[Baypiggies] History of python naming

Guido van Rossum guido at python.org
Tue Feb 5 02:44:52 CET 2008

On Feb 4, 2008 5:24 PM, Shannon -jj Behrens <jjinux at gmail.com> wrote:
> On Feb 4, 2008 3:54 PM, Guido van Rossum <guido at python.org> wrote:
> > On Feb 4, 2008 3:46 PM, Jeff Younker <jeff at drinktomi.com> wrote:
> > > Does anyone have pointers to the history of python naming
> > > conventions?   I'm curious how the std library ended up with
> > > outliers like StringIO.StringIO.
> >
> > It was simply me experimenting with different ways of doing things,
> > way, way long ago. And perhaps not pushing back enough on alternative
> > naming conventions used by early contributors. PEP 8 came much, much
> > later.
> The killer for me was camelCaseMethod names.  I seem to remember them
> being allowed by the style guide seven years ago, and I have a bunch
> of legacy code using them.  I remember when I first read the style
> guide again and noticed that they were not allowed.  I was like,
> "Crap!  How am I supposed to change the public API of open source
> code?!?" *sigh*

You don't, of course; at least not until you have  amajor API design
on hand anyways. The PEP allows for historical deviations; it's mainly
meant to prevent more of such deviations to happen.

--Guido van Rossum (home page: http://www.python.org/~guido/)

More information about the Baypiggies mailing list