elasticsearch - ElasticSearch-整个文档中的词组匹配？不只是一个特定领域

Question

有什么方法可以在整个文档中使用 flex match_phrase吗？不只是一个特定 Realm 。
我们希望用户能够输入带引号的搜索词，并在文档中的任何位置进行词组匹配。

{
    "size": 20,
    "from": 0,
    "query": {
        "match_phrase": {
            "my_column_name": "I want to search for this exact phrase"
        }
    }
}

目前，我只发现了特定 Realm 的词组匹配。我必须指定字段以在其中进行短语匹配。
我们的文档有数百个字段，因此我认为在每个match_phrase查询中手动输入600多个字段并不可行。生成的JSON将非常庞大。

Answer 1

您可以使用带有类型短语的multi-match query在每个字段上运行match_phrase查询，并使用最佳字段中的_score。参见词组和词组前缀。

If no fields are provided, the multi_match query defaults to the index.query.default_field index settings, which in turn defaults to *. This extracts all fields in the mapping that are eligible to term queries and filters the metadata fields. All extracted fields are then combined to build a query.

添加带有索引数据，搜索查询和搜索结果的工作示例
索引数据:

{
    "name":"John",
    "cost":55,
    "title":"Will Smith"
}
{
    "name":"Will Smith",
    "cost":55,
    "title":"book"
}

搜索查询:

{
  "query": {
    "multi_match": {
      "query": "Will Smith",
      "type": "phrase"
    }
  }
}

搜索结果:

"hits": [
      {
        "_index": "64519840",
        "_type": "_doc",
        "_id": "1",
        "_score": 1.2199391,
        "_source": {
          "name": "Will Smith",
          "cost": 55,
          "title": "book"
        }
      },
      {
        "_index": "64519840",
        "_type": "_doc",
        "_id": "2",
        "_score": 1.2199391,
        "_source": {
          "name": "John",
          "cost": 55,
          "title": "Will Smith"
        }
      }
    ]