repo_util.py

import json

import requests

# only care about manifest list or v2 manifests
manifest_header = ' '.join([
    'application/vnd.docker.distribution.manifest.list.v2+json',
    'application/vnd.docker.distribution.manifest.v2+json'])


def scan_repo(repo, tags=None):

    # get auth token
    s = requests.Session()
    token = s.get('https://auth.docker.io/token?service=registry.docker.io' +
                  '&scope=repository:%s:pull' % repo).json()['token']
    s.headers.update({'Authorization': 'Bearer %s' % token})

    # get all tag names if not specified
    if tags is None:
        tags = s.get('https://registry-1.docker.io/v2/%s/tags/list' % repo) \
                .json()
        tags = tags['tags']
    if len(tags) == 0:
        raise 'no tags? giving up'

    # get manifest list (TODO) or manifest
    data = {}
    s.headers.update({'Accept': manifest_header})
    for tag in tags:
        print 'scanning: %s:%s' % (repo, tag)
        manifest = s.get('https://registry-1.docker.io/v2/%s/manifests/%s' %
                         (repo, tag)).json()
        for blobsum in manifest['fsLayers']:
            print blobsum['blobSum']
        with open('manifest.json', 'w') as f:
            f.write(json.dumps(manifest, indent=4))

if __name__ == '__main__':
    scan_repo('library/ruby', ['latest'])