jterstriep/twapture repository overview

⁠Twapture - capturing tweets using Twitter's streaming API

Twapture provides a flexible tool for capturing tweets from Twitter's streaming API. The stream can be filtered based on topics (keywords and phrases) and/or location (bounding box). If no filters are specified the entire public data stream is sampled.

With no options, twapture dumps all incoming tweets to stdout in human readable format. But, twapture uses Python's excellent logging tool to record incoming tweets to file. By default, a WatchedFileHandler is used so that an external process can manage file rotation, but a TimedRotatingFileHandling can be specified to allow twapture to self-manage its record files.

Twapture can be used as a command line tool to explore the twitter stream, but it is primarily designed to work in concert with supervisord to provide a robust, continuously running capture service.

The twapture package includes 3 custom classes; MultiplexListener, StatusEncoder, and TwitterStreamStats. MultiplexListener allows a series of filtering functions to be performed on each status update. StatusEncoder helps format a status for inclusion in files or outputing to another service. TwitterStreamStats provides a MultiplexListener filter that collects statistics on the incoming stream.

⁠Installion

git clone https://github.com/jterstriep/twapture.git
sudo python setup.py install

⁠To Do

• Write documentation
• Add package to PyPI

⁠Resources

https://dev.twitter.com/overview/api⁠

http://www.tweepy.org/⁠

http://objectpath.org/reference.html⁠