GatherNews 0.2.1

Gathers unstructured News data into a SQLite3 db

Gathers unstructured news data and commits it to a SQLite3 database.The goal for GatherNews is to quickly and simply capture news data.

GatherNews allows you to specify which News sites you want to capture by providing the RSS link in “feeds_list.txt” like this:

http://feeds.reuters.com/Reuters/worldNews
http://rss.cnn.com/rss/money_latest.rss

You can then gather the news using the load_db() method:

>>> # Create new tables if any new RSS feed addresses have been added
>>> # Populate all tables with RSS news feeds
>>> # Remove duplicate entries
>>> capture_feeds.load_db()

The examples folder contains working code for each module.

Features

• Creates tables with a predefined schema

• Populates each table in the SQLite3 database with articles

• Removes duplicate articles from each table

• Includes bug fixes for issues raised about version 0.1.0

• Faster than version 0.1.0, see the benchmarks!

Installation

To install GatherNews use pip:

$ pip install gathernews

Testing

After installation, you can launch the test suite from outside the source directory (you will need to have nosetests installed):

$ nosetests --exe GatherNews

Bug Resolution

In GatherNews 0.1.0, a bug was introduced that does not allow you to add new RSS feeds to the ‘feeds_list.txt’ after your initial call of the create_tables() method. Before using version 0.2.0 on a preexisting database call the following method.

>>> path = "path/to_your/existing_sqlite3_db/
>>> from gathernews.gRSS import CaptureFeeds
>>> capture_feeds = CaptureFeeds(path)
>>> capture_feeds.fix_create_table_bug()

If this gives you any problems, email me directly ([e-mail]). Thanks.

Documentation

Documentation is available at http://gathernews.readthedocs.org/en/latest/

Contribute

1. Issue tracker is here: https://github.com/Bonza-Times/GatherNews/issues

2. Fork it!