Python 如何从弹性搜索中获取唯一的分数搜索结果

我试图显示弹性搜索中唯一的“描述”行。我正在尝试获取多个具有相同描述的重复行中的一行。我不想聚合，因为我还需要来自其他列的其他信息。下面的代码是我试图实现的，但没有实现

  res = esconnection.search(index='data', body={
        # "query": {
        #     "match": {"description": query_input}
        # },
        # "size": 30

        "query": {
            "multi_match": {
                "description": query_input
            }
        },
        "aggs": {
            "top-descriptions": {
                "terms": {
                    "field": "description"
                },
                "aggs": {
                    "top_description_hits": {
                        "top_hits": {
                            "sort": [
                                {
                                    "_score": {
                                        "order": "desc"
                                    }
                                }
                            ],
                            "size": 1
                        }
                    }
                }
            }

        }
    })
    return res["hits"]["hits"]

可用于对字段上的文档进行分组

允许根据字段值折叠搜索结果。这个 折叠是通过仅选择排序靠前的文档来完成的 折叠键。例如，下面的查询检索最佳tweet 对于每个用户，按喜欢的数量对其进行排序

资料

我有三份描述为“棕色狐狸”的文件和两份描述为“棕色狗”的文件

查询：

结果:

仅返回2个文档。

---

字段折叠提供了“内部点击”等功能：如果您想查看组下的文档。使用排序，您可以决定显示哪个文档。

Thanksss alot@jaspreet。这正是我几天来一直在寻找的：）

[
      {
        "_index" : "index4",
        "_type" : "_doc",
        "_id" : "P1lTjHEBF99yL6wF31iA",
        "_score" : 1.0,
        "_source" : {
          "description" : "brown fox"
        }
      },
      {
        "_index" : "index4",
        "_type" : "_doc",
        "_id" : "QFlTjHEBF99yL6wF8liO",
        "_score" : 1.0,
        "_source" : {
          "description" : "brown fox"
        }
      },
      {
        "_index" : "index4",
        "_type" : "_doc",
        "_id" : "QVlTjHEBF99yL6wF91gU",
        "_score" : 1.0,
        "_source" : {
          "description" : "brown fox"
        }
      },
      {
        "_index" : "index4",
        "_type" : "_doc",
        "_id" : "QllUjHEBF99yL6wFFFh5",
        "_score" : 1.0,
        "_source" : {
          "description" : "brown dog"
        }
      },
      {
        "_index" : "index4",
        "_type" : "_doc",
        "_id" : "Q1lUjHEBF99yL6wFGFhQ",
        "_score" : 1.0,
        "_source" : {
          "description" : "brown dog"
        }
      }
    ]

{
  "query": {
    "match": {
      "description": {
        "query": "brown"
      }
    }
  },
  "collapse": {
    "field": "description.keyword" --> notice keyword
  }
}

"hits" : [
      {
        "_index" : "index4",
        "_type" : "_doc",
        "_id" : "P1lTjHEBF99yL6wF31iA",
        "_score" : 0.087011375,
        "_source" : {
          "description" : "brown fox"
        },
        "fields" : {
          "description.keyword" : [
            "brown fox"
          ]
        }
      },
      {
        "_index" : "index4",
        "_type" : "_doc",
        "_id" : "QllUjHEBF99yL6wFFFh5",
        "_score" : 0.087011375,
        "_source" : {
          "description" : "brown dog"
        },
        "fields" : {
          "description.keyword" : [
            "brown dog"
          ]
        }
      }
    ]