What were the thoughts behind replacement_imp.py
1 view
Skip to first unread message
Mark Jones
Jul 22, 2009, 12:19:27 AM7/22/09
to app-engine-django
This code is breaking other things, like wingdbstub.py for debugging.
I think it may also be responsible for breaking the Debug Probe as
well. Does it have to be this do nothing? Why is it written this
way?
<pre>
"""This file acts as a very minimal replacement for the 'imp' module.
It contains only what Django expects to use and does not actually
implement the
same functionality as the real 'imp' module.
"""
def find_module(name, path=None):
"""Django needs imp.find_module, but it works fine if nothing is
found."""
raise ImportError
</pre>
I think it may also be responsible for breaking the Debug Probe as
well. Does it have to be this do nothing? Why is it written this
way?
<pre>
"""This file acts as a very minimal replacement for the 'imp' module.
It contains only what Django expects to use and does not actually
implement the
same functionality as the real 'imp' module.
"""
def find_module(name, path=None):
"""Django needs imp.find_module, but it works fine if nothing is
found."""
raise ImportError
</pre>
Matt Brown
Jul 22, 2009, 3:59:30 AM7/22/09
to app-engi...@googlegroups.com
On Wed, Jul 22, 2009 at 6:19 AM, Mark Jones<mark...@gmail.com> wrote:
>
> This code is breaking other things, like wingdbstub.py for debugging.
> I think it may also be responsible for breaking the Debug Probe as
> well. Does it have to be this do nothing? Why is it written this
> way?
>
> This code is breaking other things, like wingdbstub.py for debugging.
> I think it may also be responsible for breaking the Debug Probe as
> well. Does it have to be this do nothing? Why is it written this
> way?
No, it can do more if it needs to.
It's written the way it is at the moment because I couldn't find
anything in the bits of Django that I was trying to make work that
actually needed to use imp.
If you'd like to contribute a more functional implementation it would
be much appreciated I'm sure.
Cheers
--
Matt Brown
Mark Jones
Jul 29, 2009, 10:36:52 PM7/29/09
to app-engine-django
I've found that app-engine-patch seems to be a fuller implementation
of the Django backend. Also, they have some nice ideas that should be
considered for stock Django regarding auto url configuration.
The one thing they seem to have problems with is ancestor/parent
entity groups.
On Jul 22, 2:59 am, Matt Brown <mattbrown...@gmail.com> wrote:
of the Django backend. Also, they have some nice ideas that should be
considered for stock Django regarding auto url configuration.
The one thing they seem to have problems with is ancestor/parent
entity groups.
On Jul 22, 2:59 am, Matt Brown <mattbrown...@gmail.com> wrote:
Matt Brown
Aug 4, 2009, 7:05:55 AM8/4/09
to app-engi...@googlegroups.com
On Thu, Jul 30, 2009 at 3:36 AM, Mark Jones<mark...@gmail.com> wrote:
>
> I've found that app-engine-patch seems to be a fuller implementation
> of the Django backend. Also, they have some nice ideas that should be
> considered for stock Django regarding auto url configuration.
>
> The one thing they seem to have problems with is ancestor/parent
> entity groups.
>
> I've found that app-engine-patch seems to be a fuller implementation
> of the Django backend. Also, they have some nice ideas that should be
> considered for stock Django regarding auto url configuration.
>
> The one thing they seem to have problems with is ancestor/parent
> entity groups.
Waldemar and the others behind app-engine-patch have certainly done
some excellent work in replacing more of the Django backends.
It's unlikely that those changes will be merged into
google-app-engine-django unless one of the app-engine-patch developers
wants to submit them to us for inclusion.
Your best option at this point may just be to use app-engine-patch directly.
Cheers
--
Matt Brown
ma...@mattb.net.nz
Mob +353 86 608 7117 www.mattb.net.nz
Reply all
Reply to author
Forward
0 new messages