/tools

The Standard Ebooks toolset for producing our ebook files.

Primary LanguagePythonOtherNOASSERTION

About

A collection of tools Standard Ebooks uses to produce its ebooks, including basic setup of ebooks, text processing, and build tools.

Installing this toolset using pipx makes the se command line executable available. Its various commands are described below, or you can use se help to list them.

Installation

The toolset requires Python >= 3.6.

To install the toolset locally for development and debugging, see Installation for Developers.

Optionally, install Ace and the se build --check command will automatically run it as part of the checking process.

Ubuntu 20.04 (Trusty) users

# Install some pre-flight dependencies.
sudo apt install -y calibre default-jre git python3-dev python3-pip python3-venv

# Install pipx.
python3 -m pip install --user pipx
python3 -m pipx ensurepath

# Install the toolset.
pipx install standardebooks

Optional: Install shell completions

# Install ZSH completions.
sudo ln -s $HOME/.local/pipx/venvs/standardebooks/lib/python3.*/site-packages/se/completions/zsh/_se /usr/share/zsh/vendor-completions/_se && hash -rf && compinit

# Install Bash completions.
sudo ln -s $HOME/.local/pipx/venvs/standardebooks/lib/python3.*/site-packages/se/completions/bash/se /usr/share/bash-completion/completions/se

# Install Fish completions.
sudo ln -s $HOME/.local/pipx/venvs/standardebooks/lib/python3.*/site-packages/se/completions/fish/se $HOME/.config/fish/completions/se.fish

Fedora users

# Install some pre-flight dependencies.
sudo dnf install calibre git java-1.8.0-openjdk python3-devel

# Install pipx.
python3 -m pip install --user pipx
python3 -m pipx ensurepath

# Install the toolset.
pipx install standardebooks

Optional: Install shell completions

# Install ZSH completions.
sudo ln -s $HOME/.local/pipx/venvs/standardebooks/lib/python3.*/site-packages/se/completions/zsh/_se /usr/share/zsh/vendor-completions/_se && hash -rf && compinit

# Install Bash completions.
sudo ln -s $HOME/.local/pipx/venvs/standardebooks/lib/python3.*/site-packages/se/completions/bash/se /usr/share/bash-completion/completions/se

# Install Fish completions.
sudo ln -s $HOME/.local/pipx/venvs/standardebooks/lib/python3.*/site-packages/se/completions/fish/se $HOME/.config/fish/completions/se.fish

macOS users (up to macOS 11)

These instructions were tested on macOS 10.15 to 11, on Intel macs.

  1. Install the Homebrew package manager. Or, if you already have it installed, make sure it’s up to date:

    brew update

  2. Install dependencies:

    # Install some pre-flight dependencies.
brew install cairo calibre git openjdk pipx python
pipx ensurepath
sudo ln -sfn /usr/local/opt/openjdk/libexec/openjdk.jdk /Library/Java/JavaVirtualMachines/openjdk.jdk

# Install the toolset.
pipx install standardebooks

# Optional: Bash users who have set up bash-completion via brew can install tab completion.
ln -s $HOME/.local/pipx/venvs/standardebooks/lib/python3.*/site-packages/se/completions/bash/se $(brew --prefix)/etc/bash_completion.d/se

# Optional: Fish users can install tab completion.
ln -s $HOME/.local/pipx/venvs/standardebooks/lib/python3.*/site-packages/se/completions/fish/se $HOME/.config/fish/completions/se.fish

OpenBSD 6.6 Users

These instructions were tested on OpenBSD 6.6, but may also work on the 6.5 release as well.

  1. Create a text file to feed into pkg_add called ~/standard-ebooks-packages. It should contain the following:

    py3-pip--
py3-virtualenv--
py3-gitdb--
jdk--%11
calibre--
git--

  2. Install dependencies using doas pkg_add -ivl ~/standard-ebooks-packages. Follow linking instructions provided by pkg_add to save keystrokes, unless you want to have multiple python versions and pip versions. In my case, I ran doas ln -sf /usr/local/bin/pip3.7 /usr/local/bin/pip.

  3. Add ~/.local/bin to your path.

  4. Run pip install --user pipx

  5. If you’re using ksh from base and have already added ~/.local/bin, you can skip pipx ensurepath because this step is for bash users.

  6. The rest of the process is similar to that used on other platforms:

    # Install the toolset.
pipx install standardebooks

Installation for developers

If you want to work on the toolset source, it’s helpful to tell pipx to install the package in “editable” mode. This will allow you to edit the source of the package live and see changes immediately, without having to uninstall and re-install the package.

To do that, follow the general installation instructions above; but instead of doing pipx install standardebooks, do the following:

git clone https://github.com/standardebooks/tools.git
pipx install --editable ./tools

Now the se binary is in your path, and any edits you make to source files in the tools/ directory are immediately reflected when executing the binary.

Running commands on the entire corpus

As a developer, it’s often useful to run an se command like se lint or se build on the entire corpus for testing purposes. This can be very time-consuming in a regular invocation (like se lint /path/to/ebook/repos/*), because each argument is processed sequentially. Instead of waiting for a single invocation to process all of its arguments sequentially, use GNU Parallel to start multiple invocations in parallel, with each one processing a single argument. For example:

# Slow, each argument is processed in sequence
se lint /path/to/ebook/repos/*

# Fast, multiple invocations each process a single argument in parallel
export COLUMNS; parallel --keep-order se lint ::: /path/to/ebook/repos/*

The toolset tries to detect when it’s being invoked from parallel, and it adjusts its output to accomodate.

We export COLUMNS because se lint needs to know the width of the terminal so that it can format its tabular output correctly. We pass the --keep-order flag to output results in the order we passed them in, which is useful if comparing the results of multiple runs.

Linting with pylint and mypy

Before we can use pylint or mypy on the toolset source, we have to inject them into the venv pipx created for the standardebooks package:

pipx inject standardebooks pylint==2.8.2 mypy==0.812

Then make sure to call the pylint and mypy binaries that pipx installed in the standardebooks venv, not any other globally-installed binaries:

cd /path/to/tools/repo
$HOME/.local/pipx/venvs/standardebooks/bin/pylint se

Testing with pytest

Similar to pylint, the pytest command can be injected into the venv pipx created for the standardebooks package:

pipx inject standardebooks pytest==6.2.4

The tests are executed by calling pytest from the top level or your tools repo:

cd /path/to/tools/repo
$HOME/.local/pipx/venvs/standardebooks/bin/pytest

Adding tests

Tests are added under the tests directory. Most of the tests are based around the idea of having “golden” output files. Each command is run against a set of input files and then the resulting output files are compared against the resulting golden files. The test fails if the output files do not match the golden files. The data files can be found in the tests/data directory.

A custom test flag --save-golden-files has been added to automatically update the the golden files for the tests (in an out directory for the command).

The usual test development process is:

  1. Update in files with new test data and/or change the command implementation.
  2. Run pytest and see some tests fail.
  3. Run pytest --save-golden-files and then diff the data directory to ensure that the out files are as expected.
  4. Commit changes (including new out contents).

Another custom test flag --save-new-draft is also available. This flag is used to update the book skeleton, generated by the se create-draft command, that is used as input for the other tests. Whenever the draft contents change (e.g. edits to the core.css file) the tests/data/draft files should be updated by calling pytest --save-new-draft.

Code style

  • In general we follow a relaxed version of PEP 8. In particular, we use tabs instead of spaces, and there is no line length limit.

  • Always use the regex module instead of the re module.

Help wanted

We need volunteers to take the lead on the following goals:

  • Add more test cases to the test framework.

  • Figure out if it’s possible to install Bash/ZSH completions using setup.py, without root; this may not be possible?

  • Writing installation instructions for Bash and ZSH completions for MacOS.

  • Currently we install the whole Calibre package, which is very big, but it’s only used to convert epub to azw3. Can we inline Calibre’s azw3 conversion code in the ./vendor/ directory, to avoid having to install the entire package as a big dependency?

Tool descriptions

  • se british2american

    Try to convert British quote style to American quote style in DIRECTORY/src/epub/text/.

    Quotes must already be typogrified using the se typogrify tool.

    This script isn’t perfect; proofreading is required, especially near closing quotes near to em-dashes.

  • se build

    Build an ebook from a Standard Ebook source directory.

  • se build-images

    Build ebook cover and titlepage images in a Standard Ebook source directory and place the output in DIRECTORY/src/epub/images/.

  • se build-manifest

    Generate the <manifest> element for the given Standard Ebooks source directory and write it to the ebook’s metadata file.

  • se build-spine

    Generate the <spine> element for the given Standard Ebooks source directory and write it to the ebook’s metadata file.

  • se build-title

    Generate the title of an XHTML file based on its headings and update the file’s <title> element.

  • se build-toc

    Generate the table of contents for the ebook’s source directory and update the ToC file.

  • se clean

    Prettify and canonicalize individual XHTML, SVG, or CSS files, or all XHTML, SVG, or CSS files in a source directory.

  • se compare-versions

    Use Firefox to render and compare XHTML files in an ebook repository. Run on a dirty repository to visually compare the repository’s dirty state with its clean state. If a file renders differently, place screenshots of the new, original, and diff (if available) renderings in the current working directory. A file called diff.html is created to allow for side-by-side comparisons of original and new files.

  • se create-draft

    Create a skeleton of a new Standard Ebook.

  • se dec2roman

    Convert a decimal number to a Roman numeral.

  • se extract-ebook

    Extract an .epub, .mobi, or .azw3 ebook into ./FILENAME.extracted/ or a target directory.

  • se find-mismatched-dashes

    Find words with mismatched dashes in a set of XHTML files. For example, extra-physical in one file and extraphysical in another.

  • se find-mismatched-diacritics

    Find words with mismatched diacritics in a set of XHTML files. For example, cafe in one file and café in another.

  • se find-unusual-characters

    Find characters outside a nominal expected range in a set of XHTML files. This can be useful to find transcription mistakes and mojibake.

  • se help

    List available SE commands.

  • se hyphenate

    Insert soft hyphens at syllable breaks in an XHTML file.

  • se interactive-replace

    Perform an interactive search and replace on a list of files using Python-flavored regex. The view is scrolled using the arrow keys, with alt to scroll by page in any direction. Basic Emacs (default) or Vim style navigation is available. The following actions are possible: (y) Accept replacement. (n) Reject replacement. (a) Accept all remaining replacements in this file. (r) Reject all remaining replacements in this file. (c) Center on match. (q) Save this file and quit.

  • se lint

    Check for various Standard Ebooks style errors.

  • se make-url-safe

    Make a string URL-safe.

  • se modernize-spelling

    Modernize spelling of some archaic words, and replace words that may be archaically compounded with a dash to a more modern spelling. For example, replace ash-tray with ashtray.

  • se prepare-release

    Calculate work word count, insert release date if not yet set, and update modified date and revision number.

  • se recompose-epub

    Recompose a Standard Ebooks source directory into a single HTML5 file, and print to standard output.

  • se renumber-endnotes

    Renumber all endnotes and noterefs sequentially from the beginning.

  • se roman2dec

    Convert a Roman numeral to a decimal number.

  • se semanticate

    Apply some scriptable semantics rules from the Standard Ebooks semantics manual to a Standard Ebook source directory.

  • se shift-endnotes

    Increment or decrement the specified endnote and all following endnotes by 1 or a specified amount.

  • se split-file

    Split an XHTML file into many files at all instances of <!--se:split-->, and include a header template for each file.

  • se titlecase

    Convert a string to titlecase.

  • se typogrify

    Apply some scriptable typography rules from the Standard Ebooks typography manual to a Standard Ebook source directory.

  • se unicode-names

    Display Unicode code points, descriptions, and links to more details for each character in a string. Useful for differentiating between different flavors of spaces, dashes, and invisible characters like word joiners.

  • se word-count

    Count the number of words in an HTML file and optionally categorize by length.

  • se xpath

    Print the results of an xpath expression evaluated against a set of XHTML files. The default namespace is removed.

What a Standard Ebooks source directory looks like

Many of these tools act on Standard Ebooks source directories. Such directories have a consistent minimal structure:

.
|-images/
|--cover.jpg
|--cover.source.jpg
|--cover.svg
|--titlepage.svg
|-src/
|--epub/
|---css/
|----core.css
|----local.css
|----se.css
|---images/
|----cover.svg
|----titlepage.svg
|---text/
|----colophon.xhtml
|----imprint.xhtml
|----titlepage.xhtml
|----uncopyright.xhtml
|---content.opf
|---onix.xml
|---toc.xhtml
|--META-INF/
|---container.xml
|--mimetype
|-LICENSE.md

./images/ contains source images for the cover and titlepages, as well as ebook-specific source images. Source images should be in their maximum available resolution, then compressed and placed in ./src/epub/images/ for distribution.

./src/epub/ contains the actual epub files.