ferrero-opentext/Python-Version/scripts/a1_to_a2_download.py
DJP 65f2c9c68e Add Box file deletion, email notifications, and log rotation for A2→A3
Major Features Added:

1. Delete Files from Box After Upload
   - After successful DAM upload, delete file from Box
   - Prevents reprocessing same files
   - Keeps Box folder clean
   - Only deletes on success (keeps on failure for retry)

2. Email Notification for Each Upload
   - New template: a2_to_a3_file_uploaded
   - Sends email immediately after each successful upload
   - Includes: filename, clean filename, asset ID, tracking ID
   - Don't wait for "all done" - notify per file
   - Recipients: configured in .env (REPORT_EMAILS)

3. Log Rotation for Both Scripts
   - Uses RotatingFileHandler
   - Max file size: 10MB per log file
   - Backup count: 28 files (approximately 1 month)
   - Auto-rotates when log reaches 10MB
   - Keeps logs/a1_to_a2.log (current)
   - Backups: logs/a1_to_a2.log.1, .2, .3, etc.
   - Automatically deletes logs older than 28 rotations
   - Applied to both A1→A2 and A2→A3 scripts

Flow Changes:
A2→A3 now:
1. Poll Box folder
2. Find V2 files
3. Download from Box
4. Upload to DAM
5. Delete from Box  NEW
6. Send email notification  NEW
7. Store derivative record
8. Exit

Log Management:
- Active logs: ~10MB max
- Rotated backups: 28 files = ~280MB total
- Automatic cleanup (no manual intervention needed)
- 1 week of detailed logs + 3 weeks of backups

Database:
- Added dam_asset_id and upload_status columns to derivative_assets
- Fixed store_derivative_asset() to use existing schema columns

🤖 Generated with Claude Code

Co-Authored-By: Claude <noreply@anthropic.com>
2025-10-30 19:41:02 -04:00

315 lines
11 KiB
Python
Executable file

#!/usr/bin/env python3
"""
A1→A2 Master Asset Downloader
Polls DAM for campaigns with status A1, downloads master assets, uploads to Box
Updates status to A2 only when ALL assets successfully processed
Compatible with Python 3.6+
"""
import sys
import os
import time
import logging
# Add shared library to path
sys.path.insert(0, os.path.join(os.path.dirname(__file__), '..'))
from shared.config_loader import load_config, load_field_mappings
from shared.dam_client import DAMClient
from shared.box_client import BoxClient
from shared.database import Database
from shared.notifier import Notifier
# Setup logging with rotation
from logging.handlers import RotatingFileHandler
# Create logs directory if it doesn't exist
os.makedirs('logs', exist_ok=True)
os.makedirs('logs/backup', exist_ok=True)
# Configure logging with rotation
# Keep 1 week of active logs (7 days * 10MB = 70MB)
# Backup rotates keep 4 weeks (28 backups * 10MB = 280MB total)
log_handler = RotatingFileHandler(
'logs/a1_to_a2.log',
maxBytes=10*1024*1024, # 10MB per file
backupCount=28 # Keep 28 rotated files (approximately 1 month)
)
log_handler.setLevel(logging.INFO)
log_handler.setFormatter(logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s'))
console_handler = logging.StreamHandler()
console_handler.setLevel(logging.INFO)
console_handler.setFormatter(logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s'))
logging.basicConfig(
level=logging.INFO,
handlers=[log_handler, console_handler]
)
logger = logging.getLogger('A1toA2')
def process_campaign(campaign, dam, box, db, notifier, config):
"""
Process single campaign - download all master assets
Returns:
dict with success, processed_count, failed_count
"""
campaign_id = campaign['asset_id']
campaign_name = campaign['campaign_name']
campaign_number = campaign.get('campaign_id', 'N/A')
logger.info("=" * 60)
logger.info("Processing campaign: {} ({})".format(campaign_name, campaign_number))
logger.info("=" * 60)
try:
# Get master assets
master_assets = dam.get_master_assets(campaign_id)
total_assets = len(master_assets)
logger.info("Found {} master assets".format(total_assets))
if total_assets == 0:
logger.warning("No master assets found, skipping campaign")
return {'success': False, 'processed': 0, 'failed': 0}
# Track results
processed_assets = []
failed_assets = []
# Get Final Assets folder for upload directory
final_folder_id = dam.find_final_assets_folder(campaign_id)
if not final_folder_id:
logger.error("Final Assets folder not found")
return {'success': False, 'processed': 0, 'failed': total_assets}
# Process each asset
for asset in master_assets:
asset_id = asset['asset_id']
asset_name = asset.get('name', 'unknown')
try:
logger.info("Processing: {}".format(asset_name))
# 1. Download from DAM
file_path = dam.download_asset(
asset_id,
output_dir='temp/downloads/{}'.format(campaign_id)
)
# 2. Generate tracking ID
tracking_id = db.generate_unique_tracking_id()
# 3. Upload to Box (use campaign_number for folder name, not asset_id)
box_result = box.upload_with_tracking_id(
file_path=file_path,
campaign_id=campaign_number, # Use C000000078, not hex asset_id
campaign_name=campaign_name,
tracking_id=tracking_id
)
# 4. Store in database with FULL metadata
db_result = db.store_master_asset(
tracking_id=tracking_id,
opentext_id=asset_id,
asset_data=asset,
box_file_id=box_result['file_id'],
box_url=box_result['url'],
upload_folder_id=final_folder_id
)
if db_result['success']:
processed_assets.append({
'asset_id': asset_id,
'asset_name': asset_name,
'tracking_id': tracking_id,
'box_file_id': box_result['file_id'],
'box_url': box_result['url']
})
logger.info("✓ Success: {}{}".format(asset_name, tracking_id))
else:
raise Exception("Database storage failed")
# Clean up temp file
os.remove(file_path)
except Exception as e:
logger.error("✗ Failed: {} - {}".format(asset_name, str(e)))
failed_assets.append({
'asset_id': asset_id,
'asset_name': asset_name,
'error': str(e)
})
# CHECK: All assets processed successfully?
all_done = len(processed_assets) == total_assets
logger.info("")
logger.info("Campaign {} Results:".format(campaign_id))
logger.info(" Total: {}".format(total_assets))
logger.info(" Successful: {}".format(len(processed_assets)))
logger.info(" Failed: {}".format(len(failed_assets)))
logger.info(" All Done: {}".format("YES" if all_done else "NO"))
logger.info("")
if all_done:
# ALL assets processed - update status
logger.info("All assets processed - Updating status A1 → A2")
status_result = dam.update_campaign_status(campaign_id, 'A2')
if status_result['success']:
logger.info("✓ Status updated successfully")
# Send webhook notification
if config['webhooks']['campaign_status_update']['enabled']:
logger.info("Sending campaign status webhook...")
notifier.send_webhook(
url=config['webhooks']['campaign_status_update']['url'],
payload={
'campaign_id': campaign_id,
'campaign_number': campaign_number,
'campaign_name': campaign_name,
'old_status': 'A1',
'new_status': 'A2',
'asset_count': len(processed_assets),
'processed_assets': processed_assets,
'timestamp': int(time.time())
}
)
# Send success email
notifier.send_email(
template_name='a1_to_a2_complete',
recipients=config['notifications']['recipients']['success'],
data={
'campaign_name': campaign_name,
'campaign_id': campaign_id,
'campaign_number': campaign_number,
'asset_count': len(processed_assets)
}
)
return {'success': True, 'processed': len(processed_assets), 'failed': 0}
else:
logger.error("✗ Status update failed: {}".format(status_result.get('error')))
# Don't send success notification if status update failed
return {'success': False, 'processed': len(processed_assets), 'failed': 0}
else:
# NOT all done - some failed
logger.warning("Campaign incomplete - NOT updating status (remains A1)")
# Send partial completion email
notifier.send_email(
template_name='a1_to_a2_partial',
recipients=config['notifications']['recipients']['errors'],
data={
'campaign_name': campaign_name,
'campaign_id': campaign_id,
'total_assets': total_assets,
'successful': len(processed_assets),
'failed': len(failed_assets),
'failed_assets': failed_assets
}
)
return {'success': False, 'processed': len(processed_assets), 'failed': len(failed_assets)}
except Exception as e:
logger.error("Campaign processing failed: {}".format(str(e)))
return {'success': False, 'processed': 0, 'failed': total_assets}
def main():
"""Main polling loop"""
logger.info("=" * 60)
logger.info("Ferrero A1→A2 Master Asset Downloader Starting")
logger.info("=" * 60)
# Load configuration
config = load_config('config/config.yaml')
# Initialize clients
dam = DAMClient(config)
box = BoxClient(config)
db = Database(config)
notifier = Notifier(config)
# Test connections
logger.info("Testing connections...")
if not dam.test_connection():
logger.error("DAM connection failed - exiting")
sys.exit(1)
if not box.test_connection():
logger.error("Box connection failed - exiting")
sys.exit(1)
if not db.test_connection():
logger.error("Database connection failed - exiting")
sys.exit(1)
logger.info("All connections OK")
logger.info("")
# SINGLE RUN MODE - Process ONE campaign and exit
# Cron will run this script every 5 minutes, processing one campaign at a time
try:
logger.info("Searching for A1 campaigns...")
# Search for campaigns with status A1
campaigns = dam.search_campaigns(status='A1')
if not campaigns:
logger.info("No A1 campaigns found - exiting")
db.close()
sys.exit(0)
# Process ONLY THE FIRST campaign
campaign = campaigns[0]
logger.info("Found {} A1 campaigns - processing first one only".format(len(campaigns)))
logger.info("")
result = process_campaign(campaign, dam, box, db, notifier, config)
if result['success']:
logger.info("")
logger.info("=" * 60)
logger.info("✓ Campaign completed successfully")
logger.info(" Processed: {} assets".format(result['processed']))
logger.info(" Status updated: A1 → A2")
logger.info("=" * 60)
db.close()
sys.exit(0)
else:
logger.warning("")
logger.warning("=" * 60)
logger.warning("✗ Campaign incomplete or failed")
logger.warning(" Processed: {} assets".format(result['processed']))
logger.warning(" Failed: {} assets".format(result['failed']))
logger.warning(" Status NOT updated (remains A1)")
logger.warning("=" * 60)
db.close()
sys.exit(1)
except Exception as e:
logger.critical("Script error: {}".format(str(e)))
# Send critical error notification
notifier.send_email(
template_name='upload_failed',
recipients=config['notifications']['recipients']['critical'],
data={
'filename': 'A1→A2 Script',
'tracking_id': 'N/A',
'error': str(e)
}
)
db.close()
sys.exit(1)
if __name__ == '__main__':
main()