Question

假设您拥有普通会员和高级会员资格的电影订阅服务。

以下是用户活动生成的数据样本，并作为文档存储在集合中：

[
    {
        "eventType": "sessionInfo",
        "userType": "premium",
        "sessionGroupID": 1
    },
    {
        "eventType": "mediaPlay",
        "productSKU": "starwars",
        "sessionGroupID": 1,
        "elapsed": 200
    },
    {
        "eventType": "sessionInfo",
        "userType": "premium",
        "sessionGroupID": 2
    },
    {
        "eventType": "mediaPlay",
        "productSKU": "xmen",
        "sessionGroupID": 2,
        "elapsed": 500
    },
    {
        "eventType": "sessionInfo",
        "userType": "normal",
        "sessionGroupID": 3
    },
    {
        "eventType": "mediaPlay",
        "productSKU": "xmen",
        "sessionGroupID": 3,
        "elapsed": 10
    },
    {
        "eventType": "sessionInfo",
        "userType": "normal",
        "sessionGroupID": 4
    },
    {
        "eventType": "mediaPlay",
        "productSKU": "xmen",
        "sessionGroupID": 4,
        "elapsed": 100
    },
    {
        "eventType": "sessionInfo",
        "userType": "normal",
        "sessionGroupID": 5
    },
    {
        "eventType": "mediaPlay",
        "productSKU": "xmen",
        "sessionGroupID": 5,
        "elapsed": 5
    },
    {
        "eventType": "mediaPlay",
        "productSKU": "starwars",
        "sessionGroupID": 5,
        "elapsed": 25
    }
]

您可以看到有两个“eventTypes”：

“sessionInfo”文档，包含整个公共信息用户会话
“mediaPlay”文件，用于存储多少秒电影被观看了。

（每个“mediaPlay”事件都包含sessionGroupID，因此它可以与该会话相关联。）

问题＃1：

鉴于总共有数千万份文件，您如何编写一个总计每部电影的已用观看时间的查询，按用户类型分组？

所需的查询结果：

premium users - total of "elapsed":
    xmen: 500
    starwars: 200

normal users - total of "elapsed":
    xmen: 115
    starwars: 25

问题＃2：

如果数据没有针对此类查询进行最佳结构化，那么理想的结构是什么？

例如，将“mediaInfo”文档中的“mediaPlay”事件嵌套为嵌套数组会更好吗？

喜欢这个吗？

[
    {
        "eventType": "sessionInfo",
        "userType": "premium",
        "sessionGroupID": 1,
        "viewLog": [
            {
                "eventType": "mediaPlay",
                "productSKU": "starwars",
                "sessionGroupID": 1,
                "elapsed": 200
            }
        ]
    },
    {
        "eventType": "sessionInfo",
        "userType": "premium",
        "sessionGroupID": 2,
        "viewLog": [
            {
                "eventType": "mediaPlay",
                "productSKU": "xmen",
                "sessionGroupID": 2,
                "elapsed": 500
            }
        ]
    },
    {
        "eventType": "sessionInfo",
        "userType": "normal",
        "sessionGroupID": 3,
        "viewLog": [
            {
                "eventType": "mediaPlay",
                "productSKU": "xmen",
                "sessionGroupID": 3,
                "elapsed": 10
            }
        ]
    },
    {
        "eventType": "sessionInfo",
        "userType": "normal",
        "sessionGroupID": 4,
        "viewLog": [
            {
                "eventType": "mediaPlay",
                "productSKU": "xmen",
                "sessionGroupID": 4,
                "elapsed": 100
            }
        ]
    },
    {
        "eventType": "sessionInfo",
        "userType": "normal",
        "sessionGroupID": 5,
        "viewLog": [
            {
                "eventType": "mediaPlay",
                "productSKU": "xmen",
                "sessionGroupID": 5,
                "elapsed": 5
            },
            {
                "eventType": "mediaPlay",
                "productSKU": "starwars",
                "sessionGroupID": 5,
                "elapsed": 25
            }
        ]
    }
]

感谢您提供任何指导和建议！

Answer 1

以下查询迭代集合并收集按userTypes分组的所有会话ID。然后它创建一个子查询，它遍历集合并收集所有电影以及eventType为“mediaPlay”且收集的会话包含sessionGroupID的已用时间总和。

@@coll是bind parameter，其中包含您的收藏名称。

FOR doc IN @@coll
  FILTER doc.eventType == "sessionInfo"
  COLLECT userTypes = doc.userType INTO sessions = doc.sessionGroupID
  RETURN {
    "userTypes" : userTypes,
    "movies" : (
      FOR event IN @@coll
        FILTER event.sessionGroupID IN sessions
        FILTER event.eventType == "mediaPlay"
        COLLECT movie = event.productSKU INTO elapsed = event.elapsed
        RETURN { "movie" : movie, "elapsed" : SUM(elapsed) }
      )
  }

此查询的结果是：

[
  {
    "userTypes": "normal",
    "movies": [
      {
        "movie": "starwars",
        "elapsed": 25
      },
      {
        "movie": "xmen",
        "elapsed": 115
      }
    ]
  },
  {
    "userTypes": "premium",
    "movies": [
      {
        "movie": "starwars",
        "elapsed": 200
      },
      {
        "movie": "xmen",
        "elapsed": 500
      }
    ]
  }
]

关于你的第二个问题。嵌套数组/对象不会优化此查询，但您应该将数据拆分为两个集合。每个eventType一个（例如，将事件命名为eventType sessionInfo和mediaPlay）。这减少了所需过滤器语句的数量，更重要的是，它允许您通过sessionInfos和mediaPlay单独查询，从而大大提高您的性能。

查询将如下所示：

FOR doc IN sessionInfo
  COLLECT userTypes = doc.userType INTO sessions = doc.sessionGroupID
  RETURN {
    "userTypes" : userTypes,
    "movies" : (
      FOR event IN mediaPlay
        FILTER event.sessionGroupID IN sessions
        COLLECT movie = event.productSKU INTO elapsed = event.elapsed
        RETURN { "movie" : movie, "elapsed" : SUM(elapsed) }
      )
  }

是否可以在ArangoDB中编写查询以聚合已连接文档中的值？

问题＃1：

问题＃2：

1 个答案: