On Sep 22, 2009, at 10:22 PM, Ian Bicking wrote:
I propose we switch primarily to "native" strings: str on both
Python 2 and 3.
[...]
All [...] headers will be treated as Latin1.
I like this. I think it would be "cleaner" to use bytes for all these
things, but it's not really important. Given that headers and such are
generally going to be ASCII, this works fine, is easy to implement,
and is convenient for users.
I then propose that we eliminate SCRIPT_NAME and PATH_INFO. Instead
we have:
wsgi.script_name
wsgi.path_info (I'm not entirely set on these names)
IMO it would be better to continue providing SCRIPT_NAME/PATH_INFO, as
latin-1 decoded strings -- the same as the other keys, but mark them
deprecated, and *also* provide wsgi.script_name / wsgi.path_info with
the semantics you describe.
I'll also note that it's generally possible to create wsgi.script_name
and wsgi.path_info in a CGI wrapper, given only SCRIPT_NAME and
PATH_INFO. You merely have to assume that all '/' characters are real
slashes, and should remain slashes instead of being quoted into %2F.
This is guaranteed true for most servers out there (they will forbid
invoking a CGI with %2Fs in the url.) So nobody should have any issues
implementing the WSGI-container side of this proposal.
James
_______________________________________________
Web-SIG mailing list
Web-SIG@python.org
Web SIG: http://www.python.org/sigs/web-sig
Unsubscribe:
http://mail.python.org/mailman/options/web-sig/archive%40mail-archive.com
