In Python, any variable can be re-bound at will -- and modules don't let you define special methods such as an instance's __setattr__ to stop attribute re-binding. Easy solution (in Python 2.1 and up): use an instance as "module"...
| Python |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 | # Put in const.py...:
class _const:
class ConstError(TypeError): pass
def __setattr__(self,name,value):
if self.__dict__.has_key(name):
raise self.ConstError, "Can't rebind const(%s)"%name
self.__dict__[name]=value
import sys
sys.modules[__name__]=_const()
# that's all -- now any client-code can
import const
# and bind an attribute ONCE:
const.magic = 23
# but NOT re-bind it:
const.magic = 88 # raises const.ConstError
# you may also want to add the obvious __delattr__
|
Discussion
In Python 2.1 and up, no check is made any more to force entries in sys.modules to be actually module objects. You can install an instance object there and take advantage of its attribute-access special methods (e.g., as in this snippet, to prevent rebindings; probably more useful, to synthesize attributes on the fly in a __getattr__; whatever...), while still having client-code get the thing with "import whatever". You may also choose to see this as a more Pythonic 'singleton' pattern, or, rather, idiom:-) (but, regarding singleton, also see http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/66531).


Comments
good for typechecking too?
Don't get carried away, though. This is clever, but keep in mind that all the standard modules implement constants this way:
This is much clearer to me. (It runs faster, too.)
Jason Orendorff, What prevents someone from re-declaring your variables?
The fact that they are all-uppercase and that the programmer knows that means they are intended to be constants.
If the programmer is a moron, you're screwed anyway. I find all this tendency to try and create libraries/frameworks/utils that can not possibly be used incorrectly a little wrong-headed. Just follow conventions, document, and hope for the best.
except: Just make sure that you don't follow that philosophy when you are creating your end product. Your end-user should be able to crash the product. They should, instead, get warnings, if anything.
Good philosophy. This philosophy is the reason I like python. No need for restrictive private functions and stuff. It should be clear to people that this function is not supposed to be called, but if they REALLY want to, they can do it. And yes, this has nothing to do with the end user, which should of course be protected against h(im|er)self.
Error after running the module. After typing at prompt >>> : import const I get this error: Traceback (most recent call last): File "", line 1, in -toplevel- import const ImportError: No module named const
I also tried >>> import _const (because definition has _const (not const)) - but the same error.
So how does it work ?
THIS IS THE CODE that I RAN: class const: class ConstError(TypeError): pass def __setattr__(self,name,value): if self.__dict__.has_key(name): raise self.ConstError, "Can't rebind const(%s)"%name self.__dict__[name]=value import sys sys.modules[__name__]=_const()
-thanks-cs-
Did you create a file called const.py containing lines 2 through 9, inclusive? Did you place the file somewhere where the import would see it?
Sign in to comment