Sample script to fix ContentType of object stored in AWS S3.

fix_content_type_s3.py

import boto3

s3 = boto3.client('s3')

bucket_name = '<bucket name>'
valid_content_type = ['image/jpeg', 'image/png', 'image/gif']


def is_valid_header(bucket, key):
    try:
        response = s3.head_object(Bucket=bucket, Key=key)
        if 'ContentType' in response and response['ContentType'] in valid_content_type:
            return True
        else:
            return False
    except Exception as e:
        print(e)
        return True


if __name__ == "__main__":
    response = s3.list_objects_v2(
        Bucket=bucket_name,
        MaxKeys=800, # Scan 800 file at a time limit is 1000
        Prefix='<file name prifix>'
    )
    Y = 0

    while True:
        Y = Y + 800
        NextContinuationToken = response['NextContinuationToken']
        objects = response['Contents']
        for object in objects:
            key = object['Key']

            if not is_valid_header(bucket_name, key):
                resp = s3.copy_object(
                    ACL='public-read',
                    Bucket=bucket_name,
                    ContentType=valid_content_type[1],
                    CopySource={'Bucket': bucket_name, 'Key': key},
                    Key=key,
                    MetadataDirective = 'REPLACE'
                )
                print(resp)

        print("Processed {0} images".format(Y))

        response = s3.list_objects_v2(
            Bucket=bucket_name,
            MaxKeys=800,
            Prefix='ng/profile/profile_',
            ContinuationToken=NextContinuationToken
        )