Question

这里的示例文档：

{
  "_id": "doc_1",
  "play_count": 1,
  "meta": {
    "ancestors": [
      {"content_type": "Folder"},
      {"content_type": "SerieContainer",
        "_id": "super_doc_1"
      }
    ]
  }
}


{
  "_id": "doc_2",
  "play_count": 10,
  "meta": {
    "ancestors": [
      {"content_type": "Folder"},
      {"content_type": "SerieContainer",
        "_id": "super_doc_1"
      }
    ]
  }
}


{
  "_id": "doc_3",
  "play_count": 100,
  "meta": {
    "ancestors": [
      {"content_type": "Folder"},
      {"content_type": "SerieContainer",
        "_id": "super_doc_1"
      }
    ]
  }
}


{
  "_id": "doc_4",
  "play_count": 500,
  "meta": {
    "ancestors": [
      {"content_type": "Folder"},
      {"content_type": "SerieContainer",
        "_id": "super_doc_2"
      }
    ]
  }
}


{
  "_id": "doc_5",
  "play_count": 5,
  "meta": {
    "ancestors": [
      {"content_type": "Folder"},
      {"content_type": "SerieContainer",
        "_id": "super_doc_2"
      }
    ]
  }
}

是否可以按_id内容＆＃34; content_type＆＃34;字段等于＆＃34; SerieContainer＆＃34;然后得到＆＃34; play_count＆＃34;他们的领域？

例如：

super_doc_1: 111 (doc_1, doc_2, doc_3)
super_doc_2: 505 (doc_4, doc_5)

Answer 1

确保ancestors的类型为nested，以便存在关联。所以你可以得到你的mapping内容如下。我刚刚给出了array类型：

"ancestors": {
          "type": "nested",
          "include_in_parent": true, 
          "properties": {
            "content_type": {
              "index": "not_analyzed",
              "type": "string"
            },
            "_id": {
              "index": "not_analyzed",
              "type": "string"
            }
          }
        }

然后，也许您可以使用aggs这样的groupby查询来{ "aggs": { "ancestors": { "nested": { "path": "item.meta.ancestors" }, "aggs": { "id": { "terms": { "field": "item.meta.ancestors.id" }, "aggs": { "content_type": { "terms": { "field": "item.meta.tags.content_type" } } } } } } } }使用ID和内容类型。

pattern

因此，您可以使用ResultSet来匹配您的内容类型或ID字段以进行过滤。这SO可以派上用场。希望它有所帮助！

按字段的字段分组

1 个答案: