spb/zyquo-atlas Public License

1/*2 * Copyright (c) 2010 Arc90 Inc3 *4 * Licensed under the Apache License, Version 2.0 (the "License");5 * you may not use this file except in compliance with the License.6 * You may obtain a copy of the License at7 *8 *     http://www.apache.org/licenses/LICENSE-2.09 *10 * Unless required by applicable law or agreed to in writing, software11 * distributed under the License is distributed on an "AS IS" BASIS,12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.13 * See the License for the specific language governing permissions and14 * limitations under the License.15 */1617/*18 * This code is heavily based on Arc90's readability.js (1.7.1) script19 * available at: http://code.google.com/p/arc90labs-readability20 */2122var REGEXPS = {23  // NOTE: These two regular expressions are duplicated in24  // Readability.js. Please keep both copies in sync.25  unlikelyCandidates:26    /-ad-|ai2html|banner|breadcrumbs|combx|comment|community|cover-wrap|disqus|extra|footer|gdpr|header|legends|menu|related|remark|replies|rss|shoutbox|sidebar|skyscraper|social|sponsor|supplemental|ad-break|agegate|pagination|pager|popup|yom-remote/i,27  okMaybeItsACandidate: /and|article|body|column|content|main|mathjax|shadow/i,28};2930function isNodeVisible(node) {31  // Have to null-check node.style and node.className.includes to deal with SVG and MathML nodes.32  return (33    (!node.style || node.style.display != "none") &&34    !node.hasAttribute("hidden") &&35    //check for "fallback-image" so that wikimedia math images are displayed36    (!node.hasAttribute("aria-hidden") ||37      node.getAttribute("aria-hidden") != "true" ||38      (node.className &&39        node.className.includes &&40        node.className.includes("fallback-image")))41  );42}4344/**45 * Decides whether or not the document is reader-able without parsing the whole thing.46 * @param {Object} options Configuration object.47 * @param {number} [options.minContentLength=140] The minimum node content length used to decide if the document is readerable.48 * @param {number} [options.minScore=20] The minumum cumulated 'score' used to determine if the document is readerable.49 * @param {Function} [options.visibilityChecker=isNodeVisible] The function used to determine if a node is visible.50 * @return {boolean} Whether or not we suspect Readability.parse() will suceeed at returning an article object.51 */52function isProbablyReaderable(doc, options = {}) {53  // For backward compatibility reasons 'options' can either be a configuration object or the function used54  // to determine if a node is visible.55  if (typeof options == "function") {56    options = { visibilityChecker: options };57  }5859  var defaultOptions = {60    minScore: 20,61    minContentLength: 140,62    visibilityChecker: isNodeVisible,63  };64  options = Object.assign(defaultOptions, options);6566  var nodes = doc.querySelectorAll("p, pre, article");6768  // Get <div> nodes which have <br> node(s) and append them into the `nodes` variable.69  // Some articles' DOM structures might look like70  // <div>71  //   Sentences<br>72  //   <br>73  //   Sentences<br>74  // </div>75  var brNodes = doc.querySelectorAll("div > br");76  if (brNodes.length) {77    var set = new Set(nodes);78    [].forEach.call(brNodes, function (node) {79      set.add(node.parentNode);80    });81    nodes = Array.from(set);82  }8384  var score = 0;85  // This is a little cheeky, we use the accumulator 'score' to decide what to return from86  // this callback:87  return [].some.call(nodes, function (node) {88    if (!options.visibilityChecker(node)) {89      return false;90    }9192    var matchString = node.className + " " + node.id;93    if (94      REGEXPS.unlikelyCandidates.test(matchString) &&95      !REGEXPS.okMaybeItsACandidate.test(matchString)96    ) {97      return false;98    }99100    if (node.matches("li p")) {101      return false;102    }103104    var textContentLength = node.textContent.trim().length;105    if (textContentLength < options.minContentLength) {106      return false;107    }108109    score += Math.sqrt(textContentLength - options.minContentLength);110111    if (score > options.minScore) {112      return true;113    }114    return false;115  });116}117118if (typeof module === "object") {119  /* eslint-disable-next-line no-redeclare */120  /* global module */121  module.exports = isProbablyReaderable;122}123