elasticsearch searchAPI 之 search template 查询模板

最近在公司又用到了elasticsearch，也用到了查询模板，顺便写篇文章记录一下查询模板的使用。

以1个需求为例讲解es模板的使用：

页面上某个按钮在一段时间内的点击次数统计，并且可以以小时，天，月为单位进行汇总，并且需要去重。

创建索引，只定义3个字段，user_id, user_name和create_time:

-POST /$ES/event_index

{
  "mappings": {
    "event": {
      "_ttl": {
        "enabled": false
      },
      "_timestamp": {
        "enabled": true,
        "format": "yyyy-MM-dd HH:mm:ss"
      },
      "properties": {
        "user_id": {
          "type": "string",
          "store": "no",
          "index": "not_analyzed"
        },
        "create_time": {
          "type": "date",
          "store": "no",
          "index": "not_analyzed",
          "format": "yyyy-MM-dd HH:mm:ss"
        },
        "user_name": {
          "type": "string",
          "store": "no"
        }
      }
    }
  }
}

定义对应的查询模板，模板名字stats，使用了Cardinality和DateHistogram这两个Aggregation
，其中Date Histogram嵌套在Cardinality里。在定义模板的时候，{ { } } 的表示是个参数，需要调用模板的时候传递进来:

  -POST /$ES/_search/template/stats
{
    "template": {
        "query": {
            "bool": {
                "must": [
                    {
                        "range": {
                            "create_time": {
                                "gte": "{{earliest}}",
                                "lte": "{{latest}}"
                            }
                        }
                    }
                ]
            }
        },
        "size": 0,
        "aggs": {
            "stats_data": {
                "date_histogram": {
                    "field": "create_time",
                    "interval": "{{interval}}"
                },
                "aggs": {
                    "time": {
                        "cardinality": {
                            "field": "user_id"
                        }
                    }
                }
            }
        }
    }
}

Cardinality Aggregation的作用就是类似sql中的distinct，去重。

Date Histogram Aggregation的作用是根据时间进行统计。内部有个interval属性表面统计的范畴。

下面加几条数据到event_index里：

-POST $ES/event_index/event
{
    "user_id": "1",
    "user_name": "format1",
    "create_time": "2015-11-07 12:00:00"
}

-POST $ES/event_index/event
{
    "user_id": "2",
    "user_name": "format2",
    "create_time": "2015-11-07 13:30:00"
}

-POST $ES/event_index/event
{
    "user_id": "3",
    "user_name": "format3",
    "create_time": "2015-11-07 13:30:00"
}

-POST $ES/event_index/event
{
    "user_id": "1",
    "user_name": "format1",
    "create_time": "2015-11-07 13:50:00"
}

-POST $ES/event_index/event
{
    "user_id": "1",
    "user_name": "format1",
    "create_time": "2015-11-07 13:55:00"
}

11-07 12-13点有1条数据，1个用户
11-07 13-14点有4条数据，3个用户

使用模板查询：

curl -XGET "$ES/event_index/_search/template" -d'{
  "template": { "id": "stats" }, 
  "params": { "earliest": "2015-11-07 00:00:00", "latest": "2015-11-07 23:59:59", "interval": "hour" }
}'    

结果：

{
    "took": 3,
    "timed_out": false,
    "_shards": {
        "total": 5,
        "successful": 5,
        "failed": 0
    },
    "hits": {
        "total": 5,
        "max_score": 0,
        "hits": []
    },
    "aggregations": {
        "stats_data": {
            "buckets": [
                {
                    "key_as_string": "2015-11-07 12:00:00",
                    "key": 1446897600000,
                    "doc_count": 1,
                    "time": {
                        "value": 1
                    }
                },
                {
                    "key_as_string": "2015-11-07 13:00:00",
                    "key": 1446901200000,
                    "doc_count": 4,
                    "time": {
                        "value": 3
                    }
                }
            ]
        }
    }
}

12点-13点的只有1条数据，1个用户。13-14点的有4条数据，3个用户。

以天(day)统计：

curl -XGET "$ES/event_index/_search/template" -d'{
  "template": { "id": "stats" }, 
  "params": { "earliest": "2015-11-07 00:00:00", "latest": "2015-11-07 23:59:59", "interval": "day" }
}'    

结果：

{
    "took": 4,
    "timed_out": false,
    "_shards": {
        "total": 5,
        "successful": 5,
        "failed": 0
    },
    "hits": {
        "total": 5,
        "max_score": 0,
        "hits": []
    },
    "aggregations": {
        "stats_data": {
            "buckets": [
                {
                    "key_as_string": "2015-11-07 00:00:00",
                    "key": 1446854400000,
                    "doc_count": 5,
                    "time": {
                        "value": 3
                    }
                }
            ]
        }
    }
}

11-07这一天有5条数据，3个用户。

本文只是简单说明了es查询模板的使用，也简单使用了2个aggregation。更多内容可以去官网查看相关资料。