Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #57946 > unrolled thread

Re: stacked decorators and consolidating

Back to article view | Back to comp.lang.python

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: stacked decorators and consolidating Tim Chase <python.list@tim.thechases.com> - 2013-10-29 12:54 -0500
    Re: stacked decorators and consolidating Gregory Ewing <greg.ewing@canterbury.ac.nz> - 2013-10-30 11:46 +1300

#57946 — Re: stacked decorators and consolidating

On 2013-10-29 17:42, MRAB wrote:
> If you apply the stacked decorators you get:
> 
>      myfun = dec1(args1)(dec2(args2)(dec3(args3)(myfun)))
> 
> If you apply dec_all you get:
> 
>      myfun = dec1(args1)(dec2(args2)(dec3(args3)))(myfun)
> 
> See the difference? You need the lambda to fix that.

In this case, they happen to be CherryPy decorators:

  @cherrypy.expose()
  @cherrypy.tools.json_in()
  @cherrypy.tools.json_out()
  def myfunc(...): pass

I'd have figured they would be associative, making the result end up
the same either way, but apparently not.  Thanks for helping shed
some light on the subtle difference.

-tkc

[toc] | [next] | [standalone]

#57978

Tim Chase wrote:
> I'd have figured they would be associative, making the result end up
> the same either way, but apparently not.

They're not associative because function application
is not associative: f(g(x)) is not the same thing as
f(g)(x).

-- 
Greg

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.python

csiph-web