Damnation!

Moshe Zadka moshez at math.huji.ac.il
Sun May 21 06:11:33 CEST 2000


On Sat, 20 May 2000, Hamish Lawson wrote:

> The issue is essentially about enforcing a naming rule that
> disallows names that differ only in case. My idea would be that
> namespaces would still be fundamentally case-sensitive, but you
> could have the interpreter additionally enforce the above naming
> rule. Whether or not this enforcement is the default behaviour,
> it can always be turned off.

Note: this is my last post on the subject.
No, that won't do at all: half the modules in the world will still not be
tested with this switch, so you'll still have to run without it a lot of
the time. Once you begin running without it, you'll miss the mistakes in
your own modules, and turning it back on would mean to fix 10-20 modules
before you can continue debugging. It isn't similar to -t and -tt, since
once there is a .pyc, these have no effect, while the naming switch will
still be needed.

It's much better to say "the language is case-insensitive" and to have the
tools correct case as you write.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com





More information about the Python-list mailing list