Scrape all photos from a public Facebook page.

fb-dl.py

#!/usr/bin/env python

############################### README ###############################
# External dependencies:
# * libmagic
# * python-dateutil
# * python-magic
# * requests
#
# The shell command call at the end was tested with GNU date from GNU
# coreutils. Other implementations might not work (for instance, BSD
# date is not compatible). However, you may comment out the last
# section without affecting the download.
#
# The "last_update_timestamp" is reserved for the update feature,
# which is not implemented yet.
######################################################################

import commands
import json
import os
import sys
import urllib

import dateutil.parser
import magic
import requests

########################### CUSTOMIZE THIS ###########################
page_id = "567045406742962"
access_token = "app_id|app_secret" # see https://developers.facebook.com/docs/facebook-login/access-tokens#apptokens
dest = os.path.expanduser("~/img/sns/apink-official-facebook")
website_title = "apink-official-facebook"
######################################################################

if not os.path.exists(dest):
    os.makedirs(dest)

# read last update time, if it is available
last_update_record = dest + "/last_update_timestamp"
if os.path.exists(last_update_record):
    f = open(last_update_record, "r")
    last_update_timestamp = f.readline()
    f.close()
    last_update_time = dateutil.parser.parse(last_update_timestamp)
else:
    last_update_time = dateutil.parser.parse("1970-01-01T00:00+00:00")

# this function makes an API call with only an access_token (which
# could be just app-id|app-secret)
def fb_public_call(endpoint, params, access_token):
    params["access_token"] = access_token
    response = requests.get("https://graph.facebook.com/" + endpoint,
                            params=params)
    return response.json()

# this function downloads a photo
# return codes are defined below
SUCCESS = 0
FAILED_DOWNLOAD = 1
UNRECOGNIZED_MIME = 2
OLD_PHOTO = 255 # photo older than last update time
def handle_photo(photo, album_id):
    # print information
    photo_id = photo["id"]
    time = dateutil.parser.parse(photo["created_time"])
    if time < last_update_time:
        return OLD_PHOTO
    time_print = time.strftime("%b %d, %Y")
    time_full = time.strftime("%Y%m%d%H%M%S")
    original_image = photo["images"][0]
    height = original_image["height"]
    width = original_image["width"]
    format_string = "date: %s   id: %s   size: %sx%s"
    print format_string % (time_print, photo_id, width,
                           height)
    # download file
    source_uri = original_image["source"]
    filename = time_full + "-" + website_title + "-" + \
               album_id + "-" + photo_id
    filepath = dest + "/" + filename
    urllib.urlretrieve(source_uri, filepath)
    # identify mime type and attach extension
    if os.path.exists(filepath):
        mime = magic.from_file(filepath, mime=True)
        if mime == "image/gif":
            newfilepath = filepath + ".gif"
        elif mime == "image/jpeg":
            newfilepath = filepath + ".jpg"
        elif mime == "image/png":
            newfilepath = filepath + ".png"
        else:
            err = filepath + ": error: " + \
                  "unrecgonized image type\n"
            sys.stderr.write(err)
            return UNRECOGNIZED_MIME
        os.rename(filepath, newfilepath)
        return SUCCESS
    else:
        # donwload failed for whatever reason
        err = "error: " + filename + " failed to " + \
              "downloaded from " + source_uri + "\n"
        sys.stderr.write(err)
        return FAILED_DOWNLOAD

# this function handles an album, i.e., download newly added photos
# since the last update
def handle_album(album):
    # print album info
    album_id = album["id"]
    format_string = "downloading album \"%s\" " + \
                    "(album id: %s; photo count: %s)"
    print format_string % (album["name"], album_id,
                           album["count"])
    print "-" * 80
    # retrieve photos in the album
    photos_response = fb_public_call(album["id"] + "/photos",
                                     params, access_token)
    while True:
        for photo in photos_response["data"]:
            if handle_photo(photo, album_id) == OLD_PHOTO:
                # already encountered old photo in this album
                # no need to look further into the past
                print
                return

        if "next" in photos_response["paging"]:
            next_uri = photos_response["paging"]["next"]
            photos_response = requests.get(next_uri).json()
        else:
            break
    print
    
params = {}
# retrieve albums
albums_response = fb_public_call(page_id + "/albums", params,
                                 access_token);
while True:
    for album in albums_response["data"]:
        handle_album(album)

    if "next" in albums_response["paging"]:
        next_uri = albums_response["paging"]["next"]
        albums_response = requests.get(next_uri).json()
    else:
        break

# update feature yet to be implemented
# create a file "last_update_timestamp" for future use
f = open(last_update_record, "w")
f.write(commands.getoutput("date -u --iso-8601=seconds"))
f.close()

looks like it maybe an issue on a facebook "Company page" here's an example: http://findmyfbid.com/success/222407964522124

Yup, same community pages as well

Does this work on any pages anymore? Didn't seem to work on page ID 394278667320706 or other relatively public personal pages. Just trying to debug for Shane's use case.

@shaneholloman @leotam Hey guys. First, commenting on a gist is not the best way to get help. In case you don't know, there are no notifications for comments on gists, so if I haven't been looking through my older gists by chance and saw a greater-than-zero number of comments here, I would have never known.

Secondly, regarding the issues you're having: this script is more than a year old, and I quitted scraping Facebook stuff at least nine months ago. Things have changed. Just try the REST requests with Facebook's API Explorer, or a REST client like Paw, or just good ol' curl, and you'll see that /222407964522124/albums returns 400, and although /394278667320706/albums works, using the album endpoint on the albums returned just give me a truncated response with only created_time, name and id, without actual info of photos. Same goes for a few other public pages I tried. So there's nothing I can do; the simplest access token isn't good enough anymore.

What I would suggest is to do a full blown OAuth. I'm not going to try, so I won't guarantee success.

Since this thing was broken, I built my own. It's still a work in progress but it currently downloads photos according to albums.

How come I need an access token for that ? It's a public page

not running

Traceback (most recent call last):
  File "fb-dl.py", line 140, in <module>
    handle_album(album)
  File "fb-dl.py", line 114, in handle_album
    album["count"])
KeyError: 'count'

@Mouradif: Since it's the only way you're getting access to Facebook's API.

I have a problem:
Traceback (most recent call last):
File "fb-dl.py", line 138, in
for album in albums_response["data"]:
KeyError: 'data'

What can I do?