This project aims to implement a cli wrapper for basic operations againts twitter's REST api. Results are store into an sqlite database.
Clone this repository and run npm install to install dependencies. Create a new database using the provided schema (sqlite/schema.sqlite). If you have installed the sqlite cli tool this command should create the databse sqlite3 DATABASENAME < sqlite/schema.sqlite.
Create a new app at https://apps.twitter.com/. Copy and rename auth.json.example to auth.json and insert the keys and tokens from your new app.
To prepare the sentiment detection add a folder sentistrength to the project. Copy the SentiStrength jar file to this folder and rename the jar to sentistrength.jar. Put any sentiment lookup file into a subfolder data.
Run node index.js <COMMAND>
| Command | Description |
|---|---|
timeline |
Retrieve your current home timeline |
user $USER |
Retrieve the timeline from the user whose twitter screenname is $USER |
import $USER_DATA |
Update meta information in user table. Can only be used with the sqlite option. $USER_DATA must point to an json formatted file with an array user objects containing screen_name, party and origin properties. |
| Options | Description |
|---|---|
--count |
Number of tweets to be retrieved |
--sqlite |
Store retrieved tweets into a sqlite database. If not value is given, results are stored in tweets.sqlite |
--html |
Resolves urls in tweets and stores html content in database. |
--sentistrength |
Augment results with sentiment scores (See above: Sentiment Detection). |
--auth |
Path to oauth file. If this option is ommitted, oauth information will be read from auth.json |
The code in this repository is provide under MIT license. No twitter content is stored in or published through this repository. Make sure that your usage of this software complies to Twitters Developer Agreement. Other licences may be applied to dependcies use to run this software.