[Python-Dev] Re: [Patches] [Patch #101175] Fix slight bug in the Ref manual docs on listcomprehensions

[Greg Ewing]

> Guido hates it *because* it's almost certainly an error.

Yes, I know what you meant. I was just trying to point out
that, as far as I can see, it's only Guido's *opinion* that
it's almost certainly an error.

Let n1 be the number of times that [x if y] appears in some
program and the programmer actually meant to write something
else. Let n2 be the number of times [x if y] appears and
the programmer really meant it.

Now, I agree that n1+n2 will probably be a very small number.
But from that alone it doesn't follow that a given instance
of [x if y] is probably an error. That is only true if
n1 is much greater than n2, and in the absence of any
experience, there's no reason to believe that.

> A zip with no arguments has no universe to zip over; a listcomp without
> iterators has no universe to build on... The *intent* here is to
> supply a flexible and highly expressive way to build lists out of other
> sequences; no other sequences, use something else.

That's a reasonable argument. It might even convince me if
I think about it some more. I'll think about it some more.

> if you choose not to do the work anymore, you took yourself out of the
> loop.

You're absolutely right. I'll shut up now.

(By the way, I think your mail must have gone astray, Tim --
I don't recall ever being offered ownership of a PEP, whatever
that might entail.)

Greg Ewing, Computer Science Dept, +--------------------------------------+
University of Canterbury,	   | A citizen of NewZealandCorp, a	  |
Christchurch, New Zealand	   | wholly-owned subsidiary of USA Inc.  |
greg@cosc.canterbury.ac.nz	   +--------------------------------------+