@datagraphics/baker
is a build tool by and for the Los Angeles Times. The Times uses it to build the static pages published at latimes.com/projects. You can use it however you'd like.
An example of how The Times puts the package to use is available at datadesk/baker-example-page-template.
Requirements
Installation
npm install -D @datagraphics/baker
What is Baker and why do you use it?
Baker is a development environment that can be converted into a static website that’s ready for the web. With a minimal amount of HTML, CSS and JavaScript, you can publish a project. The Los Angeles Times uses Baker to write custom code for projects that aren’t possible within the rigid templates of our content management system.
Does anyone else use Baker?
Yes. Here are some examples of Baker is use outside of the Los Angeles Times.
If you know of other examples, please add them to the list.
How does Baker work?
Baker brings together a bunch of different technologies.
The HTML templating is powered by Nunjucks, giving us a Jinja2-like experience for organizing and creating our HTML pages. This is also very similar to the templating language used in Django.
CSS styles are written using the preprocessor Sass. Sass enhances CSS by adding features that don't exist in CSS yet like nesting, mixins, inheritance and other tricks. Baker also uses the postprocessor called Autoprefixer, which automatically adds special prefixes to our CSS for browser support. (--webkit
, --moz
, etc.)
JavaScript is bundled using Rollup, which makes it possible for us to write modern JavaScript that gets optimized and prepared in a way that makes it load as fast as possible for our users. Code we write is passed through a JavaScript compiler called Babel, which rewrites our code to make sure it works in all the browsers we support.
Data imports, powered by quaff, allow for easily imported structured data files into templates, which is useful for making data visualizations.
How do I get started using it?
The repository at github.com/datadesk/baker-example-page-template is our premade starter that comes with HTML, styles and scripts ready for experimentation. It also includes GitHub Actions that can deploy staging and production version of your page. It works after only minimal configuration. You could customize it to match the look and feel of your site.
Contributing
Fork the repository and clone it locally. The enter the code directory and install the package's dependencies.
npm install
Branch off. Make any changes. Preview them with the test server.
npm start
Run our tests.
npm run build
Once they pass, your changes should be briefly documented in the CHANGELOG.md
file under the [Unreleased]
header. Depending on the type of change you are making, you may need to add a new subheader as defined by Keep a Changelog. For example, if you are changing how a feature works, you may need to add a ### [Changed]
subhead.
Commit. Submit a pull request.
Releasing
This package is distributed using npm. To publish a new release, you will need to have an npmjs account with ownership of the @datagraphics/baker namespace.
Next you should use npm's version command to up the version number. You have to decide if you're a major, minor or patch release. If you're unsure, review the standards defined at semver.org. Then run one of the commands below. The code will be updated appropriately.
npm version major
npm version minor
npm version patch
Rename the [Unreleased]
section of the CHANGELOG.md
with the same version number. Commit.
git add CHANGELOG.md
git commit -m "Updated CHANGELOG"
Release the new version of the package.
npm publish
Push your work to GitHub, including tag created by the npm version
command.
git push origin main --tags
Create a new release on GitHub at github.com/datadesk/baker/releases with the same version number. Paste the changelog entry into the post as a bullet list.