PEP8 79 char max

[Joshua Landau]

On 29 July 2013 22:34, Steven D'Aprano <steve+comp.lang.python at pearwood.info
> wrote:

> On Mon, 29 Jul 2013 15:18:59 -0500, Ed Leafe wrote:
>
> > On Jul 29, 2013, at 3:08 PM, Joel Goldstick <joel.goldstick at gmail.com>
> > wrote:
> >> Not performance, but human readability
> >
> >       IMO, this isn't always the case. There are many lines of code
> that are
> >       broken up to meet the 79 character limit, and as a result become
> much
> >       less readable.
>
> Speaking of readability, what's with the indentation of your post? The
> leading tab plays havoc with my newsreader's word-wrapping.
>
> Breaking lines to fit in 79 characters should almost always be perfectly
> readable, if you break it at natural code units rather than at random
> places. E.g. I have a code snippet that looks like this:
>
> [....whatever...]
> else:
>     completer = completer.Completer(
>                     bindings=(r'"\C-xo": overwrite-mode',
>                               r'"\C-xd": dump-functions',
>                               )
>                             )
>
> I'm not entirely happy with the placement of the closing brackets, but by
> breaking the line at the arguments to Completer, and then putting one
> binding per line, I think it is perfectly readable. And much more
> readable than (say) this:
>
>
> else:
>     completer = completer.Completer(bindings=
>         (r'"\C-xo": overwrite-mode', r'"\C-xd": dump-functions',))
>

But less readable to me than:

    completer = completer.Completer(bindings=[r'"\C-xo": overwrite-mode',
r'"\C-xd": dump-functions'])

Personal preference.

As far as I can tell, that's pretty much the longest line I have in my
> personal code base, possibly excepting unit tests with long lists of
> data. I simply don't write deeply nested classes and functions unless I
> absolutely need to.


I'd go for:

    completer = completer.Completer(bindings=[
  r'"\C-xo": overwrite-mode',
  r'"\C-xd": dump-functions'
    ])

although possibly drop the "bindings=" if possible. "[]" is less ambiguous
a construct than "()"¹ and the "balance" of the code is better my way if
such ephemeral ideas as that count.

Anyway, the point I'm trying to make is that *line length is a personal
thing*. There are two rules:

1) Stick with what other people on the team are doing, if relevant
2) Don't be stupid

The rest is your choice. Some people like 80 character limits, but I've
consistently preferred "whatever you think" as a better rule.

¹ As in there are fewer possible uses, so it's quicker to know what you're
using it for
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20130729/5c7d5263/attachment.html>