页面树结构

2017-11-09 ApacheCN 开源组织,第二期邀请成员活动,一起走的更远 : http://www.apachecn.org/member/209.html


MachineLearning 优酷地址 : http://i.youku.com/apachecn

转至元数据结尾
转至元数据起始

Range Aggregation

A multi-bucket value source based aggregation that enables the user to define a set of ranges-每个代表一个bucket。在聚合过程中,从每个文件中提取的值将根据每个存储区范围进行检查,并对相关/匹配文档进行“bucket”,值得注意的是,注意,此聚合包含了from值,不包括每个范围的to的值。

例子:

{
    "aggs" : {
        "price_ranges" : {
            "range" : {
                "field" : "price",
                "ranges" : [
                    { "to" : 50 },
                    { "from" : 50, "to" : 100 },
                    { "from" : 100 }
                ]
            }
        }
    }
}

响应结果:

{
    ...

    "aggregations": {
        "price_ranges" : {
            "buckets": [
                {
                    "to": 50,
                    "doc_count": 2
                },
                {
                    "from": 50,
                    "to": 100,
                    "doc_count": 4
                },
                {
                    "from": 100,
                    "doc_count": 4
                }
            ]
        }
    }
}


Keyed Response

将keyed标志设置为true会将一个惟一的字符串键与每个bucket关联起来,并将范围作为散列而不是数组返回:

{
    "aggs" : {
        "price_ranges" : {
            "range" : {
                "field" : "price",
                "keyed" : true,
                "ranges" : [
                    { "to" : 50 },
                    { "from" : 50, "to" : 100 },
                    { "from" : 100 }
                ]
            }
        }
    }
}

响应结果:

{
    ...

    "aggregations": {
        "price_ranges" : {
            "buckets": {
                "*-50.0": {
                    "to": 50,
                    "doc_count": 2
                },
                "50.0-100.0": {
                    "from": 50,
                    "to": 100,
                    "doc_count": 4
                },
                "100.0-*": {
                    "from": 100,
                    "doc_count": 4
                }
            }
        }
    }
}

可以为每个范围自定义key:

{
    "aggs" : {
        "price_ranges" : {
            "range" : {
                "field" : "price",
                "keyed" : true,
                "ranges" : [
                    { "key" : "cheap", "to" : 50 },
                    { "key" : "average", "from" : 50, "to" : 100 },
                    { "key" : "expensive", "from" : 100 }
                ]
            }
        }
    }
}


Script

{
    "aggs" : {
        "price_ranges" : {
            "range" : {
                "script" : {
                    "lang": "painless",
                    "inline": "doc['price'].value"
                },
                "ranges" : [
                    { "to" : 50 },
                    { "from" : 50, "to" : 100 },
                    { "from" : 100 }
                ]
            }
        }
    }
}

这将把script参数解释为具有painless脚本语言和no script参数的inline脚本。要使用文件脚本,请使用以下语法

{
    "aggs" : {
        "price_ranges" : {
            "range" : {
                "script" : {
                    "file": "my_script",
                    "params": {
                        "field": "price"
                    }
                },
                "ranges" : [
                    { "to" : 50 },
                    { "from" : 50, "to" : 100 },
                    { "from" : 100 }
                ]
            }
        }
    }
}

对于索引脚本,用id参数替换file参数。

Value Script

假设产品价格是美元,但我们希望得到欧元的价格区间,我们可以使用value script在聚合之前转换价格(假设汇率是0.8

{
    "aggs" : {
        "price_ranges" : {
            "range" : {
                "field" : "price",
                "script" : {
                    "lang": "painless",
                    "inline": "_value * params.conversion_rate",
                    "params" : {
                        "conversion_rate" : 0.8
                    }
                },
                "ranges" : [
                    { "to" : 35 },
                    { "from" : 35, "to" : 70 },
                    { "from" : 70 }
                ]
            }
        }
    }
}


Sub Aggregations

下面的示例不仅将文档“bucket”到不同的bucket,而且计算每个价格范围内的价格统计

{
    "aggs" : {
        "price_ranges" : {
            "range" : {
                "field" : "price",
                "ranges" : [
                    { "to" : 50 },
                    { "from" : 50, "to" : 100 },
                    { "from" : 100 }
                ]
            },
            "aggs" : {
                "price_stats" : {
                    "stats" : { "field" : "price" }
                }
            }
        }
    }
}

响应结果:

{
    "aggregations": {
        "price_ranges" : {
            "buckets": [
                {
                    "to": 50,
                    "doc_count": 2,
                    "price_stats": {
                        "count": 2,
                        "min": 20,
                        "max": 47,
                        "avg": 33.5,
                        "sum": 67
                    }
                },
                {
                    "from": 50,
                    "to": 100,
                    "doc_count": 4,
                    "price_stats": {
                        "count": 4,
                        "min": 60,
                        "max": 98,
                        "avg": 82.5,
                        "sum": 330
                    }
                },
                {
                    "from": 100,
                    "doc_count": 4,
                    "price_stats": {
                        "count": 4,
                        "min": 134,
                        "max": 367,
                        "avg": 216,
                        "sum": 864
                    }
                }
            ]
        }
    }
}

如果sub aggregation(子集合)也基于与范围聚合相同的值(如上例中的统计信息聚合),则可以省略其值的定义。以下内容将返回与上述相同的结果

{
    "aggs" : {
        "price_ranges" : {
            "range" : {
                "field" : "price",
                "ranges" : [
                    { "to" : 50 },
                    { "from" : 50, "to" : 100 },
                    { "from" : 100 }
                ]
            },
            "aggs" : {
                "price_stats" : {
                    "stats" : {} #1
                }
            }
        }
    }
}

#1  我们不需要指定price,因为我们默认从父范围聚合“继承”它

  • 无标签