Medium Post

step1.py

import requests

headers = {"Authorization" : "Bearer [INSERT AIRTABLE API KEY HERE]"}
params = (
    ('maxRecords', '100'),
    ('view', 'Grid view'))
URL = 'https://api.airtable.com/v0/appgaUkvDLKnYOmdA/URLS%20to%20Process'

response = requests.get(URL, params = params, headers = headers)

airtable_dictionary = response.json()
urls = []
for entry in range(0, len(airtable_dictionary['records'])):
    url = airtable_dictionary['records'][entry]['fields']['URL']
    urls.append(url)

step2.py

import newspaper

article = newspaper.Article(urls[0])
article.download()
article.parse()
article.nlp()

step3.py

def publish_date(article):
    from datetime import datetime 
    dtime = article.publish_date

    if dtime != None:
        strftime = dtime.strftime("%Y-%m-%d")
        date = strftime

    if dtime == None:
        strtime = article.meta_data['article']['published']
        date = strtime[0:10]
    
    return date
  
def determine_media(article):
    media = article.is_media_news()
    if media == False:
        media = "False"
    else:
        media = "True"
    return media
 
def determine_opinion(article):
    if 'opinion' in article.meta_data['article'].keys():
        opinion = article.meta_data['article']['opinion']
    else:
        opinion = "Uncategorized"
    return opinion

step4.py

upload_dict = {
        "fields": {
            "URL" : article.url,
            "Title" : article.title,
            "Date" : publish_date(article),
            "Author(s)" : article.authors,
            "Site Name" : article.meta_data['og']['site_name'],
            "Media" : determine_media(article),
            "Opinion" : determine_opinion(article),
            "Keywords" : article.keywords,
            "Site Twitter" : article.meta_data['twitter']['site'],
            "Text" : article.text,
            },
        "typecast" : True
        }