get_content_formats.py

import json
from jsonpath_ng import jsonpath, parse

def get_content_formats(filename):

  content_formats = dict()

  image = parse("$..image")
  video = parse("$..embedUrl")
  local_business = parse("$..address")
  review = parse("$..review")
  top_story = parse("$..publisher")
  faq = parse("$..acceptedAnswer")
  job = parse("$..employmentType")

  with open(filename,"r") as f:
    data = json.load(f)

    content_formats["image"] = [match.value for match in image.find(data)]
    content_formats["video"] = [match.value for match in video.find(data)]
    content_formats["local_business"] = [match.value for match in local_business.find(data)]
    content_formats["review"] = [match.value for match in review.find(data)]
    content_formats["top_story"] = [match.value for match in top_story.find(data)]
    content_formats["faq"] = [match.value for match in faq.find(data)]
    content_formats["job"] = [match.value for match in job.find(data)]

    return content_formats