195 lines
10 KiB
JavaScript
195 lines
10 KiB
JavaScript
import fs from 'node:fs/promises';
|
|
import path from 'node:path';
|
|
import { createHash } from 'node:crypto';
|
|
import fg from 'fast-glob';
|
|
|
|
const readText = async (file) => fs.readFile(file, 'utf8');
|
|
|
|
function hrefValues(html, rel) {
|
|
const values = [];
|
|
const pattern = new RegExp(`<link\\b[^>]*\\brel=["']${rel}["'][^>]*\\bhref=["']([^"']+)["']`, 'gi');
|
|
for (const match of html.matchAll(pattern)) values.push(match[1]);
|
|
return values;
|
|
}
|
|
|
|
function routeFromHtml(file, root) {
|
|
const relative = path.relative(root, file).replaceAll(path.sep, '/');
|
|
if (relative === 'index.html') return '/';
|
|
if (relative.endsWith('/index.html')) return `/${relative.slice(0, -'/index.html'.length)}`;
|
|
return `/${relative.replace(/\.html$/, '')}`;
|
|
}
|
|
|
|
function formActions(html) {
|
|
return [...html.matchAll(/<form\b([^>]*)>/gi)].map((match) =>
|
|
match[1].match(/\baction=["']([^"']*)["']/i)?.[1] ?? null);
|
|
}
|
|
|
|
function rootRelativeUrls(html) {
|
|
const urls = [];
|
|
for (const match of html.matchAll(/\b(?:href|action)=["']([^"']+)["']/gi)) urls.push(match[1]);
|
|
return urls;
|
|
}
|
|
|
|
function hasTerminalSlash(url) {
|
|
if (!url.startsWith('/') || url.startsWith('//')) return false;
|
|
const pathname = url.split(/[?#]/, 1)[0];
|
|
return pathname !== '/' && pathname.endsWith('/');
|
|
}
|
|
|
|
const formAProductionUrls = new Map([
|
|
['/contact', 'https://form.jotform.com/262946078314059?formContext=contact&language=en'],
|
|
['/es/contact', 'https://form.jotform.com/262946078314059?formContext=contact&language=es'],
|
|
['/lp/aba-therapy', 'https://form.jotform.com/262946078314059?formContext=landing&language=en'],
|
|
['/lp/aba-insurance', 'https://form.jotform.com/262946078314059?formContext=landing&language=en'],
|
|
['/lp/aba-ahcccs', 'https://form.jotform.com/262946078314059?formContext=landing&language=en'],
|
|
['/lp/aba-private-pay', 'https://form.jotform.com/262946078314059?formContext=landing&language=en'],
|
|
['/lp/early-intervention', 'https://form.jotform.com/262946078314059?formContext=landing&language=en'],
|
|
]);
|
|
|
|
function iframeSources(html) {
|
|
// Astro correctly escapes query separators in serialized HTML attributes.
|
|
// Normalize that representation before comparing the fixed allowlisted URL.
|
|
return [...html.matchAll(/<iframe\b[^>]*\bsrc=["']([^"']+)["'][^>]*>/gi)].map((match) => match[1].replaceAll('&', '&'));
|
|
}
|
|
|
|
function formAAdapters(html) {
|
|
return [...html.matchAll(/<section\b[^>]*\bdata-jotform-form-a\b[^>]*>/gi)].map((match) => match[0]);
|
|
}
|
|
|
|
function unsafeSignals(html, allowApprovedFormA = false) {
|
|
const signals = [];
|
|
if (!allowApprovedFormA && /<iframe\b[^>]*\bsrc=["'][^"']*jotform/i.test(html)) signals.push('active Jotform iframe');
|
|
if (!allowApprovedFormA && /262946078314059/i.test(html)) signals.push('isolated-test Form A identifier');
|
|
if (/<script\b[^>]*\bsrc=["'][^"']*(?:jotform|googletagmanager|google-analytics|doubleclick)/i.test(html)) signals.push('active vendor tracking/form script');
|
|
if (!allowApprovedFormA && /data-(?:form|forms)-mode=["']live["']/i.test(html)) signals.push('live form mode');
|
|
if (/api\.azinstitute4autism\.com/i.test(html)) signals.push('production API origin');
|
|
if (/AIza[0-9A-Za-z_-]{20,}|-----BEGIN (?:RSA|OPENSSH|EC) PRIVATE KEY-----|(?:api[_-]?key|secret|password)\s*[:=]/i.test(html)) signals.push('credential-like content');
|
|
return signals;
|
|
}
|
|
|
|
export async function verifyArtifact(root, mode = 'staging') {
|
|
const formAEnabled = mode === 'isolated-test' || mode === 'production';
|
|
const production = mode === 'production';
|
|
const stagingPolicy = mode === 'staging' || mode === 'isolated-test';
|
|
const htmlFiles = (await fg('**/*.html', { cwd: root, absolute: true })).sort();
|
|
const routes = [];
|
|
const canonicalMap = new Map();
|
|
const errors = [];
|
|
const warnings = [];
|
|
let hreflangLinks = 0;
|
|
let activeForms = 0;
|
|
let secretSignals = [];
|
|
let formAIframes = 0;
|
|
|
|
for (const file of htmlFiles) {
|
|
const html = await readText(file);
|
|
const route = routeFromHtml(file, root);
|
|
const robots = html.match(/<meta\b[^>]*\bname=["']robots["'][^>]*\bcontent=["']([^"']*)["']/i)?.[1] ?? null;
|
|
const canonical = html.match(/<link\b[^>]*\brel=["']canonical["'][^>]*\bhref=["']([^"']+)["']/i)?.[1] ?? null;
|
|
const title = html.match(/<title>([^<]*)<\/title>/i)?.[1]?.trim() ?? '';
|
|
const actions = formActions(html);
|
|
const nonSearchActions = actions.filter((action) => action && action !== '/search');
|
|
const trailingSlashUrls = rootRelativeUrls(html).filter(hasTerminalSlash);
|
|
const expectedFormAUrl = formAEnabled ? formAProductionUrls.get(route) : undefined;
|
|
const signals = unsafeSignals(html, formAEnabled);
|
|
const jotformFrames = iframeSources(html).filter((src) => {
|
|
try {
|
|
return new URL(src).origin === 'https://form.jotform.com';
|
|
} catch {
|
|
return false;
|
|
}
|
|
});
|
|
const adapters = formAAdapters(html);
|
|
if (!title) errors.push(`${route}: missing title`);
|
|
if (!canonical) errors.push(`${route}: missing canonical`);
|
|
if (!robots) errors.push(`${route}: missing robots directive`);
|
|
if (trailingSlashUrls.length) errors.push(`${route}: root-relative URL(s) end in a slash: ${trailingSlashUrls.join(', ')}`);
|
|
if (stagingPolicy && robots !== 'noindex,nofollow') errors.push(`${route}: staging robots is ${robots ?? 'missing'}`);
|
|
if (production && robots === 'noindex,nofollow') errors.push(`${route}: production artifact is noindex`);
|
|
if (canonical) {
|
|
const prior = canonicalMap.get(canonical);
|
|
if (prior) errors.push(`${route}: duplicate canonical ${canonical} (also ${prior})`);
|
|
canonicalMap.set(canonical, route);
|
|
}
|
|
hreflangLinks += hrefValues(html, 'alternate').filter((href) => href).length;
|
|
if (signals.length) secretSignals.push({ route, signals });
|
|
if (formAEnabled) {
|
|
if (expectedFormAUrl) {
|
|
if (jotformFrames.length !== 1 || jotformFrames[0] !== expectedFormAUrl) errors.push(`${route}: missing or invalid approved Form A iframe URL`);
|
|
const expectedMode = production ? 'live' : 'isolated-test';
|
|
const expectedEnvironment = production ? 'production' : 'isolated-test';
|
|
if (adapters.length !== 1 || !adapters[0].includes(`data-form-mode="${expectedMode}"`) || !adapters[0].includes(`data-form-environment="${expectedEnvironment}"`)) {
|
|
errors.push(`${route}: Form A adapter lacks the approved ${expectedMode} ${expectedEnvironment} gate`);
|
|
}
|
|
formAIframes += jotformFrames.length;
|
|
} else if (jotformFrames.length || adapters.length || /262946078314059/i.test(html)) {
|
|
errors.push(`${route}: unauthorized Form A iframe or identifier in production artifact`);
|
|
}
|
|
}
|
|
if (nonSearchActions.length) activeForms += nonSearchActions.length;
|
|
routes.push({ route, title, robots, canonical, nativeForms: actions.length, nonSearchActions: nonSearchActions.length });
|
|
}
|
|
|
|
const sitemapFile = path.join(root, 'sitemap.xml');
|
|
const sitemap = await fs.readFile(sitemapFile, 'utf8').catch(() => '');
|
|
const sitemapRoutes = [...sitemap.matchAll(/<loc>([^<]+)<\/loc>/gi)].map((match) => match[1]);
|
|
const reviewIndex = htmlFiles.some((file) => routeFromHtml(file, root) === '/lp');
|
|
if (stagingPolicy) {
|
|
if (sitemapRoutes.length) errors.push(`staging sitemap contains ${sitemapRoutes.length} URLs`);
|
|
if (hreflangLinks) errors.push(`staging emits ${hreflangLinks} hreflang links`);
|
|
if (!reviewIndex) errors.push('staging /lp review index is missing');
|
|
} else {
|
|
if (!sitemapRoutes.length) errors.push('indexing-enabled sitemap is empty');
|
|
if (reviewIndex) errors.push('indexing-enabled artifact contains staging-only /lp review index');
|
|
}
|
|
if (activeForms) errors.push(`found ${activeForms} non-search native form action(s)`);
|
|
if (secretSignals.length) errors.push(`unsafe artifact signals: ${JSON.stringify(secretSignals)}`);
|
|
|
|
const files = (await fg('**/*', { cwd: root, absolute: true, onlyFiles: true })).sort();
|
|
const formAResourceFiles = [];
|
|
const digest = createHash('sha256');
|
|
for (const file of files) {
|
|
const relative = path.relative(root, file).replaceAll(path.sep, '/');
|
|
if (/\.(?:html?|js|mjs|css|json|xml|txt)$/i.test(relative)) {
|
|
const contents = await fs.readFile(file, 'utf8');
|
|
if (/262946078314059/i.test(contents)) formAResourceFiles.push(relative);
|
|
}
|
|
digest.update(relative);
|
|
digest.update('\0');
|
|
digest.update(createHash('sha256').update(await fs.readFile(file)).digest('hex'));
|
|
digest.update('\n');
|
|
}
|
|
if (formAEnabled) {
|
|
const expectedFormAFiles = new Set([...formAProductionUrls.keys()].map((route) => route === '/' ? 'index.html' : `${route.slice(1)}/index.html`));
|
|
const unexpectedFormAFiles = formAResourceFiles.filter((file) => !expectedFormAFiles.has(file));
|
|
if (unexpectedFormAFiles.length) errors.push(`Form A identifier appears outside approved production routes: ${unexpectedFormAFiles.join(', ')}`);
|
|
if (formAIframes !== formAProductionUrls.size) errors.push(`production artifact has ${formAIframes} approved Form A iframe(s), expected ${formAProductionUrls.size}`);
|
|
} else if (formAResourceFiles.length) {
|
|
errors.push(`isolated-test Form A identifier appears in artifact: ${formAResourceFiles.join(', ')}`);
|
|
}
|
|
return {
|
|
mode,
|
|
htmlRoutes: htmlFiles.length,
|
|
sitemapUrls: sitemapRoutes.length,
|
|
hreflangLinks,
|
|
reviewIndex,
|
|
activeNativeFormActions: activeForms,
|
|
formAIframes,
|
|
artifactFiles: files.length,
|
|
artifactSha256: digest.digest('hex'),
|
|
routes,
|
|
errors,
|
|
warnings,
|
|
passed: errors.length === 0,
|
|
};
|
|
}
|
|
|
|
if (process.argv[1] && path.resolve(process.argv[1]) === path.resolve(new URL(import.meta.url).pathname)) {
|
|
const mode = process.argv.find((arg) => arg.startsWith('--mode='))?.slice('--mode='.length) ?? 'staging';
|
|
const root = process.argv.find((arg) => arg.startsWith('--root='))?.slice('--root='.length) ?? 'dist';
|
|
if (!['staging', 'indexing', 'isolated-test', 'production'].includes(mode)) throw new Error(`Unsupported mode: ${mode}`);
|
|
const result = await verifyArtifact(path.resolve(root), mode);
|
|
console.log(JSON.stringify(result, null, 2));
|
|
if (!result.passed) process.exitCode = 1;
|
|
}
|