I figure out that this is a nice way to access objects directly by the
request URL, no need to use the .load extension, and I got the same output
as I expect in shell.
example:
#########################################################################
def wrapper():
div = DIV('wrapper',BR(),HR())
otherdiv = DIV('This is another div')
def function1():
div = DIV('function 1',BR(),HR())
return locals()
#### This code below should be encapsulated ####
from inspect import isfunction as __isfunction
if not request.args(0) in locals():
return locals()
elif __isfunction(locals().get(request.args(0))):
return locals().get(request.args(0),locals)()
else:
return locals().get(request.args(0),locals)
#########################################################################
When I call: http://127.0.0.1:8000/testapp/default/wrapper
I got normal web2py app returning the objects div and otherdiv (with
scaffold layout)
When I call: http://127.0.0.1:8000/testapp/default/wrapper/function1
I got function 1 div as expected
and when I call
http://127.0.0.1:8000/testapp/default/wrapper/div
I got the DIV object as HTML <div>wrapper<br /><hr /></div>
also for: http://127.0.0.1:8000/testapp/default/wrapper/otherdiv
<div>This is another div</div>
I found this usefull to return pure HTML callbacks, no need to use .load and
I got the same what web2py shows in shell with no aditional blank line.
Can I rely on that kind of behaviour, or this way of doing can change in the
future,? (perhaps for security reasons)
