seo/scripts/user_approval.py

#!/usr/bin/env python3
"""
User Approval Mechanism for SEO Recommendations
Allows users to review and approve recommendations from CSV files.
"""

import csv
import json
import logging
import sys
from pathlib import Path
from typing import Dict, List, Optional
from datetime import datetime
from config import Config

# Setup logging
logging.basicConfig(
    level=logging.INFO,
    format='%(asctime)s - %(levelname)s - %(message)s'
)
logger = logging.getLogger(__name__)


class UserApprovalSystem:
    """System for reviewing and approving SEO recommendations."""

    def __init__(self):
        """Initialize the approval system."""
        self.output_dir = Path(__file__).parent.parent / 'output'
        self.approved_recommendations = []
        self.rejected_recommendations = []
        self.pending_recommendations = []

    def load_recommendations_from_csv(self, csv_file: str) -> List[Dict]:
        """Load recommendations from CSV file."""
        recommendations = []
        
        if not Path(csv_file).exists():
            logger.error(f"CSV file not found: {csv_file}")
            return recommendations

        try:
            with open(csv_file, 'r', encoding='utf-8') as f:
                reader = csv.DictReader(f)
                for row in reader:
                    recommendations.append(dict(row))
            
            logger.info(f"Loaded {len(recommendations)} recommendations from {csv_file}")
            return recommendations
        except Exception as e:
            logger.error(f"Error loading CSV: {e}")
            return recommendations

    def display_recommendation(self, recommendation: Dict, index: int, total: int):
        """Display a single recommendation for user review."""
        print(f"\n{'='*80}")
        print(f"RECOMMENDATION {index}/{total}")
        print(f"{'='*80}")
        
        # Display different fields depending on the type of recommendation
        if 'post_title' in recommendation:
            print(f"Post Title: {recommendation.get('post_title', 'N/A')}")
            print(f"Post ID: {recommendation.get('post_id', 'N/A')}")
            print(f"Site: {recommendation.get('site', 'N/A')}")
            print(f"Current Categories: {recommendation.get('current_categories', 'N/A')}")
            print(f"Proposed Category: {recommendation.get('proposed_category', 'N/A')}")
            print(f"Proposed Site: {recommendation.get('proposed_site', 'N/A')}")
            print(f"Reason: {recommendation.get('reason', 'N/A')}")
            print(f"Confidence: {recommendation.get('confidence', 'N/A')}")
            print(f"Content Preview: {recommendation.get('content_preview', 'N/A')[:100]}...")
        elif 'title' in recommendation:
            print(f"Post Title: {recommendation.get('title', 'N/A')}")
            print(f"Post ID: {recommendation.get('post_id', 'N/A')}")
            print(f"Site: {recommendation.get('site', 'N/A')}")
            print(f"Decision: {recommendation.get('decision', 'N/A')}")
            print(f"Recommended Category: {recommendation.get('recommended_category', 'N/A')}")
            print(f"Reason: {recommendation.get('reason', 'N/A')}")
            print(f"Priority: {recommendation.get('priority', 'N/A')}")
            print(f"AI Notes: {recommendation.get('ai_notes', 'N/A')}")
        else:
            # Generic display for other types of recommendations
            for key, value in recommendation.items():
                print(f"{key.replace('_', ' ').title()}: {value}")

    def get_user_choice(self) -> str:
        """Get user's approval choice."""
        while True:
            print(f"\nOptions:")
            print(f"  'y' or 'yes' - Approve this recommendation")
            print(f"  'n' or 'no' - Reject this recommendation") 
            print(f"  's' or 'skip' - Skip this recommendation for later review")
            print(f"  'q' or 'quit' - Quit and save current progress")
            
            choice = input(f"\nEnter your choice: ").strip().lower()
            
            if choice in ['y', 'yes']:
                return 'approved'
            elif choice in ['n', 'no']:
                return 'rejected'
            elif choice in ['s', 'skip']:
                return 'pending'
            elif choice in ['q', 'quit']:
                return 'quit'
            else:
                print("Invalid choice. Please enter 'y', 'n', 's', or 'q'.")

    def review_recommendations(self, recommendations: List[Dict], title: str = "Recommendations"):
        """Review recommendations with user interaction."""
        print(f"\n{'='*80}")
        print(f"REVIEWING {title.upper()}")
        print(f"Total recommendations to review: {len(recommendations)}")
        print(f"{'='*80}")
        
        for i, recommendation in enumerate(recommendations, 1):
            self.display_recommendation(recommendation, i, len(recommendations))
            
            choice = self.get_user_choice()
            
            if choice == 'quit':
                logger.info("User chose to quit. Saving progress...")
                break
            elif choice == 'approved':
                recommendation['status'] = 'approved'
                self.approved_recommendations.append(recommendation)
                logger.info(f"Approved recommendation {i}")
            elif choice == 'rejected':
                recommendation['status'] = 'rejected'
                self.rejected_recommendations.append(recommendation)
                logger.info(f"Rejected recommendation {i}")
            elif choice == 'pending':
                recommendation['status'] = 'pending_review'
                self.pending_recommendations.append(recommendation)
                logger.info(f"Skipped recommendation {i} for later review")

    def export_approved_recommendations(self, filename_suffix: str = "") -> str:
        """Export approved recommendations to CSV."""
        if not self.approved_recommendations:
            logger.info("No approved recommendations to export")
            return ""

        timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
        filename = f"approved_recommendations_{timestamp}{filename_suffix}.csv"
        csv_file = self.output_dir / filename

        # Get all unique fieldnames from recommendations
        fieldnames = set()
        for rec in self.approved_recommendations:
            fieldnames.update(rec.keys())
        fieldnames = sorted(list(fieldnames))

        with open(csv_file, 'w', newline='', encoding='utf-8') as f:
            writer = csv.DictWriter(f, fieldnames=fieldnames)
            writer.writeheader()
            writer.writerows(self.approved_recommendations)

        logger.info(f"Exported {len(self.approved_recommendations)} approved recommendations to: {csv_file}")
        return str(csv_file)

    def export_rejected_recommendations(self, filename_suffix: str = "") -> str:
        """Export rejected recommendations to CSV."""
        if not self.rejected_recommendations:
            logger.info("No rejected recommendations to export")
            return ""

        timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
        filename = f"rejected_recommendations_{timestamp}{filename_suffix}.csv"
        csv_file = self.output_dir / filename

        # Get all unique fieldnames from recommendations
        fieldnames = set()
        for rec in self.rejected_recommendations:
            fieldnames.update(rec.keys())
        fieldnames = sorted(list(fieldnames))

        with open(csv_file, 'w', newline='', encoding='utf-8') as f:
            writer = csv.DictWriter(f, fieldnames=fieldnames)
            writer.writeheader()
            writer.writerows(self.rejected_recommendations)

        logger.info(f"Exported {len(self.rejected_recommendations)} rejected recommendations to: {csv_file}")
        return str(csv_file)

    def export_pending_recommendations(self, filename_suffix: str = "") -> str:
        """Export pending recommendations to CSV."""
        if not self.pending_recommendations:
            logger.info("No pending recommendations to export")
            return ""

        timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
        filename = f"pending_recommendations_{timestamp}{filename_suffix}.csv"
        csv_file = self.output_dir / filename

        # Get all unique fieldnames from recommendations
        fieldnames = set()
        for rec in self.pending_recommendations:
            fieldnames.update(rec.keys())
        fieldnames = sorted(list(fieldnames))

        with open(csv_file, 'w', newline='', encoding='utf-8') as f:
            writer = csv.DictWriter(f, fieldnames=fieldnames)
            writer.writeheader()
            writer.writerows(self.pending_recommendations)

        logger.info(f"Exported {len(self.pending_recommendations)} pending recommendations to: {csv_file}")
        return str(csv_file)

    def run_interactive_approval(self, csv_files: List[str]):
        """Run interactive approval process for multiple CSV files."""
        logger.info("="*70)
        logger.info("USER APPROVAL SYSTEM FOR SEO RECOMMENDATIONS")
        logger.info("="*70)
        
        for csv_file in csv_files:
            logger.info(f"\nLoading recommendations from: {csv_file}")
            recommendations = self.load_recommendations_from_csv(csv_file)
            
            if not recommendations:
                logger.warning(f"No recommendations found in {csv_file}, skipping...")
                continue
            
            # Get the filename without path for the title
            filename = Path(csv_file).stem
            self.review_recommendations(recommendations, title=filename)
        
        # Export results
        logger.info("\n" + "="*70)
        logger.info("EXPORTING RESULTS")
        logger.info("="*70)
        
        approved_file = self.export_approved_recommendations()
        rejected_file = self.export_rejected_recommendations()
        pending_file = self.export_pending_recommendations()
        
        # Summary
        logger.info(f"\n{'─'*70}")
        logger.info("APPROVAL SUMMARY:")
        logger.info(f"  Approved: {len(self.approved_recommendations)}")
        logger.info(f"  Rejected: {len(self.rejected_recommendations)}")
        logger.info(f"  Pending: {len(self.pending_recommendations)}")
        logger.info(f"{'─'*70}")
        
        if approved_file:
            logger.info(f"\nApproved recommendations saved to: {approved_file}")
        if rejected_file:
            logger.info(f"Rejected recommendations saved to: {rejected_file}")
        if pending_file:
            logger.info(f"Pending recommendations saved to: {pending_file}")
        
        logger.info(f"\n✓ Approval process complete!")

    def run_auto_approval(self, csv_files: List[str], auto_approve_threshold: float = 0.8):
        """Auto-approve recommendations based on confidence threshold."""
        logger.info("="*70)
        logger.info("AUTO APPROVAL SYSTEM FOR SEO RECOMMENDATIONS")
        logger.info("="*70)
        logger.info(f"Auto-approval threshold: {auto_approve_threshold}")
        
        all_recommendations = []
        for csv_file in csv_files:
            logger.info(f"\nLoading recommendations from: {csv_file}")
            recommendations = self.load_recommendations_from_csv(csv_file)
            all_recommendations.extend(recommendations)
        
        approved_count = 0
        rejected_count = 0
        
        for rec in all_recommendations:
            # Check if there's a confidence field and if it meets the threshold
            confidence_str = rec.get('confidence', 'Low').lower()
            confidence_value = 0.0
            
            if confidence_str == 'high':
                confidence_value = 0.9
            elif confidence_str == 'medium':
                confidence_value = 0.6
            elif confidence_str == 'low':
                confidence_value = 0.3
            else:
                # Try to parse as numeric value if possible
                try:
                    confidence_value = float(confidence_str)
                except ValueError:
                    confidence_value = 0.3  # Default to low
            
            if confidence_value >= auto_approve_threshold:
                rec['status'] = 'auto_approved'
                self.approved_recommendations.append(rec)
                approved_count += 1
            else:
                rec['status'] = 'auto_rejected'
                self.rejected_recommendations.append(rec)
                rejected_count += 1
        
        # Export results
        logger.info("\n" + "="*70)
        logger.info("EXPORTING AUTO-APPROVAL RESULTS")
        logger.info("="*70)
        
        approved_file = self.export_approved_recommendations("_auto")
        rejected_file = self.export_rejected_recommendations("_auto")
        
        # Summary
        logger.info(f"\n{'─'*70}")
        logger.info("AUTO APPROVAL SUMMARY:")
        logger.info(f"  Auto-approved: {approved_count}")
        logger.info(f"  Auto-rejected: {rejected_count}")
        logger.info(f"{'─'*70}")
        
        if approved_file:
            logger.info(f"\nAuto-approved recommendations saved to: {approved_file}")
        if rejected_file:
            logger.info(f"Auto-rejected recommendations saved to: {rejected_file}")
        
        logger.info(f"\n✓ Auto-approval process complete!")


def main():
    """Main entry point."""
    import argparse

    parser = argparse.ArgumentParser(
        description='Review and approve SEO recommendations'
    )
    parser.add_argument(
        'csv_files',
        nargs='+',
        help='CSV files containing recommendations to review'
    )
    parser.add_argument(
        '--auto',
        action='store_true',
        help='Run auto-approval mode instead of interactive mode'
    )
    parser.add_argument(
        '--threshold',
        type=float,
        default=0.8,
        help='Confidence threshold for auto-approval (default: 0.8)'
    )

    args = parser.parse_args()

    approval_system = UserApprovalSystem()
    
    if args.auto:
        approval_system.run_auto_approval(args.csv_files, args.threshold)
    else:
        approval_system.run_interactive_approval(args.csv_files)


if __name__ == '__main__':
    main()