procurement-ingest/test_local.py

#!/usr/bin/env python3
"""
Local test script - parses sample WO emails through the deterministic template
parser, falling back to Claude on Bedrock for anything the templates don't cover.

Usage:
    # AWS credentials with bedrock:InvokeModel are only needed if a sample falls
    # back to the AI extractor; pure-template samples parse fully offline.
    python test_local.py
"""

import json
import sys
from pathlib import Path

sys.path.insert(0, str(Path(__file__).parent / "lambdas" / "wo" / "email_processor"))

from handler import parse_raw_email, extract_with_bedrock
from template_parser import try_deterministic_parse


def main():
    samples_dir = Path(__file__).parent / "samples"
    eml_files = list(samples_dir.glob("*.eml"))

    if not eml_files:
        print("No .eml files found in samples/")
        return

    for eml_path in eml_files:
        print(f"\n{'=' * 80}")
        print(f"FILE: {eml_path.name}")
        print(f"{'=' * 80}")

        raw = eml_path.read_bytes()
        email_data = parse_raw_email(raw)

        print(f"Subject: {email_data['subject']}")
        print(f"From:    {email_data['sender']}")
        print(f"Date:    {email_data['date']}")
        print(f"Body preview: {email_data['body'][:200]}...")
        print()

        parsed, method, template_id, reason = try_deterministic_parse(email_data)
        if parsed is None:
            print(f"Template miss ({template_id}/{reason}); sending to Bedrock...")
            parsed = extract_with_bedrock(email_data)
            method = "ai_fallback"
        print(f"Parse method: {method} (template={template_id}, reason={reason})")
        print(json.dumps(parsed, indent=2))


if __name__ == "__main__":
    main()