Understanding requests



Im’ building a tiwtter based website, wich “only” shows results of certain querys and store some data to play with, also renders some twits to original twitter format,

I’m seeing that my api limits are reached almost in a second, and I dont know if this is 'cos what I want to do is impossible or is cos I’m doing something wrong.

We dont post or delete nothing to/from twitter, we only retrieve information and “play” with it to show like we want to.

We have a couple “streaming” searchers, one importer and one renderer, with all this, we only search and retrieve information, any of them is oauth, is there any way, idea or suggestion to improve the api limits?

And. exactly what is a api request done and counted?



How many requests are you issuing to the API in an hour?


this is my main question, a search is a request?

to copy a twit is a request?

streaming twits searching a couple of hashtags is a request?


The REST API has defined rate limits around GET-based requests. 150 GET-based unauthenticated requests per hour per IP address. 350 GET-based authenticated requests per hour per user per IP address. Each GET-based call to a REST API endpoint counts as one request.

For the Search API, each query that you send is a request. The rate limit for the Search API is unpublished and we recommend keeping well bellow 1 query per second.

The Streaming API has two different kinds of rate limits – one is the rate limit of how many tweets you can be served in a “moment” in respect to the overall fidelity of the Firehose (1%). The other is a rate limit on how often you can reconnect to the streaming API – these are more practical rate limits to protect against connection churn mostly.

Finally, there are “allowances” for accounts on Twitter. These aren’t really API limits but limits associated with accounts and the activities they can take on Twitter – how many tweets a user can issue, how many DMs, how many favs they can perform, and so on.



Then there is no way to improve this limits? I’ve seen that whitelisting is not an option. Hard…


I am sorry for being dense but I am no computer programmer and am still struggling to understand exactly what constitutes a request.

My plan is to have 5 or so lists which each might generate 10 tweets an hour being checked every hour by a Wordpress plugin.

How many requests is this likely to use up?

Of course I did not write the plugin so I have no idea how many GET-based requests are made in such a situation.


Assuming the plugin only makes one API request per list per check, that would be 5 requests / hour. If the request is unauthenticated, this will count against the IP address the wordpress server is running on.


Aduh sayateh teu ngarti bahasana maklum nuju di ajar biar tidak ketinggalan zaman kata anak sekarangmeh betul apa benar… hapunteun ah ka sadayana nya… wkwkwk

closed #9

closed #10