ttd_output.py

#
# Usage: ttd.py <jobid>
# 
# This program send ttd_id list with segment_name to The Trade Desk
# by thier APIs.
# Query result of <jobid> requires ttd_id and segment_name.
# 
#
import os
import sys
import tdclient
import glob
import time
import csv

### TD Setting ###
apikey = 'TD_MASTER_APIKEY'

### TTD Setting ###
ad_id = 'xxxx'
secret = 'xxxxx'
endpoint = 'http://tok-data.adsrvr.org/data/advertiser'

### Config Parameter ###
lines_per_file = 1000  # Adjust this parameter to make body size of API within 4MB.
sleep_per_api = 5 # API call interval adjsutment between performance and stability.


# Functions
def get_ttdid_list():
    jobid = sys.argv[1];
    print('jobid')
    print(jobid)
    with tdclient.Client(apikey) as client:
        # get job result, expecting ttd_id and segment_name columns
        ids = client.job_result(jobid)
        # write ids to local file
        f = open('ids.csv', 'w')
        for x in ids:
            f.write(str(x[0]) + "," + str(x[1]) + "\n")
        f.close()

def split_file():
    in_file_name = 'ids.csv'
    out_file_name_template = 'splitted_%d'
    
    max_lines = lines_per_file
    
    split_index = 1
    line_index = 1
    out_file = open(out_file_name_template % (split_index,), "w")
    in_file = open(in_file_name)
    line = in_file.readline()
    while line:
        if line_index > max_lines:
            print("Starting file: %d" % split_index)
            out_file.close()
            split_index = split_index + 1
            line_index = 1
            out_file = open(out_file_name_template % (split_index,), "w")
        out_file.write(line)
        line_index = line_index + 1
        line = in_file.readline()
    
    out_file.close()
    in_file.close()

def make_body(ids):
    body = '{"AdvertiserId":"' + ad_id + '","Items":['
    body_id = ''
    for x in ids:
        body_id += '{"TDID":"' + x[0] + '","Data":[{"Name":"' + x[1] + '"}]},'
    body_id = body_id.rstrip(",")
    body += body_id
    body += ']}'
    return body

def make_digest(message, key):
    import hashlib
    import hmac
    import base64
    key = bytes(key, 'UTF-8')
    message = bytes(message, 'UTF-8')
    digester = hmac.new(key, message, hashlib.sha1)
    signature1 = digester.digest()
    signature2 = base64.standard_b64encode(signature1)    
    return str(signature2, 'UTF-8')

def send_body(signature, body):
    import requests
    headers = {
        'TtdSignature': signature,
        'Content-Length': str(len(body))
    }
    response = requests.post(endpoint, headers=headers, data=body)
    print(response)
    print(response.text)

def remove_glob(pathname, recursive=True):
    for p in glob.glob(pathname, recursive=recursive):
        if os.path.isfile(p):
            os.remove(p)
  
# Main
args = sys.argv
if(len(args) != 2):
    print("Usage: python {} <jobid>".format(args[0]))
    quit()

get_ttdid_list()
split_file()

files = glob.glob('splitted_*')
for file in files:
    csv_file = open(file, "r", encoding="utf_8", errors="", newline="\n" )
    f = csv.reader(csv_file, delimiter=",", doublequote=True, lineterminator="\n", quotechar='"', skipinitialspace=True)
    ids = []
    for x in f:
        ids.append(x)
    csv_file.close
    #print(ids)
    
    body = make_body(ids)
    #print('body')
    #print(body)
    print('length')
    print(str(len(body)))
    
    signature = make_digest(body, secret)
    
    print('signature')
    print(signature)
    
    send_body(signature, body)
    time.sleep(sleep_per_api)

remove_glob('splitted_*')