Bug report

Bug description:

Issue

email.quoprimime.decode() strips only one character when removing a trailing line ending that wasn't in the original input. When eol='\r\n' (2 characters), only the \n is removed, leaving a stray \r.

The bug is in this line:

decoded = decoded[:-1]

It should be decoded[:-len(eol)].

Reproducer

import email.quoprimime as qp
print(repr(qp.decode('abc', eol='\r\n')))  # 'abc\r' — expected 'abc'
print(repr(qp.decode('a\nb', eol='\r\n')))  # 'a\r\nb\r' — trailing \r is spurious

CPython versions tested on:

CPython main branch

Operating systems tested on:

No response

Linked PRs

• gh-145831: email.quoprimime: decode() leaves stray \r when eol='\r\n' #145832
• [3.13] gh-145831: email.quoprimime: decode() leaves stray \r when eol='\r\n' (GH-145832) #148311
• [3.14] gh-145831: email.quoprimime: decode() leaves stray \r when eol='\r\n' (GH-145832) #148312