
在早期我们还需要不断地拼凑最大limits来获取交易数据

但是币安已经提供了下载链接，理论上我们只需要清洗一下数据就ok了，现在获取数据已经不用那么麻烦了,一分钟获取了一年的，效率还是很高的

![](images/%7BB6E79FE7-A100-4084-919B-60564478A7AC%7D.png)

```
import os
import zipfile
import pandas as pd
import requests

BASE_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))

def get_binance_kline(symbol, time_interval, date, file_path):
    """
    获取k线数据
    :param symbol:
    :param time_interval: 1s, 1m, 5m, 15m, 30m, 1h, 2h, 4h, 6h, 8h, 12h, 1d
    :param date: xxxx-xx 2021-01
    :param file_path:
    :return:
    """
    url = f"https://data.binance.vision/data/spot/monthly/klines/{symbol.upper()}/{time_interval}/{symbol.upper()}-{time_interval}-{date}.zip"
    response = requests.get(url)

    # 检查文件夹是否存在
    if not os.path.exists(file_path):
        os.makedirs(file_path)

    with open(f"{file_path}/{symbol.upper()}-{time_interval}-{date}.zip", "wb") as f:
        f.write(response.content)


def unzip_file(zip_src, dst_dir):
    """
    解压缩文件
    :param zip_src: 压缩文件路径
    :param dst_dir: 解压缩路径
    :return:
    """
    r = zipfile.is_zipfile(zip_src)
    if r:
        fz = zipfile.ZipFile(zip_src, 'r')
        for file in fz.namelist():
            fz.extract(file, dst_dir)
    else:
        print('This is not zip')


# 处理数据
def clear_data(timestamp):
    for file in os.listdir(f"{BASE_DIR}/data/binance/{timestamp}"):
        if file.endswith(".zip"):
            unzip_file(f"{BASE_DIR}/data/binance/{timestamp}/{file}", f"{BASE_DIR}/data/binance/{timestamp}")
            os.remove(f"{BASE_DIR}/data/binance/{timestamp}/{file}")

    # 合并csv
    for file in os.listdir(f"{BASE_DIR}/data/binance/{timestamp}"):
        if file.endswith(".csv"):
            os.system(f"cat {BASE_DIR}/data/binance/{timestamp}/{file} >> {BASE_DIR}/data/binance/{timestamp}/all.csv")
            os.remove(f"{BASE_DIR}/data/binance/{timestamp}/{file}")

    df = pd.read_csv(f"{BASE_DIR}/data/binance/{timestamp}/all.csv", names=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11],
                     header=None)
    if df.empty:
        pass
    df.rename(columns={0: 'MTS', 1: 'open', 2: 'high', 3: 'low',
                       4: 'close', 5: 'volume'}, inplace=True)
    df['candle_begin_time'] = pd.to_datetime(df['MTS'], unit='ms')
    df = df[['candle_begin_time', 'open', 'high', 'low', 'close', 'volume']].copy()
    # 去重、排序
    df.drop_duplicates(subset=['candle_begin_time'], keep='last', inplace=True)
    df.sort_values('candle_begin_time', inplace=True)
    df.reset_index(drop=True, inplace=True)

    df.to_csv(f"{BASE_DIR}/data/binance/{timestamp}/all.csv", index=False)
    file_path = f"{BASE_DIR}/data/binance/{timestamp}/all.csv"

    return file_path

if __name__ == '__main__':
    symbol = "BTCUSDT"
    time_interval = "5m"
    start_time = "2021-01"
    end_time = "2022-02"

    # 获取当前时间戳做为文件夹名
    current_timestamp = int(time.time())

    # 遍历月份
    start_year = int(start_time.split("-")[0])
    start_month = int(start_time.split("-")[1])
    end_year = int(end_time.split("-")[0])
    end_month = int(end_time.split("-")[1])
    for year in range(start_year, end_year + 1):
        for month in range(1, 13):
            if year == start_year and month < start_month:
                continue
            if year == end_year and month > end_month:
                continue
            date = f"{year}-{month:02d}"
            logger.info(f"symbol:{symbol},date: {date}")
            get_binance_kline(symbol, time_interval, date, f"{BASE_DIR}/data/binance/{current_timestamp}")

    # 清洗数据
    file_path = clear_data(current_timestamp)

    return flie_path
```
