Is there any limit, which restricts the number of times I can fetch comments of posts from a facebook page?
I'm trying to fetch comments from a facebook page with a long term access token.
But when I'm performing the get request for the first time, I'm getting the below error:
Error in callAPI(url = url, token = token) :
(#32) Page request limit reached
Called from: callAPI(url = url, token = token)
I've been requesting data from this page with an older token, for the past one year. Now, I've generated a new token and getting the below error with my first GET request.
My maximum request per day is not more than 1000.
Language Used: R Programming
Package Used: Rfacebook
Note: Not talking of the 4800 requests for a user within 24hrs.
The issue is resolved by holding the request job for one or two days.
Once it's reset, I'm able to fetch data again.
Related
I use some self developed script to get info about my posts and followers. Recently I discovered that Instagram changed rate limits for my app. I see following info in headers:
x-ratelimit-limit: 200
I learned Instagram blog and page with changes but found no mentions about any changes since Oct 1, 2017.
Is this new limit related only to my token or to entire platform? Could you check your token?
Yes It is, Although no official communication.
I also experienced that.
I found this in their documentation.
The Instagram API uses the same rate limiting as the Graph API (200 calls per user per hour) with one exception: the /media/comments edge limits writes to 60 writes per user per hour. Please refer to the Graph API's rate limiting documentation for more information.
https://developers.facebook.com/docs/instagram-api/overview/#rate-limiting
https://techcrunch.com/2018/04/02/instagram-api-limit/
Hope this will be helpful.
Update:-
Further on 4th April few more APIs deprecated by Instagram.
https://www.instagram.com/developer/changelog/
Some perspective:
This 200 per hour call has been in the Graph API documentation for at least a month (according to wayback machine).
Facebook must have panicked over the weekend and pressed the "require Graph API permissions" button, thus our current situation. :\
Yes,
The limit has been reduced.
Here are some reference links for detailed documentation:
Instagram API limit reduced to 200 from 5000
https://techcrunch.com/2018/04/02/instagram-api-limit/
https://developers.facebook.com/docs/instagram-api/overview/#rate-limiting
Hope it helps
Thanks
Here's our answer: https://www.instagram.com/developer/changelog/
The API deprecated immediately. So really the limits probably don't matter that much anymore. I really don't see myself building another app based on their API...
Here is The New Instagram's Public Api Update , unfortunately Limit rates are worthless for some Features in Instagram .
April 4, 2018
The following endpoints are deprecated immediately:
Follows and Relationships
GET /users/self/follows
GET /users/self/followed-by
GET /users/self/requested-by
GET /users/{user-id}/relationship
POST /users/{user-id}/relationship
Commenting on Public Content
GET /media/{media-id}/comments (Deprecation not relevant when fetching comments for self media)
POST /media/{media-id}/comments
DEL /media/{media-id}/comments/comment-id
Media
GET /media/{media-id}
GET /media/shortcode/{shortcode}
Likes
GET /media/{media-id}/likes
POST /media/{media-id}/likes
DEL /media/{media-id}/likes
User Search
GET /users/{user-id}
GET /users/{user-id}/media/recent
GET /users/self/media/liked
GET /users/search
Subscriptions - to receive notifications when media is posted
Some information on Public Content returned through hashtag and location search will be removed - Name, Bio, Comments, Commenters, Follower Count, Following Count, Post Count, and Profile Picture
I have to create an webpage where i need to show all the events from the past or upcoming ones from an facebook page. I have to retrive the title, description, image, time and place of every single one and make a box with it.
After that, i have to retrive all the interested people.
I'm a beginner trying to make big things! Thank you very much.
This isnt a simple problem - I have just done this so will give you a start.
You need to use Facebooks Graph API to get the data, you will need to read up on the Graph API and how to use it. Documentation can be found here. Its worth noting due to a security breach the API isn't offering all of its services to new developers currently but you can get most - read about this here.
You can query the API using any language but I suggest do it with python and use the requests module. Alternatively you could use this SDK which makes things a lot easier, trust me! With the SDK you need to get a user access token which you can get from here.
SDK
With the SDK its possible to do this
import facebook
graph = facebook.GraphAPI(access_token="your_token", version="2.12")
event = graph.get_object(id='event_id',
fields='attending_count,declined_count')
print(event['attending_count'])
print(event['declined_count'])
You can add more fields to the fields list. However you can use any language you choose to request the data, as you can just construct your own URL's.
Graph API & Requests
This is how I did it in python using requests to query the API.
import requests, json
#Make a requests session to fetch data from URL's
session = requests.Session()
session.mount('https://', requests.adapters.HTTPAdapter(max_retries=10))
#API base URL
baseUrl = 'https://graph.facebook.com/v2.11/'
#Fields you want to get
fields = ['id', 'name', 'start_time', 'end_time', 'description', 'place']
#replace CLIENT_ID and CLIENT_SECRET with ones from a facebook app (make one)
tokenPath= 'oauth/access_token?client_id={0}&client_secret={1}&&'\
'grant_type=client_credentials'.format(CLIENT_ID, CLIENT_SECRET)
#Get the generated token
token = session.get(baseUrl + tokenPath).json()['access_token']
event = session.get(baseUrl, params={
"ids": "event_id",
"fields": ",".join(fields),
"access_token": token,
}
).json()
print(event["name"])
print(event["place"])
You can make a facebook app here which is where the CLIENT_ID & CLIENT_SECRET come from.
Why would you use URL requests and not the SDK?
Using the SDK the generated user access tokens expire every day, its possible to extend them but only for 60 days. If you want to use the SDK it has a method built in which allows you to extend the token, in order to extend the token you also need to create a facebook app. Its also possible to extend tokens using this website, though I cant vouch for its security - it could potentially access and exploit your data.
To extend the token with SDK:
import facebook
graph = facebook.GraphAPI(user_access_short_lived_token)
extended_token = graph.extend_access_token(app_id, app_secret)
print(extended_token)
Using the URL approach, you create a client access token at runtime (read about user access token vs client access token here) and the token is valid so long as your app is active. So if you are going to use this a lot or want to create it and not have to change/extend tokens then use the URL approach.
Hope this gets you started.
So recently there was a update to the Facebook Graph API and now pretty much everything requires a access_token to retrieve any type of data.
Previously you could get the number of page likes by accessing the graph like so:
http://graph.facebook.com/{page-name}
But now if you try it says you need to have a access token because of a recent update to the API. Now the issue i am having is i cant access the likes even with an access token this is the response i am getting:
Request
http://graph.facebook.com/{page-name}/?access_token={access_token}
Response
array(2) {
["name"] "Page Name"
["id"] "Page Id"
}
Now at first i thought the access token wasn't being generated with the correct scopes but i am 99% sure you only need read_stream to pull that type of data.
Second thought is that they have removed the ability to acces likes of any page and you need a page token to receive that type of data or finally i am missing something incredibly small here and its still achievable?
So my question is can you still get the number of likes of any page using the Facebook Graph API and if the answer is yes how do you do it?
any help appreciated
This can help you! Just add ?fields=likes after the page name/id. You can use any access token for fetching data!
graph.facebook.com/{page_name}?fields=likes&access_token={token}
The Graph API v2.4 reduces the number of fields in default responses.
https://developers.facebook.com/blog/post/2015/07/08/graph-api-v2.4/
Fewer default fields for faster performance: To help improve performance on mobile network connections, we've reduced the number of fields that the API returns by default. You should now use the ?fields=field1,field2 syntax to declare all the fields you want the API to return.
If you do ?fields=likes it should show up.
You can do like this:
https://graph.facebook.com/v2.4/{page_id}/fields=likes
Currently i am using following FQL query :-
https://api.facebook.com/method/fql.query?query=select post_id,likes FROM stream WHERE source_id=XXXXXXXX LIMIT 0,10000&access_token=XXXXXXXXXXXXX&format=json
I have two Facebook Accounts / pages & able to download data using above API. Able to retrieve posts & post likes count.
But, from last three days, for one Facebook account above FQL query is not working. it is returning message :- "error": "Request failed" . For other account it is working fine.
For each facebook Account / pages i have generated separate Access Tokens.
But, if i update limit from : -
Limit 0,50
Limit 50,100
then it is working & returning posts from page but not returning all post which i have previously getting from Limit 0,10000
Please help me, if any one have idea about this issue ?
Thanks
Facebook will cancel queries which take too long or too many resources to execute. In general, I would never use LIMITs which are that high. There's a high probability that they will fail, and you can't really incfuence the query execution "load" other than setting the LIMIT to a reasonable number (which means implementing a paging mechanism).
We are to create a statistics module for our news portal which involves crawling like and share count for each article (~7000 urls and continuously increasing).
The statistics is to be regenerated at least daily and without user iteraction.
We tried using fql without access_token (as it would involve user iteraction):
select ... from link_stat where url = ...
select ... from link_stat where url IN (...,...,...)
same with multiquery
But we are getting 403 Forbidden response after a few requests and the actual ip "banned" (all other requests without access_token result in 403) for some time.
Is there a proper way of collecting like and share counts for the particullar website?
A quick way to query facebook using FQL is to install the "fbcmd" app. This app lets you access the facebook and execute FQL statements from command line. It needs authorization once, during installation. After that, you can execute any number of queries.