a875477ef3
build(deps): bump minimist from 1.2.3 to 1.2.6 |
||
---|---|---|
bin | ||
lib | ||
test | ||
.gitignore | ||
.npmignore | ||
.travis.yml | ||
README.md | ||
cli.js | ||
contributing.md | ||
index.js | ||
package-lock.json | ||
package.json |
README.md
electron-docs-linter
Parse and validate Electron's API documentation.
Installation
npm install electron-docs-linter --save
CLI Usage
To lint the docs:
electron-docs-linter path/to/electron/docs
If errors are found, they are printed to STDERR and the process exits un-gracefully.
To lint the docs and save the generated JSON schema to a file:
electron-docs-linter docs/api --version=1.2.3 --outfile=api.json
Programmatic Usage
The module exports a function that parses markdown docs in a given directory, then returns a JSON representation of all the APIs.
const lint = require('electron-docs-linter')
const docPath = './test/fixtures/electron/docs/api'
const targetVersion = '1.2.3' // the soon-to-be-released version of electron
lint(docPath, targetVersion).then(function (apis) {
// `apis` is an array of API objects. To find one:
var win = apis.find(api => api.name === 'BrowserWindow')
// The array also has a string key for each API name, so
// you can access APIs like this too:
win = apis.BrowserWindow
win.events.length
// => 25
win.events[0]
// {
// "name": "page-title-updated",
// "description": "Emitted when the document...",
// "returns": [
// {
// "name": "event",
// "type": "Event"
// }
// ]
// }
win.instanceMethods[20]
// {
// name: 'setSize',
// signature: '(width, height[, animate])'
// }
})
How It Works
The linter starts with a list of all the API names as seed data.
Each API's structure is inferred by parsing its raw markdown documentation from the electron repo. The electron-docs module abstracts away the challenges of fetching file contents in bulk.
Electron's API documentation adheres to Electron Coding Style and the Electron Styleguide, so its content can be programmatically parsed. To make the content easy to parse, the raw markdown is converted to HTML using marky-markdown-lite, which returns a cheerio DOM object that can be queried and traversed using familiar CSS selectors.
The result is an array of API objects. The following metadata is included for each API, where appropriate:
- name
- description
- type (Class or Module)
- process (main, renderer, or both)
- methods
- events
- static methods (aka class methods)
- instance events
- instance methods
- instance properties
- website URL
- GitHub repository URL
Related Things and Prior Art
- https://github.com/atom/autocomplete-atom-api
- https://kapeli.com/docsets#dashDocset
- issue: Publish the public API as JSON
- https://raw.githubusercontent.com/atom/autocomplete-atom-api/master/completions.json
- devdocs.io
- Node.js - About this Documentation
TypeScript Definitions
A lot of people want an up-to-date TypeScript definition file for Electron.
- https://github.com/MarshallOfSound/Electron-DefinitelyTyped (WIP)
- https://github.com/electron/electron/issues/4875
- https://www.npmjs.com/package/@types/electron
- https://github.com/DefinitelyTyped/DefinitelyTyped/tree/master/github-electron
- https://github.com/RyanCavanaugh/dts-dom - A DOM library for generation TypeScript declaration (.d.ts) files
- https://github.com/lbovet/typson - Converts TypeScript to JSON-schema
- https://github.com/lbovet/docson - Documentation for your JSON types
Dependencies
- cheerio: Tiny, fast, and elegant implementation of core jQuery designed specifically for the server
- clean-deep: Remove falsy, empty or nullable values from objects
- decamelize: Convert a camelized string into a lowercased one with a custom separator: unicornRainbow → unicorn_rainbow
- dedent: An ES6 string tag that strips indentation from multi-line strings
- electron-docs: Fetch Electron documentation as raw markdown strings
- keyed-array: Recursively add named keys to arrays of objects
- lodash.pick: The lodash method
_.pick
exported as a module. - lodash.sum: The lodash method
_.sum
exported as a module. - marky-markdown-lite: A version of marky-markdown that does less
- minimist: parse argument options
- ora: Elegant terminal spinner
- path-exists: Check if a path exists
- pify: Promisify a callback-style function
- revalidator: A cross-browser / node.js validator powered by JSON Schema
- semver: The semantic version parser used by npm.
- to-markdown: HTML-to-Markdown converter
Dev Dependencies
- chai: BDD/TDD assertion library for node.js and the browser. Test framework agnostic.
- mocha: simple, flexible, fun test framework
- rimraf: A deep deletion module for node (like
rm -rf
) - standard: JavaScript Standard Style
License
MIT