Adding a new statement to Python

[motter]

http://eli.thegreenplace.net/2010/06/30/python-internals-adding-a-new-statement-to-python/

[ghc]

I would have liked to see how easy it is to add a new statement to pypy for comparison's sake.

[kingkilr → ghc]

It's going to have all these same steps, but they'll mostly be in python. I'll try to write a tutorial on this at some point.

[perfunctory]

Very useful post I think. Not to depreciate Python the language, but it might be interesting to compare "adding a new statement to Python" with "adding a new statement to Lisp". Especially for people struggling to grasp Lisp's benefits.

[gaius → perfunctory]

Or Forth.

http://www.yosefk.com/blog/my-history-with-forth-stack-machi...

my mind was immediately blown away by the following passage right at the top of system.fth, the part of pForth implemented in Forth on top of the C interpreter:

    : (   41 word drop ; immediate
    ( That was the definition for the comment word. )
    ( Now we can add comments to what we are doing! )

[Kototama → perfunctory]

Here, adding a 'unless' statement:

    (defmacro unless [expr & body] `(if ~expr nil (do ~@body)))

Taken from: http://stackoverflow.com/questions/2977882/defining-clojure-...

Macro in Lisp are essentially functions taking code as argument and returning code, but they are executed at compile-time.

Edit: replaced with the correct version, sorry for the error.

[roryokane → Kototama]

According to that question's answer and comments, that definition is buggy (making it not quite a fair comparison). Here is a less buggy, more flexible version, taken from a comment:

  (defmacro unless [expr & body] `(if ~expr nil (do ~@body)))

[davidw → perfunctory]

Or adding a new statement to Tcl:

http://wiki.tcl.tk/917

[gaius → davidw]

Or BASIC ;-) http://en.wikipedia.org/wiki/DOS_Wedge

On a more serious note, the ability to add constructs to the language is not nearly as unique to LISP as Lispers think it is.

[draegtun → perfunctory]

Or Io:

  unless := method(call evalArgAt(0) ifFalse(call evalArgAt(1)))

Usage example...

  unless(1 == 2, writeln("Tis false"))

[motter]

I was reminded of this by the post on the goto statement -- having used the AST module recently in a limited capacity[1], I was really impressed by how easy it was to use. There's a good pycon talk for those interested specifically in this part of the process[2], and you can also see the full grammar[3].

[1] http://blueprintforge.com/blog/2012/02/27/static-modificatio...

[2] http://blip.tv/pycon-us-videos-2009-2010-2011/pycon-2011-wha...

[3] http://docs.python.org/library/ast.html#abstract-grammar

[thomaslee]

Here's my original paper on the topic in case folks are interested:

http://tomlee.co/wp-content/uploads/2008/12/python-language-...

[drstrangevibes]

How well do you know the Sutras?

Special cases aren't special enough to break the rules. ... There should be one-- and preferably only one --obvious way to do it.

[eliben → drstrangevibes]

Did you even try to read the linked article :) ?

Here's the second section:

A language-advocacy digression

------------------------------

This article doesn’t attempt to suggest the addition of an until statement to Python. Although I think such a statement would make some code clearer, and this article displays how easy it is to add, I completely respect Python’s philosophy of minimalism. All I’m trying to do here, really, is gain some insight into the inner workings of Python.

[drstrangevibes → eliben]

point taken