python file automatic running after saving in vscode

Question

This is a very weried situation, in vscode, I edited my code in the root directory and saved it as test_distribution.py. After I run this file in vscode terminal. Everytime I edit and save this file, it gets run automatically.

So I moved this file to the subfolder of this project, but I found out it still get run when I edit and save this file. So I changed the file name to distribution_test.py and add logging and save this file again. It turns out it will still run when I edit and save this file.

How should I figure out why this is happening? What kind of info should I log out in order to figure out this issue?

As you can see the 1300000_bake.xlsx is generated by this python file, I even moved this python file to a different folder and changed the name of it as I stated above, I am very curious how to find out which program is running this

Pics as follow:

for store_id in store_ids:
df = pd.DataFrame(columns=columns)
for query_date in date_range:
    logging.info(f"提高计算速度, 使用常规kind_code_cate1类型")
    kind_cate1s = normal_cate1
    for kind_cate1 in kind_cate1s:
        for setting in [dev_setting, prod_setting]:
            eng = conn(setting)
            delivery_cycle = pd.read_sql(
                f"select delivery_cycle from dim_mty_goods where kind_code_cate1={kind_cate1} and can_be_ordered=1",
                eng,
            )
            delivery_cycle = np.unique(delivery_cycle["delivery_cycle"].values)
            delivery_cycle = min(delivery_cycle)  # 只有冷藏存在多个值, 1, 2, 3 取1, 其他均为单一值
            start_time = time.time()
            logging.info(
                f"开始计算, store_id: {store_id}, date: {query_date}, kind_cate1: {kind_cate1}, dev 环境: {setting == dev_setting}"
            )
            logging.info(
                f"bash command: {create_bash(query_date, kind_cate1, store_id, setting == dev_setting)}"
            )
            status = os.system(
                create_bash(
                    query_date=query_date,
                    kind_cate1=kind_cate1,
                    store_id=store_id,
                    dev_setting=(setting == dev_setting),
                )
            )
            elapsed = time.time() - start_time
            if status != 0:
                logging.info(
                    f'计算出现问题, store_id: {store_id}, date: {query_date}, kind code cate1: {kind_cate1}, 环境: {setting["db"]}'
                )
                break
            logging.info(f"返回耗时: {elapsed}")
            if elapsed < 60:
                calculate_time = elapsed
            elif setting == dev_setting:
                calculate_time = pd.DataFrame()
                while calculate_time.empty:
                    calculate_time = pd.read_sql(
                        f'select cal_time from calculate_time where dt = "{str(query_date)}" and store_id={store_id} and kind_code_cate1={kind_cate1}',
                        eng,
                    )
                    # 等待写库
                    time.sleep(0.1)
                    logging.warning(
                        f"未查询到计算时间, store_id: {store_id}, date: {query_date}, kind_code_cate1: {kind_cate1}"
                    )
                calculate_time = min(calculate_time["cal_time"].values)
            else:
                calculate_time = elapsed
            logging.info(f"计算时间, {calculate_time}")
            arrival_sale_day = [
                "sale_predict_qtty_t1",
                "sale_predict_qtty_t2",
                "sale_predict_qtty_t3",
            ][delivery_cycle]
            dev_prod = int(setting["db"] == "marsboard_dev")
            df_temp = pd.DataFrame()
            while df_temp.empty:
                df_temp = pd.read_sql(
                    f'select store_id, dt, kind_code_cate1, {arrival_sale_day}, perdict_stock_qtty, predict_num, goods_id, goods_name from order_list where store_id={store_id} and kind_code_cate1={kind_cate1} and dt="{str(query_date)}"',
                    eng,
                )
                logging.warning(f"未查询到order list 里的信息")
                if pd.read_sql(
                    f'select store_id, dt, kind_code_cate1, {arrival_sale_day}, perdict_stock_qtty, predict_num, goods_id, goods_name from order_list where store_id={store_id} and kind_code_cate1={kind_cate1} and dt="{str(query_date)}"',
                    prod_eng,
                ).empty:
                    logging.info(f"生产环境本品类亦无数据, {kind_cate1}")
                    break
                time.sleep(0.1)
            df_temp["dev_prod"] = dev_prod
            df_temp["calculate_time"] = float(calculate_time)
            df_temp.columns = columns

            df = pd.concat([df_temp, df], axis=0, ignore_index=True)
            print(df)
df.to_excel(f"{store_id}_bake.xlsx", index=False)

Answer 1

Just to confirm, the root cause in my case was that I was importing the module in my test files using PyTest. It seems that with every new update, the module is imported again dynamically. So, adding things like previously suggested as

if __name__ == '__main':
   fun()

Will help. You can also add the following to settings.json of VSCODE:

"python.testing.autoTestDiscoverOnSaveEnabled": false

Answer 2

I have the same issue. It's now fixed. To do my test I used pytest and it was able to read my file when it started with test_. Because I'm stupid I named it a random filename test_func.py and in this one, I wrote "from main.py import *". So when I saved, and used pytest to run my test, my main file was run too.

To fix this I just renamed my file in teste-func.py and it works.

Sorry my English is terrible. I hope my comment is helpful.

Answer 3

I avoid this issue by adding

if __name__ == '__main':
   fun()

I guess this file will be import or load or something else, then it gets run. Anyway, this is not important.

Answer 4

Perhaps , You have mistakenly installed an extension 'Run on Save' . I think that's causing the Problem.