[Python-Dev] Unifying Long Integers and Integers: baseint (original) (raw)

Skip Montanaro skip at pobox.com
Thu Aug 12 04:50:09 CEST 2004 

Greg> I can enivisage a future in which 'string' means unicode string,
Greg> and byte sequences are called something else entirely, such as
Greg> 'bytevector' or 'bytearray' (or maybe just 'array' :-).

After another couple minutes of thought:

1. Maybe in 2.5 introduce a "bytes" builtin as a synonym for "str" and
   recommend its use when the intent is an arbitrary sequence of bytes.
   Add 'b' as a string literal prefix to generate a bytes object (which
   will really just be a string).  At the same time, start raising
   exceptions when non-ascii string literals are used without a coding
   cookie.  (Warnings are already issued for that.)

2. In 2.6, make str a synonym for unicode, remove basestring (or also
   make it a synonym for unicode), make bytes its own distinct (mutable?
   immutable?) type, and have b"..." literals generate bytes objects.

Then again, maybe this can't happen until Py3K. Before then we could do the following though:

1. Make bytes a synonuym for str.

2. Warn about the use of bytes as a variable name.

3. Introduce b"..." literals as a synonym for current string literals,
   and have them *not* generate warnings if non-ascii characters were
   used in them without a coding cookie.

PEP time?

Skip

More information about the Python-Dev mailing list