Skip to main content
The Newspaper4k extractor is a fast, general-purpose content extractor optimized for news articles and blog posts. It automatically identifies article structure without configuration.

​
Overview

Best for:
  • News websites
  • Blog posts
  • Magazine articles
  • Clean, semantic HTML
Strengths:
  • Fast extraction
  • Automatic article detection
  • Extracts metadata (author, date, keywords, images)
  • No configuration needed
Limitations:
  • Requires semantic HTML structure
  • May struggle with unusual layouts
  • Not suitable for non-article content (e-commerce, listings)

​
Configuration

Enable in EXTRACTOR_ORDER: 
{
  "settings": {
    "EXTRACTOR_ORDER": ["newspaper", "trafilatura"]
  }
}
No additional configuration required.

​
Extracted Fields

​
Standard Fields

​
title
string
Article title (extracted from <h1>, <title>, or meta tags)Falls back to title_hint if extraction fails.
​
content
string
Main article text (cleaned, no ads/navigation)Validation: Min 100 characters
​
author
string
Author name(s) (comma-separated if multiple)Extracted from:
  • <meta name="author">
  • rel="author"
  • Common author class names
​
published_date
datetime
Publication dateExtracted from:
  • <time> elements
  • <meta property="article:published_time">
  • URL patterns

​
Metadata Fields

​
metadata.top_image
string
Main article image URLExtracted from:
  • <meta property="og:image">
  • <img> tags within article
​
metadata.keywords
string[]
Extracted keywords (auto-generated from content)
​
metadata.summary
string
Auto-generated article summary

​
Example Output

ScrapedArticle(
    url="https://www.bbc.co.uk/news/articles/abc123",
    title="Breaking News: Major Event Occurs",
    content="Full article text here...",
    author="John Smith, Jane Doe",
    published_date=datetime(2024, 2, 24, 10, 30),
    source="newspaper4k",
    metadata={
        "top_image": "https://example.com/image.jpg",
        "keywords": ["news", "event", "breaking"],
        "summary": "A brief summary of the article..."
    }
)

​
Fallback Behavior

Newspaper fails when:
  • Title extraction fails and no title_hint provided
  • Content extraction returns empty or < 100 characters
  • HTML structure is non-semantic (divs without semantic classes)
Fallback to next extractor: 
"EXTRACTOR_ORDER": ["newspaper", "trafilatura"]
If newspaper fails, trafilatura is tried next.

​
Title Hints

The spider can provide a title_hint from other sources: Example: Link text from parent page 
<a href="/article/123">This becomes title_hint</a>
If newspaper fails to extract title, it uses the hint.

​
Performance

Speed: Fast (processes HTML in-memory) Recommended concurrency: 
{
  "settings": {
    "EXTRACTOR_ORDER": ["newspaper", "trafilatura"],
    "CONCURRENT_REQUESTS": 16,
    "DOWNLOAD_DELAY": 1
  }
}

​
When to Use

​
Use Newspaper When:

Well-structured news websites
Blog platforms (WordPress, Medium, Ghost)
Semantic HTML with proper tags
Need automatic metadata extraction

​
Don’t Use Newspaper When:

E-commerce sites (use custom extractors)
Forums or discussion boards (use callbacks)
Heavily customized layouts (use custom selectors)
JavaScript-rendered content (use playwright first)

​
Debugging

Check if newspaper succeeded: 
scrapai show spider 1 --project proj
Look for source: newspaper4k in output. Test on sample page: 
# Download page HTML
scrapai inspect https://example.com/article --project proj

# Analyze structure
scrapai analyze data/proj/spider/analysis/page.html

# Run extraction test
scrapai crawl spider --limit 1 --project proj
If extraction fails:
  1. Check if content is in semantic tags (<article>, <main>, <p>)
  2. Try trafilatura as fallback
  3. Use custom selectors if generic extractors fail

​
Comparison with Trafilatura

FeatureNewspaperTrafilatura
SpeedFastFast
AccuracyGoodExcellent
MetadataRich (keywords, summary, images)Basic (description, tags)
Best ForNews sitesAny text content
FallbackUse together: ["newspaper", "trafilatura"]

​
Configuration Examples

​
News Site (Primary Strategy)

{
  "name": "bbc_co_uk",
  "source_url": "https://www.bbc.co.uk/",
  "allowed_domains": ["bbc.co.uk"],
  "start_urls": ["https://www.bbc.co.uk/news"],
  "rules": [
    {
      "allow": ["/news/articles/.*"],
      "callback": "parse_article"
    }
  ],
  "settings": {
    "EXTRACTOR_ORDER": ["newspaper", "trafilatura"],
    "DOWNLOAD_DELAY": 1,
    "CONCURRENT_REQUESTS": 16
  }
}

​
Blog with Custom Fallback

{
  "settings": {
    "EXTRACTOR_ORDER": ["newspaper", "custom"],
    "CUSTOM_SELECTORS": {
      "title": "h1.entry-title",
      "content": "div.entry-content",
      "author": "a.author-link",
      "date": "time.published"
    }
  }
}
Behavior:
  • Try newspaper first (works for most WordPress blogs)
  • Fallback to custom selectors for non-standard templates