Get mapping from DrugBank ID to drug name

id2name.py

execfile('/home/bioage/utils/utils.py')
import xmltodict #https://github.com/martinblech/xmltodict

drugbank_xml_file = "path/to/xml/file"
drugbank_json_file = "path/to/json/file" # save json here
drugbank_csv_file = "path/to/csv/file" # save csv here

# Convert XML to JSON and save
drugbank_dict = xmltodict.parse(read_file(drugbank_xml_file)) # takes around 6 minutes
write_file(drugbank_json_file, json.dumps(drugbank_dict))

# Save drug id and name to CSV
def parse_drug(drug_dict):
    return dict(
        name = str(drug_dict['name'].encode('utf-8')), # need to clean some non-ascii characters (around 6 drugs have this problems)
        id = tryfuncs(
            drug_dict['drugbank-id'],
            [lambda x: x['#text'], # id contained in dict
             lambda x: x[0]['#text']])) # id contained in dict, itself in a list

id2name = pd.DataFrame(map(parse_drug, drugbank_dict['drugbank']['drug']))
id2name.to_csv(drugbank_csv_file, index = False)