Re: Encodings of string literals; explicit codepoint escapes?

On 8/31/07, Robin Stocker <[e-mail]> wrote:
> Note that in Python 3, the u"" notation will be gone and all strings
> will be unicode strings by default [1]. If what you want is a string of
> bytes, this is what the new bytes type is for. I like this approach and
> think this will help get rid of the confusion of having the same type
> for character strings and byte strings.
>
> [1] http://www.artima.com/weblogs/viewpost.jsp?thread=157004

I've argued in the past that this is an unnecessary distinction. I
still hold to this. All strings are bytes. Period. Characters are what
you get out of those bytes when you view them through a particular
"lens" (e.g., an encoding). It should be easy to specify strings in
various encodings, including the empty (e.g., "binary") encoding.

It is possible to view a standard US-ASCII (7-bit) string through the
UTF-8 lens and it's valid in either one.

-austin
-- 
Austin Ziegler * [e-mail] * http://www.halostatue.ca/
               * [e-mail] * http://www.halostatue.ca/feed/
               * [e-mail]