我需要知道我在做什么错....因为它要求我输入页面名称，然后输入我的Facebook访问令牌，但这会给我这个错误
**
URL https://graph.facebook.com/v2.6/marvelstudios/posts/?fields=message,name错误 ，id，comments.limit（0）.summary（true），shares，reactions.limit（0）.summary（true）＆limit = 100＆access_token = ****************** **********： 2019-11-05 18：30：59.650767
**
import urllib.request as urllib
import json
import datetime
import csv
import time
# """
# INPUTS:
# text: a string
# OUTPUTS:
# a better formatted string
# """
def unicode_normalize(text):
return text.translate({ 0x2018:0x27,0x2019:0x27,0x201C:0x22,0x201D:0x22,0xa0:0x20 }).encode('utf-8')
# """
# INPUTS:
# url: a request url
# OUTPUTS:
# the data returned by calling that url
# """
def request_data_from_url(url):
req = urllib.Request(url)
success = False
while success is False:
try:
#open the url
response = urllib.urlopen(req)
#200 is the success code for http
if response.getcode() == 200:
success = True
except Exception :
#if we didn't get a success,then print the error and wait 5 seconds before trying again
print ("OOPS")
time.sleep(5)
print ("Error for URL %s: %s" % (url,datetime.datetime.now()))
print ("retrying...")
#return the contents of the response
return response.read()
# """
# INPUTS:
# page_id: the unique id for the facebook page you are trying to scrape
# access_token: authentication proving that you have a valid facebook account
# OUTPUTS:
# a python dictionary of the data on your requested page
# """
def get_facebook_page_data(page_id,access_token):
website = "https://graph.facebook.com/v2.6/"
location = "%s/posts/" % page_id
#the .limit(0).summary(true) is used to get a summarized count of all the ...
#...comments and reactions instead of getting each individual one
fields = "?fields=message,name,id," + \
"comments.limit(0).summary(true),shares," + \
"reactions.limit(0).summary(true)"
authentication = "&limit=100&access_token=%s" % (access_token)
request_url = website + location + fields + authentication
#converts facebook's response to a python dictionary to easier manipulate later
data = json.loads(request_data_from_url(request_url))
return data
# """
# INPUTS:
# post: information about a single post on the facebook page
# access_token: authentication proving that you have a valid facebook account
# OUTPUTS:
# a list with the requested fields for this post
# """
def process_post(post,access_token):
post_id = post['id']
post_message = '' if 'message' not in post.keys() else \
unicode_normalize(post['message'])
post_type = post['type']
#for datetime info,we need a few extra steps
#first convert the given datetime into the format we want
post_published = datetime.datetime.strptime(
post['created_time'],'%Y-%m-%dT%H:%M:%S+0000')
#then account for the time difference between the returned time and my time zone
post_published = post_published + \
datetime.timedelta(hours=-2)
#last,convert the datetime into a string in a format convenient for spreadsheets
post_published = post_published.strftime(
'%Y-%m-%d %H:%M:%S')
num_reactions = 0 if 'reactions' not in post else \
post['reactions']['summary']['total_count']
num_comments = 0 if 'comments' not in post else \
post['comments']['summary']['total_count']
num_shares = 0 if 'shares' not in post else post['shares']['count']
#here we call a separate API for information about reactions based on the post's post_id
#but only if this post is afer the day when reactions first appeared on facebook
reactions = get_reactions_for_post(post_id,access_token) if \
post_published > '2016-02-24 00:00:00' else {}
num_likes = 0 if 'like' not in reactions else \
reactions['like']['summary']['total_count']
#if this post is from before reactions existed,then simply set the number of likes ...
#...equal to the total number of reactions
num_likes = num_reactions if post_published < '2016-02-24 00:00:00' \
else num_likes
#function to get total number of reactions from the reactions dictionary above
def get_num_total_reactions(reaction_type,reactions):
if reaction_type not in reactions:
return 0
else:
return reactions[reaction_type]['summary']['total_count']
#get counts of all reactions
num_loves = get_num_total_reactions('love',reactions)
num_wows = get_num_total_reactions('wow',reactions)
num_hahas = get_num_total_reactions('haha',reactions)
num_sads = get_num_total_reactions('sad',reactions)
num_angrys = get_num_total_reactions('angry',reactions)
#return a list of all the fields we asked for
return (post_id,post_message,post_type,post_published,num_reactions,num_comments,num_shares,num_likes,num_loves,num_wows,num_hahas,num_sads,num_angrys)
# """
# INPUTS:
# post_id: a post id corresponding to a particular post
# access_token: authentication proving that you have a valid facebook account
# OUTPUTS:
# a python dictionary of information about the reactions associated to this post
# """
def get_reactions_for_post(post_id,access_token):
website = "https://graph.facebook.com/v2.6"
location = "/%s" % post_id
#here we ask for the number of reactions of each time associated with this post
reactions = "/?fields=" \
"reactions.type(LIKE).limit(0).summary(total_count).as(like)" \
",reactions.type(LOVE).limit(0).summary(total_count).as(love)" \
",reactions.type(WOW).limit(0).summary(total_count).as(wow)" \
",reactions.type(HAHA).limit(0).summary(total_count).as(haha)" \
",reactions.type(SAD).limit(0).summary(total_count).as(sad)" \
",reactions.type(ANGRY).limit(0).summary(total_count).as(angry)"
authentication = "&access_token=%s" % access_token
request_url = website + location + reactions + authentication
# retrieve data and store in python dictionary
data = json.loads(request_data_from_url(request_url))
return data
# """
# INPUTS:
# page_id: the unique id for the facebook page you are trying to scrape
# access_token: authentication proving that you have a valid facebook account
# OUTPUTS:
# nothing,simply prints how many posts were processed and how long it took
# """
def scrape_facebook_page(page_id,access_token):
#open up a csv (comma separated values) file to write data to
with open('%s_facebook_posts.csv' % page_id,'w') as file:
#let w represent our file
w = csv.writer(file)
#write the header row
w.writerow(["post_id","post_message","post_type","post_published","num_reactions","num_comments","num_shares","num_likes","num_loves","num_wows","num_hahas","num_sads","num_angrys"])
has_next_page = True
num_processed = 0
scrape_starttime = datetime.datetime.now()
print ( "Scraping %s Facebook Page: %s\n" % (page_id,scrape_starttime))
#get first batch of posts
posts = get_facebook_page_data(page_id,access_token)
#while there is another page of posts to process
while has_next_page:
#we just limit to 200 posts for simplicity,if you want all the posts,just remove this
if num_processed == 200:
break
#for each individual post in our retrieved posts ...
for post in posts['data']:
#...get post info and write to our spreadsheet
w.writerow(process_post(post,access_token))
num_processed += 1
#if there is a next page of posts to get,then get next page to process
if 'paging' in posts.keys():
posts = json.loads(request_data_from_url(
posts['paging']['next']))
#otherwise,we are done!
else:
has_next_page = False
print ("Completed!\n%s posts Processed in %s" % \
(num_processed,datetime.datetime.now() - scrape_starttime))
page_id = input("Please Paste Public Page Name:")
access_token = input("Please Paste Your access Token:")
if __name__ == '__main__':
scrape_facebook_page(page_id,access_token)