0

我想在 rethinkdb python 中做两个字段计数数据。示例：我的表是.. 通道表

{
"channel_name":  "channel01" ,
"id":  "58115c5b-af1f-4b1a-b572-20611ba34ee5" ,
"userid":  "b9936a2f-6cea-41ef-a7f8-1812c3192112" ,
"visibility":  "Public"
}

{
"channel_name":  "channel02" ,
"id":  "1852fac4-3056-46d9-9dfa-2cd969872daa" ,
"userid":  "b9936a2f-6cea-41ef-a7f8-1812c3192112" ,
"visibility":  "Public"
}

{
"channel_name":  "channel03" ,
"id":  "4304fa69-173b-4284-9bf8-d9078bb018f8" ,
"userid":  "ef5e924e-9cab-44a1-a381-16b35d0d7578" ,
"visibility":  "Public"
}

事件表

{
"content":  "Welcome to channel01" ,
"id":  "cc9bb13e-07e1-4ec9-b1ec-4546b8f8feda" ,
"parent":  "channel01" ,
"type":  "message" ,
"user": {
      "user_id":  "b9936a2f-6cea-41ef-a7f8-1812c3192112" ,
               "user_name":  "wymh" 
    } 
}

{
"content":  "hello" ,
"id":  "288847a9-ffb5-4531-bdc3-ddb3effbd993" ,
"parent":  "channel02" ,
"type":  "message" ,
"user": {
      "user_id":  "b9936a2f-6cea-41ef-a7f8-1812c3192112" ,
      "user_name":  "wymh" 
    } 
} 

{
"content":  "hi" ,
"id":  "3232f906-d9f1-43af-a594-5da82376f296" ,
"parent":  "channel02" ,
"type":  "message" ,
"user": {
      "user_id":  "ef5e924e-9cab-44a1-a381-16b35d0d7578" ,
      "user_name":  "waiyan" 
    } 
} 

{
"content":  "Nice to meet you" ,
"id":  "77b7afb3-af98-47a0-84bf-9ac8bd1fe14c" ,
"parent":  "channel02" ,
"type":  "message" ,
"user": {
      "user_id":  "b9936a2f-6cea-41ef-a7f8-1812c3192112" ,
      "user_name":  "wymh" 
    } 
} 

{
"content":  "hi" ,
"id":  "31068384-4734-4733-a7f7-677edd21e557" ,
"parent":  "channel03" ,
"type":  "message" ,
"user": {
      "user_id":  "ef5e924e-9cab-44a1-a381-16b35d0d7578" ,
      "user_name":  "waiyan" 
    } 
}

{
"content":  "hi" ,
"id":  "0b535878-3cac-4619-9c48-f0985a7ee587" ,
"parent":  "channel03" ,
"type":  "message" ,
"user": {
      "user_id":  "7e485c95-f0b6-45f7-8af2-d93bde0ea8a3" ,
      "user_name":  "user01" 
    } 
}

我的查询是：

channel_count =(yield r.table("channel").filter(r.row['visibility'] != 'private').inner_join(
              r.table("events"),
              lambda channel, events:
              (channel["channel_name"] == events["parent"])
          ).zip().group('channel_name').count().run())

我的查询输出数据是：

  {u'channel01': 1, u'channel02': 3, u'channel03': 2}

我想得到以下结果：

{'channel01': 1, 'userid': 1}
{'channel02': 3, 'userid': 2}
{'channel03': 2, 'userid': 2}

请帮助我如何编写 python Rethinkdb 查询？

4

1 回答 1

1

如果您正在寻找每个频道的唯一用户 ID 数和事件数，我会执行以下操作（我将其隔开以使其更具可读性 IMO）...

>>> events_per_channel = (
        r
        .table("channel")
        .filter(r.row['visibility'] != 'private')
        .inner_join(
            r.table("events"),
            lambda channel,
            events: (channel["channel_name"] == events["parent"])
        )
        .zip()
        .map(
            {
                'channel_name': r.row['channel_name'],
                'user_id': r.row['user']['user_id']
            }
        )
        .group('channel_name')
        .count()
        .ungroup()
        .map(
            lambda x:
            {
                'channel_name': x['group'],
                'events': x['reduction'],
                'uniq_user_ids': (
                    r
                    .table('events')
                    .filter(
                        {
                            'parent': x['group']
                        }
                    )
                    .coerce_to('array')
                    .map(lambda y: y['user']['user_id'])
                    .distinct()
                    .count()
               )
            }
    )
    .run()
)
>>> print json.dumps(events_per_channel, indent=4)
[
    {
        "channel_name": "channel01", 
        "events": 1, 
        "uniq_user_ids": 1
    }, 
    {
        "channel_name": "channel02", 
        "events": 3, 
        "uniq_user_ids": 2
    }, 
    {
        "channel_name": "channel03", 
        "events": 2, 
        "uniq_user_ids": 2
    }
]

基本上我所做的是使用嵌套查询每个频道的唯一用户数以及使用取消分组，因此我可以使用减少关键字从组中获取原始数据。输出与您要求的不完全一样，因为您要求的是使用通道的名称作为键，使用值作为事件的数量。我所做的是为每个值创建一个键名。如果这不是你真正的意思，请告诉我。

于 2015-09-13T23:54:50.683 回答