gistfile1.py

import time
from datetime import datetime

from lxml import etree

def time_handler(s):
    t = time.strptime(s, "%m/%d/%Y %I:%M:%S %p")
    return time.strftime("%Y-%m-%dT%H:%M:%S%Z", t)

def feed_parser(tree):
    nodes = []
    
    for job in tree:

        if job[0] == 'start':
            node = {"Jobid": job[1].attrib['Jobid']}
            for child in job[1].getchildren():
            
                if child.tag == "JobCategories":
                    val = child.getchildren()[0].getchildren()[0].text.strip()
                elif child.tag in ["JobPostDate", "JobExpireDate"]:
                    val = time_handler(child.text.strip())
                else:
                    val = child.text.strip()
                    
                node[unicode(child.tag.strip())] = unicode(val)
            
            nodes.append(node)
        else:
            continue

    return nodes
    
ktype = {"kelly_job": {"properties":
                       {"DivisionDesc": {"type": "string", "store": "yes"},
                        "Jobid": {"type": "string", "store": "yes"},
                        "JobTitle": {"type": "string", "store": "yes"},
                        "Salary": {"type": "float", "store": "yes"},
                        "JobBody": {"type": "string", "store": "yes"},
                        "PhyCity": {"type": "string", "store": "yes"},
                        "PhyState": {"type": "string", "store": "yes"},
                        "PhyCountryCode": {"type": "string", "store": "yes"},
                        "PhyPostalCode": {"type": "string", "store": "yes"},
                        "JobPostDate": {"type": "date", "store": "yes"},
                        "JobExpireDate": {"type": "date", "store": "yes"},
                        "ApplyOnlineURL": {"type": "string", "store": "yes"},
                        "JobCategories": {"type": "string", "store": "yes"}}}}

if __name__ == "__main__":
    from pyelasticsearch import ElasticSearch

    kn = "kelly_job"
    idx = "kelly"
    conn = ElasticSearch("http://localhost:9200/")
    conn.create_index(idx)
    conn.put_mapping(kn, ktype)
    XML = etree.parse("xml2.xml")
    walk = etree.iterwalk(XML, events=("start", "end"), tag="Job")

    for node in feed_parser(walk):
        conn.index(node, idx, kn)