Finance News Aggregator
Table of Contents
Overview
Current Version: 0.1.3
The finnews library collects news articles from 8 financial news providers via their RSS feeds. It provides a clean facade pattern — create one News() client and access any provider through properties.
Features:
- 8 providers with ~84 feed methods
- Enum-based topic selection for CNBC and MarketWatch
- Automatic retry with exponential backoff
- User-Agent rotation to avoid bot detection
- Structured
NewsArticle / NewsFeed dataclasses with Jupyter HTML rendering
- Secure XML parsing via
defusedxml
Installation
pip install fin-news
Upgrade to the latest version:
pip install --upgrade fin-news
Quick Start
from finnews import News
# Create the main client.
news_client = News()
# Fetch top CNBC news.
cnbc = news_client.cnbc
top_news = cnbc.news_feed(topic='top_news')
# Each item is a dictionary with title, link, description, etc.
for article in top_news:
print(article['title'])
Providers
CNBC
cnbc = news_client.cnbc
# List all available topic keys.
print(cnbc.topics)
# Fetch by topic string.
articles = cnbc.news_feed(topic='top_news')
investing = cnbc.investing_feeds(topic='investing')
blogs = cnbc.blogs(topic='charting_asia')
tv = cnbc.videos_and_tv(topic='top_video')
# Fetch all feeds at once.
all_feeds = cnbc.all_feeds()
MarketWatch
market_watch = news_client.market_watch
# List all available topic keys.
print(market_watch.topics)
# Fetch by topic string.
top = market_watch.top_stories()
headlines = market_watch.real_time_headlines()
bulletins = market_watch.bulletins()
pulse = market_watch.market_pulse()
NASDAQ
nasdaq = news_client.nasdaq
# List all available feed methods.
print(nasdaq.feeds)
# Fetch feeds.
commodities = nasdaq.commodities_feed()
crypto = nasdaq.cryptocurrency_feed()
ai_news = nasdaq.artificial_intelligence_feed()
earnings = nasdaq.earnings_feed()
# Fetch by ticker symbol.
ticker_news = nasdaq.ticker_feed(ticker_symbol='AAPL')
S&P Global
sp_global = news_client.sp_global
# List all available feed methods.
print(sp_global.feeds)
# Fetch feeds.
research = sp_global.research()
indices = sp_global.all_indices()
commentary = sp_global.market_commentary()
CNN Finance
cnn = news_client.cnn_finance
# List all available feed methods.
print(cnn.feeds)
# Fetch feeds.
top_stories = cnn.top_stories()
markets = cnn.markets()
technology = cnn.technology()
economy = cnn.economy()
Seeking Alpha
seeking_alpha = news_client.seeking_alpha
# List all available feed methods.
print(seeking_alpha.feeds)
# Fetch feeds.
latest = seeking_alpha.latest_articles()
popular = seeking_alpha.most_popular_articles()
# Fetch by ticker or sector.
aapl = seeking_alpha.stocks(ticker='AAPL')
tech = seeking_alpha.sectors(sector='technology')
Wall Street Journal
wsj = news_client.wsj
# List all available feed methods.
print(wsj.feeds)
# Fetch feeds.
market_news = wsj.market_news()
world = wsj.world_news()
opinions = wsj.opinions()
Yahoo Finance
yahoo = news_client.yahoo_finance
# List all available feed methods.
print(yahoo.feeds)
# Fetch general news.
news = yahoo.news()
# Fetch headlines for specific symbols.
headlines = yahoo.headlines(symbols=['AAPL', 'MSFT'])
Structured Models
Raw dictionaries can be converted to NewsArticle and NewsFeed dataclasses for IDE autocomplete and Jupyter rendering:
from finnews import News, NewsArticle, NewsFeed
news_client = News()
raw = news_client.cnbc.news_feed(topic='top_news')
# Single article
article = NewsArticle.from_dict(raw[0], source='cnbc')
print(article.title, article.link)
# Full feed collection
feed = NewsFeed.from_dicts(raw, source='cnbc')
print(len(feed), "articles")
# In Jupyter notebooks, articles and feeds render as HTML automatically.
feed # displays an HTML table
Enums
CNBC and MarketWatch support enum-based topic selection:
from finnews import News
from finnews.news_enum import CNBCTopNews, MarketWatch as MWEnum
news_client = News()
# CNBC with enum.
articles = news_client.cnbc.news_feed(topic=CNBCTopNews.REAL_ESTATE)
# MarketWatch with enum.
articles = news_client.market_watch.top_stories()
Available CNBC enums: CNBCTopNews, CNBCInvesting, CNBCBlogs, CNBCTVVideoAndTV, CNBCTVProgramsEurope, CNBCTVProgramsAsia
API Reference
Entry Point
| Class |
Description |
News() |
Main facade — access all providers via properties |
Provider Properties
| Property |
Class |
Topics/Feeds |
news.cnbc |
CNBC |
.topics — 55 topic keys |
news.market_watch |
MarketWatch |
.topics — 4 topic keys |
news.nasdaq |
NASDAQ |
.feeds — 22 feed methods |
news.sp_global |
SPGlobal |
.feeds — 14 feed methods |
news.cnn_finance |
CNNFinance |
.feeds — 22 feed methods |
news.seeking_alpha |
SeekingAlpha |
.feeds — 13 feed methods |
news.wsj |
WallStreetJournal |
.feeds — 6 feed methods |
news.yahoo_finance |
YahooFinance |
.feeds — 2 feed methods |
Data Models
| Class |
Description |
NewsArticle |
Dataclass: title, link, description, published, source, extra |
NewsFeed |
Collection of NewsArticle — iterable, indexable, len() support |
Exceptions
| Exception |
Description |
FinnewsError |
Base exception for all finnews errors |
InvalidTopicError |
Raised for unknown topic keys (also a KeyError) |
FeedRequestError |
HTTP request failures |
FeedParseError |
XML parsing failures |
Utility
| Method |
Description |
news.save_to_file(content, file_name) |
Save results to samples/responses/<file_name>.jsonc |
Comparison with Alternatives
| Feature |
fin-news |
feedparser |
gnews |
newscatcher |
| Finance-focused providers |
8 providers |
Any RSS |
Google News |
News API |
| Topic-based feeds |
Yes |
N/A |
Yes |
Yes |
| Keyword search |
No |
N/A |
Yes |
Yes |
| Structured models |
Yes (NewsArticle) |
Yes |
Yes |
Yes |
| Enum topic selection |
Yes |
No |
No |
No |
| Retry with backoff |
Yes |
N/A |
Yes |
No |
| Jupyter rendering |
Yes |
No |
No |
No |
| Async support |
Yes |
No |
No |
No |
| No API key required |
Yes |
Yes |
Yes |
No |
Contributing
See CONTRIBUTING.md for development setup, testing, and PR guidelines.
Support
Patreon: Help support this project by donating to the Patreon Page.
YouTube: Watch more content on Sigma Coding.
评论