如何在Python中提取文本特定内容？附日志文本处理需求

Here's a practical approach to pull out the notes from your log entries and save them to a new file. I'll cover two methods—one using basic string manipulation (great if your logs have a rock-solid consistent format) and another with regex (more flexible for slight variations in log structure).

Method 1: Basic String Splitting (Consistent Log Structure) #

If your logs always follow the exact pattern you shared (where notes end right before the username starting with u), this simple split method gets the job done quickly:

# Open input log file and output file in one go (cleaner resource handling)
with open('input_logs.txt', 'r') as input_file, open('extracted_notes.txt', 'w') as output_file:
    for line in input_file:
        # Skip empty lines to avoid clutter in output
        if not line.strip():
            continue
        # Check if the line has the "Notes: " section we need
        if "Notes: " in line:
            # Split the line to get everything after "Notes: "
            notes_section = line.split("Notes: ")[1]
            # Split again at the start of the username (assuming it starts with 'u')
            # This isolates the notes text from the trailing username/date/time
            notes_text = notes_section.split(" u")[0].strip()
            # Write the cleaned note to the output file
            output_file.write(f"{notes_text}\n")

Method 2: Regular Expression (Flexible Log Format) #

If your logs might have minor variations (like different username prefixes or slight date formatting shifts), regex is more robust. This pattern targets the notes content specifically, ignoring the trailing metadata:

import re

# Regex pattern to capture notes between "Notes: " and the trailing username/date/time block
note_pattern = re.compile(r'Notes: (.*?)\s+\w+\s+\d{2}\s+\w{3}\s+\d{4}\s+\d{2}:\d{2}:\d{2}')

with open('input_logs.txt', 'r') as input_file, open('extracted_notes.txt', 'w') as output_file:
    for line in input_file:
        match = note_pattern.search(line)
        if match:
            # Grab the captured notes text and trim any extra whitespace
            notes_text = match.group(1).strip()
            output_file.write(f"{notes_text}\n")

Quick Breakdown of the Regex Pattern: #

• Notes: : Matches the exact phrase to kick off our capture
• (.*?): Non-greedy capture of all characters until we hit the next metadata block
• \s+\w+: Catches the username (a word with no spaces)
• \d{2}\s+\w{3}\s+\d{4}: Matches dates like 21 Mar 2018
• \d{2}:\d{2}:\d{2}: Matches timestamps like 10:38:46

Testing with Your Sample Logs: #

For your example lines:

This offense was closed with reason: Non-Issue. Notes: There is no suspicious situation. u0t9231 21 Mar 2018 10:38:46
This offense was closed with reason: Non-Issue. Notes: Expected traffic. u0t9231 21 Mar 2018 14:20:11

Both methods will output these lines to extracted_notes.txt:

There is no suspicious situation.
Expected traffic.

Just swap out input_logs.txt with your actual log file path, and extracted_notes.txt with your preferred output file name.

内容的提问来源于stack exchange，提问作者Kirchoff