求助：修复Python正则表达式跨行捕获BEGIN与PAUSE间内容的问题

Got it, let's sort out your regex issue step by step. The problem with your current code comes from two key mistakes: wrong flag usage and incorrect parameter order in re.findall().

What Was Wrong #

1. Incorrect parameter order: You passed re.MULTILINE as the second argument to re.findall(), but the correct order is re.findall(pattern, string, flags). So your flag wasn't being applied at all!
2. Wrong flag choice: re.MULTILINE only changes how ^ and $ work (matching start/end of each line instead of the whole string). It doesn't make the . character match newline characters, which is what you need for multiline capture.

The Fixed Solution #

Use the re.DOTALL (or its alias re.S) flag instead—this makes . match all characters including newlines. Here's the corrected code:

import re

# Your multiline input example
input_text = """START
BEGIN
PROGRAM STARTED
PROGRAM IN PROGRESS
PAUSE
STOP"""

# Correct regex with proper flags and parameter order
matches = re.findall('(?<=BEGIN)(.*?)(?=PAUSE)', input_text, re.DOTALL)

# Optional: Clean up whitespace around the matched content
clean_matches = [match.strip() for match in matches]
print(clean_matches)
# Output: ['PROGRAM STARTED\nPROGRAM IN PROGRESS']

Breakdown of the Fix #

• re.DOTALL: Ensures the .*? pattern spans across line breaks, capturing everything between BEGIN and PAUSE regardless of newlines.
• Non-greedy .*?: Makes sure we stop at the first occurrence of PAUSE instead of matching all the way to the last one (which would happen with greedy .*).
• Parameter order: We now pass the input string as the second argument, flags as the third—so the regex engine actually uses the DOTALL flag.

Optional Adjustment (Trim Whitespace) #

If you want to automatically ignore leading/trailing newlines or spaces between BEGIN and PAUSE, tweak the regex to account for whitespace:

matches = re.findall('(?<=BEGIN\s*)(.*?)(?=\s*PAUSE)', input_text, re.DOTALL)

This will skip any spaces or newlines right after BEGIN and right before PAUSE, giving you a cleaner match without needing to strip() later.

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