single-file-jsdom.js 4.7 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152
  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, Buffer */
  24. const fs = require("fs");
  25. const crypto = require("crypto");
  26. const jsdom = require("jsdom");
  27. const dataUri = require("strong-data-uri");
  28. const iconv = require("iconv-lite");
  29. const request = require("request-promise-native");
  30. const { JSDOM } = jsdom;
  31. const SCRIPTS = [
  32. "./lib/single-file/util/doc-util-core.js",
  33. "./lib/single-file/util/doc-helper.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/single-file-core.js",
  40. "./lib/single-file/modules/html-minifier.js",
  41. "./lib/single-file/modules/css-fonts-minifier.js",
  42. "./lib/single-file/modules/css-fonts-alt-minifier.js",
  43. "./lib/single-file/modules/css-matched-rules.js",
  44. "./lib/single-file/modules/css-medias-alt-minifier.js",
  45. "./lib/single-file/modules/css-rules-minifier.js",
  46. "./lib/single-file/modules/html-images-alt-minifier.js",
  47. "./lib/single-file/modules/html-serializer.js",
  48. ];
  49. SCRIPTS.forEach(scriptPath => eval(fs.readFileSync(scriptPath).toString()));
  50. const docHelper = this.docHelper;
  51. const modules = {
  52. docHelper: docHelper,
  53. srcsetParser: this.srcsetParser,
  54. cssMinifier: this.cssMinifier,
  55. htmlMinifier: this.htmlMinifier,
  56. serializer: this.serializer,
  57. fontsMinifier: this.fontsMinifier.getInstance(this.cssTree, this.fontPropertyParser, docHelper),
  58. fontsAltMinifier: this.fontsAltMinifier.getInstance(this.cssTree),
  59. cssRulesMinifier: this.cssRulesMinifier.getInstance(this.cssTree),
  60. matchedRules: this.matchedRules.getInstance(this.cssTree),
  61. mediasMinifier: this.mediasMinifier.getInstance(this.cssTree, this.mediaQueryParser),
  62. imagesAltMinifier: this.imagesAltMinifier.getInstance(this.srcsetParser)
  63. };
  64. const domUtil = {
  65. getResourceContent,
  66. parseDocContent,
  67. parseSVGContent,
  68. isValidFontUrl,
  69. getContentSize,
  70. digestText
  71. };
  72. exports.getClass = () => {
  73. const DocUtil = this.DocUtilCore.getClass(modules, domUtil);
  74. return this.SingleFileCore.getClass(DocUtil, this.cssTree);
  75. };
  76. function parseDocContent(content, baseURI) {
  77. const doc = (new JSDOM(content, {
  78. contentType: "text/html"
  79. })).window.document;
  80. let baseElement = doc.querySelector("base");
  81. if (!baseElement || !baseElement.getAttribute("href")) {
  82. if (baseElement) {
  83. baseElement.remove();
  84. }
  85. baseElement = doc.createElement("base");
  86. baseElement.setAttribute("href", baseURI);
  87. doc.head.insertBefore(baseElement, doc.head.firstChild);
  88. }
  89. return doc;
  90. }
  91. function parseSVGContent(content) {
  92. return (new JSDOM(content, {
  93. contentType: "image/svg+xml"
  94. })).window.document;
  95. }
  96. async function digestText(algo, text) {
  97. const hash = crypto.createHash(algo.replace("-", "").toLowerCase());
  98. hash.update(text, "utf-8");
  99. return hash.digest("hex");
  100. }
  101. function getContentSize(content) {
  102. return Buffer.byteLength(content, "utf-8");
  103. }
  104. function isValidFontUrl(/* urlFunction */) {
  105. // TODO?
  106. return true;
  107. }
  108. async function getResourceContent(resourceURL, options) {
  109. let resourceContent;
  110. resourceContent = await request({
  111. method: "GET",
  112. uri: resourceURL,
  113. resolveWithFullResponse: true,
  114. encoding: null,
  115. headers: {
  116. "User-Agent": options.userAgent
  117. }
  118. });
  119. return {
  120. getUrl() {
  121. // TODO
  122. return resourceURL;
  123. },
  124. getContentType() {
  125. return resourceContent.headers["content-type"];
  126. },
  127. getStatusCode() {
  128. return resourceContent.statusCode;
  129. },
  130. getSize() {
  131. return resourceContent.body.byteLength;
  132. },
  133. getText(charset) {
  134. return iconv.decode(resourceContent.body, charset);
  135. },
  136. async getDataUri(contentType) {
  137. return dataUri.encode(resourceContent.body, contentType || this.getContentType());
  138. }
  139. };
  140. }