An interface for viewing regulations data. This project combines all of the data from a parsed regulation and generates navigable, accessible HTML, complete with associated information.
This repository is part of a larger project. To read about it, please see http://cfpb.github.io/eregulations/.
- Navigability: Instead of the monolithic text usually used to represent regulations, proper indentation, spacing, etc., this presents the text in a clean, readable form.
- Inline Interpretations: Associated interpretations are displayed right along side the paragraph/section/appendix they are interpreting.
- Definitions: Defined terms are highlighted in the regulation text; clicking them opens the term's definition in the sidebar.
- Internal Citations: Citations to other parts of the same regulation are clickable links, allowing easy navigation within a regulation
- Alternate Versions of the Regulation: Quick access to versions of the regulation by browsing effective dates, or typing in an arbitrary date is provided.
- Comparing Versions: A "Diff-view" is provided to highlight additions and deletions between versions of the regulation.
- Section-by-Section Analysis: Analysis for regulation paragraphs/sections/appendices are provided in the sidebar. This analysis comes from Federal Register notice preambles.
- Additional Formatting Concerns: Tables, subscripts, notes, and even source code found within a regulation are rendered as appropriate markup.
- Responsive Design: The application design is responsive, adjusting to the device and screen size of the user.
This application lives in two worlds, roughly translating to a Python, Django app, and a Backbone, Javascript app, which communicate through the Django templates. To get started quickly, we provide a default, pre-compiled version of the Javascript assets.
Requirements are retrieved and/or build automatically via buildout (see below).
- zc-buildout - Tool used for building the application and handling dependencies
- lxml - Used to convert strings into XML for processing
- requests - Client library for reading data from an API
If running tests:
- mock - makes constructing mock objects/functions easy
- django-nose - provides nosetest as a test runner
- nose-exclude - allows certain directories to be excluded from nose
- nose-testconfig - pass configuration data to tests; used to configure selenium tests
- coverage - provides test-coverage metrics
- selenium - functional testing via a headless web browser
Buildout is a simple tool for building and distributing python applications quickly. We use it to get a version of the API up and running without needing all of the fuss usually associated with setting up Django. Just run
$ pip install zc.buildout
$ buildout
After downloading the internet, you'll notice that some helpful scripts are
located in bin
, including bin/django
and bin/test
. The
latter will run our test suite while the former is equivalent to running
manage.py in a traditional Django environment.
With that, you can start the development server:
$ ./bin/django runserver
For most tweaks, you will simply need to run the Sphinx documentation builder again.
$ ./bin/sphinx-build -b dirhtml -d docs/_build/doctrees/ docs/ docs/_build/dirhtml/
The output will be in docs/_build/dirhtml
.
If you are adding new modules, you may need to re-run the skeleton build script first:
$ rm docs/regulations*.rst
$ ./bin/sphinx-apidoc -F -o docs regulations
The application code in JavaScript uses Backbone.js as a foundation, though in some non-standard ways. If you plan to do work on this layer, it is recommended that you acquaint yourself with this starter documentation.
The front end of the site uses a number of JavaScript libraries and frameworks to create the unique experience of navigating and reading a regulation, as you can see at http://consumerfinance.gov/eregulations. If you'd like to modify the JavaScript layer, you should set up the build and testing environment.
If you run the application with env = "built"
in your local_settings.py
and would like to use the UI as it ships with this project, you can skip this.
The application's UI itself uses a number of dependencies that you can see in package.json and bower.json. To start, we are going to be concerned with the foundations of the environment:
- npm, a package manager to install dependencies in the development environment: https://npmjs.org/
- Grunt, a task runner that modules to build and run tests depend on: http://gruntjs.com/
- Bower, a utility to fetch dependencies for the: UI http://bower.io/
First we will need npm. npm ships with Node.js. If you don't already have it installed, there are a few ways to get it.
- You can grab and install a binary or installer from http://nodejs.org/download/
- If you're on OS X, you can use Homebrew if that's your thing
- If you are using Ubuntu, the default apt-get package is out of date. Do:
sudo add-apt-repository ppa:chris-lea/node.js
sudo apt-get update
sudo apt-get install nodejs
If you receive an error about add-apt-repository
not being found, do:
sudo apt-get install python-software-properties
You will need to install the Grunt command line interface and Bower globally using npm.
cd regulations-site
npm install -g grunt-cli bower
The rest of the dependencies you will need are managed by Bower and npm. Do:
npm install
bower install
The UI uses RequireJS for script loading and modularity in the UI code. The configuration for RequireJS modules is needed in a couple of different locations for testing and codebase compression. There is a script in the repository that takes care of most of this for you.
Run ./require.sh
from the root of the repository.
If you need to make changes to modules in your instance of the application, edit require.paths.json and require.shim.json accordingly. These files map to the paths
and shim
objects in the RequireJS configuration object. For more information: http://requirejs.org/docs/api.html#config
In the root of the repository, copy example-config.json
to config.json
and edit as necessary. Grunt depends on these settings to carry out tasks.
testURL
is the environment that you would like tests to run on.frontEndPath
is the path to the root of your codebase where thecss/
andjs/
directories are.testPath
is the path to the functional test specs.
There are a number of tasks configured in Gruntfile.js. On the last lines, you will find tasks that group subtasks into common goals. Running grunt build
will run unit, functional and lint tests, and compress static assets. Its recommended that you run this task before deploying changes.
The Grunt build will run a suite of Selenium tests written in Python and a small suite of Mocha.js unit tests. All tests run in Sauce Labs. These tests run as part of the grunt build
tasks. To use these, a little extra environment setup is required.
After you create a Sauce Labs account:
- In your bash config (probably
~/.bash_profile
), define two variables:$SAUCE_USERNAME
and$SAUCE_ACCESS_KEY
which house your username and access key from Sauce Labs. - If you want to test a local or otherwise not publically available environment, download and run Sauce Connect. If you do need Sauce Connect, you will need to start it before running tests/Grunt builds.
- Be sure that the Django server is running in the environment you want to test.
- They also require having the environment serving data from
dummy_api/
. To start the dummy API, from the root of your repo, run./dummy_api/start.sh 0.0.0.0:8282
. - The tests run using nose. If you wish to run the tests outside of the Grunt environment, you may by running
nosetests regulations/uitests/*.py
from the root of the repo.
- To run unit tests individually:
regulations/static/regulations/js/unittests/sauce_unit_tests.sh http://url.of/test/site
from the root of the repo. - Unit tests do not require running the dummy API.
- You may also run the unit tests locally with no additional configuration by loading the following URL in a web browser:
http://your.site/static/regulations/js/unittests/runner.html
.