webdriver-chrome.js 5.5 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139
  1. /*
  2. * Copyright 2010-2019 Gildas Lormeau
  3. * contact : gildas.lormeau <at> gmail.com
  4. *
  5. * This file is part of SingleFile.
  6. *
  7. * The code in this file is free software: you can redistribute it and/or
  8. * modify it under the terms of the GNU Affero General Public License
  9. * (GNU AGPL) as published by the Free Software Foundation, either version 3
  10. * of the License, or (at your option) any later version.
  11. *
  12. * The code in this file is distributed in the hope that it will be useful,
  13. * but WITHOUT ANY WARRANTY; without even the implied warranty of
  14. * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero
  15. * General Public License for more details.
  16. *
  17. * As additional permission under GNU AGPL version 3 section 7, you may
  18. * distribute UNMODIFIED VERSIONS OF THIS file without the copy of the GNU
  19. * AGPL normally required by section 4, provided you include this license
  20. * notice and a URL through which recipients can access the Corresponding
  21. * Source.
  22. */
  23. /* global require, exports */
  24. const fs = require("fs");
  25. const chrome = require("selenium-webdriver/chrome");
  26. const { Builder } = require("selenium-webdriver");
  27. const SCRIPTS = [
  28. "../../lib/hooks/hooks-frame.js",
  29. "../../lib/frame-tree/frame-tree.js",
  30. "../../lib/lazy/content/content-lazy-loader.js",
  31. "../../lib/single-file/util/doc-util.js",
  32. "../../lib/single-file/util/doc-helper.js",
  33. "../../lib/single-file/util/timeout.js",
  34. "../../lib/single-file/vendor/css-tree.js",
  35. "../../lib/single-file/vendor/html-srcset-parser.js",
  36. "../../lib/single-file/vendor/css-minifier.js",
  37. "../../lib/single-file/vendor/css-font-property-parser.js",
  38. "../../lib/single-file/vendor/css-media-query-parser.js",
  39. "../../lib/single-file/modules/html-minifier.js",
  40. "../../lib/single-file/modules/css-fonts-minifier.js",
  41. "../../lib/single-file/modules/css-fonts-alt-minifier.js",
  42. "../../lib/single-file/modules/css-matched-rules.js",
  43. "../../lib/single-file/modules/css-medias-alt-minifier.js",
  44. "../../lib/single-file/modules/css-rules-minifier.js",
  45. "../../lib/single-file/modules/html-images-alt-minifier.js",
  46. "../../lib/single-file/modules/html-serializer.js",
  47. "../../lib/single-file/single-file-core.js",
  48. "../../lib/single-file/single-file-browser.js"
  49. ];
  50. exports.getPageData = async options => {
  51. const RESOLVED_CONTENTS = {
  52. "lib/lazy/web/web-lazy-loader-before.js": fs.readFileSync(require.resolve("../../lib/lazy/web/web-lazy-loader-before.js")).toString(),
  53. "lib/lazy/web/web-lazy-loader-after.js": fs.readFileSync(require.resolve("../../lib/lazy/web/web-lazy-loader-after.js")).toString()
  54. };
  55. let driver;
  56. try {
  57. const builder = new Builder();
  58. const chromeOptions = new chrome.Options();
  59. if (options.browserHeadless === undefined || options.browserHeadless) {
  60. chromeOptions.headless();
  61. }
  62. if (options.browserDisableWebSecurity === undefined || options.browserDisableWebSecurity) {
  63. chromeOptions.addArguments("--disable-web-security");
  64. }
  65. if (options.browserBypassCSP === undefined || options.browserBypassCSP) {
  66. chromeOptions.addExtensions([require.resolve("./extensions/signed/bypass_csp-0.0.2-fx.xpi")]);
  67. }
  68. if (options.userAgent) {
  69. await chromeOptions.addArguments("--user-agent=" + JSON.stringify(options.userAgent));
  70. }
  71. if (options.browserExecutablePath) {
  72. chromeOptions.setChromeBinaryPath(options.browserExecutablePath);
  73. }
  74. builder.setChromeOptions(chromeOptions);
  75. driver = await builder.forBrowser("chrome").build();
  76. driver.manage().setTimeouts({ script: null, pageLoad: null, implicit: null });
  77. if (options.browserWidth && options.browserHeight) {
  78. const window = driver.manage().window();
  79. if (window.setRect) {
  80. window.setRect(options.browserHeight, options.browserWidth);
  81. } else if (window.setSize) {
  82. window.setSize(options.browserWidth, options.browserHeight);
  83. }
  84. }
  85. await driver.get(options.url);
  86. let scripts = SCRIPTS.map(scriptPath => fs.readFileSync(require.resolve(scriptPath)).toString()).join("\n");
  87. scripts += "\nlazyLoader.getScriptContent = " + (function (path) { return (RESOLVED_CONTENTS)[path]; }).toString().replace("RESOLVED_CONTENTS", JSON.stringify(RESOLVED_CONTENTS)) + ";";
  88. const mainWindowHandle = driver.getWindowHandle();
  89. const windowHandles = await driver.getAllWindowHandles();
  90. await Promise.all(windowHandles.map(async windowHandle => {
  91. await driver.switchTo().window(windowHandle);
  92. driver.executeScript(scripts);
  93. }));
  94. await driver.switchTo().window(mainWindowHandle);
  95. const result = await driver.executeAsyncScript(getPageDataScript(), options);
  96. if (result.error) {
  97. throw result.error;
  98. } else {
  99. return result.pageData;
  100. }
  101. } finally {
  102. if (driver) {
  103. driver.quit();
  104. }
  105. }
  106. };
  107. function getPageDataScript() {
  108. return `
  109. const [options, callback] = arguments;
  110. getPageData()
  111. .then(pageData => callback({ pageData }))
  112. .catch(error => callback({ error: error.toString() }));
  113. async function getPageData() {
  114. options.insertSingleFileComment = true;
  115. const preInitializationPromises = [];
  116. if (!options.saveRawPage) {
  117. if (!options.removeFrames) {
  118. preInitializationPromises.push(frameTree.getAsync(options));
  119. }
  120. if (options.loadDeferredImages) {
  121. preInitializationPromises.push(lazyLoader.process(options));
  122. }
  123. }
  124. [options.framesData] = await Promise.all(preInitializationPromises);
  125. options.doc = document;
  126. options.win = window;
  127. const SingleFile = SingleFileBrowser.getClass();
  128. const singleFile = new SingleFile(options);
  129. await singleFile.run();
  130. return await singleFile.getPageData();
  131. }
  132. `;
  133. }