MS WINDOWS: input() swallows KeyboardInterrupt in Python 3.3

[Drekin]

BPO	17619
Nosy	@jcea, @vstinner, @tjguk
Files	input-ctrlc.patch

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = None
closed_at = <Date 2013-04-03.14:05:57.371>
created_at = <Date 2013-04-02.16:35:51.491>
labels = ['type-bug']
title = 'MS WINDOWS: input() swallows KeyboardInterrupt in Python 3.3'
updated_at = <Date 2013-04-03.14:05:57.371>
user = 'https://bugs.python.org/Drekin'

bugs.python.org fields:

activity = <Date 2013-04-03.14:05:57.371>
actor = 'sbt'
assignee = 'none'
closed = True
closed_date = <Date 2013-04-03.14:05:57.371>
closer = 'sbt'
components = []
creation = <Date 2013-04-02.16:35:51.491>
creator = 'Drekin'
dependencies = []
files = ['29666']
hgrepos = []
issue_num = 17619
keywords = ['patch']
message_count = 13.0
messages = ['185845', '185873', '185877', '185878', '185899', '185900', '185903', '185906', '185908', '185912', '185918', '185920', '185921']
nosy_count = 6.0
nosy_names = ['jcea', 'vstinner', 'tim.golden', 'python-dev', 'sbt', 'Drekin']
pr_nums = []
priority = 'normal'
resolution = 'fixed'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue17619'
versions = ['Python 3.3']

[Drekin]

At least on Windows, input() doesn't raise KeyboardInterrupt when Ctrl-C is hit in Python 3.3 (it does in Python 3.2).

3.3:
>>> x = input() # Ctrl-C hit - no exception raised
>>> x
NameError

3.2:
>>> x = input() # Ctrl-C hit
KeyboardInterrupt

[jcea]

Seems to be windows specific. It works fine on Linux and Solaris.

[vstinner]

It may be related to the following commit:

changeset: 77861:9523c122d6fc
user: Tim Golden <mail@timgolden.me.uk>
date: Fri Jun 29 18:39:26 2012 +0100
files: Parser/myreadline.c
description:
Issue bpo-1677: Handle better a race condition between the interactive interpreter and
the Ctrl-C signal handler on Windows

[vstinner]

@Drekin: What is your Windows version? How did you start Python?

[Drekin]

I have Windows Vista Business SP2, 64-bit, (6.0.6002). If I start Python any way (from console, from shortcut, via py.exe or not) the same behavior occurs.

[Drekin]

However it raises the exception correctly in IDLE.

The same behavior occured on Windows 7 Home Premium SP1, 64-bit, (6.1.7601).

[tjguk]

That's because IDLE uses a completely different input loop from the
console interpreter.

I'll try to get to this but I'm chock-a-block with other work at the
moment. If anyone else wants to dig, please do so. if the worst came to
the worst we could back out my previous changeset but that would still
leave us with an intermittent race-condition.

[Drekin]

There is related weird behavior:

>>> try:
...     input()
... except KeyboardInterrupt:
...     print("exception occured")
...
# Ctrl-C hit
Traceback (most recent call last):
  File "<stdin>", line 2, in <module>
KeyboardInterrupt

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "<stdin>", line 2, in <module>
KeyboardInterrupt
>>>

Seems like handling of Ctrl-C is kind of “out of sync”.

[sbt]

Maybe this is related to

http://bugs.python.org/issue13673

which causes PyTraceback_Print() to fail if a signal is received but PyErr_CheckSignals() has not been called.

Note that wrapping in "try: ... except: raise" makes a traceback appear:

>>> try: input()
... except: raise
...
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
KeyboardInterrupt

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
KeyboardInterrupt

[sbt]

Adding PyErr_CheckSignals() after PyOS_Readline() in builtin_input() seems to fix the problem.

[tjguk]

+1

Richard - are you in a position to commit / push?

[python-dev]

New changeset 241cd716bf5f by Richard Oudkerk in branch '3.3':
Issue bpo-17619: Make input() check for Ctrl-C correctly on Windows.
http://hg.python.org/cpython/rev/241cd716bf5f

[sbt]

Richard - are you in a position to commit / push?

Done.