Check arXiv metadata

10.08.2018.2.py

import boto3, configparser, os, botocore

def download_file(key):
    """
    Downloads given filename from source bucket to destination directory.

    Parameters
    ----------
    key : str
        Name of file to download
    """

    # Ensure src directory exists 
    if not os.path.isdir('src'):
        os.makedirs('src')

    # Download file
    print('\nDownloading s3://arxiv/{} to {}...'.format(key, key))

    try:
        s3resource.meta.client.download_file(
            Bucket='arxiv', 
            Key=key,  # name of file to download from
            Filename=key,  # path to file to download to
            ExtraArgs={'RequestPayer':'requester'})
    except botocore.exceptions.ClientError as e:
        if e.response['Error']['Code'] == "404":
            print('ERROR: ' + key + " does not exist in arxiv bucket")

if __name__ == '__main__':
    """Runs if script is called on command line"""

    # Download manifest file to current directory
    download_file('src/arXiv_src_manifest.xml')