[Python] elasticsearch bulk insert contain _id

 

 elasticserch에 데이터를 넣으면 UUID를 이용하여 자동으로 생성해주지만 실제 RDB와 연동하여 사용할 경우 ID를 PK과 값과

같이 관리할 일이 있다.

 

---

 

Sample Dataframe

 

import numpy as np
import pandas as pd
from datetime import datetime
from elasticsearch import Elasticsearch
from elasticsearch import helpers



productId = ['AAA001', 'AAA002', 'AAA003' , 'AAA004', 'AAA005']
price = [15000, 11000, 21000, 25000, 14500]
soldDay = [datetime(2020, 1, 2, 0, 0),
           datetime(2020, 1, 3, 0, 0),
           datetime(2020, 1, 5, 0, 0),
           datetime(2020, 1, 7, 0, 0),
           datetime(2020, 1, 9, 0, 0)]

df = pd.DataFrame([ x for x in zip(productId, price, soldDay)], columns=['productId','price','soldDay'])
df

 

 

 

 

bulk insert

 

es = Elasticsearch(host=' ', port=' ')

data = [
  {
    "_index": "product",
    "_type": "product",
    "_id": x[0],
    "_source": {
        "price": x[1],
        "soldDay": x[2],
        "timestamp": datetime.today().date()}
  }
    for x in zip(df['productId'],df['price'],df['soldDay'])
]

helpers.bulk(es, data)

 

 

 

 

 

productId 컬럼이 _id로 등록 

 

 

 

 

_id 'AAA001' 로 검색

 

 

res = es.get(index="product", doc_type='product', id='AAA001')
print(res['_source'])



{'price': 15000, 'soldDay': '2020-01-02T00:00:00', 'timestamp': '2020-03-04'}