补全客户数据

import csv
import time

import requests

API = "https://handelsregister.ai/api/v1/fetch-organization"
HEADERS = {"x-api-key": "YOUR_API_KEY", "Accept": "application/json"}
PER_MINUTE = 60  # fetch-organization 的限制


def lookup(row):
    # 登记法院加登记号是唯一的；否则用名称和城市
    q = row["register"] or f'{row["name"]} {row["city"]}'
    headers = {**HEADERS, "Idempotency-Key": f'customer-{row["id"]}'}
    for attempt in range(5):
        response = requests.get(API, params={"q": q}, headers=headers, timeout=120)
        if response.status_code == 429:
            time.sleep(int(response.headers.get("Retry-After", "10")))
        elif response.status_code in (408, 500, 502, 503, 504):
            time.sleep(30 * (attempt + 1))
        else:
            break
    return response


spent = 0
with open("customers.csv", newline="") as source, open("customers_enriched.csv", "w", newline="") as target:
    reader = csv.DictReader(source)
    writer = csv.DictWriter(target, fieldnames=reader.fieldnames + ["entity_id", "register_court", "register_number"])
    writer.writeheader()
    for row in reader:
        response = lookup(row)
        if response.status_code == 402:
            print("积分不足，运行已停止")
            break
        data = response.json()
        spent += data.get("meta", {}).get("request_credit_cost", 0)
        if response.status_code == 200:
            registration = data.get("registration") or {}
            row["entity_id"] = data["entity_id"]
            row["register_court"] = registration.get("court")
            row["register_number"] = f'{registration.get("register_type")} {registration.get("register_number")}'
        writer.writerow(row)
        time.sleep(60 / PER_MINUTE)

print(f"已用：{spent} 积分")