如何在 Lambda 函数中读取 S3 文件（在 python 中）

Question

I am trying to read a file from S3, which has the following content stored in it:

   {"empID":{"n":"7"},"name":{"s":"NewEntry"}}
   {"empID":{"n":"3"},"name":{"s":"manish"}}
   {"empID":{"n":"2"},"name":{"s":"mandeep"}}
   {"empID":{"n":"4"},"name":{"s":"Vikas"}}
   {"empID":{"n":"1"},"name":{"s":"babbar"}}

I want to iterate over each and every object and do some some processing on them.

I am taking reference from this code:

import json
import boto3
s3_obj =boto3.client('s3')

s3_clientobj = s3_obj.get_object(Bucket='dane-fetterman-bucket', Key='mydata.json')
s3_clientdata = s3_clientobj['Body'].read().decode('utf-8')

print("printing s3_clientdata")
print(s3_clientdata)
print(type(s3_clientdata))


s3clientlist=json.loads(s3_clientdata)
print("json loaded data")
print(s3clientlist)
print(type(s3clientlist))

but there is not any "Body" attribute in the file. Can i get some points to do the desired stuff.

Answer 1

The issue is that the file actually contains individual JSON on each line, rather than being a complete JSON object itself.

Therefore, the program needs to process each line independently:

import json
import boto3

s3_client = boto3.client('s3')

s3_clientobj = s3_client.get_object(Bucket='my-bucket', Key='mydata.json')

for line in s3_clientobj['Body'].iter_lines():
    object = json.loads(line)
    print(f"ID: {object['empID']['n']} Name: {object['name']['s']}")

Alternatively, you could download the whole object to disk, then just use normal for line in open('file'): syntax.

See also: Read a file line by line from S3 using boto?