/Twitter_Crawler

Easy-to-use tool for mining tweets with Twitter API. Support both twitter API version 1 and academic API

Primary LanguagePythonMIT LicenseMIT

Twitter_Crawler

Easy-to-use tool for mining tweets with Twitter API. Support both Twitter API version 1 and academic API.

It will collect tweets based on your need, and store the result either in a json file (supported only for academic API) or in a mongo database (supported for both academic API and Twitter API version 1).

Multi-process will be used to accelerate collection speed with Twitter API version 1.

Author: Hongbo Fang

Contact: fanghongdoublebo@gmail.com (Please contact me directly or post on issue if you want a new feature or find a bug)

Get started

  • Download tw_crawler_release.py to your local machine
  • Import Twitter_Crawler_Version2 (for Twitter academic API ) or Twitter_Crawler (for Twitter API version 1) in your python script
  • You will need Twitter Academic API Bearer token to use Twitter_Crawler_Version2. get it here.
  • You will need Twitter API version 1 to use Twitter_Crawler. get it here

Examples

Here are some examples to quickly get started!

Academic API (Twitter_Crawler_Version2)

  • Task 1: Collect all tweets posted by given users, and save it in a json file.

    from tw_crawler_release import Twitter_Crawler_Version2

tw_crawler = Twitter_Crawler_Version2('YOUR TWITTER API BEARER_TOKEN')
tw_crawler.crawl_tweets_from_user(['1169594598760562689'], # list of users to collect tweets from, user specified by their user id (in str) 
                                      './save_result.json', # the json file directory to which you want to save your result
                                      save_format = 'json' # indicate it's stored in json file
                                      )

    The result is stored as a list of tweets collected.

  • Task 2: Collect all tweets contain given keywords, and save it in a json file.

    from tw_crawler_release import Twitter_Crawler_Version2

tw_crawler = Twitter_Crawler_Version2('YOUR TWITTER API BEARER_TOKEN')
tw_crawler.crawl_tweets_contain_keyword(['Hello World'], # list of keywords as query 
                                      './save_result.json', # the json file directory to which you want to save your result
                                      save_format = 'json' # indicate it's stored in json file
                                      )

    The result is stored as a list of tweets collected.

  • Task 3: Collect all tweets in reply to given tweets, and save it in a json file.

 from tw_crawler_release import Twitter_Crawler_Version2
 
 tw_crawler = Twitter_Crawler_Version2('YOUR TWITTER API BEARER_TOKEN')
 tw_crawler.crawl_tweets_replyto_tweet(['1507319500684898317'], # list of tweets 
                                       './save_result.json', # the json file directory to which you want to save your result
                                       save_format = 'json' # indicate it's stored in json file
                                       )

Twitter API version 1 (Twitter_Crawler)

Come up soon

Advanced features

  • Specify the post date range of tweet collected with academic API.

     tw_crawler.crawl_tweets_from_user(['1169594598760562689'], # list of users to collect tweets from, user specified by their user id (in str) 
                                       './save_result.json', # the json file directory to which you want to save your result; the result is stored as a list of tweets collected 
                                       save_format = 'json', # indicate it's stored in json file
                                       start = '2006-03-21T00:00:00Z', 
                                       end = '2008-03-21T00:00:00Z', 
                                       )

    The parameter start or end is a string of time, and it has be in the format of "yyyy-mm-ddThh-mm-ssZ"

  • Specify the kind of information to be retrieved for collected tweets

     tw_crawler.crawl_tweets_from_user(['1169594598760562689'], # list of users to collect tweets from, user specified by their user id (in str) 
                                       './save_result.json', # the json file directory to which you want to save your result; the result is stored as a list of tweets collected 
                                       save_format = 'json', # indicate it's stored in json file
                                       field_list = 'author_id,public_metrics,in_reply_to_user_id,created_at,entities,referenced_tweets,geo,lang' 
                                       )

    The parameter field_list is a string of field seperated by comma, the current one shown is the default. See the complete list of fields here.

  • Save the collected tweets into a mongo database instead of a json file

   import pymongo
   from tw_crawler_release import Twitter_Crawler_Version2
   
   
   client = pymongo.MongoClient(host='localhost', username = 'YOUR MONGO DB USERNAME', 
       password = 'YOUR MONGO DB PASSWORD', authSource = 'Authorized database', port=27017)
   db = client.twitter # a selected database (for example, mine is twitter)
   tw_crawler = Twitter_Crawler_Version2('YOUR TWITTER API BEARER_TOKEN', db)
   
   tw_crawler.crawl_tweets_from_user(['1169594598760562689'], # list of users to collect tweets from, user specified by their user id (in str) 
                                         'mongocollection', # mongo collection to store the collected tweets
                                         save_format = 'mongo' # indicate it's stored in mongo file
                                         )