client.py

"""
Basic Python client for AMR.
http://ipscience-help.thomsonreuters.com/LAMRService/WebServicesOverviewGroup/overview.html
"""

from itertools import zip_longest
import os
import xml.etree.ElementTree as ET
import requests

try:
    USER = os.environ['WOS_USER']
    PASSWORD = os.environ['WOS_PASSWORD']
except KeyError:
    raise Exception("Unable to read WOS_USER and WOS_PASSWORD environment variables.")

AMR_URL = "https://ws.isiknowledge.com/cps/xrpc"
BATCH_SIZE = 50
THROTTLE_CAP = 2500  # Max number of records you may request per minute


def grouper(iterable, n, fillvalue=None):
    """
    Group iterable into n sized chunks.
    See: http://stackoverflow.com/a/312644/758157
    """
    args = [iter(iterable)] * n
    return zip_longest(*args, fillvalue=fillvalue)


def read(raw_in):
    ns = {'isi': 'http://www.isinet.com/xrpc41'}
    raw = ET.fromstring(raw_in)
    out = {}
    recs = raw.findall('isi:fn/isi:map/isi:map', ns)
    if len(recs) < 1:
        print('\n\nERROR: The AMR API did not return any records. Response text:')
        print(raw_in)
    else:
        for cite in recs:
            cite_key = cite.attrib['name']
            meta = {}
            for val in cite.findall('isi:map/isi:val', ns):
                meta[val.attrib['name']] = val.text
                if val.attrib['name'] == 'message':
                    print('\n\nMessage from AMR: {}'.format(val.text))
            out[cite_key] = meta
    return out


def get(request_xml):
    headers = {'Content-Type': 'application/xml'}
    response = requests.post(AMR_URL, data=request_xml, headers=headers)
    # parse into xml
    xml = read(response.text)
    return xml