| 1 |
// Copyright 2017 The Closure Library Authors. All Rights Reserved. |
| 2 |
// |
| 3 |
// Licensed under the Apache License, Version 2.0 (the "License"); |
| 4 |
// you may not use this file except in compliance with the License. |
| 5 |
// You may obtain a copy of the License at |
| 6 |
// |
| 7 |
// http://www.apache.org/licenses/LICENSE-2.0 |
| 8 |
// |
| 9 |
// Unless required by applicable law or agreed to in writing, software |
| 10 |
// distributed under the License is distributed on an "AS-IS" BASIS, |
| 11 |
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 12 |
// See the License for the specific language governing permissions and |
| 13 |
// limitations under the License. |
| 14 |
|
| 15 |
|
| 16 |
/** |
| 17 |
* @fileoverview Contains utility methods to extract text content from HTML. |
| 18 |
* @supported IE 10+, Chrome 26+, Firefox 22+, Safari 7.1+, Opera 15+ |
| 19 |
*/ |
| 20 |
|
| 21 |
goog.provide('goog.html.textExtractor'); |
| 22 |
|
| 23 |
goog.require('goog.array'); |
| 24 |
goog.require('goog.dom.TagName'); |
| 25 |
goog.require('goog.html.sanitizer.HtmlSanitizer'); |
| 26 |
goog.require('goog.object'); |
| 27 |
goog.require('goog.userAgent'); |
| 28 |
|
| 29 |
|
| 30 |
/** |
| 31 |
* Safely extracts text from an untrusted HTML string using the HtmlSanitizer. |
| 32 |
* Compared to goog.html.utils.stripHtmlTags, it tries to be smarter about |
| 33 |
* printing newlines between blocks and leave out textual content that would not |
| 34 |
* be displayed to the user (such as SCRIPT and STYLE tags). |
| 35 |
* @param {string} html The untrusted HTML string. |
| 36 |
* @return {string} |
| 37 |
*/ |
| 38 |
// TODO(pelizzi): consider an optional bool parameter to also extract the text |
| 39 |
// content of alt attributes and such. |
| 40 |
goog.html.textExtractor.extractTextContent = function(html) { |
| 41 |
if (!goog.html.textExtractor.isSupported()) { |
| 42 |
return ''; |
| 43 |
} |
| 44 |
// Disable all attributes except style to protect against DOM clobbering. |
| 45 |
var sanitizer = new goog.html.sanitizer.HtmlSanitizer.Builder() |
| 46 |
.onlyAllowAttributes(['style']) |
| 47 |
.allowCssStyles() |
| 48 |
.build(); |
| 49 |
// The default policy of the sanitizer strips the content of tags such as |
| 50 |
// SCRIPT and STYLE, whose non-textual content would otherwise end up in the |
| 51 |
// extracted text. |
| 52 |
var sanitizedNodes = sanitizer.sanitizeToDomNode(html); |
| 53 |
// textContent and innerText do not handle spacing between block elements |
| 54 |
// properly. We need to reimplement a similar algorithm ourselves and account |
| 55 |
// for spacing between block elements. |
| 56 |
return goog.html.textExtractor.extractTextContentFromNode_(sanitizedNodes) |
| 57 |
.trim(); |
| 58 |
}; |
| 59 |
|
| 60 |
|
| 61 |
/** |
| 62 |
* Recursively extract text from the supplied DOM node and its descendants. |
| 63 |
* @param {!Node} node |
| 64 |
* @return {string} |
| 65 |
* @private |
| 66 |
*/ |
| 67 |
goog.html.textExtractor.extractTextContentFromNode_ = function(node) { |
| 68 |
switch (node.nodeType) { |
| 69 |
case Node.ELEMENT_NODE: |
| 70 |
var element = /** @type {!Element} */ (node); |
| 71 |
if (element.tagName == goog.dom.TagName.BR) { |
| 72 |
return '\n'; |
| 73 |
} |
| 74 |
var result = goog.array |
| 75 |
.map( |
| 76 |
node.childNodes, |
| 77 |
goog.html.textExtractor.extractTextContentFromNode_) |
| 78 |
.join(''); |
| 79 |
if (goog.html.textExtractor.isBlockElement_(element)) { |
| 80 |
result = '\n' + result + '\n'; |
| 81 |
} |
| 82 |
return result; |
| 83 |
case Node.TEXT_NODE: |
| 84 |
return node.nodeValue.replace(/\s+/g, ' ').trim(); |
| 85 |
default: |
| 86 |
return ''; |
| 87 |
} |
| 88 |
}; |
| 89 |
|
| 90 |
|
| 91 |
/** |
| 92 |
* A set of block elements. |
| 93 |
* @private @const {!Object<!goog.dom.TagName, boolean>} |
| 94 |
*/ |
| 95 |
goog.html.textExtractor.BLOCK_ELEMENTS_ = goog.object.createSet( |
| 96 |
goog.dom.TagName.ADDRESS, goog.dom.TagName.BLOCKQUOTE, |
| 97 |
goog.dom.TagName.CENTER, goog.dom.TagName.DIV, goog.dom.TagName.DL, |
| 98 |
goog.dom.TagName.FIELDSET, goog.dom.TagName.FORM, goog.dom.TagName.H1, |
| 99 |
goog.dom.TagName.H2, goog.dom.TagName.H3, goog.dom.TagName.H4, |
| 100 |
goog.dom.TagName.H5, goog.dom.TagName.H6, goog.dom.TagName.HR, |
| 101 |
goog.dom.TagName.OL, goog.dom.TagName.P, goog.dom.TagName.PRE, |
| 102 |
goog.dom.TagName.TABLE, goog.dom.TagName.UL); |
| 103 |
|
| 104 |
|
| 105 |
/** |
| 106 |
* Returns true whether this is a block element, i.e. the browser would visually |
| 107 |
* separate the text content from the text content of the previous node. |
| 108 |
* @param {!Element} element |
| 109 |
* @return {boolean} |
| 110 |
* @private |
| 111 |
*/ |
| 112 |
goog.html.textExtractor.isBlockElement_ = function(element) { |
| 113 |
return element.style.display == 'block' || |
| 114 |
goog.html.textExtractor.BLOCK_ELEMENTS_.hasOwnProperty(element.tagName); |
| 115 |
}; |
| 116 |
|
| 117 |
|
| 118 |
/** |
| 119 |
* Whether the browser supports the text extractor. The extractor depends on the |
| 120 |
* HTML Sanitizer, which only supports IE starting from version 10. |
| 121 |
* Visible for testing. |
| 122 |
* @return {boolean} |
| 123 |
* @package |
| 124 |
*/ |
| 125 |
goog.html.textExtractor.isSupported = function() { |
| 126 |
return !goog.userAgent.IE || goog.userAgent.isVersionOrHigher(10); |
| 127 |
}; |