@ -268,6 +283,8 @@ def use_skip_page(text_only, skip_pdf, output_pdf, output_text):
         f.write('[skipped page]')
 
     if not text_only:
+        with suppress(FileNotFoundError):
+            os.remove(output_pdf)  # In case it was partially created
         os.symlink(skip_pdf, output_pdf)
         return
 
@@ -280,13 +297,18 @@ def use_skip_page(text_only, skip_pdf, output_pdf, output_text):
     with open(output_pdf, 'wb') as out:
         pdf_out = pypdf.PdfFileWriter()
         w, h = page0.mediaBox.getWidth(), page0.mediaBox.getHeight()
+        # If skip page has a /Rotate key, replicate the rotation
+        rotation = int(page0.get('/Rotate', 0))
+        if rotation % 180 == 90:
+            w, h = h, w
         pdf_out.addBlankPage(w, h)
         pdf_out.write(out)
 
 
 def generate_pdf(*, input_image, skip_pdf, output_pdf, output_text,
                  language: list, engine_mode, text_only: bool,
-                 tessconfig: list, timeout: float, pagesegmode: int, log):
+                 tessconfig: list, timeout: float, pagesegmode: int,
+                 user_words, user_patterns, log):
     '''Use Tesseract to render a PDF.
 
     input_image -- image to analyze
@@ -309,22 +331,29 @@ def generate_pdf(*, input_image, skip_pdf, output_pdf, output_text,
     if text_only:
         args_tesseract.extend(['-c', 'textonly_pdf=1'])
 
+    if user_words:
+        args_tesseract.extend(['--user-words', user_words])
+
+    if user_patterns:
+        args_tesseract.extend(['--user-patterns', user_patterns])
+
     prefix = os.path.splitext(output_pdf)[0]  # Tesseract appends suffixes
 
-    # Reminder: test suite tesseract spoofers will break after any changes
+    # Reminder: test suite tesseract spoofers might break after any changes
     # to the number of order parameters here
+
     args_tesseract.extend([
         input_image,
         prefix,
-        'txt',
         'pdf',
+        'txt'
     ] + tessconfig)
 
     try:
         log.debug(args_tesseract)
         stdout = check_output(
-            args_tesseract, close_fds=True, stderr=STDOUT,
-            universal_newlines=True, timeout=timeout)
+            args_tesseract, stderr=STDOUT,
+            timeout=timeout)
         if os.path.exists(prefix + '.txt'):
             shutil.move(prefix + '.txt', output_text)
     except TimeoutExpired:
@@ -332,10 +361,10 @@ def generate_pdf(*, input_image, skip_pdf, output_pdf, output_text,
         use_skip_page(text_only, skip_pdf, output_pdf, output_text)
     except CalledProcessError as e:
         tesseract_log_output(log, e.output, input_image)
-        if 'read_params_file: parameter not found' in e.output:
+        if b'read_params_file: parameter not found' in e.output:
             raise TesseractConfigError() from e
 
-        if 'Image too large' in e.output:
+        if b'Image too large' in e.output:
             use_skip_page(text_only, skip_pdf, output_pdf, output_text)
             return
         raise e from e

src/ocrmypdf/exec/unpaper.py

@@ -1,27 +1,30 @@
-#!/usr/bin/env python3
 # © 2015 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/>.
+
 # unpaper documentation:
 # https://github.com/Flameeyes/unpaper/blob/master/doc/basic-concepts.md
 
-from subprocess import CalledProcessError, STDOUT, check_output, check_call
+from subprocess import CalledProcessError, STDOUT, check_output
 from tempfile import NamedTemporaryFile
 import sys
 import os
 from functools import lru_cache
 from ..exceptions import MissingDependencyError
-from . import get_program
-
-
-@lru_cache(maxsize=1)
-def version():
-    args_unpaper = [
-        get_program('unpaper'),
-        '--version'
-    ]
-    version = check_output(
-        args_unpaper, close_fds=True, universal_newlines=True,
-        stderr=STDOUT, timeout=5)
-    return version.strip()
+from . import get_version
 
 
 try:
@@ -31,9 +34,14 @@ except ImportError:
     raise
 
 
+@lru_cache(maxsize=1)
+def version():
+    return get_version('unpaper')
+
+
 def run(input_file, output_file, dpi, log, mode_args):
     args_unpaper = [
-        get_program('unpaper'),
+        'unpaper',
         '-v',
         '--dpi', str(dpi)
     ] + mode_args

src/ocrmypdf/helpers.py

@@ -0,0 +1,169 @@
+# © 2016 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/>.
+
+from functools import partial
+from collections.abc import Iterable
+from contextlib import suppress, contextmanager
+from pathlib import Path
+import sys
+import os
+import multiprocessing
+
+
+def re_symlink(input_file, soft_link_name, log=None):
+    """
+    Helper function: relinks soft symbolic link if necessary
+    """
+
+    if log is None:
+        prdebug = partial(print, file=sys.stderr)
+    else:
+        prdebug = log.debug
+
+    # Guard against soft linking to oneself
+    if input_file == soft_link_name:
+        prdebug("Warning: No symbolic link made. You are using " +
+                "the original data directory as the working directory.")
+        return
+
+    # Soft link already exists: delete for relink?
+    if os.path.lexists(soft_link_name):
+        # do not delete or overwrite real (non-soft link) file
+        if not os.path.islink(soft_link_name):
+            raise FileExistsError(
+                "%s exists and is not a link" % soft_link_name)
+        try:
+            os.unlink(soft_link_name)
+        except OSError:
+            prdebug("Can't unlink %s" % (soft_link_name))
+
+    if not os.path.exists(input_file):
+        raise FileNotFoundError(
+            "trying to create a broken symlink to %s" % input_file)
+
+    prdebug("os.symlink(%s, %s)" % (input_file, soft_link_name))
+
+    # Create symbolic link using absolute path
+    os.symlink(
+        os.path.abspath(input_file),
+        soft_link_name
+    )
+
+
+def is_iterable_notstr(thing):
+    return isinstance(thing, Iterable) and not isinstance(thing, str)
+
+
+def page_number(input_file):
+    "Get one-based page number implied by filename (000002.pdf -> 2)"
+    return int(os.path.basename(input_file)[0:6])
+
+
+def available_cpu_count():
+    try:
+        return multiprocessing.cpu_count()
+    except NotImplementedError:
+        pass
+
+    try:
+        import psutil
+        return psutil.cpu_count()
+    except (ImportError, AttributeError):
+        pass
+
+    complain(
+        "Could not get CPU count.  Assuming one (1) CPU."
+        "Use -j N to set manually.")
+    return 1
+
+
+def is_file_writable(test_file):
+    """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():
+        # Python 3.5 does not accept parameters for Path.resolve() and behaves
+        # as if strict=True (throws an exception on failure). Python 3.6 
+        # defaults to strict=False. This implements strict=False like behavior
+        # for Python 3.5.
+        if sys.version_info[0:2] <= (3, 5):
+            p = Path(os.path.realpath(str(p)))
+        else:
+            p = p.resolve(strict=False)
+
+    # p.is_file() throws an exception in some cases
+    if p.exists() and p.is_file():
+        return os.access(
+            str(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
+
+
+if sys.version_info[0:2] <= (3, 5):
+    def universal_open(p, *args, **kwargs):
+        "Work around Python 3.5's inability to open(pathlib.Path())"
+        try:
+            return p.open(*args, **kwargs)
+        except AttributeError:
+            return open(p, *args, **kwargs)
+
+
+    def fspath(path):
+        import pathlib
+        '''https://www.python.org/dev/peps/pep-0519/#os'''
+        if isinstance(path, (str, bytes)):
+            return path
+
+        # Work from the object's type to match method resolution of other magic
+        # methods.
+        path_type = type(path)
+        try:
+            path = path_type.__fspath__(path)
+        except AttributeError:
+            # Added for Python 3.5 support.
+            if isinstance(path, pathlib.Path):
+                return str(path)
+            elif hasattr(path_type, '__fspath__'):
+                raise
+        else:
+            if isinstance(path, (str, bytes)):
+                return path
+            else:
+                raise TypeError("expected __fspath__() to return str or bytes, "
+                                "not " + type(path).__name__)
+
+        raise TypeError(
+            "expected str, bytes, pathlib.Path or os.PathLike object, not "
+            + path_type.__name__)
+
+else:
+    universal_open = open
+    fspath = os.fspath

src/ocrmypdf/hocrtransform.py

@@ -0,0 +1,347 @@
+#!/usr/bin/env python3
+#
+# Copyright (c) 2010, Jonathan Brinley
+#   Original version from: https://github.com/jbrinley/HocrConverter
+#
+# Copyright (c) 2013-14, Julien Pfefferkorn
+#   Modifications
+#
+# Copyright (c) 2015-16, James R. Barlow
+#   Set text to transparent
+#
+# 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.
+
+from reportlab.pdfgen.canvas import Canvas
+from reportlab.lib.units import inch
+from xml.etree import ElementTree
+from PIL import Image
+from collections import namedtuple
+from math import atan, sin, cos
+import re
+import argparse
+
+
+Rect = namedtuple('Rect', ['x1', 'y1', 'x2', 'y2'])
+
+
+class HocrTransformError(Exception):
+    pass
+
+
+class HocrTransform():
+
+    """
+    A class for converting documents from the hOCR format.
+    For details of the hOCR format, see:
+    http://kba.cloud/hocr-spec/
+    """
+
+    box_pattern = re.compile(r'bbox((\s+\d+){4})')
+    baseline_pattern = re.compile(r'''
+        baseline \s+ 
+        ([\-\+]?\d*\.?\d*) \s+  # +/- decimal float
+        ([\-\+]?\d+)            # +/- int''', re.VERBOSE)
+    ligatures = str.maketrans({
+        'ff': 'ff',
+        'ffi': 'f‌f‌i',
+        'ffl': 'f‌f‌l',
+        'fi': 'fi',
+        'fl': 'fl',
+    })
+
+    def __init__(self, hocrFileName, dpi):
+        self.dpi = dpi
+        self.hocr = ElementTree.parse(hocrFileName)
+
+        # if the hOCR file has a namespace, ElementTree requires its use to
+        # find elements
+        matches = re.match(r'({.*})html', self.hocr.getroot().tag)
+        self.xmlns = ''
+        if matches:
+            self.xmlns = matches.group(1)
+
+        # get dimension in pt (not pixel!!!!) of the OCRed image
+        self.width, self.height = None, None
+        for div in self.hocr.findall(
+                ".//%sdiv[@class='ocr_page']" % (self.xmlns)):
+            coords = self.element_coordinates(div)
+            pt_coords = self.pt_from_pixel(coords)
+            self.width = pt_coords.x2 - pt_coords.x1
+            self.height = pt_coords.y2 - pt_coords.y1
+            # there shouldn't be more than one, and if there is, we don't want
+            # it
+            break
+        if self.width is None or self.height is None:
+            raise HocrTransformError("hocr file is missing page dimensions")
+
+    def __str__(self):
+        """
+        Return the textual content of the HTML body
+        """
+        if self.hocr is None:
+            return ''
+        body = self.hocr.find(".//%sbody" % (self.xmlns))
+        if body:
+            return self._get_element_text(body)
+        else:
+            return ''
+
+    def _get_element_text(self, element):
+        """
+        Return the textual content of the element and its children
+        """
+        text = ''
+        if element.text is not None:
+            text += element.text
+        for child in element.getchildren():
+            text += self._get_element_text(child)
+        if element.tail is not None:
+            text += element.tail
+        return text
+
+    @classmethod
+    def element_coordinates(cls, element):
+        """
+        Returns a tuple containing the coordinates of the bounding box around
+        an element
+        """
+        out = (0, 0, 0, 0)
+        if 'title' in element.attrib:
+            matches = cls.box_pattern.search(element.attrib['title'])
+            if matches:
+                coords = matches.group(1).split()
+                out = Rect._make(int(coords[n]) for n in range(4))
+        return out
+
+    @classmethod
+    def baseline(cls, element):
+        """
+        Returns a tuple containing the baseline slope and intercept.
+        """
+        if 'title' in element.attrib:
+            matches = cls.baseline_pattern.search(element.attrib['title'])
+            if matches:
+                return float(matches.group(1)), int(matches.group(2))
+        return (0, 0) 
+
+    def pt_from_pixel(self, pxl):
+        """
+        Returns the quantity in PDF units (pt) given quantity in pixels
+        """
+        return Rect._make(
+            (c / self.dpi * inch) for c in pxl)
+
+    @classmethod
+    def replace_unsupported_chars(cls, s):
+        """
+        Given an input string, returns the corresponding string that:
+        - is available in the helvetica facetype
+        - does not contain any ligature (to allow easy search in the PDF file)
+        """
+        return s.translate(cls.ligatures)
+
+    def to_pdf(self, outFileName, imageFileName=None, showBoundingboxes=False,
+               fontname="Helvetica", invisibleText=False, interwordSpaces=False):
+        """
+        Creates a PDF file with an image superimposed on top of the text.
+        Text is positioned according to the bounding box of the lines in
+        the hOCR file.
+        The image need not be identical to the image used to create the hOCR
+        file.
+        It can have a lower resolution, different color mode, etc.
+        """
+        # create the PDF file
+        # page size in points (1/72 in.)
+        pdf = Canvas(
+            outFileName, pagesize=(self.width, self.height), pageCompression=1)
+
+        # draw bounding box for each paragraph
+        # light blue for bounding box of paragraph
+        pdf.setStrokeColorRGB(0, 1, 1)
+        # light blue for bounding box of paragraph
+        pdf.setFillColorRGB(0, 1, 1)
+        pdf.setLineWidth(0)		# no line for bounding box
+        for elem in self.hocr.findall(
+                ".//%sp[@class='%s']" % (self.xmlns, "ocr_par")):
+
+            elemtxt = self._get_element_text(elem).rstrip()
+            if len(elemtxt) == 0:
+                continue
+
+            pxl_coords = self.element_coordinates(elem)
+            pt = self.pt_from_pixel(pxl_coords)
+
+            # draw the bbox border
+            if showBoundingboxes:
+                pdf.rect(
+                    pt.x1, self.height - pt.y2, pt.x2 - pt.x1, pt.y2 - pt.y1,
+                    fill=1)
+                    
+        found_lines = False
+        for line in self.hocr.findall(
+                ".//%sspan[@class='%s']" % (self.xmlns, "ocr_line")):
+            found_lines = True
+            self._do_line(pdf, line, "ocrx_word", fontname, invisibleText, 
+                          interwordSpaces, showBoundingboxes)
+
+        if not found_lines:
+            # Tesseract did not report any lines (just words)
+            root = self.hocr.find(".//%sdiv[@class='%s']" % (self.xmlns, "ocr_page"))
+            self._do_line(pdf, root, "ocrx_word", fontname, invisibleText,
+                          interwordSpaces, showBoundingboxes)
+        # put the image on the page, scaled to fill the page
+        if imageFileName is not None:
+            pdf.drawImage(imageFileName, 0, 0,
+                          width=self.width, height=self.height)
+
+        # finish up the page and save it
+        pdf.showPage()
+        pdf.save()
+
+
+    @classmethod
+    def polyval(cls, poly, x):
+        return x * poly[0] + poly[1]
+
+
+    def _do_line(self, pdf, line, elemclass, fontname, invisibleText, 
+                 interwordSpaces, showBoundingboxes):
+        pxl_line_coords = self.element_coordinates(line)
+        line_box = self.pt_from_pixel(pxl_line_coords)
+        line_height = line_box.y2 - line_box.y1
+
+        slope, pxl_intercept = self.baseline(line)
+        if abs(slope) < 0.005:
+            slope = 0.0
+        angle = atan(slope)
+        cos_a, sin_a = cos(angle), sin(angle)
+
+        text = pdf.beginText()
+        intercept = pxl_intercept / self.dpi * inch
+
+        # Don't allow the font to break out of the bounding box. Division by
+        # cos_a accounts for extra clearance between the glyph's vertical axis
+        # on a sloped baseline and the edge of the bounding box.
+        fontsize = (line_height - abs(intercept)) / cos_a
+        text.setFont(fontname, fontsize)
+        if invisibleText:
+            text.setTextRenderMode(3)  # Invisible (indicates OCR text)
+
+        # Intercept is normally negative, so this places it above the bottom
+        # of the line box
+        baseline_y2 = self.height - (line_box.y2 + intercept)
+
+        if showBoundingboxes:
+            # draw the baseline in magenta, dashed
+            pdf.setDash()
+            pdf.setStrokeColorRGB(0.95, 0.65, 0.95)
+            pdf.setLineWidth(0.5)
+            # negate slope because it is defined as a rise/run in pixel
+            # coordinates and page coordinates have the y axis flipped
+            pdf.line(line_box.x1, 
+                     baseline_y2,
+                     line_box.x2,
+                     self.polyval((-slope, baseline_y2), 
+                                  line_box.x2 - line_box.x1))
+            # light green for bounding box of word/line
+            pdf.setDash(6, 3)
+            pdf.setStrokeColorRGB(1, 0, 0)
+
+        text.setTextTransform(
+            cos_a, -sin_a, sin_a, cos_a,
+            line_box.x1, baseline_y2
+        )
+        pdf.setFillColorRGB(0, 0, 0)  # text in black
+
+        elements = line.findall(
+                ".//%sspan[@class='%s']" % (self.xmlns, elemclass))
+        for elem in elements:
+            elemtxt = self._get_element_text(elem).strip()
+            elemtxt = self.replace_unsupported_chars(elemtxt)
+            if elemtxt == '':
+                continue
+
+            pxl_coords = self.element_coordinates(elem)
+            box = self.pt_from_pixel(pxl_coords)
+            if interwordSpaces:
+                # if  `--interword-spaces` is true, append a space
+                # to the end of each text element to allow simpler PDF viewers 
+                # such as PDF.js to better recognize words in search and copy 
+                # and paste. Do not remove space from last word in line, even
+                # though it would look better, because it will interfere with
+                # naive text extraction. \n does not work either.
+                elemtxt += ' '
+                box = Rect._make((
+                    box.x1, 
+                    line_box.y1,
+                    box.x2 + pdf.stringWidth(' ', fontname, line_height), 
+                    line_box.y2))
+            box_width = box.x2 - box.x1
+            font_width = pdf.stringWidth(elemtxt, fontname, fontsize)
+
+            # draw the bbox border
+            if showBoundingboxes:
+                pdf.rect(
+                    box.x1,
+                    self.height - line_box.y2,
+                    box_width,
+                    line_height,
+                    fill=0)
+
+            # Adjust relative position of cursor
+            # This is equivalent to:
+            #   text.setTextOrigin(pt.x1, self.height - line_box.y2)
+            # but the former generates a full text reposition matrix (Tm) in the
+            # content stream while this issues a "offset" (Td) command.
+            # .moveCursor() is relative to start of the text line, where the
+            # "text line" means whatever reportlab defines it as. Do not use
+            # use .getCursor(), since moveCursor() rather unintuitively plans 
+            # its moves relative to .getStartOfLine().
+            # For skewed lines, in the text transform we set up a rotated
+            # coordinate system, so we don't have to account for the 
+            # incremental offset. Surprisingly most PDF viewers can handle this.
+            cursor = text.getStartOfLine()
+            dx = box.x1 - cursor[0]
+            dy = baseline_y2 - cursor[1]            
+            text.moveCursor(dx, dy)
+
+            text.setHorizScale(100 * box_width / font_width)
+            text.textOut(elemtxt)
+        pdf.drawText(text)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description='Convert hocr file to PDF')
+    parser.add_argument('-b', '--boundingboxes', action="store_true",
+                        default=False, help='Show bounding boxes borders')
+    parser.add_argument('-r', '--resolution', type=int,
+                        default=300,
+                        help='Resolution of the image that was OCRed')
+    parser.add_argument('-i', '--image', default=None,
+                        help='Path to the image to be placed above the text')
+    parser.add_argument('--interword-spaces', action='store_true',
+                         default=False, help='Add spaces between words')
+    parser.add_argument('hocrfile', help='Path to the hocr file to be parsed')
+    parser.add_argument(
+        'outputfile', help='Path to the PDF file to be generated')
+    args = parser.parse_args()
+
+    hocr = HocrTransform(args.hocrfile, args.resolution)
+    hocr.to_pdf(args.outputfile, args.image, args.boundingboxes, interwordSpaces=args.interword_spaces)

src/ocrmypdf/leptonica.py

@@ -3,6 +3,21 @@
 #
 # © 2013-16: jbarlow83 from Github (https://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/>.
+#
 # Python FFI wrapper for Leptonica library
 
 import argparse
@@ -14,6 +29,7 @@ from ctypes.util import find_library
 from .lib._leptonica import ffi
 from functools import lru_cache
 from enum import Enum
+from .helpers import fspath
 
 lept = ffi.dlopen(find_library('lept'))
 
@@ -208,26 +224,27 @@ class Pix:
             return 'P'
 
     @classmethod
-    def read(cls, filename):
+    def read(cls, path):
         """Load an image file into a PIX object.
 
         Leptonica can load TIFF, PNM (PBM, PGM, PPM), PNG, and JPEG.  If
         loading fails then the object will wrap a C null pointer.
         """
+        filename = fspath(path)
         with LeptonicaErrorTrap():
-            return cls(lept.pixRead(
-                filename.encode(sys.getfilesystemencoding())))
+            return cls(lept.pixRead(os.fsencode(filename)))
 
     def write_implied_format(
-            self, filename, jpeg_quality=0, jpeg_progressive=0):
+            self, path, jpeg_quality=0, jpeg_progressive=0):
         """Write pix to the filename, with the extension indicating format.
 
         jpeg_quality -- quality (iff JPEG; 1 - 100, 0 for default)
         jpeg_progressive -- (iff JPEG; 0 for baseline seq., 1 for progressive)
         """
+        filename = fspath(path)
         with LeptonicaErrorTrap():
             lept.pixWriteImpliedFormat(
-                filename.encode(sys.getfilesystemencoding()),
+                os.fsencode(filename),
                 self._pix, jpeg_quality, jpeg_progressive)
 
     def topil(self):
@@ -564,35 +581,4 @@ if __name__ == '__main__':
     parser_deskew.set_defaults(func=deskew)
 
     args = parser.parse_args()
-
-    if get_leptonica_version() != u'leptonica-1.69':
-        print("Unexpected leptonica version: %s" % getLeptonicaVersion())
-
-    args.func(args)
-
-
-def test_skew_angle():
-    from PIL import Image, ImageDraw
-    from tempfile import NamedTemporaryFile
-
-    im = Image.new(mode='1', size=(1000, 1000), color=1)
-
-    draw = ImageDraw.Draw(im)
-    for n in range(20):
-        draw.line([(50, 25 + 50*n), (950, 25 + 50*n)], width=1)
-    del draw
-
-    test_angles = [0.1 * ang for ang in range(1, 10)] + \
-                  [float(ang) for ang in range(1, 7)]
-    test_angles += [-ang for ang in test_angles]
-    test_angles = sorted(test_angles)
-
-    for rotate_angle in test_angles:
-        rotated_im = im.rotate(rotate_angle)
-        with NamedTemporaryFile(prefix='lept-skew', suffix='.png', delete=True) as tmpfile:
-            rotated_im.save(tmpfile)
-            pix = pixRead(tmpfile.name)
-            angle, confidence = pixFindSkew(pix)
-            print('{0} {1}  {2}'.format(rotate_angle, angle, confidence), file=sys.stderr)
-
-
+    args.func(args)

src/ocrmypdf/lib/__init__.py

@@ -0,0 +1,28 @@
+# © 2017 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/>.
+
+"""Bindings to external libraries"""
+
+import os as _os
+
+try:
+    import fitz
+except ImportError:
+    fitz = None
+
+if _os.environ.get('_OCRMYPDF_NO_FITZ'):
+    fitz = None

src/ocrmypdf/lib/_leptonica.py

@@ -0,0 +1,10 @@
+# auto-generated file
+import _cffi_backend
+
+ffi = _cffi_backend.FFI('ocrmypdf.lib._leptonica',
+    _version = 0x2601,
+    _types = b'\x00\x00\x0F\x0D\x00\x00\xBB\x03\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\xBE\x03\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x01\x11\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x01\x11\x00\x00\xBA\x03\x00\x00\x0F\x03\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x01\x11\x00\x00\x01\x11\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x01\x11\x00\x00\x01\x11\x00\x00\x01\x11\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x01\x11\x00\x00\x01\x11\x00\x00\x01\x11\x00\x00\x0D\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x01\x11\x00\x00\x01\x11\x00\x00\x01\x11\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x01\x11\x00\x00\x01\x11\x00\x00\x0D\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x01\x11\x00\x00\x01\x11\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x01\x11\x00\x00\x01\x11\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x0D\x01\x00\x00\x02\x03\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x01\x11\x00\x00\x0D\x01\x00\x00\x0D\x01\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x01\x11\x00\x00\x07\x01\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x01\x11\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x01\x11\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\x08\x11\x00\x00\x00\x0F\x00\x00\x01\x0D\x00\x00\xC7\x03\x00\x00\x1C\x01\x00\x00\x00\x0F\x00\x00\xBD\x0D\x00\x00\x00\x0F\x00\x00\x49\x0D\x00\x00\x00\x0F\x00\x00\x02\x0D\x00\x00\x01\x11\x00\x00\x01\x11\x00\x00\x1F\x03\x00\x00\x00\x0F\x00\x00\x02\x0D\x00\x00\x01\x11\x00\x00\x01\x11\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x6E\x11\x00\x00\x6E\x11\x00\x00\x6E\x11\x00\x00\x00\x0F\x00\x00\x02\x0D\x00\x00\x01\x11\x00\x00\x6E\x11\x00\x00\x6E\x11\x00\x00\x00\x0F\x00\x00\x02\x0D\x00\x00\x01\x11\x00\x00\x49\x11\x00\x00\x49\x11\x00\x00\x00\x0F\x00\x00\x02\x0D\x00\x00\x01\x11\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x0D\x01\x00\x00\x07\x01\x00\x00\x49\x11\x00\x00\x00\x0F\x00\x00\x02\x0D\x00\x00\x01\x11\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x6E\x11\x00\x00\x6E\x11\x00\x00\x00\x0F\x00\x00\x02\x0D\x00\x00\x01\x11\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x0D\x01\x00\x00\x01\x03\x00\x00\x9D\x11\x00\x00\x00\x0F\x00\x00\x02\x0D\x00\x00\x01\x11\x00\x00\xC6\x03\x00\x00\x65\x03\x00\x00\x00\x0F\x00\x00\x02\x0D\x00\x00\x08\x11\x00\x00\x01\x11\x00\x00\x07\x01\x00\x00\x07\x01\x00\x00\x00\x0F\x00\x00\x02\x0D\x00\x00\xC4\x03\x00\x00\xA3\x11\x00\x00\x01\x11\x00\x00\x0D\x01\x00\x00\x00\x0F\x00\x00\xCA\x0D\x00\x00\x10\x11\x00\x00\x00\x0F\x00\x00\xCA\x0D\x00\x00\x9D\x11\x00\x00\x00\x0F\x00\x00\xCA\x0D\x00\x00\xCA\x03\x00\x00\x00\x0F\x00\x00\x00\x09\x00\x00\x01\x09\x00\x00\x02\x09\x00\x00\xBE\x03\x00\x00\x02\x01\x00\x00\x0E\x01\x00\x00\x0B\x01\x00\x00\x05\x01\x00\x00\x03\x01\x00\x00\xBC\x03\x00\x00\xC5\x03\x00\x00\x04\x01\x00\x00\xC7\x03\x00\x00\x08\x01\x00\x00\x0C\x01\x00\x00\x06\x01\x00\x00\x00\x01',
+    _globals = (b'\x00\x00\xB1\x23boxDestroy',0,b'\x00\x00\x67\x23getLeptonicaVersion',0,b'\x00\x00\xB7\x23lept_free',0,b'\x00\x00\x69\x23makePixelSumTab8',0,b'\x00\x00\x16\x23pixAnd',0,b'\x00\x00\x23\x23pixBackgroundNorm',0,b'\x00\x00\x1B\x23pixCleanBackgroundToWhite',0,b'\x00\x00\x0D\x23pixClipRectangle',0,b'\x00\x00\x8D\x23pixColorFraction',0,b'\x00\x00\x59\x23pixColorMagnitude',0,b'\x00\x00\x0A\x23pixConvertRGBToLuminance',0,b'\x00\x00\x6B\x23pixCorrelationBinary',0,b'\x00\x00\x80\x23pixCountPixels',0,b'\x00\x00\x63\x23pixDeserializeFromMemory',0,b'\x00\x00\x50\x23pixDeskew',0,b'\x00\x00\xB4\x23pixDestroy',0,b'\x00\x00\x0A\x23pixEndianByteSwapNew',0,b'\x00\x00\x00\x23pixFindPageForeground',0,b'\x00\x00\x7B\x23pixFindSkew',0,b'\x00\x00\x2F\x23pixGammaTRC',0,b'\x00\x00\x70\x23pixGetAverageMaskedRGB',0,b'\x00\x00\x36\x23pixGlobalNormRGB',0,b'\x00\x00\x12\x23pixInvert',0,b'\x00\x00\x54\x23pixMaskOverColorPixels',0,b'\x00\x00\x85\x23pixNumSignificantGrayColors',0,b'\x00\x00\x96\x23pixOtsuAdaptiveThreshold',0,b'\x00\x00\x3E\x23pixOtsuThreshOnBackgroundNorm',0,b'\x00\x00\x60\x23pixRead',0,b'\x00\x00\x50\x23pixRemoveColormap',0,b'\x00\x00\x12\x23pixRotate180',0,b'\x00\x00\x50\x23pixRotateOrth',0,b'\x00\x00\x4B\x23pixScale',0,b'\x00\x00\xA0\x23pixSerializeToMemory',0,b'\x00\x00\xA5\x23pixWriteImpliedFormat',0,b'\x00\x00\xAB\x23pixWriteMemPng',0),
+    _struct_unions = ((b'\x00\x00\x00\xBA\x00\x00\x00\x02Box',b'\x00\x00\x02\x11x',b'\x00\x00\x02\x11y',b'\x00\x00\x02\x11w',b'\x00\x00\x02\x11h',b'\x00\x00\xC7\x11refcount'),(b'\x00\x00\x00\xBB\x00\x00\x00\x02Pix',b'\x00\x00\xC7\x11w',b'\x00\x00\xC7\x11h',b'\x00\x00\xC7\x11d',b'\x00\x00\xC7\x11spp',b'\x00\x00\xC7\x11wpl',b'\x00\x00\xC7\x11refcount',b'\x00\x00\x02\x11xres',b'\x00\x00\x02\x11yres',b'\x00\x00\x02\x11informat',b'\x00\x00\x02\x11special',b'\x00\x00\xBD\x11text',b'\x00\x00\xC3\x11colormap',b'\x00\x00\xC6\x11data'),(b'\x00\x00\x00\xBC\x00\x00\x00\x02PixColormap',b'\x00\x00\xB8\x11array',b'\x00\x00\x02\x11depth',b'\x00\x00\x02\x11nalloc',b'\x00\x00\x02\x11n')),
+    _typenames = (b'\x00\x00\x00\xBABOX',b'\x00\x00\x00\xBBPIX',b'\x00\x00\x00\xBCPIXCMAP',b'\x00\x00\x00\x1Fl_float32',b'\x00\x00\x00\xBFl_float64',b'\x00\x00\x00\xC1l_int16',b'\x00\x00\x00\x02l_int32',b'\x00\x00\x00\xC0l_int64',b'\x00\x00\x00\xC2l_int8',b'\x00\x00\x00\xC9l_uint16',b'\x00\x00\x00\xC7l_uint32',b'\x00\x00\x00\xC8l_uint64',b'\x00\x00\x00\xC5l_uint8'),
+)

src/ocrmypdf/lib/compile_leptonica.py

@@ -1,3 +1,21 @@
+#!/usr/bin/env python3
+# © 2017 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/>.
+
 from cffi import FFI
 
 ffibuilder = FFI()
@@ -181,10 +199,39 @@ pixColorMagnitude(PIX     *pixs,
                   l_int32  bwhite,
                   l_int32  type);
 
-void
+PIX *
+pixMaskOverColorPixels(PIX     *pixs,
+                       l_int32  threshdiff,
+                       l_int32  mindist);
+
+l_int32
+pixGetAverageMaskedRGB(PIX        *pixs,
+                       PIX        *pixm,
+                       l_int32     x,
+                       l_int32     y,
+                       l_int32     factor,
+                       l_int32     type,
+                       l_float32  *prval,
+                       l_float32  *pgval,
+                       l_float32  *pbval);
+
+PIX * 
+pixGlobalNormRGB(PIX * 	pixd,
+                 PIX * 	pixs,
+                 l_int32 	rval,
+                 l_int32 	gval,
+                 l_int32 	bval,
+                 l_int32 	mapval); 
+
+PIX *
+pixInvert(PIX * pixd,
+          PIX * pixs);
+
+void                 
 boxDestroy(BOX  **pbox);
 
-void lept_free(void *ptr);
+void
+lept_free(void *ptr);  
 """)
 
 

src/ocrmypdf/pdfa.py

@@ -1,10 +1,24 @@
-#!/usr/bin/env python3
 # © 2015 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/>.
+
 # Generate a PDFA_def.ps file for Ghostscript >= 9.14
 
 from string import Template
-import codecs
+from binascii import hexlify
 import pkg_resources
 import PyPDF2 as pypdf
 
@@ -79,7 +93,7 @@ def encode_text_string(s: str) -> str:
         return ''
 
     utf16_bytes = s.encode('utf-16be')
-    ascii_hex_bytes = codecs.encode(b'\xfe\xff' + utf16_bytes, 'hex')
+    ascii_hex_bytes = hexlify(b'\xfe\xff' + utf16_bytes)
     ascii_hex_str = ascii_hex_bytes.decode('ascii').lower()
     return ascii_hex_str
 

src/ocrmypdf/pdfinfo.py

@@ -0,0 +1,783 @@
+#!/usr/bin/env python3
+# © 2015 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/>.
+
+from subprocess import Popen, PIPE
+from decimal import Decimal
+from math import hypot, isclose
+import re
+import sys
+from collections import namedtuple
+from collections.abc import MutableMapping, Mapping
+import warnings
+from pathlib import Path
+from enum import Enum
+from contextlib import contextmanager
+
+import PyPDF2 as pypdf
+from .lib import fitz
+from .helpers import universal_open, fspath
+
+
+
+matrix_mult = pypdf.pdf.utils.matrixMultiply
+
+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')
+
+
+FRIENDLY_COLORSPACE = {
+    '/DeviceGray': Colorspace.gray,
+    '/CalGray': Colorspace.gray,
+    '/DeviceRGB': Colorspace.rgb,
+    '/CalRGB': Colorspace.rgb,
+    '/DeviceCMYK': Colorspace.cmyk,
+    '/Lab': Colorspace.lab,
+    '/ICCBased': Colorspace.icc,
+    '/Indexed': Colorspace.index,
+    '/Separation': Colorspace.sep,
+    '/DeviceN': Colorspace.devn,
+    '/Pattern': Colorspace.pattern,
+    '/G': Colorspace.gray,  # Abbreviations permitted in inline images
+    '/RGB': Colorspace.rgb,
+    '/CMYK': Colorspace.cmyk,
+    '/I': Colorspace.index,
+}
+
+FRIENDLY_ENCODING = {
+    '/CCITTFaxDecode': Encoding.ccitt,
+    '/DCTDecode': Encoding.jpeg,
+    '/JPXDecode': Encoding.jpeg2000,
+    '/JBIG2Decode': Encoding.jbig2,
+    '/CCF': Encoding.ccitt,  # Abbreviations permitted in inline images
+    '/DCT': Encoding.jpeg,
+    '/AHx': Encoding.asciihex,
+    '/A85': Encoding.ascii85,
+    '/LZW': Encoding.lzw,
+    '/Fl': Encoding.flate,
+    '/RL': Encoding.runlength
+}
+
+FRIENDLY_COMP = {
+    Colorspace.gray: 1,
+    Colorspace.rgb: 3,
+    Colorspace.cmyk: 4,
+    Colorspace.lab: 3,
+    Colorspace.index: 1
+}
+
+
+UNIT_SQUARE = (1.0, 0.0, 0.0, 1.0, 0.0, 0.0)
+
+
+def _matrix_from_shorthand(shorthand):
+    """Convert from PDF matrix shorthand to full matrix
+
+    PDF 1.7 spec defines a shorthand for describing the entries of a matrix
+    since the last column is always (0, 0, 1).
+    """
+
+    a, b, c, d, e, f = map(float, shorthand)
+    return ((a, b, 0),
+            (c, d, 0),
+            (e, f, 1))
+
+
+def _shorthand_from_matrix(matrix):
+    """Convert from transformation matrix to PDF shorthand."""
+    a, b = matrix[0][0], matrix[0][1]
+    c, d = matrix[1][0], matrix[1][1]
+    e, f = matrix[2][0], matrix[2][1]
+    return tuple(map(float, (a, b, c, d, e, f)))
+
+
+def _is_unit_square(shorthand):
+    values = map(float, shorthand)
+    pairwise = zip(values, UNIT_SQUARE)
+    return all([isclose(a, b, rel_tol=1e-3) for a, b in pairwise])
+
+XobjectSettings = namedtuple('XobjectSettings',
+    ['name', 'shorthand', 'stack_depth'])
+
+InlineSettings = namedtuple('InlineSettings',
+    ['settings', 'shorthand', 'stack_depth'])
+
+ContentsInfo = namedtuple('ContentsInfo', 
+    ['xobject_settings', 'inline_images', 'found_text'])
+
+
+def _normalize_stack(operations):
+    """Fix runs of qQ's in the stack
+
+    For some reason PyPDF2 converts runs of qqq, QQ, QQQq, etc. into single
+    operations.  Break this silliness up and issue each stack operation
+    individually so we don't lose count.
+
+    """
+    for operands, command in operations:
+        if re.match(br'Q*q+$', command):   # Zero or more Q, one or more q
+            for char in command:           # Split into individual bytes
+                yield ([], bytes([char]))  # Yield individual bytes
+        else:
+            yield (operands, command)
+
+
+def _interpret_contents(contentstream, initial_shorthand=UNIT_SQUARE):
+    """Interpret the PDF content stream
+
+    The stack represents the state of the PDF graphics stack.  We are only
+    interested in the current transformation matrix (CTM) so we only track
+    this object; a full implementation would need to track many other items.
+
+    The CTM is initialized to the mapping from user space to device space.
+    PDF units are 1/72".  In a PDF viewer or printer this matrix is initialized
+    to the transformation to device space.  For example if set to
+    (1/72, 0, 0, 1/72, 0, 0) then all units would be calculated in inches.
+
+    Images are always considered to be (0, 0) -> (1, 1).  Before drawing an
+    image there should be a 'cm' that sets up an image coordinate system
+    where drawing from (0, 0) -> (1, 1) will draw on the desired area of the
+    page.
+
+    PDF units suit our needs so we initialize ctm to the identity matrix.
+
+    PyPDF2 replaces inline images with a fake "INLINE IMAGE" operator.
+
+    """
+
+    operations = contentstream.operations
+    stack = []
+    ctm = _matrix_from_shorthand(initial_shorthand)
+    xobject_settings = []
+    inline_images = []
+    found_text = False
+
+    for n, op in enumerate(_normalize_stack(operations)):
+        operands, command = op
+        if command == b'q':
+            stack.append(ctm)
+            if len(stack) > 32:
+                raise RuntimeError(
+                    "PDF graphics stack overflow, command %i" % n)
+        elif command == b'Q':
+            try:
+                ctm = stack.pop()
+            except IndexError:
+                raise RuntimeError(
+                    "PDF graphics stack underflow, command %i" % n)
+        elif command == b'cm':
+            ctm = matrix_mult(
+                _matrix_from_shorthand(operands), ctm)
+        elif command == b'Do':
+            image_name = operands[0]
+            settings = XobjectSettings(
+                name=image_name, shorthand=_shorthand_from_matrix(ctm),
+                stack_depth=len(stack))
+            xobject_settings.append(settings)
+        elif command == b'INLINE IMAGE':
+            settings = operands['settings']
+            inline = InlineSettings(
+                settings=settings, shorthand=_shorthand_from_matrix(ctm),
+                stack_depth=len(stack))
+            inline_images.append(inline)
+        elif command in (b'Tj', b'TJ', b'"', b"'"):
+            found_text = True
+
+
+    return ContentsInfo(
+        xobject_settings=xobject_settings,
+        inline_images=inline_images,
+        found_text=found_text)
+
+
+def _get_dpi(ctm_shorthand, image_size):
+    """Given the transformation matrix and image size, find the image DPI.
+
+    PDFs do not include image resolution information within image data.
+    Instead, the PDF page content stream describes the location where the
+    image will be rasterized, and the effective resolution is the ratio of the
+    pixel size to raster target size.
+
+    Normally a scanned PDF has the paper size set appropriately but this is
+    not guaranteed. The most common case is a cropped image will change the
+    page size (/CropBox) without altering the page content stream. That means
+    it is not sufficient to assume that the image fills the page, even though
+    that is the most common case.
+
+    A PDF image may be scaled (always), cropped, translated, rotated in place
+    to an arbitrary angle (rarely) and skewed. Only equal area mappings can
+    be expressed, that is, it is not necessary to consider distortions where
+    the effective DPI varies with position.
+
+    To determine the image scale, transform an offset axis vector v0 (0, 0),
+    width-axis vector v0 (1, 0), height-axis vector vh (0, 1) with the matrix,
+    which gives the dimensions of the image in PDF units. From there we can
+    compare to actual image dimensions. PDF uses
+    row vector * matrix_tranposed unlike the traditional
+    matrix * column vector.
+
+    The offset, width and height vectors can be combined in a matrix and
+    multiplied by the transform matrix. Then we want to calculated
+        magnitude(width_vector - offset_vector)
+    and
+        magnitude(height_vector - offset_vector)
+
+    When the above is worked out algebraically, the effect of translation
+    cancels out, and the vector magnitudes become functions of the nonzero
+    transformation matrix indices. The results of the derivation are used
+    in this code.
+
+    pdfimages -list does calculate the DPI in some way that is not completely
+    naive, but it does not get the DPI of rotated images right, so cannot be
+    used anymore to validate this. Photoshop works, or using Acrobat to
+    rotate the image back to normal.
+
+    It does not matter if the image is partially cropped, or even out of the
+    /MediaBox.
+
+    """
+
+    a, b, c, d, _, _ = ctm_shorthand
+
+    # Calculate the width and height of the image in PDF units
+    image_drawn_width = hypot(a, b)
+    image_drawn_height = hypot(c, d)
+
+    # The scale of the image is pixels per unit of default user space (1/72")
+    scale_w = image_size[0] / image_drawn_width
+    scale_h = image_size[1] / image_drawn_height
+
+    # DPI = scale * 72
+    dpi_w = scale_w * 72.0
+    dpi_h = scale_h * 72.0
+
+    return dpi_w, dpi_h
+
+
+class ImageInfo:
+    DPI_PREC = Decimal('1.000')
+
+    def __init__(self, *, name='', pdfimage=None, inline=None,
+                 shorthand=None):
+
+        self._name = name
+        self._shorthand = shorthand
+        if inline:
+            # Fixme does not work for inline images with non abbreviated
+            # fields
+            self._origin = 'inline'
+            self._width = inline.settings['/W']
+            self._height = inline.settings['/H']
+            self._bpc = inline.settings.get('/BPC', 8)
+            try:
+                self._color = FRIENDLY_COLORSPACE[inline.settings['/CS']]
+            except Exception:
+                self._color = '-'
+            self._comp = FRIENDLY_COMP.get(self._color, '?')
+            if '/F' in inline.settings:
+                filter_ = inline.settings['/F']
+                if isinstance(filter_, pypdf.generic.ArrayObject):
+                    filter_ = filter_[0]
+                self._enc = FRIENDLY_ENCODING.get(filter_, 'image')
+            else:
+                self._enc = 'image'
+        elif pdfimage:
+            self._origin = 'xobject'
+            self._width = pdfimage['/Width']
+            self._height = pdfimage['/Height']
+            if '/BitsPerComponent' in pdfimage:
+                self._bpc = pdfimage['/BitsPerComponent']
+            else:
+                self._bpc = 8
+
+            # Fixme: this is incorrectly treats explicit masks as stencil masks,
+            # but good enough for now. Explicit masks have /ImageMask true but are
+            # never called for in content stream, instead are drawn as a /Mask on
+            # other images. For our purposes finding out the details of /Mask
+            # will seldom matter.
+            if '/ImageMask' in pdfimage:
+                self._type = 'stencil' if pdfimage['/ImageMask'].value \
+                    else 'image'
+            else:
+                self._type = 'image'
+            if '/Filter' in pdfimage:
+                filter_ = pdfimage['/Filter']
+                if isinstance(filter_, pypdf.generic.ArrayObject):
+                    filter_ = filter_[0]
+                self._enc = FRIENDLY_ENCODING.get(filter_, 'image')
+            else:
+                self._enc = 'image'
+            if '/ColorSpace' in pdfimage:
+                cs = pdfimage['/ColorSpace']
+                if isinstance(cs, pypdf.generic.ArrayObject):
+                    cs = cs[0]
+                self._color = FRIENDLY_COLORSPACE.get(cs, '-')
+            else:
+                self._color = FRIENDLY_COLORSPACE[Colorspace.jpeg2000] \
+                    if self._enc == Encoding.jpeg2000 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, 'jbig2'):
+                self._comp = FRIENDLY_COMP[Colorspace.gray]
+
+    @property
+    def name(self):
+        return self._name
+
+    @property
+    def type_(self):
+        return self._type
+
+    @property
+    def width(self):
+        return self._width
+
+    @property
+    def height(self):
+        return self._height
+
+    @property
+    def bpc(self):
+        return self._bpc
+
+    @property
+    def color(self):
+        return self._color
+
+    @property
+    def comp(self):
+        return self._comp
+
+    @property
+    def enc(self):
+        return self._enc
+
+    @property
+    def xres(self):
+        return _get_dpi(self._shorthand, (self._width, self._height))[0]
+
+    @property
+    def yres(self):
+        return _get_dpi(self._shorthand, (self._width, self._height))[1]
+
+    def __getitem__(self, item):
+        warnings.warn("ImageInfo.__getitem__", DeprecationWarning)
+        if item in ('name', 'width', 'height', 'bpc', 'color', 'comp', 'enc'):
+            return getattr(self, item)
+        elif item == 'dpi_w':
+            return Decimal(self.xres).quantize(self.DPI_PREC)
+        elif item == 'dpi_h':
+            return Decimal(self.yres).quantize(self.DPI_PREC)
+        elif item == 'dpi':
+            return Decimal(self.xres * self.yres).sqrt().quantize(
+                self.DPI_PREC)
+        else:
+            raise KeyError(item)
+
+    def __repr__(self):
+        class_locals = {attr: getattr(self, attr, None) for attr in dir(self)
+                        if not attr.startswith('_')}
+        return (
+            "<ImageInfo '{name}' {type_} {width}x{height} {color} "
+            "{comp} {bpc} {enc} {xres}x{yres}>").format(**class_locals)
+
+
+def _find_inline_images(contentsinfo):
+    "Find inline images in the contentstream"
+
+    for n, inline in enumerate(contentsinfo.inline_images):
+        yield ImageInfo(name='inline-%02d' % n, shorthand=inline.shorthand,
+                        inline=inline)
+
+
+def _image_xobjects(container):
+    """Search for all XObject-based images in the container
+
+    Usually the container is a page, but it could also be a Form XObject
+    that contains images. Filter out the Form XObjects which are dealt with
+    elsewhere.
+
+    Generate a sequence of tuples (image, xobj container), where container,
+    where xobj is the name of the object and image is the object itself,
+    since the object does not know its own name.
+
+    """
+
+    if '/Resources' not in container:
+        return
+    resources = container['/Resources']
+    if '/XObject' not in resources:
+        return
+    for xobj in resources['/XObject']:
+        candidate = resources['/XObject'][xobj]
+        if candidate['/Subtype'] == '/Image':
+            pdfimage = candidate
+            yield (pdfimage, xobj)
+
+
+def _find_regular_images(container, contentsinfo):
+    """Find images stored in the container's /Resources /XObject
+
+    Usually the container is a page, but it could also be a Form XObject
+    that contains images.
+
+    Generates images with their DPI at time of drawing.
+
+    """
+
+    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 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
+                # gets drawn into a square of 1x1 PDF units (or 1/72",
+                # or 0.35 mm).  The equivalent DPI will be >100,000.  Exclude
+                # these from our DPI calculation for the page.
+                continue
+
+            yield ImageInfo(name=draw.name, pdfimage=pdfimage, shorthand=
+                            draw.shorthand)
+
+
+def _find_form_xobject_images(pdf, container, contentsinfo):
+    """Find any images that are in Form XObjects in the container
+
+    The container may be a page, or a parent Form XObject.
+
+    """
+    if '/Resources' not in container:
+        return
+    resources = container['/Resources']
+    if '/XObject' not in resources:
+        return
+    for xobj in resources['/XObject']:
+        candidate = resources['/XObject'][xobj]
+        if candidate['/Subtype'] != '/Form':
+            continue
+
+        form_xobject = candidate
+        for settings in contentsinfo.xobject_settings:
+            if settings.name != xobj:
+                continue
+
+            # Find images once for each time this Form XObject is drawn.
+            # This could be optimized to cache the multiple drawing events
+            # but in practice both Form XObjects and multiple drawing of the
+            # same object are both very rare.
+            ctm_shorthand = settings.shorthand
+            yield from _find_images(
+                pdf=pdf, container=form_xobject, shorthand=ctm_shorthand)
+
+
+def _find_images(*, pdf, container, shorthand=None):
+    """Find all individual instances of images drawn in the container
+
+    Usually the container is a page, but it may also be a Form XObject.
+
+    On a typical page images are stored inline or as regular images
+    in an XObject.
+
+    Form XObjects may include inline images, XObject images,
+    and recursively, other Form XObjects; and also vector drawing commands.
+
+    Every instance of an image being drawn somewhere is flattened and
+    treated as a unique image, since if the same image is drawn multiple times
+    on one page it may be drawn at differing resolutions, and our objective
+    is to find the resolution at which the page can be rastered without
+    downsampling.
+
+    """
+
+    if container.get('/Type') == '/Page' and '/Contents' in container:
+        # For a /Page the content stream is attached to the page's /Contents
+        page = container
+        contentstream = pypdf.pdf.ContentStream(page.getContents(), pdf)
+        initial_shorthand = shorthand or UNIT_SQUARE
+    elif container.get('/Type') == '/XObject' and \
+            container['/Subtype'] == '/Form':
+        # For a Form XObject that content stream is attached to the XObject
+        contentstream = pypdf.pdf.ContentStream(container, pdf)
+
+        # Set the CTM to the state it was when the "Do" operator was
+        # encountered that is drawing this instance of the Form XObject
+        ctm = _matrix_from_shorthand(shorthand or UNIT_SQUARE)
+
+        # A Form XObject may provide its own matrix to map form space into
+        # user space. Get this if one exists
+        form_matrix = _matrix_from_shorthand(
+                container.get('/Matrix', UNIT_SQUARE))
+
+        # Concatenate form matrix with CTM to ensure CTM is correct for
+        # drawing this instance of the XObject
+        ctm = matrix_mult(form_matrix, ctm)
+        initial_shorthand = _shorthand_from_matrix(ctm)
+    else:
+        return
+
+    contentsinfo = _interpret_contents(contentstream, initial_shorthand)
+
+    yield from _find_inline_images(contentsinfo)
+    yield from _find_regular_images(container, contentsinfo)
+    yield from _find_form_xobject_images(pdf, container, contentsinfo)
+
+
+def _naive_find_text(*, pdf, page):
+    if not(page.get('/Type') == '/Page' and '/Contents' in page):
+        # Not a page, or has no /Contents => no text
+        return False
+
+    # First we check the main content stream    
+    contentstream = pypdf.pdf.ContentStream(page.getContents(), pdf)
+    contentsinfo = _interpret_contents(contentstream, UNIT_SQUARE)
+    if contentsinfo.found_text:
+        return True
+
+    # Then see if there is a Form XObject with with a content stream
+    # that might have text.  For full completeness we should recursively
+    # search nested Form XObjects, as we do with images.  But that is
+    # rare.
+    if '/Resources' in page:
+        resources = page['/Resources']
+        if '/XObject' in resources:    
+            for xobj in resources['/XObject']:
+                candidate = resources['/XObject'][xobj]
+                if candidate['/Subtype'] != '/Form':
+                    continue
+                form_xobject = candidate                
+                # Content stream is attached to Form XObject dictionary
+                contentstream = pypdf.pdf.ContentStream(form_xobject, pdf)
+                sub_contentsinfo = _interpret_contents(
+                    contentstream, UNIT_SQUARE)
+                if sub_contentsinfo.found_text:
+                    return True
+    return False
+
+
+def _page_has_text(infile, pageno):
+    doc = fitz.Document(infile)
+    text = doc.getPageText(pageno)
+    if text.strip() != '':
+        return True
+    return False
+
+
+def _pdf_get_pageinfo(pdf, pageno: int, infile):
+    pageinfo = {}
+    pageinfo['pageno'] = pageno
+    pageinfo['images'] = []
+
+    if isinstance(pdf, Path):
+        pdf = pypdf.PdfFileReader(str(pdf))
+    elif isinstance(pdf, str):
+        pdf = pypdf.PdfFileReader(pdf)
+
+    page = pdf.pages[pageno]
+
+    if fitz:
+        pageinfo['has_text'] = _page_has_text(str(infile), pageno)
+    else:
+        pageinfo['has_text'] = _naive_find_text(pdf=pdf, page=page)
+
+    width_pt = page.mediaBox.getWidth()
+    height_pt = page.mediaBox.getHeight()
+
+    userunit = page.get('/UserUnit', Decimal(1.0))
+    pageinfo['userunit'] = userunit
+    pageinfo['width_inches'] = width_pt * userunit / Decimal(72.0)
+    pageinfo['height_inches'] = height_pt * userunit / Decimal(72.0)
+
+    try:
+        pageinfo['rotate'] = int(page['/Rotate'])
+    except KeyError:
+        pageinfo['rotate'] = 0
+
+    userunit_shorthand = (userunit, 0, 0, userunit, 0, 0)
+    pageinfo['images'] = [im for im in
+                          _find_images(pdf=pdf, container=page,
+                                       shorthand=userunit_shorthand)]
+    if pageinfo['images']:
+        xres = max(image['dpi_w'] for image in pageinfo['images'])
+        yres = max(image['dpi_h'] for image in pageinfo['images'])
+        pageinfo['xres'], pageinfo['yres'] = xres, yres
+        pageinfo['width_pixels'] = \
+            int(round(xres * pageinfo['width_inches']))
+        pageinfo['height_pixels'] = \
+            int(round(yres * pageinfo['height_inches']))
+
+    return pageinfo
+
+
+def _pdf_get_all_pageinfo(infile):
+    with universal_open(infile, 'rb') as f:
+        pdf = pypdf.PdfFileReader(f)
+        return [PageInfo(pdf, n, infile) for n in range(pdf.numPages)]
+
+
+class PageInfo:
+    def __init__(self, pdf, pageno, infile):
+        self._pageno = pageno
+        self._infile = infile
+        self._pageinfo = _pdf_get_pageinfo(pdf, pageno, infile)
+
+    @property
+    def pageno(self):
+        return self._pageno
+
+    @property
+    def has_text(self):
+        return self._pageinfo['has_text']
+
+    @property
+    def width_inches(self):
+        return self._pageinfo['width_inches']
+
+    @property
+    def height_inches(self):
+        return self._pageinfo['height_inches']
+
+    @property
+    def width_pixels(self):
+        return int(round(self.width_inches * self.xres))
+
+    @property
+    def height_pixels(self):
+        return int(round(self.height_inches * self.yres))
+
+    @property
+    def rotation(self):
+        return self._pageinfo.get('rotate', None)
+
+    @rotation.setter
+    def rotation(self, value):
+        if value in (0, 90, 180, 270, 360, -90, -180, -270):
+            self._pageinfo['rotate'] = value
+        else:
+            raise ValueError("rotation must be a cardinal angle")
+
+    @property
+    def images(self):
+        return self._pageinfo['images']
+
+    @property
+    def xres(self):
+        return self._pageinfo.get('xres', None)
+
+    @property
+    def yres(self):
+        return self._pageinfo.get('yres', None)
+
+    @property
+    def userunit(self):
+        return self._pageinfo.get('userunit', None)
+
+    @property
+    def min_version(self):
+        if self.userunit is not None:
+            return '1.6'
+        else:
+            return '1.5'
+
+    def __repr__(self):
+        return (
+            '<PageInfo '
+            'pageno={} {}"x{}" rotation={} res={}x{} has_text={}>').format(
+            self.pageno, self.width_inches, self.height_inches,
+            self.rotation,
+            self.xres, self.yres, self.has_text
+        )
+
+
+class PdfInfo:
+    """Get summary information about a PDF
+    
+    """
+    def __init__(self, infile):
+        self._infile = infile
+        self._pages = _pdf_get_all_pageinfo(infile)
+        if fitz:
+            self._toc = fitz.Document(fspath(infile)).getToC()
+        else:
+            self._toc = []
+
+    @property
+    def pages(self):
+        return self._pages
+
+    @property
+    def min_version(self):
+        # The minimum PDF is the maximum version that any particular page needs
+        return max(page.min_version for page in self.pages)
+
+    @property
+    def has_userunit(self):
+        return any(page.userunit != 1.0 for page in self.pages)
+
+    @property
+    def filename(self):
+        if not isinstance(self._infile, (str, Path)):
+            raise NotImplementedError("can't get filename from stream")
+        return self._infile
+
+    @property
+    def table_of_contents(self):
+        return self._toc
+
+    def __getitem__(self, item):
+        return self._pages[item]
+
+    def __len__(self):
+        return len(self._pages)
+
+    def __repr__(self):
+        return "<PdfInfo('...'), page count={}>".format(len(self))
+
+    # def __getstate__(self):
+    #     state = {'_infile': self._infile}
+    #     return state
+    #
+    # def __setstate__(self, state):
+    #     self._infile = state['_infile']
+    #     self._pages = _pdf_get_all_pageinfo(self._infile)
+
+
+def main():
+    import argparse
+    parser = argparse.ArgumentParser()
+    parser.add_argument('infile')
+    args = parser.parse_args()
+    info = _pdf_get_all_pageinfo(args.infile)
+    from pprint import pprint
+    pprint(info)
+
+
+if __name__ == '__main__':
+    main()

src/ocrmypdf/pipeline.py

@@ -1,36 +1,47 @@
-#!/usr/bin/env python3
 # © 2016 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/>.
 
 from contextlib import suppress
-from tempfile import mkdtemp
-from functools import partial
+from shutil import copyfileobj
+from pathlib import Path
 import sys
 import os
-import re
 import shutil
-import warnings
-import multiprocessing
-import atexit
-import textwrap
+import re
+
 import img2pdf
-import logging
-import argparse
-
 import PyPDF2 as pypdf
-from PIL import Image
 
+from PIL import Image
 from ruffus import formatter, regex, Pipeline, suffix
 
 from .hocrtransform import HocrTransform
-from .pageinfo import pdf_get_all_pageinfo
-from .pdfa import generate_pdfa_ps, file_claims_pdfa
+from .pdfinfo import PdfInfo, Encoding, Colorspace
+from .pdfa import generate_pdfa_ps
 from .helpers import re_symlink, is_iterable_notstr, page_number
 from .exec import ghostscript, tesseract, qpdf
-from .exceptions import *
+from .lib import fitz
+from .exceptions import PdfMergeFailedError, UnsupportedImageFormatError, \
+    DpiError, PriorOcrFoundError, InputFileError
 from . import leptonica
 from . import PROGRAM_NAME, VERSION
 
 
+
 VECTOR_PAGE_DPI = 400
 
 # -------------
@@ -50,7 +61,12 @@ class JobContext:
     """
 
     def __init__(self):
-        self.pdfinfo = []
+        self.pdfinfo = None
+        self.options = None
+        self.work_folder = None
+
+    def generate_pdfinfo(self, infile):
+        self.pdfinfo = PdfInfo(infile)
 
     def get_pdfinfo(self):
         "What we know about the input PDF"
@@ -72,8 +88,8 @@ class JobContext:
         self.work_folder = work_folder
 
 
-from multiprocessing.managers import BaseManager
-class JobContextManager(BaseManager):
+from multiprocessing.managers import SyncManager
+class JobContextManager(SyncManager):
     pass
 
 
@@ -153,6 +169,22 @@ def triage_image_file(input_file, output_file, log, options):
         raise UnsupportedImageFormatError() from e
 
 
+def _pdf_guess_version(input_file, search_window=1024):
+    """Try to find version signature at start of file.
+    
+    Not robust enough to deal with appended files.
+
+    Returns empty string if not found, indicating file is probably not PDF.
+    """
+
+    with open(input_file, 'rb') as f:
+        signature = f.read(1024)
+    m = re.search(br'%PDF-(\d\.\d)', signature)
+    if m:
+        return m.group(1)
+    return ''
+
+
 def triage(
         input_file,
         output_file,
@@ -161,14 +193,12 @@ def triage(
 
     options = context.get_options()
     try:
-        with open(input_file, 'rb') as f:
-            signature = f.read(4)
-            if signature == b'%PDF':
-                if options.image_dpi:
-                    log.warning("Argument --image-dpi ignored because the "
-                                "input file is a PDF, not an image.")
-                re_symlink(input_file, output_file, log)
-                return
+        if _pdf_guess_version(input_file):
+            if options.image_dpi:
+                log.warning("Argument --image-dpi ignored because the "
+                            "input file is a PDF, not an image.")
+            re_symlink(input_file, output_file, log)
+            return
     except EnvironmentError as e:
         log.error(e)
         raise InputFileError() from e
@@ -176,43 +206,95 @@ def triage(
     triage_image_file(input_file, output_file, log, options)
 
 
-def repair_pdf(
+def repair_and_parse_pdf(
         input_file,
         output_file,
         log,
         context):
+    options = context.get_options()
+    if not options.skip_repair:
+        log.debug("Beginning qpdf repair...")
+        qpdf.repair(input_file, output_file, log)
+        log.debug("Repair OK; beginning parse...")
+    else:
+        re_symlink(input_file, output_file, log)
+
+    pdfinfo = PdfInfo(output_file)
+
+    if pdfinfo.has_userunit and options.output_type == 'pdfa':
+        log.error(
+            "This input file uses a PDF feature that is not supported "
+            "by Ghostscript, so you cannot use --output-type=pdfa for this "
+            "file. (Specifically, it uses the PDF-1.6 /UserUnit feature to "
+            "support very large or small page sizes, and Ghostscript cannot "
+            "output these files.)  Use --output-type=pdf instead."
+        )
+        raise InputFileError()
 
-    qpdf.repair(input_file, output_file, log)
-    pdfinfo = pdf_get_all_pageinfo(output_file)
     context.set_pdfinfo(pdfinfo)
     log.debug(pdfinfo)
 
 
 def get_pageinfo(input_file, context):
-    pageno = int(os.path.basename(input_file)[0:6]) - 1
+    "Get zero-based page info implied by filename, e.g. 000002.pdf -> 1"
+    pageno = page_number(input_file) - 1
     pageinfo = context.get_pdfinfo()[pageno]
     return pageinfo
 
 
 def get_page_dpi(pageinfo, options):
     "Get the DPI when nonsquare DPI is tolerable"
-    xres = max(pageinfo.get('xres', VECTOR_PAGE_DPI), options.oversample or 0)
-    yres = max(pageinfo.get('yres', VECTOR_PAGE_DPI), options.oversample or 0)
+    xres = max(pageinfo.xres or VECTOR_PAGE_DPI, options.oversample or 0)
+    yres = max(pageinfo.yres or VECTOR_PAGE_DPI, options.oversample or 0)
     return (float(xres), float(yres))
 
 
 def get_page_square_dpi(pageinfo, options):
-    "Get the DPI when we require xres == yres"
+    "Get the DPI when we require xres == yres, scaled to physical units"
+    xres = pageinfo.xres or 0
+    yres = pageinfo.yres or 0
+    userunit = pageinfo.userunit or 1
     return float(max(
-        pageinfo.get('xres', VECTOR_PAGE_DPI),
-        pageinfo.get('yres', VECTOR_PAGE_DPI),
+        (xres * userunit) or VECTOR_PAGE_DPI,
+        (yres * userunit) or VECTOR_PAGE_DPI,
+        options.oversample or 0))
+
+
+def get_canvas_square_dpi(pageinfo, options):
+    """Get the DPI when we require xres == yres, in Postscript units"""
+    return float(max(
+        (pageinfo.xres) or VECTOR_PAGE_DPI,
+        (pageinfo.yres) or VECTOR_PAGE_DPI,
         options.oversample or 0))
 
 
 def is_ocr_required(pageinfo, log, options):
-    page = pageinfo['pageno'] + 1
+    page = pageinfo.pageno + 1
     ocr_required = True
-    if not pageinfo['images']:
+
+    if pageinfo.has_text:
+        msg = "{0:4d}: page already has text! – {1}"
+
+        if not options.force_ocr and not options.skip_text:
+            log.error(msg.format(page,
+                                 "aborting (use --force-ocr to force OCR)"))
+            raise PriorOcrFoundError()
+        elif options.force_ocr:
+            log.info(msg.format(page,
+                                "rasterizing text and running OCR anyway"))
+            ocr_required = True
+        elif options.skip_text:
+            log.info(msg.format(page,
+                                "skipping all processing on this page"))
+            ocr_required = False
+    elif not pageinfo.images and not options.lossless_reconstruction:
+        # We found a page with no images and no text. That means it may
+        # have vector art that the user wants to OCR. If we determined
+        # lossless reconstruction is not possible then we have to rasterize
+        # the image. So if OCR is being forced, take that to mean YES, go 
+        # ahead and rasterize. If not forced, then pretend there's no text
+        # on the page at all so we don't lose anything.
+        # This could be made smarter by explicitly searching for vector art.
         if options.force_ocr and options.oversample:
             # The user really wants to reprocess this file
             log.info(
@@ -231,27 +313,13 @@ def is_ocr_required(pageinfo, log, options):
         else:
             log.info(
                 "{0:4d}: page has no images - "
-                "skipping all processing on this page".format(page))
+                "skipping all processing on this page to avoid losing detail. "
+                "Use --force-ocr if you wish to perform OCR on pages that "
+                "have vector content.".format(page))
             ocr_required = False
 
-    elif pageinfo['has_text']:
-        msg = "{0:4d}: page already has text! – {1}"
-
-        if not options.force_ocr and not options.skip_text:
-            log.error(msg.format(page,
-                                 "aborting (use --force-ocr to force OCR)"))
-            raise PriorOcrFoundError()
-        elif options.force_ocr:
-            log.info(msg.format(page,
-                                "rasterizing text and running OCR anyway"))
-            ocr_required = True
-        elif options.skip_text:
-            log.info(msg.format(page,
-                                "skipping all processing on this page"))
-            ocr_required = False
-
-    if ocr_required and options.skip_big and pageinfo['images']:
-        pixel_count = pageinfo['width_pixels'] * pageinfo['height_pixels']
+    if ocr_required and options.skip_big and pageinfo.images:
+        pixel_count = pageinfo.width_pixels * pageinfo.height_pixels
         if pixel_count > (options.skip_big * 1000000):
             ocr_required = False
             log.warning(
@@ -261,7 +329,7 @@ def is_ocr_required(pageinfo, log, options):
     return ocr_required
 
 
-def split_pages(
+def pre_split_pages(
         input_files,
         output_files,
         log,
@@ -287,20 +355,45 @@ def split_pages(
 
     pdfinfo = context.get_pdfinfo()
     npages = len(pdfinfo)
-    qpdf.split_pages(input_file, work_folder, npages)
 
-    from glob import glob
-    for filename in glob(os.path.join(work_folder, '*.page.pdf')):
-        pageinfo = get_pageinfo(filename, context)
+    # Ruffus needs to see a file for any task it generates, so create
+    # empty placeholders for every page.
+    for n in range(npages):
+        page = Path(work_folder) / '{0:06d}.presplit.pdf'.format(n + 1)
+        page.touch()
 
+
+def split_page(
+        placeholder_file,
+        output_file,
+        log,
+        context):
+    pageno = page_number(placeholder_file) - 1
+    input_pdf = context.get_pdfinfo().filename
+    qpdf.extract_page(input_pdf, output_file, pageno)
+
+
+def ocr_or_skip(
+        input_files,
+        output_files,
+        log,
+        context):
+    options = context.get_options()
+    work_folder = context.get_work_folder()
+    pdfinfo =  context.get_pdfinfo()
+
+    for input_file in input_files:
+        pageno = page_number(input_file) - 1
+        pageinfo = pdfinfo[pageno]
         alt_suffix = \
             '.ocr.page.pdf' if is_ocr_required(pageinfo, log, options) \
             else '.skip.page.pdf'
+
         re_symlink(
-            filename,
+            input_file,
             os.path.join(
                 work_folder,
-                os.path.basename(filename)[0:6] + alt_suffix),
+                os.path.basename(input_file)[0:6] + alt_suffix),
             log)
 
 
@@ -309,13 +402,14 @@ def rasterize_preview(
         output_file,
         log,
         context):
+    pageinfo = get_pageinfo(input_file, context)
+    options = context.get_options()
+    canvas_dpi = get_canvas_square_dpi(pageinfo, options) / 2
+    page_dpi = get_page_square_dpi(pageinfo, options) / 2
+
     ghostscript.rasterize_pdf(
-        input_file=input_file,
-        output_file=output_file,
-        xres=200,
-        yres=200,
-        raster_device='jpeggray',
-        log=log)
+        input_file, output_file, xres=canvas_dpi, yres=canvas_dpi,
+        raster_device='jpeggray', log=log, page_dpi=(page_dpi, page_dpi))
 
 
 def orient_page(
@@ -381,9 +475,9 @@ def orient_page(
         with open(output_file, 'wb') as out:
             writer.write(out)
 
-        pageno = int(os.path.basename(page_pdf)[0:6]) - 1
+        pageno = page_number(page_pdf) - 1
         pdfinfo = context.get_pdfinfo()
-        pdfinfo[pageno]['rotated'] = orient_conf.angle
+        pdfinfo[pageno].rotation = orient_conf.angle
         context.set_pdfinfo(pdfinfo)
 
 
@@ -396,26 +490,28 @@ def rasterize_with_ghostscript(
     pageinfo = get_pageinfo(input_file, context)
 
     device = 'png16m'  # 24-bit
-    if pageinfo['images']:
-        if all(image['comp'] == 1 for image in pageinfo['images']):
-            if all(image['bpc'] == 1 for image in pageinfo['images']):
+    if pageinfo.images:
+        if all(image.comp == 1 for image in pageinfo.images):
+            if all(image.bpc == 1 for image in pageinfo.images):
                 device = 'pngmono'
-            elif all(image['bpc'] > 1 and image['color'] == 'index'
-                     for image in pageinfo['images']):
+            elif all(image.bpc > 1 and image.color == Colorspace.index
+                     for image in pageinfo.images):
                 device = 'png256'
-            elif all(image['bpc'] > 1 and image['color'] == 'gray'
-                     for image in pageinfo['images']):
+            elif all(image.bpc > 1 and image.color == Colorspace.gray
+                     for image in pageinfo.images):
                 device = 'pnggray'
 
     log.debug("Rasterize {0} with {1}".format(
               os.path.basename(input_file), device))
 
     # Produce the page image with square resolution or else deskew and OCR
-    # will not work properly
-    dpi = get_page_square_dpi(pageinfo, options)
+    # will not work properly.
+    canvas_dpi = get_canvas_square_dpi(pageinfo, options)
+    page_dpi = get_page_square_dpi(pageinfo, options)
+
     ghostscript.rasterize_pdf(
-        input_file, output_file, xres=dpi, yres=dpi, raster_device=device,
-        log=log)
+        input_file, output_file, xres=canvas_dpi, yres=canvas_dpi,
+        raster_device=device, log=log, page_dpi=(page_dpi, page_dpi))
 
 
 def preprocess_remove_background(
@@ -430,11 +526,11 @@ def preprocess_remove_background(
 
     pageinfo = get_pageinfo(input_file, context)
 
-    if any(image['bpc'] > 1 for image in pageinfo['images']):
+    if any(image.bpc > 1 for image in pageinfo.images):
         leptonica.remove_background(input_file, output_file)
     else:
         log.info("{0:4d}: background removal skipped on mono page".format(
-            pageinfo['pageno']))
+            pageinfo.pageno))
         re_symlink(input_file, output_file, log)
 
 
@@ -475,7 +571,7 @@ def select_ocr_image(
         infiles,
         output_file,
         log,
-        contenxt):
+        context):
     """Select the image we send for OCR. May not be the same as the display
     image depending on preprocessing."""
 
@@ -498,6 +594,8 @@ def ocr_tesseract_hocr(
         tessconfig=options.tesseract_config,
         timeout=options.tesseract_timeout,
         pagesegmode=options.tesseract_pagesegmode,
+        user_words=options.user_words,
+        user_patterns=options.user_patterns,
         log=log
         )
 
@@ -521,8 +619,8 @@ def select_visible_page_image(
     image = next(ii for ii in infiles if ii.endswith(image_suffix))
 
     pageinfo = get_pageinfo(image, context)
-    if pageinfo['images'] and \
-            all(im['enc'] == 'jpeg' for im in pageinfo['images']):
+    if pageinfo.images and \
+            all(im['enc'] == 'jpeg' for im in pageinfo.images):
         log.debug('{:4d}: JPEG input -> JPEG output'.format(
             page_number(image)))
         # If all images were JPEGs originally, produce a JPEG as output
@@ -568,7 +666,7 @@ def select_image_layer(
         # as it accurately describes the image. It would be possible to
         # resample the image at this stage back to non-square DPI to more
         # closely resemble the input, except that the hocr renderer does not
-        # understand non-square DPI. The tess4 renderer would be fine.
+        # understand non-square DPI. The sandwich renderer would be fine.
         dpi = get_page_square_dpi(pageinfo, options)
         layout_fun = img2pdf.get_fixed_dpi_layout_fun((dpi, dpi))
 
@@ -593,8 +691,8 @@ def render_hocr_page(
 
     hocrtransform = HocrTransform(hocr, dpi)
     hocrtransform.to_pdf(output_file, imageFileName=None,
-                         showBoundingboxes=False, invisibleText=True)
-
+                         showBoundingboxes=False, invisibleText=True,
+                         interwordSpaces=True)
 
 def flatten_groups(groups):
     for obj in groups:
@@ -618,8 +716,8 @@ def render_hocr_debug_page(
 
     hocrtransform = HocrTransform(hocr, dpi)
     hocrtransform.to_pdf(output_file, imageFileName=None,
-                         showBoundingboxes=True, invisibleText=False)
-
+                         showBoundingboxes=True, invisibleText=False,
+                         interwordSpaces=True)
 
 def combine_layers(
         infiles,
@@ -631,8 +729,8 @@ def combine_layers(
     image = next(ii for ii in flatten_groups(infiles)
                  if ii.endswith('.image-layer.pdf'))
 
-    pdf_text = pypdf.PdfFileReader(open(text, "rb"))
-    pdf_image = pypdf.PdfFileReader(open(image, "rb"))
+    pdf_text = pypdf.PdfFileReader(text)
+    pdf_image = pypdf.PdfFileReader(image)
 
     page_text = pdf_text.getPage(0)
 
@@ -677,7 +775,7 @@ def combine_layers(
 
     if rotation != 0:
         log.info("{0:4d}: rotating image layer {1} degrees".format(
-            page_number(image), rotation, tx, ty))
+            page_number(image), rotation))
 
     try:
         page_text.mergeRotatedScaledTranslatedPage(
@@ -689,6 +787,12 @@ def combine_layers(
     pdf_output = pypdf.PdfFileWriter()
     pdf_output.addPage(page_text)
 
+    # If the input was scaled, re-apply the scaling
+    pageinfo = get_pageinfo(text, context)
+    if pageinfo.userunit != 1:
+        page_text[pypdf.generic.NameObject('/UserUnit')] = pageinfo.userunit
+        pdf_output._header = b'%PDF-1.6'  # Hack header to correct version
+
     with open(output_file, "wb") as out:
         pdf_output.write(out)
 
@@ -722,6 +826,8 @@ def ocr_tesseract_and_render_pdf(
         tessconfig=options.tesseract_config,
         timeout=options.tesseract_timeout,
         pagesegmode=options.tesseract_pagesegmode,
+        user_words=options.user_words,
+        user_patterns=options.user_patterns,
         log=log)
 
 
@@ -750,6 +856,8 @@ def ocr_tesseract_textonly_pdf(
         tessconfig=options.tesseract_config,
         timeout=options.tesseract_timeout,
         pagesegmode=options.tesseract_pagesegmode,
+        user_words=options.user_words,
+        user_patterns=options.user_patterns,
         log=log)
 
 
@@ -779,10 +887,17 @@ def get_pdfmark(base_pdf, options):
     if options.subject:
         pdfmark['/Subject'] = options.subject
 
-    pdfmark['/Creator'] = '{0} {1} / Tesseract OCR{2} {3}'.format(
-            PROGRAM_NAME, VERSION,
-            '+PDF' if options.pdf_renderer == 'tesseract' else '',
-            tesseract.version())
+    if options.pdf_renderer == 'tesseract':
+        renderer_tag = 'OCR+PDF'
+    elif options.pdf_renderer == 'sandwich':
+        renderer_tag = 'OCR-PDF'
+    else:
+        renderer_tag = 'OCR'
+
+    pdfmark['/Creator'] = '{0} {1} / Tesseract {2} {3}'.format(
+        PROGRAM_NAME, VERSION,
+        renderer_tag,
+        tesseract.version())
     return pdfmark
 
 
@@ -809,12 +924,21 @@ def skip_page(
     re_symlink(input_file, output_file, log)
 
 
-def merge_pages_ghostscript(
+def _merge_pages_common(
         input_files_groups,
         output_file,
         log,
         context):
-    options = context.get_options()
+    """Determine ordered list of PDF pages to merge. Returns PDF from which
+    metadata should be drawn if present (for qpdf)."""
+
+    input_files = list(f for f in flatten_groups(input_files_groups)
+                       if not f.endswith('.txt'))
+    metadata_file = next(
+        (ii for ii in input_files if ii.endswith('.repaired.pdf')), None)
+    if metadata_file in input_files:
+        input_files.remove(metadata_file)
+
     def input_file_order(s):
         '''Sort order: All rendered pages followed
         by their debug page, if any, followed by Postscript stub.
@@ -823,18 +947,40 @@ def merge_pages_ghostscript(
         right that way.'''
         if s.endswith('.ps'):
             return 99999999
-        key = int(os.path.basename(s)[0:6]) * 10
+        key = page_number(s) * 10
         if 'debug' in os.path.basename(s):
             key += 1
         return key
 
-    input_files = (f for f in flatten_groups(input_files_groups)
-                   if not f.endswith('.txt'))
     pdf_pages = sorted(input_files, key=input_file_order)
     log.debug("Final pages: " + "\n".join(pdf_pages))
+
+    return pdf_pages, metadata_file
+
+
+def merge_pages_ghostscript(
+        input_files_groups,
+        output_file,
+        log,
+        context):
+    options = context.get_options()
+    pdf_pages, _ = _merge_pages_common(
+        input_files_groups, output_file, log, context)
+    input_pdfinfo = context.get_pdfinfo()
     ghostscript.generate_pdfa(
-        pdf_pages, output_file, options.pdfa_image_compression,
-        log, options.jobs or 1)
+        pdf_version=input_pdfinfo.min_version,
+        pdf_pages=pdf_pages,
+        output_file=output_file + '_toc.pdf',
+        compression=options.pdfa_image_compression,
+        log=log,
+        threads=options.jobs or 1,
+        pdfa_part=('1' if options.output_type == 'pdfa-1' else '2'))
+    if fitz:
+        doc = fitz.Document(output_file + '_toc.pdf')
+        doc.setToC(input_pdfinfo.table_of_contents)
+        doc.save(output_file)
+    else:
+        os.replace(output_file + '_toc.pdf', output_file)
 
 
 def merge_pages_qpdf(
@@ -843,23 +989,8 @@ def merge_pages_qpdf(
         log,
         context):
     options = context.get_options()
-
-    input_files = list(f for f in flatten_groups(input_files_groups)
-                       if not f.endswith('.txt'))
-    metadata_file = next(
-        (ii for ii in input_files if ii.endswith('.repaired.pdf')))
-    input_files.remove(metadata_file)
-
-    def input_file_order(s):
-        '''Sort order: All rendered pages followed
-        by their debug page.'''
-        key = int(os.path.basename(s)[0:6]) * 10
-        if 'debug' in os.path.basename(s):
-            key += 1
-        return key
-
-    pdf_pages = sorted(input_files, key=input_file_order)
-    log.debug("Final pages: " + "\n".join(pdf_pages))
+    pdf_pages, metadata_file = _merge_pages_common(
+        input_files_groups, output_file, log, context)
 
     reader_metadata = pypdf.PdfFileReader(metadata_file)
     pdfmark = get_pdfmark(reader_metadata, options)
@@ -868,6 +999,8 @@ def merge_pages_qpdf(
     first_page = pypdf.PdfFileReader(pdf_pages[0])
 
     writer = pypdf.PdfFileWriter()
+    # copy version from source
+    writer._header = b'%PDF-' + _pdf_guess_version(pdf_pages[0])
     writer.appendPagesFromReader(first_page)
     writer.addMetadata(pdfmark)
     writer_file = pdf_pages[0].replace('.pdf', '.metadata.pdf')
@@ -876,7 +1009,39 @@ def merge_pages_qpdf(
 
     pdf_pages[0] = writer_file
 
-    qpdf.merge(pdf_pages, output_file)
+    qpdf.merge(input_files=pdf_pages, output_file=output_file,
+               min_version=context.get_pdfinfo().min_version,
+               log=log)
+
+
+def merge_pages_mupdf(
+        input_files_groups,
+        output_file,
+        log,
+        context):
+    assert fitz
+
+    options = context.get_options()
+
+    pdf_pages, metadata_file = _merge_pages_common(
+        input_files_groups, output_file, log, context)
+
+    doc = fitz.Document()
+
+    reader_metadata = pypdf.PdfFileReader(metadata_file)
+    pdfmark = get_pdfmark(reader_metadata, options)
+    pdfmark['/Producer'] = 'PyMuPDF ' + fitz.version[0]
+    pymupdf_metadata = {k[1:].lower() : v for k, v in pdfmark.items()}
+
+    for pdf_page in pdf_pages:
+        page = fitz.open(pdf_page)
+        doc.insertPDF(page)
+
+    metadata = fitz.open(metadata_file)
+    toc = metadata.getToC(simple=False)
+    doc.setToC(toc)
+    doc.setMetadata(pymupdf_metadata)
+    doc.save(output_file, garbage=4, deflate=True)
 
 
 def merge_sidecars(
@@ -884,7 +1049,6 @@ def merge_sidecars(
         output_file,
         log,
         context):
-    options = context.get_options()
     pdfinfo = context.get_pdfinfo()
 
     txt_files = [None] * len(pdfinfo)
@@ -895,21 +1059,30 @@ def merge_sidecars(
             txt_files[idx] = infile
 
     def write_pages(stream):
-        for page_number, txt_file in enumerate(txt_files):
-            if page_number != 0:
+        for page_num, txt_file in enumerate(txt_files):
+            if page_num != 0:
                 stream.write('\f')  # Form feed between pages
             if txt_file:
-                with open(txt_file, 'r') as in_:
-                    stream.write(in_.read())
+                with open(txt_file, 'r', encoding="utf-8") as in_:
+                    txt = in_.read()
+                    # Tesseract v4 alpha started adding form feeds in 
+                    # commit aa6eb6b
+                    # No obvious way to detect what binaries will do this, so
+                    # for consistency just ignore its form feeds and insert our
+                    # own
+                    if txt.endswith('\f'):
+                        stream.write(txt[:-1])
+                    else:
+                        stream.write(txt)
             else:
                 stream.write('[OCR skipped on page {}]'.format(
-                        page_number + 1))
+                        page_num + 1))
 
     if output_file == '-':
         write_pages(sys.stdout)
         sys.stdout.flush()
     else:
-        with open(output_file, 'w', encoding='utf-8') as out:
+        with open(output_file, 'w', encoding="utf-8") as out:
             write_pages(out)
 
 
@@ -920,13 +1093,16 @@ def copy_final(
         context):
     input_file = next((ii for ii in input_files if ii.endswith('.pdf')))
 
-    if output_file == '-':
-        from shutil import copyfileobj
-        with open(input_file, 'rb') as input_stream:
+    with open(input_file, 'rb') as input_stream:
+        if output_file == '-':
             copyfileobj(input_stream, sys.stdout.buffer)
             sys.stdout.flush()
-    else:
-        shutil.copy(input_file, output_file)
+        else:
+            # At this point we overwrite the output_file specified by the user
+            # use copyfileobj because then we use open() to create the file and 
+            # get the appropriate umask, ownership, etc.
+            with open(output_file, 'wb') as output_stream:
+                copyfileobj(input_stream, output_stream)
 
 
 def build_pipeline(options, work_folder, log, context):
@@ -940,8 +1116,8 @@ def build_pipeline(options, work_folder, log, context):
         output=os.path.join(work_folder, 'origin.pdf'),
         extras=[log, context])
 
-    task_repair_pdf = main_pipeline.transform(
-        task_func=repair_pdf,
+    task_repair_and_parse_pdf = main_pipeline.transform(
+        task_func=repair_and_parse_pdf,
         input=task_triage,
         filter=suffix('.pdf'),
         output='.repaired.pdf',
@@ -949,16 +1125,31 @@ def build_pipeline(options, work_folder, log, context):
         extras=[log, context])
 
     # Split (kwargs for split seems to be broken, so pass plain args)
-    task_split_pages = main_pipeline.split(
-        split_pages,
-        task_repair_pdf,
-        os.path.join(work_folder, '*.page.pdf'),
+    task_pre_split_pages = main_pipeline.split(
+        pre_split_pages,
+        task_repair_and_parse_pdf,
+        os.path.join(work_folder, '*.presplit.pdf'),
+        extras=[log, context])
+
+    task_split_pages = main_pipeline.transform(
+        task_func=split_page,
+        input=task_pre_split_pages,
+        filter=suffix('.presplit.pdf'),
+        output='.page.pdf',
+        output_dir=work_folder,
+        extras=[log, context])
+
+    task_ocr_or_skip = main_pipeline.split(
+        ocr_or_skip,
+        task_split_pages,
+        [os.path.join(work_folder, '*.ocr.page.pdf'),
+         os.path.join(work_folder, '*.skip.page.pdf')],
         extras=[log, context])
 
     # Rasterize preview
     task_rasterize_preview = main_pipeline.transform(
         task_func=rasterize_preview,
-        input=task_split_pages,
+        input=task_ocr_or_skip,
         filter=suffix('.page.pdf'),
         output='.preview.jpg',
         output_dir=work_folder,
@@ -968,7 +1159,7 @@ def build_pipeline(options, work_folder, log, context):
     # Orient
     task_orient_page = main_pipeline.collate(
         task_func=orient_page,
-        input=[task_split_pages, task_rasterize_preview],
+        input=[task_ocr_or_skip, task_rasterize_preview],
         filter=regex(r".*/(\d{6})(\.ocr|\.skip)(?:\.page\.pdf|\.preview\.jpg)"),
         output=os.path.join(work_folder, r'\1\2.oriented.pdf'),
         extras=[log, context])
@@ -1021,8 +1212,6 @@ def build_pipeline(options, work_folder, log, context):
         extras=[log, context])
     task_ocr_tesseract_hocr.graphviz(fillcolor='"#00cc66"')
     task_ocr_tesseract_hocr.active_if(options.pdf_renderer == 'hocr')
-    if tesseract.v4():
-        task_ocr_tesseract_hocr.jobs_limit(2)  # Uses multi-core on its own
 
     task_select_visible_page_image = main_pipeline.collate(
         task_func=select_visible_page_image,
@@ -1044,7 +1233,7 @@ def build_pipeline(options, work_folder, log, context):
     task_select_image_layer.graphviz(
         fillcolor='"#00cc66"', shape='diamond')
     task_select_image_layer.active_if(
-        options.pdf_renderer == 'hocr' or options.pdf_renderer == 'tess4')
+        options.pdf_renderer == 'hocr' or options.pdf_renderer == 'sandwich')
 
     task_render_hocr_page = main_pipeline.transform(
         task_func=render_hocr_page,
@@ -1074,9 +1263,7 @@ def build_pipeline(options, work_folder, log, context):
                 os.path.join(work_folder, r'\1.text.txt')],
         extras=[log, context])
     task_ocr_tesseract_textonly_pdf.graphviz(fillcolor='"#ff69b4"')
-    task_ocr_tesseract_textonly_pdf.active_if(options.pdf_renderer == 'tess4')
-    if tesseract.v4():
-        task_ocr_tesseract_textonly_pdf.jobs_limit(2)
+    task_ocr_tesseract_textonly_pdf.active_if(options.pdf_renderer == 'sandwich')
 
     task_combine_layers = main_pipeline.collate(
         task_func=combine_layers,
@@ -1087,7 +1274,8 @@ def build_pipeline(options, work_folder, log, context):
         output=os.path.join(work_folder, r'\1.rendered.pdf'),
         extras=[log, context])
     task_combine_layers.graphviz(fillcolor='"#00cc66"')
-    task_combine_layers.active_if(options.pdf_renderer == 'hocr' or options.pdf_renderer == 'tess4')
+    task_combine_layers.active_if(options.pdf_renderer == 'hocr' or 
+                                  options.pdf_renderer == 'sandwich')
 
     # Tesseract OCR+PDF
     task_ocr_tesseract_and_render_pdf = main_pipeline.collate(
@@ -1099,17 +1287,15 @@ def build_pipeline(options, work_folder, log, context):
         extras=[log, context])
     task_ocr_tesseract_and_render_pdf.graphviz(fillcolor='"#66ccff"')
     task_ocr_tesseract_and_render_pdf.active_if(options.pdf_renderer == 'tesseract')
-    if tesseract.v4():
-        task_ocr_tesseract_and_render_pdf.jobs_limit(2)  # Uses multi-core
 
     # PDF/A
     task_generate_postscript_stub = main_pipeline.transform(
         task_func=generate_postscript_stub,
-        input=task_repair_pdf,
+        input=task_repair_and_parse_pdf,
         filter=formatter(r'\.repaired\.pdf'),
         output=os.path.join(work_folder, 'pdfa.ps'),
         extras=[log, context])
-    task_generate_postscript_stub.active_if(options.output_type == 'pdfa')
+    task_generate_postscript_stub.active_if(options.output_type.startswith('pdfa'))
 
 
     # Bypass valve
@@ -1131,7 +1317,8 @@ def build_pipeline(options, work_folder, log, context):
                task_generate_postscript_stub],
         output=os.path.join(work_folder, 'merged.pdf'),
         extras=[log, context])
-    task_merge_pages_ghostscript.active_if(options.output_type == 'pdfa')
+    task_merge_pages_ghostscript.active_if(
+        options.output_type.startswith('pdfa'))
 
     task_merge_pages_qpdf = main_pipeline.merge(
         task_func=merge_pages_qpdf,
@@ -1139,10 +1326,23 @@ def build_pipeline(options, work_folder, log, context):
                task_render_hocr_debug_page,
                task_skip_page,
                task_ocr_tesseract_and_render_pdf,
-               task_repair_pdf],
+               task_repair_and_parse_pdf],
         output=os.path.join(work_folder, 'merged.pdf'),
         extras=[log, context])
-    task_merge_pages_qpdf.active_if(options.output_type == 'pdf')
+    task_merge_pages_qpdf.active_if(
+        options.output_type == 'pdf' and not fitz)
+
+    task_merge_pages_mupdf = main_pipeline.merge(
+        task_func=merge_pages_mupdf,
+        input=[task_combine_layers,
+               task_render_hocr_debug_page,
+               task_skip_page,
+               task_ocr_tesseract_and_render_pdf,
+               task_repair_and_parse_pdf],
+        output=os.path.join(work_folder, 'merged.pdf'),
+        extras=[log, context])
+    task_merge_pages_mupdf.active_if(
+        options.output_type == 'pdf' and fitz)
 
     task_merge_sidecars = main_pipeline.merge(
         task_func=merge_sidecars,
@@ -1154,8 +1354,10 @@ def build_pipeline(options, work_folder, log, context):
     task_merge_sidecars.active_if(options.sidecar)
 
     # Finalize
-    task_copy_final = main_pipeline.merge(
+    main_pipeline.merge(
         task_func=copy_final,
-        input=[task_merge_pages_ghostscript, task_merge_pages_qpdf],
+        input=[task_merge_pages_ghostscript,
+               task_merge_pages_mupdf,
+               task_merge_pages_qpdf],
         output=options.output_file,
         extras=[log, context])

test_requirements.txt

@@ -1,4 +1,5 @@
-pytest >= 3.0
+pytest >= 3.2
 pytest-helpers-namespace
 pytest-xdist
-pytest-cov
+pytest-cov
+pytest-timeout

tests/cache/2400dpi/__-l__eng__-c__textonly_pdf=1__000001.ocr.png__000001.text__pdf__txt/stderr.bin

@@ -0,0 +1 @@
+Tesseract Open Source OCR Engine v4.0.0-beta.1 with Leptonica

tests/cache/2400dpi/__-l__eng__-c__textonly_pdf=1__000001.ocr.png__000001.text__pdf__txt/txt.bin

@@ -0,0 +1,25 @@
+a la Waterman
+
+4 ons linzen
+
+3 liter water
+
+3 uien
+
+bloem, boter
+
+2 kopjes melk
+
+laurier, kruidnagel, kerrie, zout
+
+De linzgen wassen en in -l liter kokend wa-
+ter 1 dag laten weken, 2 liter water bij
+de linzen voegen, zonder het water waarin
+ze geweekt zijn af te gieten, De helft van
+de uien bakken met laurier en Kruicdnagel.
+Alle uien, kerrie en gout bij de linzen
+voegen, Alles aan de kook brengen,. Van de
+bloem met boter en melk een papje maken en
+verder afmaken met de soep, Als de linzen
+gaar Zijn is de soep klaar.
+

tests/cache/aspect/__-l__eng__-c__textonly_pdf=1__000001.ocr.png__000001.text__pdf__txt/stderr.bin

@@ -0,0 +1,3 @@
+Tesseract Open Source OCR Engine v4.0.0-beta.1 with Leptonica
+Too few characters. Skipping this page
+OSD: Weak margin (0.00) for 24 blob text block, but using orientation anyway: 0

tests/cache/aspect/__-l__eng__-c__textonly_pdf=1__000001.ocr.png__000001.text__pdf__txt/txt.bin

@@ -0,0 +1,4 @@
+ 
+
+This should be a perfect circle:
+

tests/cache/aspect/__-l__eng__000001.image__000001.rendered__pdf__txt/stderr.bin

@@ -0,0 +1,3 @@
+Tesseract Open Source OCR Engine v4.0.0-beta.1 with Leptonica
+Too few characters. Skipping this page
+OSD: Weak margin (0.00) for 24 blob text block, but using orientation anyway: 0

tests/cache/aspect/__-l__eng__000001.image__000001.rendered__pdf__txt/txt.bin

@@ -0,0 +1,4 @@
+ 
+
+This should be a perfect circle:
+

tests/cache/aspect/__-l__eng__000001.ocr.png__000001__hocr__txt/hocr.bin

@@ -0,0 +1,27 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
+    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+ <head>
+  <title></title>
+<meta http-equiv="Content-Type" content="text/html;charset=utf-8" />
+  <meta name='ocr-system' content='tesseract 4.0.0-beta.1' />
+  <meta name='ocr-capabilities' content='ocr_page ocr_carea ocr_par ocr_line ocrx_word'/>
+</head>
+<body>
+  <div class='ocr_page' id='page_1' title='image "/var/folders/37/78_114p552q16vv6vmgm5kr00000gn/T/com.github.ocrmypdf.bp6c363s/000001.ocr.png"; bbox 0 0 1000 800; ppageno 0'>
+   <div class='ocr_carea' id='block_1_1' title="bbox 296 96 704 504">
+    <p class='ocr_par' id='par_1_1' lang='eng' title="bbox 296 96 704 504">
+     <span class='ocr_line' id='line_1_1' title="bbox 296 96 704 504; baseline 0 296; x_size 169.33333; x_descenders 42.333332; x_ascenders 42.333336"><span class='ocrx_word' id='word_1_1' title='bbox 296 96 704 504; x_wconf 95'><strong> </strong></span> 
+     </span>
+    </p>
+   </div>
+   <div class='ocr_carea' id='block_1_2' title="bbox 150 592 841 622">
+    <p class='ocr_par' id='par_1_2' lang='eng' title="bbox 150 592 841 622">
+     <span class='ocr_line' id='line_1_2' title="bbox 150 592 841 622; baseline 0 -6; x_size 30; x_descenders 6; x_ascenders 8"><span class='ocrx_word' id='word_1_2' title='bbox 150 592 230 616; x_wconf 96'><strong>This</strong></span> <span class='ocrx_word' id='word_1_3' title='bbox 260 592 384 616; x_wconf 95'><strong>should</strong></span> <span class='ocrx_word' id='word_1_4' title='bbox 413 592 449 616; x_wconf 95'><strong>be</strong></span> <span class='ocrx_word' id='word_1_5' title='bbox 479 600 493 616; x_wconf 95'><strong>a</strong></span> <span class='ocrx_word' id='word_1_6' title='bbox 523 592 668 622; x_wconf 95'><strong>perfect</strong></span> <span class='ocrx_word' id='word_1_7' title='bbox 698 592 841 616; x_wconf 55'><strong>circle:</strong></span> 
+     </span>
+    </p>
+   </div>
+  </div>
+ </body>
+</html>

tests/cache/aspect/__-l__eng__000001.ocr.png__000001__hocr__txt/stderr.bin

@@ -0,0 +1,3 @@
+Tesseract Open Source OCR Engine v4.0.0-beta.1 with Leptonica
+Too few characters. Skipping this page
+OSD: Weak margin (0.00) for 24 blob text block, but using orientation anyway: 0

tests/cache/aspect/__-l__eng__000001.ocr.png__000001__hocr__txt/txt.bin

@@ -0,0 +1,4 @@
+ 
+
+This should be a perfect circle:
+

tests/cache/cardinal/__-l__eng__-c__textonly_pdf=1__000001.ocr.png__000001.text__pdf__txt/stderr.bin

@@ -0,0 +1 @@
+Tesseract Open Source OCR Engine v4.0.0-beta.1 with Leptonica

tests/cache/cardinal/__-l__eng__-c__textonly_pdf=1__000001.ocr.png__000001.text__pdf__txt/txt.bin

@@ -0,0 +1,123 @@
+The LinnSequencer
+32 Track MIDI Sequence Recorder
+
+The LinnSequencer is a state-of-the-art composition and performance tool for the professional musician. It is
+
+extremely powerful, yet amazingly simple to learn and use. It’s many remarkable features include:
+
+¢ Operation is similar to multi-track tape recorder with PLAY, STOP, RECORD, FAST
+FORWARD, REWIND, and LOCATE controls.
+
+e Each of the 100 sequences contains 32 simultaneous, polyphonic tracks. Each track may
+be assigned to one of 16 MIDI channels. Simultaneously plays up to 16 polyphonic
+
+synthesizers!
+
+¢ Ultra-fast 3%” disk drive stores complex songs in seconds and holds over 110,000 notes
+
+per disk!
+
+¢ One or all tracks may be TRANSPOSED at the touch of a key.
+e Exclusive real-time ERASE function makes editing FAST.
+* Exclusive REPEAT function automatically repeats any held notes at a pre-selected
+
+rhythmic value.
+
+¢ TIMING CORRECTION works during playback and operates without ‘chopping’ notes.
+
+¢ Optional SMPTE time code synchronization.
+
+© Optional remote control.
+
+Recording a Sequence
+
+To record a sequence, simply press RECORD and PLAY,
+then play your MIDI keyboard in time to the Sequencer’s
+click track. When the sequence loops back around to bar 1,
+you’ ll hear what you played—only all timing errors will be
+
+corrected! (Timing correction may be adjusted or defeated).
+
+Any additional notes played will be added into the track
+— existing notes are not erased while recording!
+
+FAST FORWARD, REWIND, and LOCATE controls
+may be used at any time to quickly access any location in
+your sequence for spot-recording. To overdub a new part,
+select a different track and start recording—while you
+record, the first track will play in perfect sync (unless you
+MUTE it, or SOLO another track). In this way, up to 32
+tracks may be overdubbed! All MIDI effects are recorded
+including pitch bend, modulation, velocity, aftertouch,
+sustain pedal, and program changes!
+
+Editing
+
+To erase a wrong note, simply hold ERASE and press
+the note to be erased just before it plays in the sequence—
+when played back, it will be gone. Notes may also be
+
+added, erased, or changed using the SINGLE STEP func-
+tion. To overdub notes at specific points within a sequence,
+
+Additional Features
+
+simply use LOCATE, FAST FORWARD, or REWIND to
+find the desired bar number, then start recording.
+
+The INSERT/COPY function allows you to move bars
+from one location to another—in the same sequence or a
+different one. For example, you might insert a copy of the
+first verse between the second chorus and the bridge.
+DELETE BARS operates the same way to remove
+unwanted sections,
+
+Creating a Song
+
+One way to create a song is to record each track all the
+way through (up to 999 bars). Another way is to record
+each basic section (verse, chorus, etc.) in individual
+sequences, then use the CREATE SONG function to “chain”
+them together. CREATE SONG will then automatically
+copy all the parts into a new sequence. If desired, you can
+even set the last few bars to repeat infinitely, for a fadeout.
+
+Composition Without Compromise
+
+The technology you use should never be so complex that
+it interferes with the creative process. That’s precisely why
+the LinnSequencer is designed to let you compose, record
+and edit while devoting your undivided attention to your
+music. See your Linn dealer today for a demonstration!
+
+* Simple, easy to learn operation—the 32 character LCD display clearly guides you through all operations. If needed, the
+
+HELP button displays additional explanations.
+
+* Non-destructive recording—existing notes are not erased while recording.
+¢ Two FOOTSWITCH INPUTS may be assigned to remotely control many of the commonly used functions, including
+
+ERASE, REPEAT, PLAY/STOP, or LOCATE.
+
+¢ Iwo TRIGGER OUTPUTS may be programmed to output pulses at any selected note value.
+
+© Will sync to standard LinnDrum or Linn 9000 sync tone.
+
+© Utilizes ultra high-speed, 8 MHz 80186 16 bit computer internally for FAST operation.
+* TEMPO may be specified in BEATS-PER-MINUTE or FRAMES-PER-BEAT at 24, 25, or 30 frames per second,
+
+(even drop frame!)
+
+¢ TEMPO may be entered numerically, adjustable in tenths of a Beat-Per-Minute increments, or by tapping quarter notes
+
+on the TAP TEMPO button.
+
+¢ TEMPO CHANGES may be programmed into a sequence, with smooth transitions if desired.
+¢ Any TIME SIGNATURE may be used, and may be changed within a song.
+
+linn
+Linn Electronics, Inc.
+
+18720 Oxnard Street, Tarzana, CA 91356
+(818) 708-8131 TELEX #298949 LINN UR
+

tests/cache/cardinal/__-l__eng__-c__textonly_pdf=1__000002.ocr.png__000002.text__pdf__txt/stderr.bin

@@ -0,0 +1 @@
+Tesseract Open Source OCR Engine v4.0.0-beta.1 with Leptonica

tests/cache/cardinal/__-l__eng__-c__textonly_pdf=1__000002.ocr.png__000002.text__pdf__txt/txt.bin

@@ -0,0 +1,123 @@
+The LinnSequencer
+32 Track MIDI Sequence Recorder
+
+The LinnSequencer is a state-of-the-art composition and performance tool for the professional musician. It is
+
+extremely powerful, yet amazingly simple to learn and use. It’s many remarkable features include:
+
+¢ Operation is similar to multi-track tape recorder with PLAY, STOP, RECORD, FAST
+FORWARD, REWIND, and LOCATE controls.
+
+e Each of the 100 sequences contains 32 simultaneous, polyphonic tracks. Each track may
+be assigned to one of 16 MIDI channels. Simultaneously plays up to 16 polyphonic
+
+synthesizers!
+
+¢ Ultra-fast 3%” disk drive stores complex songs in seconds and holds over 110,000 notes
+
+per disk!
+
+¢ One or all tracks may be TRANSPOSED at the touch of a key.
+e Exclusive real-time ERASE function makes editing FAST.
+* Exclusive REPEAT function automatically repeats any held notes at a pre-selected
+
+rhythmic value.
+
+¢ TIMING CORRECTION works during playback and operates without ‘chopping’ notes.
+
+¢ Optional SMPTE time code synchronization.
+
+© Optional remote control.
+
+Recording a Sequence
+
+To record a sequence, simply press RECORD and PLAY,
+then play your MIDI keyboard in time to the Sequencer’s
+click track. When the sequence loops back around to bar 1,
+you’ ll hear what you played—only all timing errors will be
+
+corrected! (Timing correction may be adjusted or defeated).
+
+Any additional notes played will be added into the track
+— existing notes are not erased while recording!
+
+FAST FORWARD, REWIND, and LOCATE controls
+may be used at any time to quickly access any location in
+your sequence for spot-recording. To overdub a new part,
+select a different track and start recording—while you
+record, the first track will play in perfect sync (unless you
+MUTE it, or SOLO another track). In this way, up to 32
+tracks may be overdubbed! All MIDI effects are recorded
+including pitch bend, modulation, velocity, aftertouch,
+sustain pedal, and program changes!
+
+Editing
+
+To erase a wrong note, simply hold ERASE and press
+the note to be erased just before it plays in the sequence—
+when played back, it will be gone. Notes may also be
+
+added, erased, or changed using the SINGLE STEP func-
+tion. To overdub notes at specific points within a sequence,
+
+Additional Features
+
+simply use LOCATE, FAST FORWARD, or REWIND to
+find the desired bar number, then start recording.
+
+The INSERT/COPY function allows you to move bars
+from one location to another—in the same sequence or a
+different one. For example, you might insert a copy of the
+first verse between the second chorus and the bridge.
+DELETE BARS operates the same way to remove
+unwanted sections,
+
+Creating a Song
+
+One way to create a song is to record each track all the
+way through (up to 999 bars). Another way is to record
+each basic section (verse, chorus, etc.) in individual
+sequences, then use the CREATE SONG function to “chain”
+them together. CREATE SONG will then automatically
+copy all the parts into a new sequence. If desired, you can
+even set the last few bars to repeat infinitely, for a fadeout.
+
+Composition Without Compromise
+
+The technology you use should never be so complex that
+it interferes with the creative process. That’s precisely why
+the LinnSequencer is designed to let you compose, record
+and edit while devoting your undivided attention to your
+music. See your Linn dealer today for a demonstration!
+
+* Simple, easy to learn operation—the 32 character LCD display clearly guides you through all operations. If needed, the
+
+HELP button displays additional explanations.
+
+* Non-destructive recording—existing notes are not erased while recording.
+¢ Two FOOTSWITCH INPUTS may be assigned to remotely control many of the commonly used functions, including
+
+ERASE, REPEAT, PLAY/STOP, or LOCATE.
+
+¢ Iwo TRIGGER OUTPUTS may be programmed to output pulses at any selected note value.
+
+© Will sync to standard LinnDrum or Linn 9000 sync tone.
+
+© Utilizes ultra high-speed, 8 MHz 80186 16 bit computer internally for FAST operation.
+* TEMPO may be specified in BEATS-PER-MINUTE or FRAMES-PER-BEAT at 24, 25, or 30 frames per second,
+
+(even drop frame!)
+
+¢ TEMPO may be entered numerically, adjustable in tenths of a Beat-Per-Minute increments, or by tapping quarter notes
+
+on the TAP TEMPO button.
+
+¢ TEMPO CHANGES may be programmed into a sequence, with smooth transitions if desired.
+¢ Any TIME SIGNATURE may be used, and may be changed within a song.
+
+linn
+Linn Electronics, Inc.
+
+18720 Oxnard Street, Tarzana, CA 91356
+(818) 708-8131 TELEX #298949 LINN UR
+

tests/cache/cardinal/__-l__eng__-c__textonly_pdf=1__000003.ocr.png__000003.text__pdf__txt/stderr.bin

@@ -0,0 +1 @@
+Tesseract Open Source OCR Engine v4.0.0-beta.1 with Leptonica

tests/cache/cardinal/__-l__eng__-c__textonly_pdf=1__000003.ocr.png__000003.text__pdf__txt/txt.bin

@@ -0,0 +1,123 @@
+The LinnSequencer
+32 Track MIDI Sequence Recorder
+
+The LinnSequencer is a state-of-the-art composition and performance tool for the professional musician. It is
+
+extremely powerful, yet amazingly simple to learn and use. It’s many remarkable features include:
+
+¢ Operation is similar to multi-track tape recorder with PLAY, STOP, RECORD, FAST
+FORWARD, REWIND, and LOCATE controls.
+
+e Each of the 100 sequences contains 32 simultaneous, polyphonic tracks. Each track may
+be assigned to one of 16 MIDI channels. Simultaneously plays up to 16 polyphonic
+
+synthesizers!
+
+¢ Ultra-fast 3%” disk drive stores complex songs in seconds and holds over 110,000 notes
+
+per disk!
+
+¢ One or all tracks may be TRANSPOSED at the touch of a key.
+e Exclusive real-time ERASE function makes editing FAST.
+* Exclusive REPEAT function automatically repeats any held notes at a pre-selected
+
+rhythmic value.
+
+¢ TIMING CORRECTION works during playback and operates without ‘chopping’ notes.
+
+¢ Optional SMPTE time code synchronization.
+
+© Optional remote control.
+
+Recording a Sequence
+
+To record a sequence, simply press RECORD and PLAY,
+then play your MIDI keyboard in time to the Sequencer’s
+click track. When the sequence loops back around to bar 1,
+you’ ll hear what you played—only all timing errors will be
+
+corrected! (Timing correction may be adjusted or defeated).
+
+Any additional notes played will be added into the track
+— existing notes are not erased while recording!
+
+FAST FORWARD, REWIND, and LOCATE controls
+may be used at any time to quickly access any location in
+your sequence for spot-recording. To overdub a new part,
+select a different track and start recording—while you
+record, the first track will play in perfect sync (unless you
+MUTE it, or SOLO another track). In this way, up to 32
+tracks may be overdubbed! All MIDI effects are recorded
+including pitch bend, modulation, velocity, aftertouch,
+sustain pedal, and program changes!
+
+Editing
+
+To erase a wrong note, simply hold ERASE and press
+the note to be erased just before it plays in the sequence—
+when played back, it will be gone. Notes may also be
+
+added, erased, or changed using the SINGLE STEP func-
+tion. To overdub notes at specific points within a sequence,
+
+Additional Features
+
+simply use LOCATE, FAST FORWARD, or REWIND to
+find the desired bar number, then start recording.
+
+The INSERT/COPY function allows you to move bars
+from one location to another—in the same sequence or a
+different one. For example, you might insert a copy of the
+first verse between the second chorus and the bridge.
+DELETE BARS operates the same way to remove
+unwanted sections,
+
+Creating a Song
+
+One way to create a song is to record each track all the
+way through (up to 999 bars). Another way is to record
+each basic section (verse, chorus, etc.) in individual
+sequences, then use the CREATE SONG function to “chain”
+them together. CREATE SONG will then automatically
+copy all the parts into a new sequence. If desired, you can
+even set the last few bars to repeat infinitely, for a fadeout.
+
+Composition Without Compromise
+
+The technology you use should never be so complex that
+it interferes with the creative process. That’s precisely why
+the LinnSequencer is designed to let you compose, record
+and edit while devoting your undivided attention to your
+music. See your Linn dealer today for a demonstration!
+
+* Simple, easy to learn operation—the 32 character LCD display clearly guides you through all operations. If needed, the
+
+HELP button displays additional explanations.
+
+* Non-destructive recording—existing notes are not erased while recording.
+¢ Two FOOTSWITCH INPUTS may be assigned to remotely control many of the commonly used functions, including
+
+ERASE, REPEAT, PLAY/STOP, or LOCATE.
+
+¢ Iwo TRIGGER OUTPUTS may be programmed to output pulses at any selected note value.
+
+© Will sync to standard LinnDrum or Linn 9000 sync tone.
+
+© Utilizes ultra high-speed, 8 MHz 80186 16 bit computer internally for FAST operation.
+* TEMPO may be specified in BEATS-PER-MINUTE or FRAMES-PER-BEAT at 24, 25, or 30 frames per second,
+
+(even drop frame!)
+
+¢ TEMPO may be entered numerically, adjustable in tenths of a Beat-Per-Minute increments, or by tapping quarter notes
+
+on the TAP TEMPO button.
+
+¢ TEMPO CHANGES may be programmed into a sequence, with smooth transitions if desired.
+¢ Any TIME SIGNATURE may be used, and may be changed within a song.
+
+linn
+Linn Electronics, Inc.
+
+18720 Oxnard Street, Tarzana, CA 91356
+(818) 708-8131 TELEX #298949 LINN UR
+

tests/cache/cardinal/__-l__eng__-c__textonly_pdf=1__000004.ocr.png__000004.text__pdf__txt/stderr.bin

@@ -0,0 +1 @@
+Tesseract Open Source OCR Engine v4.0.0-beta.1 with Leptonica

tests/cache/cardinal/__-l__eng__-c__textonly_pdf=1__000004.ocr.png__000004.text__pdf__txt/txt.bin

@@ -0,0 +1,123 @@
+The LinnSequencer
+32 Track MIDI Sequence Recorder
+
+The LinnSequencer is a state-of-the-art composition and performance tool for the professional musician. It is
+
+extremely powerful, yet amazingly simple to learn and use. It’s many remarkable features include:
+
+¢ Operation is similar to multi-track tape recorder with PLAY, STOP, RECORD, FAST
+FORWARD, REWIND, and LOCATE controls.
+
+e Each of the 100 sequences contains 32 simultaneous, polyphonic tracks. Each track may
+be assigned to one of 16 MIDI channels. Simultaneously plays up to 16 polyphonic
+
+synthesizers!
+
+¢ Ultra-fast 3%” disk drive stores complex songs in seconds and holds over 110,000 notes
+
+per disk!
+
+¢ One or all tracks may be TRANSPOSED at the touch of a key.
+e Exclusive real-time ERASE function makes editing FAST.
+* Exclusive REPEAT function automatically repeats any held notes at a pre-selected
+
+rhythmic value.
+
+¢ TIMING CORRECTION works during playback and operates without ‘chopping’ notes.
+
+¢ Optional SMPTE time code synchronization.
+
+© Optional remote control.
+
+Recording a Sequence
+
+To record a sequence, simply press RECORD and PLAY,
+then play your MIDI keyboard in time to the Sequencer’s
+click track. When the sequence loops back around to bar 1,
+you’ ll hear what you played—only all timing errors will be
+
+corrected! (Timing correction may be adjusted or defeated).
+
+Any additional notes played will be added into the track
+— existing notes are not erased while recording!
+
+FAST FORWARD, REWIND, and LOCATE controls
+may be used at any time to quickly access any location in
+your sequence for spot-recording. To overdub a new part,
+select a different track and start recording—while you
+record, the first track will play in perfect sync (unless you
+MUTE it, or SOLO another track). In this way, up to 32
+tracks may be overdubbed! All MIDI effects are recorded
+including pitch bend, modulation, velocity, aftertouch,
+sustain pedal, and program changes!
+
+Editing
+
+To erase a wrong note, simply hold ERASE and press
+the note to be erased just before it plays in the sequence—
+when played back, it will be gone. Notes may also be
+
+added, erased, or changed using the SINGLE STEP func-
+tion. To overdub notes at specific points within a sequence,
+
+Additional Features
+
+simply use LOCATE, FAST FORWARD, or REWIND to
+find the desired bar number, then start recording.
+
+The INSERT/COPY function allows you to move bars
+from one location to another—in the same sequence or a
+different one. For example, you might insert a copy of the
+first verse between the second chorus and the bridge.
+DELETE BARS operates the same way to remove
+unwanted sections,
+
+Creating a Song
+
+One way to create a song is to record each track all the
+way through (up to 999 bars). Another way is to record
+each basic section (verse, chorus, etc.) in individual
+sequences, then use the CREATE SONG function to “chain”
+them together. CREATE SONG will then automatically
+copy all the parts into a new sequence. If desired, you can
+even set the last few bars to repeat infinitely, for a fadeout.
+
+Composition Without Compromise
+
+The technology you use should never be so complex that
+it interferes with the creative process. That’s precisely why
+the LinnSequencer is designed to let you compose, record
+and edit while devoting your undivided attention to your
+music. See your Linn dealer today for a demonstration!
+
+* Simple, easy to learn operation—the 32 character LCD display clearly guides you through all operations. If needed, the
+
+HELP button displays additional explanations.
+
+* Non-destructive recording—existing notes are not erased while recording.
+¢ Two FOOTSWITCH INPUTS may be assigned to remotely control many of the commonly used functions, including
+
+ERASE, REPEAT, PLAY/STOP, or LOCATE.
+
+¢ Iwo TRIGGER OUTPUTS may be programmed to output pulses at any selected note value.
+
+© Will sync to standard LinnDrum or Linn 9000 sync tone.
+
+© Utilizes ultra high-speed, 8 MHz 80186 16 bit computer internally for FAST operation.
+* TEMPO may be specified in BEATS-PER-MINUTE or FRAMES-PER-BEAT at 24, 25, or 30 frames per second,
+
+(even drop frame!)
+
+¢ TEMPO may be entered numerically, adjustable in tenths of a Beat-Per-Minute increments, or by tapping quarter notes
+
+on the TAP TEMPO button.
+
+¢ TEMPO CHANGES may be programmed into a sequence, with smooth transitions if desired.
+¢ Any TIME SIGNATURE may be used, and may be changed within a song.
+
+linn
+Linn Electronics, Inc.
+
+18720 Oxnard Street, Tarzana, CA 91356
+(818) 708-8131 TELEX #298949 LINN UR
+

tests/cache/cardinal/__-l__osd__--psm__0__000001.ocr.preview.jpg__stdout/stderr.bin

@@ -0,0 +1 @@
+Warning. Invalid resolution 0 dpi. Using 70 instead.

tests/cache/cardinal/__-l__osd__--psm__0__000001.ocr.preview.jpg__stdout/stdout.bin

@@ -0,0 +1,6 @@
+Page number: 0
+Orientation in degrees: 0
+Rotate: 0
+Orientation confidence: 18.91
+Script: Latin
+Script confidence: 0.97

tests/cache/cardinal/__-l__osd__--psm__0__000002.ocr.preview.jpg__stdout/stderr.bin

@@ -0,0 +1 @@
+Warning. Invalid resolution 0 dpi. Using 70 instead.

tests/cache/cardinal/__-l__osd__--psm__0__000002.ocr.preview.jpg__stdout/stdout.bin

@@ -0,0 +1,6 @@
+Page number: 0
+Orientation in degrees: 90
+Rotate: 270
+Orientation confidence: 17.21
+Script: Latin
+Script confidence: 1.06

tests/cache/cardinal/__-l__osd__--psm__0__000003.ocr.preview.jpg__stdout/stderr.bin

@@ -0,0 +1 @@
+Warning. Invalid resolution 0 dpi. Using 70 instead.

tests/cache/cardinal/__-l__osd__--psm__0__000003.ocr.preview.jpg__stdout/stdout.bin

@@ -0,0 +1,6 @@
+Page number: 0
+Orientation in degrees: 180
+Rotate: 180
+Orientation confidence: 14.79
+Script: Latin
+Script confidence: 2.87

tests/cache/cardinal/__-l__osd__--psm__0__000004.ocr.preview.jpg__stdout/stderr.bin

@@ -0,0 +1 @@
+Warning. Invalid resolution 0 dpi. Using 70 instead.

tests/cache/cardinal/__-l__osd__--psm__0__000004.ocr.preview.jpg__stdout/stdout.bin

@@ -0,0 +1,6 @@
+Page number: 0
+Orientation in degrees: 270
+Rotate: 90
+Orientation confidence: 19.19
+Script: Latin
+Script confidence: 0.58