Geek asks and answers

The problem with retrieving tweets from the timeline, as Twitter does not return "next_results"

I am trying to get tweets from the Twitter timeline using the box max_idpresented in next_results, as shown here. ( Twython Search API with the following_processes )

For most requests, I get the desired number (500+) of tweets, but for some I get less than 200 tweets.

When you look at what is returned in ' next_results, I find that after returning only 100 tweets, there is no field in the response next_results. There cannot be only 200 tweets for a given topic, because I am looking for this topic when it is trending, and on Twitter I see more than 200 tweets.

Does anyone else run into this problem. Is there a workaround?

Below is the conclusion where the second answer is missing a field 'next_results'.

TEST_PROGRAM >> Calling twitter to get tweets for  #TheWorstLieEver

{u'count': 100, u'completed_in': 0.066, u'max_id_str': u'432942833725497345', u'since_id_str': u'0', **u'next_results': u'?max_id=432540545630494719&q=%23TheWorstLieEver&count=100&include
_entities=1&result_type=mixed'**, u'refresh_url': u'?since_id=432942833725497345&q=%23TheWorstLieEver&result_type=mixed&include_entities=1', u'since_id': 0, u'query': u'%23TheWorstLieEver
', u'max_id': 432942833725497345L}

{u'count': 100, u'completed_in': 0.111, u'max_id_str': u'432540545630494719', u'since_id_str': u'0', u'refresh_url': u'?since_id=432540545630494719&q=%23TheWorstLieEver&result_type=mixe
d&include_entities=1', u'since_id': 0, u'query': u'%23TheWorstLieEver', u'max_id': 432540545630494719L}
+3
source share
1 answer

I got an answer. If you are looking for a volume, you need to use the Streaming API. The Twitter search API is focused on the relevance of the request and therefore may not return all tweets. The Twitter documentation says that before participating, it’s important to know that the search API is focused on relevance, not completeness. This means that some tweets and users may not appear in the search results. If you want to be consistent, you should use the Streaming API instead.

+1
source

More articles:


All Articles