pyfontaine analyses fonts for their language and character/glyph-set support.
It has a straightforward pythonic set definition format:
Additional definitions are downloaded from the Extensis, font-config and Unicode websites during installation, and can be updated without reinstalling.
Adding your own definitions is easy. All files in the internals directory are auto-detected, so just add definition files there.
macOS:
First, install Python and the pip python package manager. This is installed by default with homebrew python, so, install homebrew, then install the neccessary depedencies (PyICU) as follows:
brew install python icu4c pkg-config; export PATH="/usr/local/opt/icu4c/bin:/usr/local/opt/icu4c/sbin:$PATH"; export PKG_CONFIG_PATH="$PKG_CONFIG_PATH:/usr/local/opt/icu4c/lib/pkgconfig"; export CC="$(which gcc)" CXX="$(which g++)"; pip3 install --no-binary=:pyicu: pyicu;
Debian:
apt-get install libicu-dev
Install the latest release easily with pip:
pip3 install fontaine --user;
To install the latest development version:
pip3 install git+https://github.com/googlefonts/pyfontaine.git#egg=fontaine --user;
Given a list of space separated font filenames, it returns a report with some general metadata and a language support analysis:
pyfontaine font.ttf;
Given a list of space separated unicode characters, or unicode values, it returns a list of character sets that include that character:
pyfontaine 0x0061; pyfontaine ğ ø ∂;
Similarly you can find out if a font supports specific characters by also giving the filename:
pyfontaine U+C480 U+C481 font.ttf;
To output font reports in various formats:
pyfontaine --xml font.ttf; pyfontaine --json font.ttf; pyfontaine --csv font.ttf; pyfontaine --wiki font.ttf;
The --wiki format produces a MediaWiki table (example)
To only show character sets from different collections:
pyfontaine --collections all font.ttf; # default pyfontaine --collections pyfontaine font.ttf; pyfontaine --collections uniblocks font.ttf; pyfontaine --collections extensis font.ttf; pyfontaine --collections fontconfig font.ttf; pyfontaine --collections cldr font.ttf; pyfontaine --collections subsets font.ttf;
To only show specific character sets:
pyfontaine --set africaan,adobe_latin_3 font.ttf;
To print a list of all the missing unicode values from each set:
pyfontaine --missing --set adobe_latin_3 font.ttf; To output visualisations of the coverage using `Hilbert curves <http://en.wikipedia.org/wiki/Hilbert_curve>`__ (thanks for the idea, `Øyvind 'pippin' Kolås <http://github.com/hodefoting>`__!): pyfontaine --show_hilbert font.ttf; ls -l coverage_pngs/; The PNG files are stored in a new directory, ``coverage_pngs``, under the current directory.
You can update remote collections data when you are online:
pyfontaine --update-data 1;
It has a python module called fontaine
To release a new version on PyPI, create and push a new git tag with a version number following the [semver](https://www.semver.org) versioning scheme.
Then set up a ~/.pypirc
file:
[distutils] index-servers=pypi [pypi] repository = https://pypi.python.org/pypi [server-login] username = user password = password
Then run:
python setup.py build; python setup.py sdist upload;
Your contributions under the GPLv3 are welcome!
pyfontaine is a python reimplementation of Fontaine by Ed Trager, and has been made by Vitaly Volkov, Dave Crossland, Mikhail Kashkin and Felipe Sanches.
We would like to thank some upstream projects that make pyfontaine even more useful:
- Thomas Phinney for the WebINK Character Sets
- Behdad Esfabod for the font-config languages definitions
- Unicode Consortium for the Unicode Blocks
- Mac OS X requires the XCode Command Line Tools to be installed
- fonttools (common) or freetype-py (fast)
- lxml
- PyICU
- simpleHilbertCurve
- matplotlib
- tabulate
- requests