foo.py

import requests
import json
from datetime import datetime

NYT_API = "http://nytimes.com/svc/video/api/v2/video/"
MEDIA_FACTORY_API = "http://media-factory.stg.nyt.net"
PRESETS = ["1080p_hls", "720p_hls", "480p_hls", "360p_hls", "240p_hls"]

def get_creation_date(hls_rendition):
    parts = hls_rendition.split("/")
    date = datetime.strptime(parts[4]+"-"+parts[5]+"-"+parts[6], "%Y-%m-%d")
    return datetime.strftime(date, "%Y-%m-%d %H:%M:%S.0")

def get_external_id(hls_rendition):
    return hls_rendition.split("/")[7].split("_")[0]

def get_slug(hls_rendition):
    try:
        return hls_rendition.split("_1_")[1].split("_wg")[0]
    except:
        return ""

def build_payload(creation_date, external_id, slug):
    return '{"metadata":{"presets":'+json.dumps(PRESETS)+',"slug":"'+slug+'","createDate":"'+creation_date+'"},"externalId":"'+external_id+'"}'

def get_source(video_id):
    response = requests.get(NYT_API + video_id)
    renditions = json.loads(response.text)['renditions']
    for rend in renditions:
        if 'synd' in rend['type']:
            return rend['url']

    for rend in renditions:
        if '1080p_mp4' in rend['type']:
            return rend['url']

def trigger_job(source, payload):
    response = requests.post(MEDIA_FACTORY_API+"/jobs", payload)
    job_id = json.loads(response.text)['jobId']
    print "Starting", job_id, "for", source
    response = requests.post(MEDIA_FACTORY_API+"/jobs/"+job_id+"/start", json.dumps({"source": source}))
    print response.text

for line in open("hls_bad_renditions_test.txt"):
    video_id, _, hls_rendition = line.split(" ")
    creation_date = get_creation_date(hls_rendition)
    external_id = get_external_id(hls_rendition)
    slug = get_slug(hls_rendition)
    if slug:
        source = get_source(video_id)
        payload = build_payload(creation_date, external_id, slug)
        trigger_job(source, payload)