news-crawler's Introduction

news-crawler

A news crawler for BBC News, Reuters and New York Times.

Update and Clarification (2021/10/03)

For BBC: It is a news collection from BBC front pages https://www.bbc.com/news, starting from 2015/07/01/. This archive is collected by @dracos. Please refer to his website: https://dracos.co.uk/made/bbc-news-archive/archive.php.
For Reuters, they have disabled their original archive website. The new website https://www.reuters.com/news/archive has only a limited number of historical articles (starting from 2020/03/08), so I did not update codes for Reuters anymore. But I still keep the codes for Reuters as an example, in case that you want to implement your own codes for Reuters.

Requirements

python3
configobj
dateutil
requests
bs4
goose3

pip install -r requirements.txt

Architecture

xxx_crawler: the executive file to crawl news.
xxx.cfg: configurations for the crawler, including api, time range and storage path etc.
xxx_link.py: fetch download links.
xxx_article: extract content and some meta data of one news article.

Usage

BBC News

python bbc_crawler.py settings/bbc.cfg

Reuters

python reuters_crawler.py reuters.cfg

New York Times

python nytimes_crawler.py nytimes.cfg

Configuration

Modify reuters.cfg, nytimes.cfg and bbc.cfg in settings folder, the main configuration items may be start_date, end_date and path.

Notes

If other news sources need to be added, just add files as the architecture, extend the basic class in each folder. Some methods may need to be rewrote.

Recommend Projects