Handle keyboard interrupts in SSL module

[jacobs99]

BPO	751916
Nosy	@loewis
Files	ssl.diff: Patch to add interrupt checking and fixes to timeout code

^{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 = 'https://github.com/loewis'
closed_at = <Date 2003-06-28.07:40:47.000>
created_at = <Date 2003-06-10.13:36:26.000>
labels = ['extension-modules']
title = 'Handle keyboard interrupts in SSL module'
updated_at = <Date 2003-06-28.07:40:47.000>
user = 'https://bugs.python.org/jacobs99'

bugs.python.org fields:

activity = <Date 2003-06-28.07:40:47.000>
actor = 'loewis'
assignee = 'loewis'
closed = True
closed_date = None
closer = None
components = ['Extension Modules']
creation = <Date 2003-06-10.13:36:26.000>
creator = 'jacobs99'
dependencies = []
files = ['5355']
hgrepos = []
issue_num = 751916
keywords = ['patch']
message_count = 3.0
messages = ['43932', '43933', '43934']
nosy_count = 2.0
nosy_names = ['loewis', 'jacobs99']
pr_nums = []
priority = 'normal'
resolution = 'accepted'
stage = None
status = 'closed'
superseder = None
type = None
url = 'https://bugs.python.org/issue751916'
versions = ['Python 2.3']

[jacobs99]

As of Python 2.3 beta 1 (and current CVS), the
SSL module does not check to see if a keyboard
interrupt occurred after waiting on connections,
reads, and writes. This results in long
delays (seconds or even minutes) before the
KeyboardInterrupt is finally raised, causing
problematic pauses for interactive processes. This
does not happen in Python 2.2.2, though I have not
looked into what has changed.

My next step is to look into httplib, the use-case that
brought this to my attention, since some of the
buffering code is not exception safe. i.e., a
KeyboardInterrupt can cause data to be lost on
subsequent reads. I suspect that it is also possible
to lose data on due to socket timeouts as well, so I
consider this issue a distinct bug.

I did not include a new unit test, since it is a little
tricky to produce. It requires a large SSL transfer and
a network timing-sensitive threshold test to detect.

Also included in the attached patch are some error
handling fixes and cleanups related to time-out
operations that I happened to notice. This module
deserves more attention, since the timeout changes
don't look like they were integrated as carefully
as they could have been and other similar errors are
possible.

[loewis]

Logged In: YES
user_id=21627

The patch looks fine to me.

[loewis]

Logged In: YES
user_id=21627

Thanks for the patch. Applied as _ssl.c 1.12