demo_functions.py

"""
Demo script to test individual utility functions without full dependencies.
Run this to see the Lambda's utility functions in action.
"""

import re
import hashlib


def extract_email_address(from_header: str) -> str:
    """Extract clean email address from 'From' header."""
    email_match = re.search(r"<([^>]+)>|([^\s<>]+@[^\s<>]+)", from_header)
    if email_match:
        return email_match.group(1) or email_match.group(2)
    return from_header.strip()


def normalize_subject_for_thread(subject: str) -> str:
    """Normalize email subject for thread_id generation."""
    if not subject:
        return ""
    normalized = re.sub(r"^(re:|fwd:|fw:)\s*", "", subject, flags=re.IGNORECASE).strip()
    normalized = " ".join(normalized.split())
    return normalized


def generate_thread_id(sender_email: str, recipient_email: str, subject: str) -> str:
    """Generate a consistent thread_id."""
    sender_clean = sender_email.lower().strip()
    recipient_clean = recipient_email.lower().strip()
    subject_normalized = normalize_subject_for_thread(subject)
    
    emails_sorted = sorted([sender_clean, recipient_clean])
    thread_components = emails_sorted + [subject_normalized]
    thread_string = "|".join(thread_components)
    
    thread_hash = hashlib.sha256(thread_string.encode("utf-8")).hexdigest()[:16]
    subject_prefix = "".join(c for c in subject_normalized[:10] if c.isalnum()).lower()
    
    if subject_prefix:
        return f"{thread_hash}-{subject_prefix}"
    else:
        return thread_hash


def extract_budget_value(budget_str: str) -> float:
    """Extract numeric budget value from budget string."""
    if not budget_str:
        return 0.0
    
    try:
        clean_budget = budget_str.replace("$", "").replace(",", "").replace("USD", "").strip()
        budget_numbers = re.findall(r"\d+(?:\.\d{2})?", clean_budget)
        
        if budget_numbers:
            return float(budget_numbers[0])
        else:
            range_match = re.search(r"(\d+(?:\.\d{2})?)\s*[-–]\s*(\d+(?:\.\d{2})?)", clean_budget)
            if range_match:
                return float(range_match.group(1))
    except (ValueError, AttributeError):
        pass
    
    return 0.0


def determine_deal_type_from_budget(budget_str: str) -> str:
    """Determine deal type based on budget description."""
    if not budget_str:
        return "FLAT_RATE"
    
    budget_lower = budget_str.lower()
    
    if any(term in budget_lower for term in ["affiliate", "commission", "%"]):
        return "AFFILIATE"
    elif "ugc" in budget_lower:
        return "UGC"
    elif "sponsored" in budget_lower:
        return "SPONSORED_POST"
    elif "partnership" in budget_lower:
        return "BRAND_PARTNERSHIP"
    elif any(term in budget_lower for term in ["revenue", "share"]):
        return "REVENUE_SHARE"
    elif any(term in budget_lower for term in ["performance", "hybrid"]):
        return "PERFORMANCE_HYBRID"
    else:
        return "FLAT_RATE"


# ============================================================================
# DEMO: Run this file to see the functions in action
# ============================================================================

if __name__ == "__main__":
    print("=" * 70)
    print("REPFLOW LAMBDA - UTILITY FUNCTIONS DEMO")
    print("=" * 70)
    
    # Demo 1: Email Extraction
    print("\n[EMAIL] 1. EMAIL EXTRACTION")
    print("-" * 70)
    
    test_emails = [
        "John Doe <john@example.com>",
        "simple@email.com",
        '"Brand Rep" <rep@brand.com>',
        "  contact@company.com  "
    ]
    
    for email in test_emails:
        result = extract_email_address(email)
        print(f"Input:  {email!r}")
        print(f"Output: {result!r}\n")
    
    # Demo 2: Subject Normalization
    print("\n[SUBJECT] 2. SUBJECT NORMALIZATION")
    print("-" * 70)
    
    test_subjects = [
        "Re: Sponsorship Opportunity",
        "Fwd: Partnership Inquiry",
        "RE: Important  Deal",
        "Brand Partnership Q4"
    ]
    
    for subject in test_subjects:
        result = normalize_subject_for_thread(subject)
        print(f"Input:  {subject!r}")
        print(f"Output: {result!r}\n")
    
    # Demo 3: Thread ID Generation
    print("\n[THREAD] 3. THREAD ID GENERATION")
    print("-" * 70)
    
    thread_tests = [
        ("alice@example.com", "bob@example.com", "Sponsorship Opportunity"),
        ("bob@example.com", "alice@example.com", "Re: Sponsorship Opportunity"),  # Same thread!
        ("alice@example.com", "charlie@example.com", "Sponsorship Opportunity"),  # Different
    ]
    
    for sender, recipient, subject in thread_tests:
        result = generate_thread_id(sender, recipient, subject)
        print(f"From:    {sender}")
        print(f"To:      {recipient}")
        print(f"Subject: {subject}")
        print(f"Thread:  {result}\n")
    
    # Demo 4: Budget Extraction
    print("\n[BUDGET] 4. BUDGET EXTRACTION")
    print("-" * 70)
    
    test_budgets = [
        "$1,000",
        "$1,000 - $5,000",
        "Budget is $2500 USD",
        "5000 dollars",
        "Negotiable"
    ]
    
    for budget in test_budgets:
        result = extract_budget_value(budget)
        print(f"Input:  {budget!r}")
        print(f"Output: ${result:,.2f}\n")
    
    # Demo 5: Deal Type Detection
    print("\n[DEALS] 5. DEAL TYPE DETECTION")
    print("-" * 70)
    
    test_deal_types = [
        "$5000 flat rate",
        "10% affiliate commission",
        "$500 for UGC content",
        "$2000 sponsored post",
        "Long-term partnership",
        "$1000 + performance bonus"
    ]
    
    for deal_desc in test_deal_types:
        result = determine_deal_type_from_budget(deal_desc)
        print(f"Input:  {deal_desc!r}")
        print(f"Type:   {result}\n")
    
    print("=" * 70)
    print("[SUCCESS] DEMO COMPLETE!")
    print("=" * 70)
    print("\nThese are the core utility functions from the Lambda.")
    print("The full Lambda uses 4 AI agents + AWS services.")
    print("\nTo see the full analysis, read: SOLUTION.md")
    print("To run all 80 tests: pytest tests/ -v")