httpbis: Ticket #74: Character Encodings in TEXT

Link:  http://trac.tools.ietf.org/wg/httpbis/trac/ticket/74

Origin:  http://www.w3.org/mid/6.0.0.20.2.20070610165356.0a69cec0@localhost

Component: p1-messaging

RFC 2616 prescribes that headers containing non-ASCII have to use either iso-8859-1 or RFC 2047. This is unnecessarily complex and not necessarily followed by implementations or by specifications of new headers.

This issue is limited to:

See also #63, #111.

Mails

Mails by Sender (Top 10) Mails by Month

History

: comment added; component, milestone set (Fri, 04 Jan 2008 06:04:37 GMT)

: comment added; summary changed (Mon, 11 Feb 2008 23:26:18 GMT)

There was discussion of this at the APPS Area Architecture Workshop, with some disagreement as to whether it's possible to encode IRI->URI->IRI. Specific advice to IRIs may be necessary.

: comment added (Fri, 14 Mar 2008 09:44:28 GMT)

Replying to mnot@pobox.com:

There was discussion of this at the APPS Area Architecture Workshop, with some disagreement as to whether it's possible to encode IRI->URI->IRI. Specific advice to IRIs may be necessary.

Is this about round-tripping IRIs through URIs? Obviously that's not possible.

For example, consider the two IRIs:

I1: http://www.example.org/Dürst

I2: http://www.example.org/D%C3%BCrst

Both would be converted to the URI:

U: http://www.example.org/D%C3%BCrst

Now whether that disctinction is relevant of course depends on which kind of URI/IRI comparison is needed; but there are cases where it is relevant (for instance, XML namespace names using IRIs (urg!)).

(see also http://tools.ietf.org/html/rfc3987#section-3.2.1 )

: comment added; description, summary changed (Fri, 04 Apr 2008 00:31:14 GMT)

: comment added; description changed (Fri, 04 Apr 2008 00:46:41 GMT)

: comment added; description changed (Fri, 04 Apr 2008 00:51:44 GMT)

Related Information

Issues List Index