viTweet is a webservice providing visualizations to better understand Twitter conversations with their Replies and Quote Tweets that follow a selected Tweet.
| Program library | Usage |
|---|---|
| Twython 3.7 | Wrapper for Twitter API |
| Flask 1.0.2 | Web framework |
| Jinja2 2.10 | Template engine |
| D3.js v4 | Graph visualization |
Make sure to import all libraries or use the pipfile ('source/Pipfile') to install them!
- Set up a Twitter developer account if you do not already have one
- Generate your access tokens
- Enter your access tokens to the configuration file '/source/config/config.ini'
-
Use an IDE like Eclipse (others have not yet been tested by us) to run the file 'app.py'
-
Run via command line
$ cd viTweet/source/ # Change to where the directory containing 'app.py' is located
$ export FLASK_APP=app.py
$ export FLASK_ENV=development # To run with debugger
$ flask run- Open the URL indicated in the console output in your brower (usually http://127.0.0.1:5000/)
- Enter a Tweet URL or ID into the search bar
- Try the different visualization options
- List visualization
- Graph visualization
- Export and import conversations
- Export search result in different formats with the button at the top right of the results page
- Import a file (type 'flat JSON') with the uploader at the bottom of the index page
- see comments in the respective files
- see the Wiki for a more extensive documentation (in German)