按最大匹配排序字符串匹配结果

Question

我想搜索一个字符串中的所有匹配，并返回所有按最大匹配结果排序的结果，假设我有一些字符串：

var strArray = [
  "This is my number one string",
  "Another string that contains number",
  "Just for example string"
];

// Results of search "another number" should be:
var resultArrayOfIndexes = [1, 0];

到目前为止，我可以在一个字符串中进行搜索，但它会返回至少有一个匹配的所有索引，但我希望结果数组按最大匹配数排序。我的代码：

function findMatch(list, phrase) {
  var preparedList = [],
    value = "";
  if (config.get("list").match.enabled) {
    for (var i = 0, length = list.length; i < length; i += 1) {

      value = config.get("getValue")(list[i]);
      var words = phrase.split(' ');
      var listMatchArr = [];
      $.each(words, function(idx, word) {
        var W = word.replace(/[\W_]+/g, ""); // match on alphaNum chars only
        if (match(value, W) && $.inArray(i, listMatchArr) == -1) { //phrase
          preparedList.push(list[i]);
          listMatchArr.push(i);
        };
      });

    }

  } else {
    preparedList = list;
  }

  return preparedList;
}

Answer 1

我假设需要不区分大小写的搜索。

下面的代码将短语更改为单个单词的数组，然后映射列表以获取{index: 0, matches:1}形式的对象数组，然后过滤掉没有匹配的对象，然后排序，然后再次映射以仅获取索引。

​

function findMatch(list, phrase) {
  var searchTerms = phrase.toLowerCase().split(/\s+/);
  return list.map(function(v, i) {
    v = v.toLowerCase();
    return {
      index: i,
      matches: searchTerms.reduce(function(a, c) {
        return a + (v.indexOf(c) !=-1 ? 1 : 0);
      }, 0)
    };
  })
  .filter(function(v) { return v.matches > 0; })
  .sort(function(a, b) { return b.matches - a.matches; })
  .map(function(v) { return v.index; });
}

var strArray = [
"This is my number one string", "Another string that contains number","Just for example string"
];

console.log(findMatch(strArray, "another number"));

​

或者扩展以下内容，以获得与ES6功能基本相同的功能：

​

function findMatch(list, phrase) {
  var searchTerms = phrase.toLowerCase().split(/\s+/);
  return list.map(function(v, i) {
    v = v.toLowerCase();
    return {
      index: i,
      matches: searchTerms.reduce((a, c) => a + (v.includes(c) ? 1 : 0), 0)
    };
  })
  .filter(v => v.matches > 0)
  .sort((a, b) => b.matches - a.matches)
  .map(v => v.index);
}

var strArray = [
"This is my number one string", "Another string that contains number","Just for example string"
];

console.log(findMatch(strArray, "another number"));

​

Answer 2

如果您熟悉正则表达式，则可以使用正则表达式来匹配您的短语，并计算字符串中有多少个单词匹配。假设您还想知道有多少个单词被匹配，您可以将其存储为一个对象数组，其中每个对象都存储计数数和目标字符串。

var strArray = [
    "This is my number one string", "Another string that contains number", "Just for example string"
];

function findMatch(list, phrase){
    var words = phrase.split(" ");

    var pattern = "";
    var length = words.length;
    // create pattern for regex match
    for(var i = 0; i < length; i++){
        pattern += words[i];
        if(i < length-1){
            pattern += "|";
        }
    }

    var counts = [];
    var re = new RegExp(pattern,"g");
    for(var i = 0; i < list.length; i++){
        var count = (list[i].toLowerCase().match(re) || []).length;
        //add to array if matched
        if(count > 0){
            counts.push({count:count,string:list[i]});
        }
    }
    //sort by max match 
    counts.sort(function(a,b){
        return b.count-a.count;
    });

    console.log(counts);
}

findMatch(strArray, "another number");

结果将如下所示：

[ { count: 2, string: 'Another string that contains number' },
  { count: 1, string: 'This is my number one string' },
  { count: 0, string: 'Just for example string' } ]