R. David Murray added the comment:
Kim: if you can prove it is broken, please open a new issue, this one is closed
(you can reference it by just typing # followed by the issue number, and the
tracker will automatically make a link).
--
___
Python
Kim Randell added the comment:
I don't have an example case, but from reading the code it looks as though the
AbstractDigestAuthHandler has exactly the same faults as the old version of
AbstractBasicAuthHandler (i.e. using req.headers.get instead of req.get_header,
and corresponding retry
Senthil Kumaran added the comment:
This is fixed in all active versions (2.7.8+, 3.4.2? and 3.5). Thanks all!
--
resolution: - fixed
stage: patch review - resolved
status: open - closed
___
Python tracker rep...@bugs.python.org
Roundup Robot added the comment:
New changeset c1edc4e43eb1 by Senthil Kumaran in branch '2.7':
Fix Issue #8797: Raise HTTPError on failed Basic Authentication immediately.
Initial patch by Sam Bull.
http://hg.python.org/cpython/rev/c1edc4e43eb1
New changeset 30e8a8f22a2a by Senthil Kumaran in
Roundup Robot added the comment:
New changeset e0510a3bdf8f by Senthil Kumaran in branch '2.7':
Fix Issue #8797: Raise HTTPError on failed Basic Authentication immediately.
Initial patch by Sam Bull.
http://hg.python.org/cpython/rev/e0510a3bdf8f
New changeset 3435c5865cfc by Senthil Kumaran in
STINNER Victor added the comment:
test_urllib2_localnet now fails since the changeset 3435c5865cfc. Example:
http://buildbot.python.org/all/builders/x86%20RHEL%206%203.x/builds/4410/steps/test/logs/stdio
Roundup Robot added the comment:
New changeset 5336a01542e2 by Senthil Kumaran in branch '2.7':
backout changeset e0510a3bdf8f - due to buildbot failures. Ref: Issue #8797
http://hg.python.org/cpython/rev/5336a01542e2
New changeset d51e739004bc by Senthil Kumaran in branch '3.4':
backout
Senthil Kumaran added the comment:
I have backed out my changes. The buildbot failures were new as I could not
reproduce them in my local test on Mac. I will run the buildbot suite on branch
and fix it before committing.
--
___
Python tracker
Serhiy Storchaka added the comment:
May be issue22165 patch fixes this test failure.
--
nosy: +serhiy.storchaka
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue8797
___
Senthil Kumaran added the comment:
@Serhiy, yeah, I think the patch attached to issue22165 will fix one of the
buildbot failures. For this issue. there was another one with Broken PIPE which
may need a different fix.
--
___
Python tracker
Senthil Kumaran added the comment:
This bug has been open for a while and I had lost sight of it. Upon prompted
recently, I dug bit
into history and could think of a good solution.
A brief history.
1. The bug maximum recursion depth exceeded when doing Basic Authentication
was introduced in
Ole Martin Handeland pythonb...@olemartin.org added the comment:
Thanks Sam! I can confirm that your patch is working well with basic auth. I
have not tried other authentication methods, though. I'm not sure about the
unredirected headers, but your fix looks like the cleanest.
--
Changes by Antoine Pitrou pit...@free.fr:
--
resolution: fixed -
stage: committed/rejected - patch review
status: closed - open
type: crash - behavior
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue8797
Sam Bull s...@pocketuniverse.ca added the comment:
I think there's a much simpler solution to this ticket than the retry logic
that's currently in place.
The code originally avoided the infinite recursion by checking to see if the
previous request had already submitted the auth credentials
Mads Kiilerich m...@kiilerich.com added the comment:
On 08/27/2010 03:47 AM, Senthil Kumaran wrote:
I agree with you respect to the other error codes, there is already
another bug open to handle this. The reset counter is reset on success
too, in another part of the code.
FWIW: From
R. David Murray rdmur...@bitdance.com added the comment:
See also new issue 9698.
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue8797
___
___
Senthil Kumaran orsent...@gmail.com added the comment:
I checked in a modified version of reset the retry count for respnse code !=401
in the following checkins:
r84323 (py3k)
r84324 (release21-maint)
r84325 (release31-maint)
Unfortunately, we wont be able to patch the 2.6.x release. You might
Changes by Senthil Kumaran orsent...@gmail.com:
--
resolution: - fixed
stage: patch review - committed/rejected
status: open - closed
Added file: http://bugs.python.org/file18648/basic_auth.patch
___
Python tracker rep...@bugs.python.org
r84324 (release21-maint)
That should be release27-maint. :)
___
Python-bugs-list mailing list
Unsubscribe:
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
Mads Kiilerich m...@kiilerich.com added the comment:
Senthil, can you tell us why this fix is correct - and convince us that it is
the Final Fix for this issue? Not because I don't trust you, but because this
issue has a bad track record.
Some comments/questions to this patch:
Why do 401
Senthil Kumaran orsent...@gmail.com added the comment:
On Thu, Aug 26, 2010 at 3:24 PM, Mads Kiilerich rep...@bugs.python.org wrote:
Mads Kiilerich m...@kiilerich.com added the comment:
Senthil, can you tell us why this fix is correct - and convince us that it is
the Final Fix for this
Dmitry Jemerov intelliy...@gmail.com added the comment:
I've also run into this problem after upgrading to Python 2.6.6. My code, which
uses the same HTTPBasicAuthHandler instance for many requests to the same
server, worked correctly with Python 2.6.2 and broke with 2.6.6. It would be
great
Changes by Éric Araujo mer...@netwok.org:
--
resolution: fixed -
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue8797
___
___
Python-bugs-list
Changes by Antoine Pitrou pit...@free.fr:
--
stage: committed/rejected - patch review
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue8797
___
___
zenyatta j...@journey.sk added the comment:
I work in sidux and my Mercurial currently doesn't work. The python version
already contains the fix for this issue (revision 81637) and it crashes
Mercurial (authorization failed) whenever a command involves more than 5
requests to the repository.
Mads Kiilerich m...@kiilerich.com added the comment:
zenyatta: Which Mercurial version? We thought we had implemented a sufficiently
ugly workaround in Mercurial. Please file an issue in
http://mercurial.selenic.com/bts/ .
--
___
Python tracker
zenyatta j...@journey.sk added the comment:
See http://mercurial.selenic.com/bts/issue2299
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue8797
___
Mads Kiilerich m...@kiilerich.com added the comment:
FYI:
The regression was introduced in 2.6.5 with issue3819 .
This also causes problems with Mercurial pushing to google code - see
http://mercurial.selenic.com/bts/issue2179 .
IIRC google code redirects to an url which then in turn
jurjen Bos j.bos-inter...@xs4all.nl added the comment:
I would like to point out that this is not going to work if someone visits more
than 5 sites with the same authentication manager. This would have to be
documentated, at least.
We could fix this by putting the retry counter in the
R. David Murray rdmur...@bitdance.com added the comment:
+1 for putting retry control on the password manager. Probably the default
should be set to 1. If I understand correctly, it is the password manager that
would be prompting the user for a new password, if someone chose to implement
Senthil Kumaran orsent...@gmail.com added the comment:
Agree to single retry for Basic Auth. We are just dealing with
BasicAuthentication here, so that's why we did not have in the
HTTPPasswdManager. But, let me consider that viewpoint too.
--
status: closed - open
Senthil Kumaran orsent...@gmail.com added the comment:
This is fixed in r81636, r81637, r81638, r81637.
Just like the Digest Authentication, the urllib2 will try for 5 times before
failing for any authentication failure ( It won't lead to recursion).
I had a thought that Basic Authentication
Ionut Turturica jono...@yahoo.com added the comment:
Why would you waste the time and resources to test 5 times for a known to be
wrong credential ?!
This is not like in a browser where the user is presented with a dialog box 5
times and he/she can try different credentials.
I don't see the
Senthil Kumaran orsent...@gmail.com added the comment:
The point of retry is for auth failure, which can happen due to any reason. It
is not just for a wrong password.
I was thinking just doing a single verification and failing, but did find some
resources (not just browsers) which adopt this
chuchiperriman chuchiperri...@gmail.com added the comment:
The same problem for me, it tries to authenticate infinite times if the
user/password are incorrect
--
nosy: +chuchiperriman
___
Python tracker rep...@bugs.python.org
Senthil Kumaran orsent...@gmail.com added the comment:
Is the behavior you posted observable on trunk?
--
assignee: - orsenthil
nosy: +orsenthil
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue8797
Senthil Kumaran orsent...@gmail.com added the comment:
Okay, I see you have mentioned 2.6.5 in the Title.
I will try the following to reproduce this.
1) Try a Basis Auth which is working.
2) Pass an Invalid username/password and see the behavior.
--
37 matches
Mail list logo
