[Python-Dev] readd u'' literal support in 3.3? (original) (raw)

Lennart Regebro regebro at gmail.com
Fri Dec 9 15🔞33 CET 2011

On Fri, Dec 9, 2011 at 04:34, Barry Warsaw <barry at python.org> wrote:

Sorry, I don't understand this.  What does it mean to be "str in both versions"?  And why would you want that?

It means that it's a str, that is a string of bytes, in Python 2, and a str, that is a string of Unicode characters, in Python 3. There are cases where you want this, for example not all libraries will accept both str and Unicode under Python 2.

As for "Unicode in Python 2 and str in Python 3", unadorned strings with the future import in Python >= 2.6 does that just fine.

Yes, but the future import will change this for all strings, making it impossible to have a string that is a "str" in both Python 2 and Python 3. For that reason, the future import is not enough as a solution (and I suspect, one major reason why I haven't actually seen any one using it).

For most cases, using something like six's b() and u() has turned out to be a better solution. It's uglier than having u'' support in Python 3, but has the benefit that b() works also in Python 2.5.

 The problem comes when you aren't or can't be sure, i.e. you have objects that are sometimes one and sometimes the other.  Such as email headers.  In that case, you're kind of screwed.  Python 2's str type let you cheat, but not without consequences.  Those consequences are spelled "UnicodeErrors" and I'll be glad to be rid of them.

Me too.

More information about the Python-Dev mailing list