elasticsearch之python案例

config为ES环境配置

#elasticsearch环境配置
ES_HOST = "IP"
ES_PORT = 9200

1. 新增记录bulk

import time
from elasticsearch import Elasticsearch
from elasticsearch import helpers
import config
es = None

def init():
    global es
    es = Elasticsearch([config.ES_HOST], port=config.ES_PORT)

def addRecord():
    print("<======start addRecord===========>")
    actions = []
    #要插入的数据
    action = {
        "_index": "properties_demo-01_2020-08",
        "_type": "_doc",
        "_id": "dtORKXQBYFLC1S8JD5l8",
        "_score": 5.7014313,
        "_source": {
            "formatValue": "59255",
            "productId": "DEMO-01",
            "propertyName": "属性2",
            "property": "prop2",
            "numberValue": 59255,
            "deviceId": "DEMO-01-0110",
            "value": "59255",
            "timestamp": 1598425197323
        }
    }
    actions.append(action)
    str = helpers.bulk(es,actions)
    print(str)
if __name__ == '__main__':
    init()
    addRecord()

1. 删除记录（delete）

# coding=utf-8

import config
from elasticsearch import Elasticsearch

es = None
def init():
    global es
    es = Elasticsearch([config.ES_HOST], port=config.ES_PORT)

def esDel():
    print("<====start del=====>")
    #删除所需三要素
    indexname='properties_demo-01_2020-08'
    doctype='_doc'
    id='dtORKXQBYFLC1S8JD5l8'
    res = es.delete(index=indexname,doc_type=doctype,id=id)
    print(res)
    print("<====end del=====>")

if __name__ == '__main__':
    init()
    esDel()

1. 查询记录（包含aggs聚合）

query下size限制hits中返回记录数

aggs下size默认为10，若超过10需手动设置

# coding=utf-8

import config
from elasticsearch import Elasticsearch

es = None

def init():
    global es
    es = Elasticsearch([config.ES_HOST], port=config.ES_PORT)

def esSearch():
    index = 'properties_'  + '*'
    deviceid =  "DEMO-01-0110"
    body = {
        "query": {
            "bool": {
                "must": [
                    {
                        "term": {"deviceId": deviceid}
                    }
                    # ,{
                    #     "term": {"property": 'prop2'}
                    # }
                ]
                # ,
                # "must_not":[
                #     {"term": {"property": 'stat'}}
                # ]
                ,
                # 时间范围过滤
                "filter": {
                    "range": {"timestamp": {"gt": 1598425197000, "lte": 1598425297000}}
                }
            }
        }
        ,
        "size":0,#query返回的hits长度，这里不需要查询明细内容设置为0
        "aggs":{
            #自定义name，解析buckets时需用到
            "property_agg":{
                "terms":{
                    "field":"property",
                    "size":20, #默认只返回10个桶，若超过10个需手动设置size
                    "order":{
                        "min_property":"desc"
                    }
                },
                "aggs": {
                    "min_property": {
                        "min": {
                            "field": "numberValue"
                        }
                    }
                    ,
                    "max_property": {
                        "max": {
                            "field": "numberValue"
                        }
                    },
                    "avg_property": {
                        "avg": {
                            "field": "numberValue"
                        }
                    },
                    "sum_property": {
                        "sum": {
                            "field": "numberValue"
                        }
                    }
                }
            }


        }
        # ,"sort": {"timestamp": {"order": "desc"}}#,
        # "collapse": {"field": "property"}
    }
    # filter_path = ['hits.hits._id', 'hits.hits._source']
    result = es.search(index=index, body=body)
    print(result['hits'])
    print(result['aggregations'])
    print(result['aggregations']['property_agg'])
    print(result['aggregations']['property_agg']['buckets'])
    buckets = result['aggregations']['property_agg']['buckets']
    for item in buckets:
        print("property:" ,item['key'],"  min_property:",item['min_property']['value'],"  avg_property",item['avg_property']['value'],"  max_property:",item['max_property']['value'],"  sum_property:",item['sum_property']['value'])

if __name__ == '__main__':
    init()
    esSearch()

1. stats聚合查询

stats包含max、min、avg、count、sum

# coding=utf-8

import config
from elasticsearch import Elasticsearch

es = None

def init():
    global es
    es = Elasticsearch([config.ES_HOST], port=config.ES_PORT)

def esSearch():
    index = 'properties_'  + '*'
    deviceid =  "DEMO-01-0110"
    body = {
        "query": {
            "bool": {
                "must": [
                    {
                        "term": {"deviceId": deviceid}
                    }
                    # ,{
                    #     "term": {"property": 'prop2'}
                    # }
                ]
                # ,
                # "must_not":[
                #     {"term": {"property": 'stat'}}
                # ]
                ,
                # 时间范围过滤
                "filter": {
                    "range": {"timestamp": {"gt": 1598425197000, "lte": 1598425297000}}
                }
            }
        }
        ,
        "size":0,#query返回的hits长度，这里不需要查询明细内容设置为0
        "aggs":{
            #自定义name，解析buckets时需用到
            "property_agg":{
                "terms":{
                    "field":"property",
                    "size":20  #默认只返回10个桶，若超过10个需手动设置size

                },
                "aggs": {
                    "stats_property": {
                        "stats": {#stats包含max、min、avg、count、sum
                            "field": "numberValue"
                        }
                    }

                }
            }


        }
        # ,"sort": {"timestamp": {"order": "desc"}}#,
        # "collapse": {"field": "property"}
    }
    # filter_path = ['hits.hits._id', 'hits.hits._source']
    result = es.search(index=index, body=body)
    print(result['hits'])
    print(result['aggregations'])
    print(result['aggregations']['property_agg'])
    print(result['aggregations']['property_agg']['buckets'])
    buckets = result['aggregations']['property_agg']['buckets']
    #解析buckets
    for item in buckets:
        print("property:" ,item['key'],"  min:",item['stats_property']['min'],"  avg",item['stats_property']['avg'],"  max:",
              item['stats_property']['max'],"  sum:",item['stats_property']['sum'],"  count:",item['stats_property']['count'])

if __name__ == '__main__':
    init()
    esSearch()