Download S3 Files with Boto

The problem is that you are downloading to a local directory that doesn't exist (media/user1). You need to either:

  • Create the directory on the local machine first
  • Just use the filename rather than a full path
  • Use the full path, but replace slashes (/) with another character -- this will ensure uniqueness of filename without having to create directories

The last option could be achieved via:

k.get_contents_to_filename(str(k.key).replace('/', '_'))

See also: Boto3 to download all files from a S3 Bucket


Downloading files using boto3 is very simple, configure your AWS credentials at system level before using this code.

client = boto3.client('s3')

// if your bucket name is mybucket and the file path is test/abc.txt
// then the Bucket='mybucket' Prefix='test'

resp = client.list_objects_v2(Bucket="<your bucket name>", Prefix="<prefix of the s3 folder>") 

for obj in resp['Contents']:
    key = obj['Key']
    //to read s3 file contents as String
    response = client.get_object(Bucket="<your bucket name>",
                         Key=key)
    print(response['Body'].read().decode('utf-8'))

    //to download the file to local
    client.download_file('<your bucket name>', key, key.replace('test',''))

replace is to locate the file in your local with s3 file name, if you don't replace it will try to save as 'test/abc.txt'.