Suggested amendment to PEP 255 (fwd)

[Gordon McMillan]

Dr. David Mertz wrote: 

>One more non-vote for a new spelling of "generator."  Despite
>protestations to the contrary by the timbot, I see the very subtle and
>late-in-the-definition-body use of 'yield' as a train wreck waiting to
>happen.  Never mind newbies, I am certain that *I* would stumble over
>the difference between generators and (real) functions all the time.

And I'm pretty sure you're underestimating yourself.

*When* might you stumble over this?

In writing one? No, you have to understand "yield" and the iterator 
protocol before you write one, and spelling it "generator" neither helps 
nor hinders in that.

In figuring out what's going on in a traceback? Not for long (and not 
likely more than once), because the traceback will already tell you that 
you're in the next method of an iterator, not in a "plain" function call.

In random sampling of code, looking for interesting functions to call? I 
just did that on the standard lib, and came up with:
 copy_reg.pickle_complex
 imaplib.IMAP4._get_response
 random.weibullvariate

my-toolbox-is-immeasurably-enriched-<wink>-ly y'rs

- Gordon