Dumping a Facebook post's comments and likes into CSV files

comments_and_likes.py

# Loading post comments and likes by Norsez Orankijanan [email protected]
# based on example by James Thornton, http://jamesthornton.com

# Facebook API Docs
# https://developers.facebook.com/docs/graph-api/using-graph-api#reading

# Get Your Facebook Access Token Here...
# https://developers.facebook.com/tools/explorer/145634995501895/?method=GET&path=me

# Before running this script...
# Set your Facebook Access Token as an environment variable in your terminal:
#export ACCESS_TOKEN="EAACEdEose0cBAHZCLwgcdvTYfXqyahiPi7zEO3qWVBf6Yde3MIYEAa0L0IhvKUzJ2djfbf0XEkUumFdo7TMBSTwScul7RXlshMwn2H60uzjtZC6DpsJgpfOAH4YTXaaKrcoWzSs89IyeMUwNaiIOoT7WQR3tVP5ybnVRjC6H5WG9Ez1973MY8BZBAs8AooZD"


import io
import json
import urllib
import urllib.request
import pprint
import time
import csv

# get Facebook access token from environment variable
ACCESS_TOKEN = "EAACEdEose0cBAHuxmvyPjbqaZAPEmJjziRHbQhPEqnHz9Sffpddz9yw43V0qms0vS5f3WvhfXFSqxCXTvyXrY1ZAFLeNZAedmrfTDZAmiqdxm7ZBerrCVAu5QgBtIZA5NLXrZAPx8LB6y9AggxTvgF1BQSUEsOHdO03dxFGYr06Fuh30IBFVLQTbCtVULJOEVQZD"
POST_ID = "1059511694183885"
# build the URL for the API endpoint
host = "https://graph.facebook.com"
params = urllib.parse.urlencode({"access_token": ACCESS_TOKEN})



def query(url):
    # open the URL and read the response
    resp = urllib.request.urlopen(url).read()

    # convert the returned JSON string to a Python datatype 
    me = json.loads(resp)

    # display the result
    #pprint.pprint(me)
    return me

def loadComments():
    path = "/v2.10/{}/comments".format(POST_ID)
    url = "{host}{path}?{params}".format(host=host, path=path, params=params)
    nextURL = url
    fname = "output_facebook_comments_" + POST_ID + "_" + str(int(time.time())) + ".csv"
    total = 0
    with io.open(fname,'w',encoding='utf8') as f:
        rowWriter = csv.writer(f, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
        while nextURL != '':
            #print("loading page: " + nextURL)
            data = query(nextURL)
            for r in data['data']:
                from_name = r['from']['name']
                from_id = r['from']['id']
                rowWriter.writerow([r['created_time'], 
                                   from_name, 
                                   from_id, 
                                   r['message'], 
                                   r['id']])
            total = total + len(data['data'])
            if 'next' in data['paging']:
                nextURL = data['paging']['next']
            else:
                nextURL = ''
            print("number of comments loaded: {}".format(total))
    print("total comments downloaded: {}".format(total))
                
def loadLikes():
    path = "/v2.10/{}/likes".format(POST_ID)
    url = "{host}{path}?{params}".format(host=host, path=path, params=params)
    nextURL = url
    fname = "output_facebook_likes_" + POST_ID + "_" + str(int(time.time())) + ".csv"
    total = 0
    with io.open(fname,'w',encoding='utf8') as f:
        rowWriter = csv.writer(f, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
        while nextURL != '':
            #print("loading page: " + nextURL)
            data = query(nextURL)
            for r in data['data']:
                rowWriter.writerow([ 
                                   r['name'], 
                                   r['id']
                ])
            total = total + len(data['data'])
            if 'next' in data['paging']:
                nextURL = data['paging']['next']
            else:
                nextURL = ''
            print("number of likes loaded: {}".format(total))
    print("total likes downloaded: {}".format(total))
                               
#
#
loadLikes()
loadComments()