Tweets number obtained in the Search API has been greatly reduced


In order to aggregate the URL that is posted on Twitter, I use the Search API.
From early yesterday, the number of tweets to get was greatly reduced.
My app request to the API every few seconds, but the same data (cache) is that from being returned.
I specify a max_id and since_id In order to avoid duplication, but the results will become 0.

It does not have over the limit of the API.
The status code returned is 200.

Query you are using is the following.
q = http
lang = ja
result_type = recent
count = 100
include_entities = true


I am experiencing the same problem using tested software that has run well for months.

Some searches return only 100 statuses while other searches only go back a few minutes or hours, not days.


After some experimenting, it looks like twitter is returning an EMPTY status portion of the JSON response every so often (about every15 - 20 requests).

I implemented a retry mechanism which appears to work well.

The problem is that an empty status node in the JSON results no longer means there are no more pages to get (unless of course you’ve reached your previously collected tweets’ highest ID).

Also, this problem seems to occur more often on some searches compared to others. I have no idea why.


This is a cache regression introduced like a week ago. See thread that I’ve opened but no one cared yet. Classic Twitter care.


I’ve opened a ticket internally to check into this issue.


We believe this issue should now be resolved. Thank you for your patience, and apologies for the inconvenience.


Thank you!