基于Pandas数据帧匹配值并迭代触发Outlook邮件发送

I see where your code is getting stuck—let's fix it to properly match keywords in your CSV, extract the right row data, and trigger emails correctly. Here's the revised code that implements your required logic:

import win32com.client as win32
import psutil
import os
import subprocess
import pandas as pd
import time

# Read CSV once (no need to read twice!)
filename = "users\\cmbp\\p4e\\test_rad_reports.csv"
df = pd.read_csv(filename, usecols=['reason', 'notes', 'email'])

def send_notification(recipient_email, subject, body):
    """Send email via Outlook with provided details"""
    outlook = win32.Dispatch('outlook.application')
    mail = outlook.CreateItem(0)
    mail.To = recipient_email
    mail.Subject = subject
    mail.Body = body
    mail.Send()  # Fixed syntax and removed unnecessary loop

def open_outlook():
    """Launch Outlook if it's not running"""
    try:
        # Use subprocess.call to avoid duplicate launches
        subprocess.call(['C:\\Program Files\\Microsoft Office\\Office16\\Outlook.exe'])
    except Exception as e:
        print(f"Outlook didn't open successfully: {str(e)}")

def check_outlook():
    """Check if Outlook is running; launch if not, then return Outlook instance"""
    outlook_running = False
    for item in psutil.pids():
        try:
            p = psutil.Process(item)
            if p.name().upper() == "OUTLOOK.EXE":
                outlook_running = True
                break
        except (psutil.NoSuchProcess, psutil.AccessDenied):
            continue  # Skip processes we can't access
    
    if not outlook_running:
        open_outlook()
        # Give Outlook a moment to launch before sending emails
        time.sleep(2)

def process_csv_and_send_emails():
    """Iterate through DataFrame, match keywords, and send notifications"""
    # Filter rows where 'reason' column contains 'reasona' (adjust match rules as needed)
    matching_rows = df[df['reason'].str.contains('reasona', case=False, na=False)]
    
    for _, row in matching_rows.iterrows():
        # Extract values from the matched row into variables
        recipient = row['email']
        reason_text = row['reason']
        notes_text = row['notes']
        
        # Customize email content
        email_subject = f"Notification for Reason: {reason_text}"
        email_body = f"Reason: {reason_text}\nNotes: {notes_text}"
        
        # Ensure Outlook is running before sending
        check_outlook()
        # Trigger email sending with extracted data
        send_notification(recipient, email_subject, email_body)
        print(f"Email sent to {recipient} successfully!")

# Run the main process
if __name__ == "__main__":
    process_csv_and_send_emails()

Key Improvements & Explanations: #

• Single CSV Read: We read the CSV once into df instead of twice—cleaner and more efficient.
• Proper Row Matching: Instead of looping over column names, we use df['reason'].str.contains() to filter rows with your target keyword. The case=False flag makes the match case-insensitive (remove it if you need exact case matching).
• Parameterized Email Function: send_notification now accepts recipient, subject, and body as parameters, avoiding messy global variable dependencies and making the function reusable.
• Fixed Outlook Check Logic: We initialize outlook_running upfront and handle edge cases like inaccessible processes. A short delay after launching Outlook ensures it's ready to send emails.
• Clear Function Separation: Each function has a single job—processing the CSV, managing Outlook state, or sending emails—making the code easier to debug and modify.
• Error Handling: Added basic exception catching to report issues with Outlook launches or process checks.

Quick Customization Tips: #

1. Adjust the keyword match in df['reason'].str.contains() to your exact target (e.g., use == 'reasona' for an exact match).
2. Modify the email_subject and email_body lines to fit your desired notification format.
3. Double-check the Outlook executable path in open_outlook() if your Office version uses a different directory.

内容的提问来源于stack exchange，提问作者Philalethes