CI/CD GEO集成
将GEO检查纳入持续交付流程

name: GEO Quality Gate
on: [push, pull_request]
jobs:
  geo-lint:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - name: Schema Validation
        run: npx schema-validator ./content/**/*.jsonld
      - name: Markdown Lint
        run: npx markdownlint ./content/**/*.md

  geo-build:
    needs: geo-lint
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - name: Build Site
        run: npm run build
      - name: Validate Structured Data
        run: npx structured-data-tester ./dist/

  geo-test:
    needs: geo-build
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - name: Fact Density Score
        run: python scripts/fact_density.py --threshold 0.6
      - name: Content Structure Check
        run: python scripts/geo_structure.py
      - name: AI Readability Test
        run: npx ai-readability-tester ./dist/

  geo-deploy:
    needs: geo-test
    if: github.ref == 'refs/heads/main'
    runs-on: ubuntu-latest
    steps:
      - name: Deploy
        run: npm run deploy
      - name: Notify AI Engines
        run: python scripts/notify_ai_engines.py
      - name: Smoke Test
        run: python scripts/smoke_test.py

#!/usr/bin/env python3
"""GEO Fact Density Scorer for CI/CD Pipeline"""
import re, sys, json

def score_fact_density(content: str) -> dict:
    # Count quantitative statements (numbers, percentages, ratios)
    numbers = len(re.findall(r'\d+\.?\d*%|\d+\.?\d*', content))
    # Count citation markers [1], [2], etc.
    citations = len(re.findall(r'\[\d+\]', content))
    # Count data structures (tables, lists with numbers)
    data_structures = content.count('<table') + content.count('<dl')
    # Calculate density ratio
    words = len(content.split())
    density = (numbers * 2 + citations * 3 + data_structures * 5) / max(words, 1)

    score = min(100, int(density * 500))
    passed = score >= 60  # threshold

    return {
        "score": score,
        "passed": passed,
        "numbers": numbers,
        "citations": citations,
        "data_structures": data_structures,
        "density_ratio": round(density, 4)
    }

if __name__ == "__main__":
    threshold = float(sys.argv[1]) if len(sys.argv) > 1 else 0.6
    content = sys.stdin.read()
    result = score_fact_density(content)
    print(json.dumps(result, indent=2))
    sys.exit(0 if result["passed"] else 1)