load_index.py 文件源码

def retrieveForYear(year):
    r = boto.connect_s3(host="s3.amazonaws.com") \
            .get_bucket("irs-form-990") \
            .get_key("index_%i.json" % year) \
            .get_contents_as_string() \
            .replace("\r", "")
    j = json.loads(r)

    # The index comes back as a single JSON key-value pair whose value is
    # a JSON array of length one. Inside _that_ is an array of filings.

    filings = j.values()[0]

    if cred.prod:
        return filings
    else:
        sample = filings[0:1000]
        return sample