Article Content Extractor for Research

You are an expert web scraper specializing in content extraction and DOM analysis.

Write a Python scraper that extracts clean article content from any news or blog page using readability heuristics. The scraper must work across different publishers without site-specific selectors.

## Requirements

### Input
- Accept {{url}} as the target page
- Handle dynamic and static HTML

### Extraction Strategy
Use content-density heuristics and semantic signals:
- **Title**: `<h1>`, `<title>`, or Open Graph / JSON-LD metadata; prefer the longest match that appears in multiple sources
- **Author**: `<meta name="author">`, JSON-LD "author", byline patterns (`class*="author"`, `rel="author"`), or text near phrases like "By" at the article top
- **Publication date**: `<time datetime>`, JSON-LD "datePublished", or `<meta property="article:published_time">`; parse to ISO 8601
- **Body text**: identify the main content container by scoring `<article>`, `<main>`, or `<div>` blocks for text density (word count ÷ tag count), paragraph depth, and link density (prefer prose over navigation); strip `<nav>`, `<aside>`, `<footer>`, elements with classes/IDs matching `ad|promo|comment|related|sidebar|cookie|subscribe|newsletter`, and hidden elements
- **Canonical URL**: `<link rel="canonical">` or fall back to the input URL

### Output Format
Return a plain-text report structured as:

```
Title: [extracted title]
Author: [name or "Unknown"]
Date: [ISO date or "Unknown"]
URL: [canonical URL]

[Body text with paragraph breaks preserved, no HTML, no boilerplate]
```

### Code Structure
- Use `requests` and `BeautifulSoup` (or `playwright` if the page requires JavaScript rendering)
- Include a `calculate_content_score(element)` function that returns a heuristic score based on text density, tag ratios, and semantic HTML
- Remove all navigation, ads, and boilerplate before extracting text
- Preserve paragraph breaks as double newlines
- Handle errors gracefully (network failures, missing fields)

### Quality Checks
- Body text must be at least 100 words to avoid extracting navigation fragments
- Strip leading/trailing whitespace and collapse multiple blank lines
- No HTML tags, image alt text, or inline links in the final body

Provide complete, production-ready code with comments explaining the heuristic scoring logic.