如何使用Python按指定周期调度不同Selenium任务？

Absolutely! Python is more than capable of handling all these scheduling scenarios for your Selenium tasks. The most robust and user-friendly tool for this is APScheduler—a flexible library that supports cron-style scheduling, interval triggers, and date-based triggers, which covers every case you’ve mentioned.

First, let’s get set up: install APScheduler and make sure you have Selenium and your browser driver (like ChromeDriver) configured properly:

pip install apscheduler selenium

Below’s a complete example that implements all your scheduling needs, plus a sample Selenium task:

Sample Selenium Task Function #

Let’s start with a simple Selenium task (you can replace this with your actual workflow):

from selenium import webdriver
from selenium.webdriver.chrome.options import Options

def run_selenium_task(task_name):
    print(f"Running {task_name}...")
    # Configure Chrome to run headless (no browser window)
    chrome_options = Options()
    chrome_options.add_argument("--headless=new")
    chrome_options.add_argument("--disable-gpu")
    
    with webdriver.Chrome(options=chrome_options) as driver:
        driver.get("https://example.com")
        # Add your actual Selenium actions here (scraping, form filling, etc.)
        print(f"{task_name} completed successfully!")

Implementing Each Scheduler #

1. Task that runs every 2 weeks #

We can use an IntervalTrigger for straightforward fixed intervals. This will run the task every 14 days starting from when the scheduler is launched:

from apscheduler.schedulers.blocking import BlockingScheduler
from apscheduler.triggers.interval import IntervalTrigger

scheduler = BlockingScheduler(timezone="Your/Timezone")  # Replace with your timezone (e.g., "America/New_York")

# Every 2 weeks (14 days)
scheduler.add_job(
    run_selenium_task,
    IntervalTrigger(days=14),
    args=["Bi-weekly Selenium Task"],
    id="bi_weekly_task"
)

2. Task that runs at the end of the 1st and 3rd week of each month #

Assuming "end of the week" means Friday (adjust to your preferred weekday), we’ll use a CronTrigger with the week parameter to target specific weeks:

from apscheduler.triggers.cron import CronTrigger

# End of 1st and 3rd week (Friday at 5 PM, for example)
scheduler.add_job(
    run_selenium_task,
    CronTrigger(
        day_of_week="fri",
        week="1,3",
        hour=17,
        minute=0,
        timezone="Your/Timezone"
    ),
    args=["1st & 3rd Week End Task"],
    id="weekly_interval_task"
)

Note: The week parameter counts weeks from the start of the month. If you need to define weeks as calendar weeks (ISO), you might need a small custom check in your task, but this works for most use cases.

3. Task that runs on the 15th and last day of each month #

Again, using CronTrigger—the L value for day targets the last day of the month:

# 15th and last day of each month (e.g., 2 AM to avoid peak traffic)
scheduler.add_job(
    run_selenium_task,
    CronTrigger(
        day="15,L",
        hour=2,
        minute=0,
        timezone="Your/Timezone"
    ),
    args=["15th & Last Day Task"],
    id="monthly_dates_task"
)

Start the Scheduler #

Finally, start the scheduler to keep it running and executing tasks:

if __name__ == "__main__":
    try:
        scheduler.start()
    except KeyboardInterrupt:
        print("Scheduler stopped manually.")

Key Notes for Production #

• Timezone: Always specify your timezone to avoid scheduling errors (use IANA timezone strings like "Europe/London").
• Headless Mode: Using headless browser mode is essential for running tasks in a server environment without a GUI.
• Persistence: For long-running schedules, consider running the script as a background service (e.g., using systemd on Linux, or Task Scheduler on Windows) or with nohup to keep it running after you log out.
• Error Handling: Add try/except blocks inside your Selenium task to catch and log errors (e.g., page load failures, element not found).

内容的提问来源于stack exchange，提问作者Dsab