news-fetch

news-fetch 1.0.0 — a lightweight, independent, explainable, confidence-aware news extraction engine for Python.

Fetch news. Know why it worked.

pip install "news-fetch==1.0.0"

from newsfetch import fetch

article = fetch("https://www.thehindu.com/...")
print(article.title)
print(article.text)
print(article.authors)
print(article.published_at)
print(article.confidence.overall)
print(article.content_source)  # e.g. json-ld.articleBody

news-fetch is not a wrapper around newspaper4k, goose, or trafilatura. It owns its HTTP client, HTML parser, multi-strategy extractors, evidence ranking, and article model.

Why news-fetch?

Feature
Independent extraction engine (JSON-LD, Open Graph, meta, semantic HTML, DOM scoring)	✅
Field-level confidence + extraction provenance	✅
Explainable sources (title_source, content_source, …)	✅
Single URL fetch() and offline extract(html)	✅
Bulk scraping (fetch_many / fetch_iter) + proxies	✅
CLI + JSONL streaming	✅
RSS / sitemap discovery	✅
Async, optional browser render, disk cache, robots.txt	✅
No API key	✅
Small core deps (lxml, requests, python-dateutil, cssselect)	✅

Architecture (conceptual)

URL / HTML
    ↓
Transport (optional)
    ↓
Document
    ↓
Evidence (JSON-LD · OG · meta · semantic · DOM · plugins)
    ↓
Rank + confidence
    ↓
Article  →  Python · JSON · JSONL · CLI

Project links

Source	Link
PyPI	pypi.org/project/news-fetch
Repository	github.com/santhoshse7en/news-fetch
Issues	github.com/santhoshse7en/news-fetch/issues
Docs	santhoshse7en.github.io/newsfetch_doc

Next: Installation · Usage · API Reference