[Python-Dev] Changes to PEP 327: Decimal data type (original) (raw)
Michael Chermside mcherm at mcherm.com
Thu Mar 18 12:47:05 EST 2004
- Previous message: [Python-Dev] A proposal has surfaced on comp.lang.python toredefine "is"
- Next message: [Python-Dev] (class) module names clarification
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
No sooner do I suggest that something is impossible:
Michael Chermside writes:
Unfortunately, there is a flaw in your methodology, and I can't think of a reasonable way to correct for it.
Then someone here goes and DOES it:
Edward Loper writes:
I ran a similar experiment, but counted the number of modules that define each type of function (since presumably most modules are internally consistent).
Good idea... I'd been considering and rejecting various dictionary based approaches for identifying multi-word identifiers, but your solution is better.
Total modules defining functions: 985 underscore+lowercase : 44% mixedCase+lowercase : 22% lowercase only : 16% InitialCaps : 10% mixedCase only : 2% underscore only : 2%
So underscore is most common; but mixedCase has a definite presence. [...] p.s., I'm definitely +1 on making a stronger statement in the style guide. Consistency is good.
Yep, +1 from me too. And if we strengthen the statement in PEP 8, then all NEW library code will be consistant (beware the wrath of the PEP 8 Police!). More importantly, people engaged in green-field Python projects will be much more likely to follow the practice. To make a good start, I hereby state my intention to stop using mixedCase and switch to under_scores instead.
-- Michael Chermside
- Previous message: [Python-Dev] A proposal has surfaced on comp.lang.python toredefine "is"
- Next message: [Python-Dev] (class) module names clarification
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]