wifi-etl/backfill.py

46 lines
1.5 KiB
Python
Raw Permalink Normal View History

import time
from datetime import date, timedelta
import psycopg2
from app.extractor.wifeed import get_access_token, extract_all_access
from app.core.config import WIFEED_CLIENT_ID, WIFEED_CLIENT_SECRET
from app.core.config import DB_HOST, DB_PORT, DB_NAME, DB_USER, DB_PASSWORD
from app.transform.merge_mac import transform_wifeed
from app.load.load_database import load
START_DATE = date(2026, 4, 21)
END_DATE = date.today() - timedelta(days=1)
SLEEP_BETWEEN = 20 # 20s entre requisições = 3 req/min, bem abaixo do limite
conn = psycopg2.connect(host=DB_HOST, port=DB_PORT, dbname=DB_NAME, user=DB_USER, password=DB_PASSWORD)
conn.autocommit = False
token = get_access_token(WIFEED_CLIENT_ID, WIFEED_CLIENT_SECRET)
time.sleep(SLEEP_BETWEEN) # pausa já após o login
current = START_DATE
while current <= END_DATE:
print(f"Processando {current}...")
try:
raw, _ = extract_all_access(token, watermark_date=current)
users = [transform_wifeed(r) for r in raw if transform_wifeed(r)]
if users:
load(conn, users, []) # sessions vazio no backfill
conn.commit()
print(f" {len(users)} usuários inseridos")
else:
print(f" sem dados")
except Exception as e:
conn.rollback()
if "429" in str(e):
print(f" Rate limit — aguardando 90s...")
time.sleep(90)
else:
print(f" ERRO em {current}: {e}")
current += timedelta(days=1)
time.sleep(SLEEP_BETWEEN)
conn.close()
print("Backfill concluído!")