Basic Python Regular expression: re.DOTALL *not* matching new lines

Question

I want to turn all occurrences of A...A into B...B for some filler between the two A's. The filler must be allowed to contain new line characters. I assumed re.DOTALL was the solution.

Here's a python script:

import re

tt1 = re.sub(r'A(?P<text>.*)A','B\g<text>B','AhiA')
print tt1 
tt1 = re.sub(r'A(?P<text>.*)A','B\g<text>B','A\nhiA')
print tt1 
tt1 = re.sub(r'A(?P<text>[.]*)A','B\g<text>B','A\nhiA')
print tt1 
tt1 = re.sub(r'A(?P<text>.*)A','B\g<text>B','A\nhiA',re.DOTALL)
print tt1 

And here's the output:

BhiB
A
hiA
A
hiA
A
hiA

What gives, and how can I replace 'A\nhiA' with 'B\nhiB'?

Answer 1

The fourth parameter to re.sub() is count (the maximum number of replace operations to be performed). re.DOTALL is 16, so you're passing a (valid) parameter in an unexpected place.

Use

re.sub(r'A(?P<text>.*)A','B\g<text>B','A\nhiA', flags=re.DOTALL)

(or place re.DOTALL in position five):

re.sub(r'A(?P<text>.*)A','B\g<text>B','A\nhiA', 0, re.DOTALL)