Merge pull request tc39#15 from nicolo-ribaudo/ipr-check-action

Add IPR check CI job

.github/workflows/ipr.yml

@@ -0,0 +1,22 @@
+name: 'ecma-262'
+
+on:
+  pull_request_target:
+    branches:
+       - '*'
+  push:
+    branches:
+       - main
+
+jobs:
+  build:
+    name: 'check IPR form'
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v3
+      - uses: actions/setup-node@v3
+      - run: node scripts/ipr-check.js tc39/source-map-spec main
+        env:
+          GH_TOKEN: ${{ secrets.GH_IPR_TOKEN }}
+          GOOGLE_API_KEY: ${{ secrets.GOOGLE_API_KEY }}

.gitignore

@@ -1,2 +1,3 @@
 .venv
 source-map.html
+.env

scripts/ipr-check.js

@@ -0,0 +1,175 @@
+#!/usr/bin/env node
+
+// Copied from https://github.com/tc39/ecma262/blob/c316ec72f6e227d88d4ef67a3e16b486c5ca356d/scripts/check-form.js,
+// with some simplifications (we do not need to handle IPR for some legacy commits)
+// - none of the current committers still need to sign the IPR form
+// - none of the current committers use multiple GitHub accounts to commit to this repository
+
+const { execSync } = require('child_process');
+
+// web URL: `https://docs.google.com/spreadsheets/d/${sheetID}/edit`
+const sheetID = '1if5bU0aV5MJ27GGKnRzyAozeKP-ILXYl5r3dzvkGFmg';
+
+const {
+	GOOGLE_API_KEY: key, // TC39 API key for google sheets
+	GH_TOKEN
+} = process.env;
+
+if (!GH_TOKEN) {
+	throw 'GH_TOKEN env var required';
+}
+if (!key) {
+	throw 'GOOGLE_API_KEY env var required';
+}
+
+const sheetData = `https://sheets.googleapis.com/v4/spreadsheets/${sheetID}/values/Sheet1!A2:A?key=${key}`;
+
+const [,, slug, branch] = process.argv;
+
+if (!slug || !branch) {
+	throw 'args required: slug, branch';
+}
+
+const sha = String(execSync(`git rev-parse --short ${branch}`)).trim();
+
+const request = async (url, method = 'GET', postData) => {
+	// adapted from https://medium.com/@gevorggalstyan/how-to-promisify-node-js-http-https-requests-76a5a58ed90c
+	const lib = url.startsWith('https://') ? require('https') : require('http');
+
+	const [h, path] = url.split('://')[1].split('/');
+	const [host, port] = h.split(':');
+
+	const params = {
+		host,
+		port: port || url.startsWith('https://') ? 443 : 80,
+		method,
+		headers: {
+			Authorization: `token ${GH_TOKEN}`,
+			'User-Agent': 'curl/7.54.0'
+		}
+	};
+
+	return new Promise((resolve, reject) => {
+		const req = lib.request(url, params, res => {
+			if (res.statusCode < 200 || res.statusCode >= 300) {
+				return reject(new Error(`Status Code: ${res.statusCode}; ${url}`));
+			}
+
+			const data = [];
+
+			res.on('data', chunk => {
+				data.push(chunk);
+			});
+
+			res.on('end', () => resolve(String(Buffer.concat(data))));
+		});
+
+		req.on('error', reject);
+
+		if (postData) {
+			req.write(postData);
+		}
+
+		req.end();
+	});
+};
+
+const perPage = 100;
+
+function getAuthorFromCommit(commitObj) {
+	if (!commitObj) {
+		return false;
+	}
+	const { author } = commitObj;
+	if (!author) {
+    throw new Error("Missing author for commit " + commitObj.sha);
+	}
+	return author.login;
+}
+
+async function getAllCommits(page = 1) {
+	const commitsURL = `https://api.github.com/repos/${slug}/commits?anon=1&per_page=${perPage}&page=${page}&sha=${sha}`;
+	const commits = await request(commitsURL).then((json) => JSON.parse(json));
+	return [...new Set([].concat(
+		commits.flatMap(x => getAuthorFromCommit(x) || []),
+		commits.length < perPage ? [] : await getAllCommits(page + 1),
+	))];
+}
+
+const authors = getAllCommits().then((authors) => {
+	const knowns = authors.filter(x => typeof x === 'string');
+	console.log(`Found ${knowns.length} authors: ${knowns.join(',')}\n`);
+	return knowns;
+});
+
+const teamURL = (team) => `https://api.github.com/orgs/tc39/teams/${team}`;
+
+function getMembers(teamID, page = 1) {
+	const memberURL = `https://api.github.com/teams/${teamID}/members?per_page=100&page=${page}`;
+	const data = request(memberURL).then((json) => JSON.parse(json));
+	return data.then((data) => {
+		if (data.length === 0) {
+			return data;
+		}
+		return getMembers(teamID, page + 1).then(nextPage => {
+			return data.concat(nextPage);
+		});
+	});
+}
+
+function handler(kind) {
+	return (data) => {
+		const names = new Set(data.map(x => x.login.toLowerCase()));
+		console.log(`Found ${names.size} ${kind}: ${[...names].join(',')}\n`);
+		return names;
+	}
+}
+
+const delegates = request(teamURL('delegates')).then((json) => JSON.parse(json)).then(data => {
+	return getMembers(data.id);
+}).then(handler('delegates'));
+
+const emeriti = request(teamURL('emeriti')).then((json) => JSON.parse(json)).then(data => {
+	return getMembers(data.id);
+}).then(handler('emeriti'));
+
+const usernames = request(sheetData).then((json) => JSON.parse(json)).then(data => {
+	if (!Array.isArray(data.values)) {
+		throw 'invalid data';
+	}
+	const usernames = new Set(
+		data.values
+			.flat(1)
+			.map(x => x.replace(/^(https?:\/\/)?github\.com\//, '').replace(/^@/, '').toLowerCase())
+			.filter(x => /^[a-z0-9_-]{1,39}$/gi.test(x))
+			.sort((a, b) => a.localeCompare(b))
+	);
+	console.log(`Found ${usernames.size} usernames: ` + [...usernames].join(',') + '\n');
+	return usernames;
+});
+
+const exceptions = new Set([
+  'jaro-sevcik', // Google empolyee
+  'jkrems', // Google employee
+  'josephschorr', // former Google empolyee
+	'sideshowbarker', // Mozilla employee
+].map(x => x.toLowerCase()));
+
+Promise.all([usernames, authors, delegates, emeriti]).then(([usernames, authors, delegates, emeriti]) => {
+	const missing = authors.filter(author => {
+		const a = author.toLowerCase();
+		const signed = usernames.has(a)
+			|| delegates.has(a)
+			|| emeriti.has(a)
+			|| exceptions.has(a);
+		return !signed;
+	});
+	if (missing.length > 0) {
+		throw `Missing ${missing.length} authors: ${missing}`;
+	} else {
+		console.log('All authors have signed the form, or are delegates or emeriti!');
+	}
+}).catch((e) => {
+	console.error(e);
+	process.exitCode = 1;
+});