On 3 Aug 2012, at 4:50 AM, Martin Weissenboeck wrote:
> Sorry, my tests were wrong. Please forget my last mail.
> I think I have mixed two versions on my computer. I have tested it again and
> now is_imperonating works as expected.
> Thank you!
That's good. I've been trying to reproduce it in a
Sorry, my tests were wrong. Please forget my last mail.
I think I have mixed two versions on my computer. I have tested it again
and now is_imperonating works as expected.
Thank you!
2012/8/3 Massimo Di Pierro
> Clearly there is a problem with that function. If auth is defined, I'd
> expect
>
>
Clearly there is a problem with that function. If auth is defined, I'd
expect
>>> a=dict()
>>> 'x' in a
False
>>> a=dict(x=1)
>>> 'x' in a
True
>>>
The problem I have is that it fails when the user is not logged in.
Perhaps this is the related to your problem?
I have a possible fix in trunk. P
Interesting results (Version 2.00.0 (2012-08-02 21:51:02) dev)
>From tools.py:
def is_impersonating(self):
return 'impersonator' in current.session.auth
The return value of is_impersonating is not False or True but None or the
whole current.session.auth as string.
A change in tools.p
I think this should be considered a bug. Than you check trunk?
On Sunday, 29 July 2012 05:37:51 UTC-5, weheh wrote:
>
> I haven't traced through all the code carefully, but is_impersonating()
> returns current.session.auth.impersonator, which is based on a cPickle. So
> you're not getting a boo
I haven't traced through all the code carefully, but is_impersonating()
returns current.session.auth.impersonator, which is based on a cPickle. So
you're not getting a boolean, as you might be led to expect from the name
of the function.
On Sunday, July 29, 2012 1:38:42 PM UTC+8, mweissen wro
6 matches
Mail list logo
