erxclau / cookie-monster

trying this again...

cookie

The bite-sized daily graphics starter.

This template repository is designed to help The Michigan Daily data team quickly start up and produce data visualizations to embed on michigandaily.com.

It has several useful features:

• MicroCMS with ArchieML
• Bundling with Vite
• D3v7 as a dependency by default
• Templating with Nunjucks
• Easy deploy to GitHub Pages
• Easy deploy to S3 bucket
• Preview screen with embed URL
• Data ingest from Google Sheets
• Easy include of ai2html output
• Ability to develop several common graphics in one repository

Using this template

Click the green "Use this template" button to create a new instance of the cookie template. Then, clone the new instance and run yarn install to initialize the project.

You can use yarn dev to start a development server and yarn build to build production-ready files into dist/.

Edit the files in src/. To create graphics, you should only be editing files within the src/graphic/ directory. You can write any markup in src/graphic/index.html, and JavaScript in src/graphic/js/graphic.js, and any styles in src/graphic/css/graphic.scss.

There are some CSS variables available to you:

• --serif for serif font (Lora)
• --sans for sans serif font (Open Sans)
• --blue, --maize, --gray, --black, --white, and --blue-{1..5} for colors

If you are embedding the graphic on the homepage, add the homepage class to the body tag. This will set the headline font to Lora, and will set the body font to Open Sans.

If you want to develop another graphic in the same instance, create a new HTML file in src/graphic that uses the base.html template. Refer to src/graphic/index.html for how to extend the template. You'll also need to create a new entry in config.json.

Fetching from Google Drive

Refer to the Google Drive fetch section in the sink README for instructions on how to set up config.json for fetching files from Google Drive.

You can import a JSON file in JS like this:

import copy from "../data/data.json";

You can import a CSV file in JS like this:

import csvfile from "../data/data.csv";

We use the @michigandaily/rollup-plugin-dsv plugin (which relies on d3-dsv) to parse CSV files (or other DSV files) into usable arrays. You can prevent automatic typing by adding a autoType=false query parameter to the end of the import path. You can also specify which columns you want imported with a columns query parameter. Refer to the transformer README for examples.

Including ai2html output

1. Use our Adobe Illustrator template to create a graphic.
2. Save the instance to the root of this repository.
3. Set the red and bold filename value under the ai2html-text section to the name of the instance file name.
4. Generate the ai2html output. It should be in src/graphic/ai2html-output.
5. In config.json, set illustrator_output_filename to the HTML output file.

Deploying to AWS S3

Refer to the AWS S3 deployment with cache invalidation section in the sink README for instructions on how to set up config.json for deploying to AWS S3.

1. Ensure sure that base in vite.config.js is routed correctly (it should probably be config.key prepended and appended by a /).
2. Run yarn sink deploy aws.

Deploying to GitHub Pages

Refer to the GitHub Pages deployment section in the sink README for instructions on how to set up config.json for deploying to GitHub Pages.

1. Ensure sure that base in vite.config.js is routed correctly (it should probably be config.key prepended and appended by a /).
2. Run yarn sink deploy github.
3. Go to Settings > Pages and check the Enforce HTTPS option. All of our sites should enforce HTTPS, so please make sure to double check this!
4. Your raw graphic will be accessible at https://michigandaily.github.io/<repository-name>/graphic/index.html.

Playwright screenshots

Running yarn build will 1) build production files and 2) trigger a process where Playwright spawns a headless Chrome browser and downloads screenshots of the graphics. If the script is unable to find a cached version of Chrome or is unable to use your default Chrome installation, it will install Chrome in a cache folder.

If you don't want yarn build to trigger the Playwright process, set COOKIE_SCREENSHOT=false in a .env file.

You can also prevent the screenshot process by setting the environment variable inline prior to a command: COOKIE_SCREENSHOT=false yarn build or COOKIE_SCREENSHOT=false yarn sink deploy github.