20 lines
1002 B
Python
20 lines
1002 B
Python
"""Download, persist and promote the official NCM dataset."""
|
|
from datetime import datetime, timezone
|
|
from pathlib import Path
|
|
import httpx
|
|
from central.app.config import settings
|
|
from central.app.datasets.ncm import parse_ncm_payload
|
|
from central.app.datasets.service import DatasetService
|
|
from central.app.db import SessionLocal
|
|
|
|
def main():
|
|
response = httpx.get(settings.ncm_dataset_url, timeout=settings.http_timeout_seconds, follow_redirects=True)
|
|
response.raise_for_status()
|
|
parsed = parse_ncm_payload(response.content)
|
|
path = Path(settings.dataset_storage_path); path.mkdir(parents=True, exist_ok=True)
|
|
(path / f"{parsed.sha256}.json").write_bytes(response.content)
|
|
with SessionLocal() as session:
|
|
dataset = DatasetService(session).import_payload(parsed, settings.ncm_dataset_url, datetime.now(timezone.utc))
|
|
print({"id": dataset.id, "status": dataset.status, "sha256": dataset.sha256, "record_count": dataset.record_count})
|
|
if __name__ == "__main__": main()
|