AWS Lake Formation 支持在嵌套的 JSON 中的列上添加 LF 标签。以下是一个示例代码：

import boto3

# Initialize the Amazon S3 resource
s3 = boto3.resource('s3')
 
# Define the target bucket name
bucket_name = 'my-bucket'
 
# Define the object key and metadata key
object_key = 'data.json'
metadata_key = 'lf-tags'
 
# Define the LF tag for a column in JSON
lf_tag = {"tag_name": "CONFIDENTIAL", "tag_type": "SECURITY_LABEL", "keys": ["nested.column"]}
 
# Load the metadata file
metadata = s3.Object(bucket_name, metadata_key).get()['Body'].read().decode('utf-8')
metadata_json = json.loads(metadata)
 
# Create a new LF tag
metadata_json['tags'].append(lf_tag)
 
# Save the metadata file
s3.Object(bucket_name, metadata_key).put(Body=json.dumps(metadata_json))
 
# Tag the column in the JSON with the LF tag
lf_client = boto3.client('lakeformation')
response = lf_client.add_lf_tags(Resource={"DatabaseName": "my-database", "TableName": "my-table"}, LFTags=[lf_tag])

注意，在以上示例代码中，“nested.column”代表要在JSON中标记的具体列。将其替换为实际的列名即可。