Stacks-Editor is a combination rich text / markdown editor that powers Stack Overflow's post editing experience.
npm install @stackoverflow/stacks-editor
<div id="editor-container"></div>
import { StacksEditor } from "@stackoverflow/stacks-editor";
// don't forget to include the styles as well
import "@stackoverflow/stacks-editor/dist/styles.css";
// include the Stacks js and css as they're not included in the bundle
import "@stackoverflow/stacks";
import "@stackoverflow/stacks/dist/css/stacks.css";
new StacksEditor(
document.querySelector("#editor-container"),
"*Your* **markdown** here"
);
<!--include Stacks -->
<link rel="stylesheet" href="path/to/node_modules/@stackoverflow/stacks/dist/css/stacks.css" />
<!-- include the bundled styles -->
<link
rel="stylesheet"
href="path/to/node_modules/@stackoverflow/stacks-editor/dist/styles.css"
/>
<div id="editor-container"></div>
<!-- highlight.js is not included in the bundle, so include it as well if you want it -->
<script src="//unpkg.com/@highlightjs/cdn-assets@latest/highlight.min.js"></script>
<!--include Stacks -->
<script src="path/to/node_modules/@stackoverflow/stacks/dist/js/stacks.min.js"></script>
<!-- include the bundle -->
<script src="path/to/node_modules/@stackoverflow/stacks-editor/dist/app.bundle.js"></script>
<!-- initialize the editor -->
<script>
new window.stacksEditor.StacksEditor(
document.querySelector("#editor-container"),
"*Your* **markdown** here",
{}
);
</script>
- Install dependencies with
npm i
- Build and start using
npm start
- Point your browser to the address listed in the output - typically http://localhost:8080/
Run all unit tests (no end-to-end tests) using
npm run test:unit
Run all end-to-end tests (written in Playwright) using
npm run test:e2e
End-to-end tests need to follow the convention of using someName.e2e.test.ts
as their filename. They'll automatically get picked up by the test runner this way.
Generate a stats.json
file for analysis using
npm run build:stats
You can upload your stats.json
file here or here for visualization. See more resources here.
We use changesets to automatize the steps necessary to publish to NPM, create GH releases and a changelog.
- Every time you do work that requires a new release to be published, add a changesets entry by running
npx chageset
and follow the instrcutions on screen. (changes that do not require a new release - e.g. changing a test file - don't need a changeset).- When opening a PR without a corresponding changeset the changesets-bot will remind you to do so. It generally makes sense to have one changeset for PR (if the PR changes do not require a new release to be published the bot message can be safely ignored)
- The release github workflow continuosly check if there are new pending changesets in the main branch, if there are it creates a GH PR (
chore(release)
see example) and continue updating it as more changesets are potentially pushed/merged to the main branch. - When we are ready to cut a release we need to simply merge the
chore(release)
PR back to main and the release github workflow will take care of publishing the changes to NPM and create a GH release for us. Thechore(release)
PR also give us an opportunity to adjust the automatically generated changelog when necessary (the entry in the changelog file is also what will end up in the GH release notes).
The release github workflow only run if the CI workflow (running linter, formatter and tests) is successful: CI is blocking accidental releases.
Despite using changesets to communicate the intent of creating releases in a more explicit way, we still follow conventional commits standards for keeping our git history easily parseable by the human eye.