PEP scepticism

phil hunt philh at comuno.freeserve.co.uk
Sat Jun 30 01:29:48 CEST 2001


On Fri, 29 Jun 2001 17:43:34 -0400, Tim Peters <tim.one at home.com> wrote:
>
>> It seems to me that optional-pass is obvious in its meaning,
>
>Sorry, I have no idea what it means.  Like you want to be able to use "pass"
>in more places, but optionally?  Or want not to have to use pass in some
>places where you must use it now?  Or ...?

It means where now you have to say:

class myClass:
   pass

or

def myfun():
   pass

when you create an initially-empty construct, you will in
future be able to say:

class MyClass:
def myfun():

and the pass keywords will be implied by the compiler. Perhaps
implied-pass is a better name?

>> but multi-liner comments aren't. There are three issues:
>>
>> 1. what should be the character sequences to begin/end multi
>> line comments
>
>What do you have against the existing method?  That is,
>
>    """This is a multi-line comment.
>    this part of the same comment.
>    so is this."""
>
>It's truly a comment:  a statement consisting of just a string is thrown
>away by the compiler, unless it's in a docstring position (is the first
>non-comment statement in a module, def or class).

I don't like this, because it just doesn't feel right to me. A
Comment should look different, syntactically, from a string literal.
Also, I suspect that docstrings are a source of inefficiency (OK,
I admit that's not a real problem when you have a short python program
and 128M of RAM).

>Or the other existing method:
>
>    # This is a multi-line comment.
>    # Although it sure looks like a sequence of
>    # single-line comments <wink>.

This is truly bad, because it makes it harder to add new lines;
you have to type "# " repeatedly when adding new comment text, which
breaks the flow of thought -- at least it dores for me.

>There's really no point answering these questions on c.l.py:  that they've
>been *asked* means the proposals are already controversial, so need PEPs to
>lay out all sides of the case.  I don't know what "optional pass" means, but
>the multi-line comment business has come up many times, so it would be
>especially good to PEP that one, in order to lay it to rest (one way or
>another) for good.

I wasn't aware that you could just plonk a string literal around anywhere
and it get ignored, thus forming a comment. In the light of that,
my multi-line comment proposal may seem a bit pointless. So here's
what I will do: write a PEP for it if at least 2 people on this
newsgroup say they support the idea.

-- 
## Philip Hunt ## philh at comuno.freeserve.co.uk ##







More information about the Python-list mailing list