On 12/23/2010 11:54 AM, John Fabiani wrote:
> On Thursday, December 23, 2010 12:39:44 am bruno desthuilliers wrote:
>> On 23 déc, 06:33, John Fabiani <jo...@jfcomputer.com> wrote:
>>
>> (snip)
>>
>> John, may I suggest that instead of trying whatever comes to mind and
>> wonder what happens, you spend some times learning Python, specially
>> the part about modules, import and the modules search path ? Your
>> problem - which has nothing to do with Django BTW - is obviously that
>> something in your PYTHONPATH shadows the stdlib modules. Launch an
>> interactive django / python shell ("./manage.py shell" command in your
>> terminal), then execute the following code:
>>
>> import sys
>> print "\n".join(sys.path)
>> import datetime
>> print datetime
>>
>> This should tell you where your modules are searched for by Python,
>> and where your datetime module is actually imported from.
>
> You can do more than suggest. I learn daily. And so today I learn where and
> how to determine where/which module is used for an import.
>
> But still your suggestion does not solve the issue at hand. Your code reports
> import datetime
>>>> print datetime
> <module 'datetime' from '/usr/lib/python2.6/lib-dynload/datetime.so'>
>
> I believe that is the correct datetime module which explains nothing! Most
> important in my mind is how did anything I did cause a difference!
>
> 1. I changed the way I import-ed datetime in my views.py. Just a simple name
> change on the theory that somewhere I had re-defined datetime. I also
> removed
> two un-used imports. At first I thought they might have been an issue but I
> could not see where they conflicted with anything and it did nothing to
> change
> the error.
>
> 2. Deleted the django models.pyc (note the pyc). On the theory that some how
> the file was corrupt.
>
> 3. Rebooted the Linux box. I hated to do this act. In the past I have
> always
> found ways around rebooting a linux box.
>
> But suddenly everything started working again!
>
> But let's assume you are correct I had added some sort of conflicting module.
>
> BTW that was my thought almost immediately! I changed the code causing the
> immediate error with the following (used only for testing):
>
> try:
> now = datetime.datetime.now()
> except:
> import datetime
> now = datetime.datetime.now()
>
> But suddenly I got a second error from the password method and again the
> error
> was 'None' has no Attribute .... The problem with that error was everything
> in the password method is contained in the same module. So if it did not
> require anything outside of the module how could it be 'None'. All these
> errors from the same module (models.py).
>
> All of this has me very concerned because I was about to deploy (after all
> the website has been tested for the last two weeks). I do believe this has
> something to do with my code (I believe there had to be some sort of conflict
> - but where).
>
> But as to your statement this has nothing to do with Django I can NOT agree.
> Like all programs Django should protect it's modules and do everything they
> can to insure the correct access to the correct imports. All frameworks need
> to protect their code. Yes it true it is hard to protect against someone
> willing to shoot them self's in the foot. But some effort is still required!
>
> Johnf
>
John:
I know it's disturbing when computers appear to behave in a
non-deterministic way. Perhaps when the system is failing you could copy
and paste the whole traceback?
Your original statement of the problem is far from clear. You said:
"""
The error first appeared as datetime has no Attribute 'None' for a line
in my
views.py
user = User.objects.create_user(c.registration_id, c.email,
request.POST['txtPassword'])
The error was from in django/contrib/auth/models.py
now = datetime.datetime.now()
"""
Presumably that line appeared in the traceback but was only the
*indirect* cause of the failure? You don't say *which* line number that
line is from the django.contrib.auth.models.py, let alone which function
or method was active when the exception occurred. While there are many
fine minds in this group you aren't really giving them much of a chance
to help you.
With more information someone might not only be able to diagnose the
error but also to help you avoid it.
regards
Steve
--
DjangoCon US 2011 Portland, OR: September 6-8 http://djangocon.us/
--
You received this message because you are subscribed to the Google Groups
"Django users" group.
To post to this group, send email to django-us...@googlegroups.com.
To unsubscribe from this group, send email to
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/django-users?hl=en.
