Question

我正在尝试为一个业务场景制定一个查询，其中我们有一个名为“types”的嵌套字段类型（即像字符串的 ArrayList）。以下是将“类型”作为字段之一的示例索引文档。

文档 1： {“类型”：[ { "标签" : "对话框", }, { "标签" : "暴力", }, { "标签": "语言", } }

文档 2： {“类型”：[ { "标签" : "对话框", } }

现在，要求是搜索查询最多匹配字段值中的一个值，即如果用户搜索“Dialog”，则它应该只返回文档 2 而不是文档 1，因为它还有其他值存在于场。基本上，它应该只获取那些与单个搜索查询值完全匹配的记录，排除字段中存在的所有其他值。

以下是映射：

{
  "media-hub-asset-metadata": {
    "mappings": {
      "dynamic": "true",
      "properties": {
        "Metadata": {
          "properties": {
            "Actors": {
              "type": "text",
              "fields": {
                "keyword": {
                  "type": "keyword",
                  "ignore_above": 256,
                  "normalizer": "lowercase_normalizer"
                },
                "ngram": {
                  "type": "text",
                  "analyzer": "ngram_tokenizer_analyzer"
                }
              }
            },
            "Types": {
              "type": "nested",
              "properties": {
                "Acronym": {
                  "type": "text",
                  "fields": {
                    "keyword": {
                      "type": "keyword",
                      "ignore_above": 256
                    }
                  }
                },
                "Display": {
                  "type": "text",
                  "fields": {
                    "keyword": {
                      "type": "keyword",
                      "ignore_above": 256
                    }
                  }
                },
                "Label": {
                  "type": "text",
                  "fields": {
                    "keyword": {
                      "type": "keyword",
                      "ignore_above": 256,
                      "normalizer": "lowercase_normalizer"
                    },
                    "ngram": {
                      "type": "text",
                      "analyzer": "ngram_tokenizer_analyzer"
                    }
                  }
                },
                "TVLabel": {
                  "type": "text",
                  "fields": {
                    "keyword": {
                      "type": "keyword",
                      "ignore_above": 256,
                      "normalizer": "lowercase_normalizer"
                    },
                    "ngram": {
                      "type": "text",
                      "analyzer": "ngram_tokenizer_analyzer"
                    }
                  }
                }
              }
            }
          }
        },
        "MetadataType": {
          "type": "text",
          "fields": {
            "keyword": {
              "type": "keyword",
              "ignore_above": 256,
              "normalizer": "lowercase_normalizer"
            },
            "ngram": {
              "type": "text",
              "analyzer": "ngram_tokenizer_analyzer"
            }
          }
        },
        "Network": {
          "type": "text",
          "fields": {
            "keyword": {
              "type": "keyword",
              "ignore_above": 256
            }
          }
        }
      }
    }
  }
}

样本索引文档：

{
  "took" : 4,
  "timed_out" : false,
  "_shards" : {
    "total" : 5,
    "successful" : 5,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 9139,
      "relation" : "eq"
    },
    "max_score" : 1.0,
    "hits" : [
      {
        "_index" : "media-hub-asset-metadata",
        "_type" : "_doc",
        "_id" : "1640655|VOD",
        "_score" : 1.0,
        "_source" : {
          "AssetId" : 1640655,
          "MaterialId" : "XMX1311",
          "Metadata" : {
            "Actors" : [
              "Owen, Clive",
              "Mueller-Stahl, Armin",
              "Watts, Naomi"
            ],
            "AirDate" : "2013-05-01T00:00:00Z",
            "ClosedCaption" : true,
            "Code" : "",
            "Types" : [
              {
                "Label" : "Dialog",
                "TVLabel" : "D"
              },
              {
                "Label" : "Violence",
                "TVLabel" : "V"
              },
              {
                "Label" : "Language",
                "TVLabel" : "L"
              }
            ]
          },
          "MetadataType" : "VOD"
        }
      }
    ]
  }
}

非常感谢任何帮助！提前致谢

Answer 1

您需要将 script_score 与 function score query 一起使用。

试试下面的查询

{
  "query": {
    "function_score": {
      "query": {
        "bool": {
          "must": [
            {
              "nested": {
                "path": "types",
                "query": {
                  "bool": {
                    "must": [
                      {
                        "match": {
                          "types.Label": "Dialog"
                        }
                      }
                    ]
                  }
                }
              }
            }
          ]
        }
      },
      "functions": [
        {
          "script_score": {
            "script": {
              "source": "params._source.containsKey('types') && params._source['types'] != null && params._source.types.size() == 1 ? 1 : 0"
            }
          }
        }
      ],
      "min_score": 0.5             // note this
    }
  }
}

搜索结果将是

"hits": [
      {
        "_index": "67594441",
        "_type": "_doc",
        "_id": "2",
        "_score": 0.53899646,
        "_source": {
          "types": [
            {
              "Label": "Dialog"
            }
          ]
        }
      ]

嵌套字段类型的弹性搜索查询

1 个答案: