Amazon Kendra支持硬删除和软删除。当删除文档时，可以通过设置documentMetadataConfiguration中的softDeleteMarker和idFieldName来执行软删除。默认情况下，这些字段包括contentAccessControl等文档元数据。因此，软删除意味着将文档中的元数据值设置为NULL，而未删除文档本身。以下是一个使用Amazon Kendra SDK Python示例，展示了如何在Kendra中执行软删除：

import boto3
from botocore.config import Config
import json

# Create a client with the Kendra endpoint URL configuration
client = boto3.client('kendra', config=Config(connect_timeout=50, read_timeout=70, retries={'max_attempts': 10}, region_name=''), endpoint_url='https://')

# Define a document and its metadata
document = {
    'Id': '123456',
    'Type': 'TXT',
    'Title': 'Sample Document',
    'Content': 'This is a sample content of a text document'
}

metadata = {
    'key1': 'val1',
    'key2': 'val2'
}

response = client.batch_put_document(
    IndexId='',
    Documents=[
        {
            'Id': document['Id'],
            'Title': document['Title'],
            'Blob': json.dumps(document),
            'Metadata': json.dumps(metadata)
        }
    ]
)

# Delete the document with softDeleteMarker
response = client.batch_delete_document(
    IndexId='',
    DocumentIdList=[
        '123456'
    ],
    DataSourceSyncJobMetricTarget={
        'DataSourceId': 'datasource-id-1',
        'MetricName': 'DOCUMENTS_DELETED'
    },
    DocumentMetadataConfiguration={
        'S3Prefix': 's3:///',
        'Name': 'CustomData',
        'Type': 'STRING_VALUE',
        'Relevance': {
            'Freshness': False
        },
        'Search': {
            'Searchable': True,
            'SearcherMetadata': {
                'DocumentIdFieldName': 'Id',
                '