On 20/02/12 08:03, Frank Millman wrote:
I don't follow all the implications of this thread, but I thought I should
mention that in python3, unicode and str are identical. If you pass 'bytes'
into etree.fromstring, you get bytes back. If you pass 'str' into
etree.fromstring, you get a (unicode) string back.
In Python 3, I think we should have String = Unicode as that distinction
becomes useless (as you point out), and there's ByteArray for binary
matters anyway.
Here's my train of thought; bear with me:
The issue here is that Unicode type may produce a str instance when data
is all ascii (because of some quirky lxml behaviour). This definitely
breaks the contract, but I can't think of any negative side effects of
this behaviour. And as Dieter points out, str is sometimes easier to
work with.
I don't want to go back to the old all-unicode behaviour just for the
sake of preserving backwards compatibility (where String = Unicode also
for Python 2). But we could be breaking it to have little to no benefit
at all.
Considering all this, my decision is to separate String and Unicode for
Python 2, as;
Explicit is better than implicit.
If you disagree, speak now or forever hold your silence :))
Best,
Burak
_______________________________________________
Soap mailing list
[email protected]
http://mail.python.org/mailman/listinfo/soap
