[Reporting/Build Chromium] Add a developer script to find the Chromium commit (#94033)

* script that finds the chromium commit

* use run from dev-utils

* update docs and inch towards build pipeline

* more docs updates

* move the scripts the root

* review feedback

scripts/chromium_version.js

@@ -0,0 +1,10 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License
+ * 2.0 and the Server Side Public License, v 1; you may not use this file except
+ * in compliance with, at your election, the Elastic License 2.0 or the Server
+ * Side Public License, v 1.
+ */
+
+require('../src/setup_node_env');
+require('../src/dev/chromium_version');

src/dev/chromium_version.ts

@@ -0,0 +1,138 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License
+ * 2.0 and the Server Side Public License, v 1; you may not use this file except
+ * in compliance with, at your election, the Elastic License 2.0 or the Server
+ * Side Public License, v 1.
+ */
+
+import { run, REPO_ROOT, ToolingLog } from '@kbn/dev-utils';
+import chalk from 'chalk';
+import cheerio from 'cheerio';
+import fs from 'fs';
+import fetch from 'node-fetch';
+import path from 'path';
+import { PackageJson } from 'type-fest';
+
+type PuppeteerRelease = string;
+type ChromiumRevision = string;
+type ChromiumCommit = string;
+
+// We forked the Puppeteer node module for Kibana,
+// So we need to translate OUR version to the official Puppeteer Release
+const forkCompatibilityMap: Record<string, PuppeteerRelease> = {
+  '5.4.1-patch.1': '5.4.1',
+};
+
+async function getPuppeteerRelease(log: ToolingLog): Promise<PuppeteerRelease> {
+  // open node_modules/puppeteer/package.json
+  const puppeteerPackageJson: PackageJson = JSON.parse(
+    fs.readFileSync(path.resolve(REPO_ROOT, 'node_modules', 'puppeteer', 'package.json'), 'utf8')
+  );
+  const { version } = puppeteerPackageJson;
+  if (version == null) {
+    throw new Error(
+      'Could not get the Puppeteer version! Check node_modules/puppteer/package.json'
+    );
+  }
+  log.info(`Kibana is using Puppeteer ${version} (${forkCompatibilityMap[version]})`);
+  return forkCompatibilityMap[version];
+}
+
+async function getChromiumRevision(
+  kibanaPuppeteerVersion: PuppeteerRelease,
+  log: ToolingLog
+): Promise<ChromiumRevision> {
+  const url = `https://raw.githubusercontent.com/puppeteer/puppeteer/v${kibanaPuppeteerVersion}/src/revisions.ts`;
+  let body: string;
+  try {
+    log.info(`Fetching code from Puppeteer source: ${url}`);
+    const rawSource = await fetch(url);
+    body = await rawSource.text();
+  } catch (err) {
+    log.error(err);
+    throw new Error(`Could not fetch ${url}. Check the URL in a browser and try again.`);
+  }
+
+  let revision: ChromiumRevision | undefined;
+  const lines = body.split('\n');
+  let cursor = lines.length;
+  while (--cursor >= 0) {
+    // look for the line of code matching `  chromium: '0123456',`
+    const test = lines[cursor].match(/^\s+chromium: '(\S+)',$/);
+    if (test != null) {
+      log.debug(`Parsed revision from source text: \`${lines[cursor]}\``);
+      [, revision] = test;
+      break;
+    }
+  }
+
+  if (revision == null) {
+    throw new Error(
+      `Could not find a Chromium revision listed in Puppeteer source! Check ${url} in a browser`
+    );
+  }
+
+  log.info(`Found Chromium revision ${revision} from Puppeteer ${kibanaPuppeteerVersion}`);
+  return revision;
+}
+
+async function getChromiumCommit(
+  revision: ChromiumRevision,
+  log: ToolingLog
+): Promise<ChromiumCommit> {
+  const url = `https://crrev.com/${revision}`;
+  log.info(`Fetching ${url}`);
+  const pageText = await fetch(url);
+  const $ = cheerio.load(await pageText.text());
+
+  // get the commit from the page title
+  let commit: ChromiumCommit | null = null;
+  const matches = $('title')
+    .text()
+    .match(/\S{40}/);
+  if (matches != null) {
+    log.debug(`Parsed commit hash from page title: \`${$('title').text()}\``);
+    [commit] = matches;
+  }
+
+  if (commit == null) {
+    throw new Error(`Could not find a Chromium commit! Check ${url} in a browser.`);
+  }
+
+  log.info(`Found Chromium commit ${commit} from revision ${revision}.`);
+  return commit;
+}
+
+run(
+  async ({
+    log,
+    flags: {
+      _: [puppeteerVersionArg],
+    },
+  }) => {
+    try {
+      let puppeteerVersion: PuppeteerRelease;
+      if (puppeteerVersionArg) {
+        puppeteerVersion = puppeteerVersionArg;
+      } else {
+        puppeteerVersion = await getPuppeteerRelease(log);
+      }
+
+      const chromiumRevision = await getChromiumRevision(puppeteerVersion, log);
+      await getChromiumCommit(chromiumRevision, log);
+    } catch (err) {
+      log.error(err);
+    }
+  },
+  {
+    description: chalk`
+      Display the Chromium git commit that correlates to a given Puppeteer release.
+
+      -  node x-pack/dev-tools/chromium_version 5.5.0  {dim # gets the Chromium commit for Puppeteer v5.5.0}
+      -  node x-pack/dev-tools/chromium_version       {dim  # gets the Chromium commit for the Kibana dependency version of Puppeteer}
+
+      You can use https://omahaproxy.appspot.com/ to look up the Chromium release that first shipped with that commit.
+    `,
+  }
+);

x-pack/build_chromium/.chromium-commit

@@ -0,0 +1 @@
+ef768c94bcb42dca4c27048615d07efadbb1c1c2

x-pack/build_chromium/README.md

@@ -44,7 +44,7 @@ gsutil cp -r gs://my-bucket/build_chromium .
 # Install the OS packages, configure the environment, download the chromium source (25GB)
 python ./build_chromium/init.sh [arch_name]
 
-# Run the build script with the path to the chromium src directory, the git commit id
+# Run the build script with the path to the chromium src directory, the git commit hash
 python ./build_chromium/build.py <commit_id> x86
 
 # OR You can build for ARM
@@ -55,16 +55,14 @@ python ./build_chromium/build.py <commit_id> arm64
 the Chromium repo to be cloned successfully. If checking out the Chromium fails
 with "early EOF" errors, the instance could be low on memory or disk space.
 
-## Getting the Commit ID
-The `build.py` script requires a commit ID of the Chromium repo. Getting `<commit_id>` can be tricky. The best technique seems to be:
-1. Create a temporary working directory and intialize yarn
-2. `yarn add puppeteer # install latest puppeter`
-3. Look through Puppeteer documentation and Changelogs to find information
-about where the "chromium revision" is located in the Puppeteer code. The code
-containing it might not be distributed in the node module.
-    - Example: https://github.com/puppeteer/puppeteer/blob/b549256/src/revisions.ts
-4. Use `https://crrev.com` and look up the revision and find the git commit info.
-    - Example: http://crrev.com/818858 leads to the git commit e62cb7e3fc7c40548cef66cdf19d270535d9350b
+## Getting the Commit Hash
+If you need to bump the version of Puppeteer, you need to get a new git commit hash for Chromium that corresponds to the Puppeteer version.
+```
+node x-pack/dev-tools/chromium_version.js [PuppeteerVersion]
+```
+
+When bumping the Puppeteer version, make sure you also update the `.chromium-commit` file with the commit hash
+for the current Chromium build, so we'll be able to construct a build pipeline for each OS (coming soon!).
 
 ## Build args
 

x-pack/plugins/reporting/server/browsers/chromium/driver_factory/index.ts

@@ -83,6 +83,12 @@ export class HeadlessChromiumDriverFactory {
         } as puppeteer.LaunchOptions);
 
         page = await browser.newPage();
+
+        // Log version info for debugging / maintenance
+        const client = await page.target().createCDPSession();
+        const versionInfo = await client.send('Browser.getVersion');
+        logger.debug(`Browser version: ${JSON.stringify(versionInfo)}`);
+
         await page.emulateTimezone(browserTimezone ?? null);
 
         // Set the default timeout for all navigation methods to the openUrl timeout (30 seconds)