如何使用 Python 和 Elasticsearch 遍历返回的 JSON 对象一名开发者在使用 Python 和 E

一名开发者在使用 Python 和 Elasticsearch 进行数据查询时遇到了问题。他想要遍历返回的 JSON 对象并打印每个电子邮件地址，但是在尝试使用 for 循环时遇到了错误：TypeError: string indices must be integers。

问题代码如下：

import json
from elasticsearch import Elasticsearch

es = Elasticsearch()

resp = es.search(index="mynewcontacts", body={"query": {"match_all": {}}})
response = json.dumps(resp)
data = json.loads(response)
#print data["hits"]["hits"][0]["_source"]["email"]
for row in data:
    print row["hits"]["hits"][0]["_source"]["email"]
return "OK"

JSON 数据示例：

{"timed_out": false, "took": 1, "_shards": {"successful": 5, "total": 5, "failed": 0}, "hits": {"max_score": 1.0, "total": 7, "hits": [{"_index": "mynewcontacts", "_type": "contact", "_score": 1.0, "_source": {"email": "sharon.zhuo@xxxxx.com.cn", "position": "Sr.Researcher", "last": "Zhuo", "first": "Sharon", "company": "Tabridge Executive Search"}, "_id": "AVYmLMlKJVSAh7zyC0xf"},
{"_index": "mynewcontacts", "_type": "contact", "_score": 1.0, "_source": {"email": "andrew.springthorpe@xxxxx.gr.jp", "position": "Vice President", "last": "Springthorpe", "first": "Andrew", "company": "SBC Group"}, "_id": "AVYmLMlRJVSAh7zyC0xg"}, {"_index": "mynewcontacts", "_type": "contact", "_score": 1.0, "_source": {"email": "mjbxxx@xxx.com", "position": "Financial Advisor", "last": "Bell", "first": "Margaret Jacqueline", "company": "Streamline"}, "_id": "AVYmLMlXJVSAh7zyC0xh"}, {"_index": "mynewcontacts", "_type": "contact", "_score": 1.0, "_source": {"email": "kokaixxx@xxxx.com", "position": "Technical Solutions Manager MMS North Asia", "last": "Okai", "first": "Kensuke", "company": "Criteo"}, "_id": "AVYmLMlfJVSAh7zyC0xi"}, {"_index": "mynewcontacts", "_type": "contact", "_score": 1.0, "_source": {"email": "mizuxxxxto@zszs.com", "position": "Sr. Strategic Account Executive", "last": "Kato", "first": "Mizuto", "company": "Twitter"}, "_id": "AVYmLMlkJVSAh7zyC0xj"}, {"_index": "mynewcontacts", "_type": "contact", "_score": 1.0, "_source": {"email": "abc@example.com", "position": "Design Manager", "last": "Okada", "first": "Kengo", "company": "ON Semiconductor"}, "_id": "AVYmLMlpJVSAh7zyC0xk"}, {"_index": "mynewcontacts", "_type": "contact", "_score": 1.0, "_source": {"email": "007@example.com", "position": "Legal Counsel", "last": "Lei", "first": "Yangzi (Karen)", "company": "Samsung China Semiconductor"}, "_id": "AVYmLMkUJVSAh7zyC0xe"}]}}

解决方案

要解决这个问题，可以采用以下两种方法：

方法一：

修改 for 循环的起始点，如下所示：

for row in data["hits"]["hits"]:
    print row["_source"]["email"]

方法二：

直接使用 Elasticsearch 的原生响应数据，无需进行 JSON 转换，如下所示：

from elasticsearch import Elasticsearch

es = Elasticsearch()

resp = es.search(index="mynewcontacts", body={"query": {"match_all": {}}})
for row in resp["hits"]["hits"]:
    print row["_source"]["email"]
return "OK"

这两种方法都可以正确地遍历返回的 JSON 对象并打印每个电子邮件地址。