S3 Multipart Upload using boto3 :

Hi, I am "using boto3 copying an S3 object to itself" in order to change the last modified date. While
response = input_s3_client.copy_object( Bucket=bucket, CopySource={'Bucket': bucket, 'Key': key}, Key=key ) has been very successful, for the file larger than 5GB, I have to use multipart upload to "upload a object to itself" in order to achieve the same goal. Apparently, AWS only gave Java and .net example for multipart upload : https://docs.aws.amazon.com/AmazonS3/latest/userguide/CopyingObjectsMPUapi.html

and I have wrote a little program based on python boto3.

import pytz
import boto3
import botocore
from boto3.s3.transfer import TransferConfig
part_size = 5 * 1024 * 1024
GB = 1024 ** 3
config = TransferConfig(multipart_threshold=5 * GB)

class S3Util:
    def update_last_modified_date(self, input_s3_client, bucket: str, key: str) -> bool:
        try:
            start_time = datetime.now(tz=pytz.UTC)
            # input_s3_client.upload_file(bucket, key, bucket, key, ExtraArgs={'ACL': 'bucket-owner-full-control'})
            # Get object metadata (used for progress tracking)
            head_object = input_s3_client.head_object(Bucket=bucket, Key=key)
            object_size = head_object['ContentLength']

            # Initiate multipart upload
            upload_id = input_s3_client.create_multipart_upload(Bucket=bucket, Key=key)[
                'UploadId']
            uploaded_parts = []

            # Upload in parts
            part_number = 1
            offset = 0
            while offset < object_size:
                # Determine part size for this iteration
                remaining_size = object_size - offset
                current_part_size = min(part_size, remaining_size)

                # Get object data for this part
                get_object_response = input_s3_client.get_object(Bucket=bucket, Key=key,
                                                                 Range=f"bytes={offset}-{offset + current_part_size - 1}")
                self.logger.info(f'body stream type: {type(get_object_response)}')
                # Upload the part
                upload_part_response = input_s3_client.upload_part(Bucket='podaac-dev-tst', Key='temp.nc',
                                                                   UploadId=upload_id, PartNumber=part_number,
                                                                   Body=get_object_response['Body'])
                uploaded_parts.append({'PartNumber': part_number, 'ETag': upload_part_response['ETag']})

                # Print progress (optional)
                self.logger.info(
                    f"Uploaded part {part_number} of {int(object_size / part_size) + 1} ({offset + current_part_size}/{object_size} bytes)")

                offset += current_part_size
                part_number += 1

            # Complete the multipart upload
            input_s3_client.complete_multipart_upload(Bucket=bucket, Key=key,
                                                      UploadId=upload_id, Parts=uploaded_parts)

            self.logger.info(f"Large object '{bucket}{key}' copied successfully")

while running the program against a 44GB file, I got UnsupportOperation(seek) error at the line while calling the input_s3_client.uploadpart and I have been struggling with it. Can somebody please help? Any suggestion would be very much appreciated.