Changeset 22


Ignore:
Timestamp:
Oct 22, 2010, 3:14:46 PM (9 years ago)
Author:
duerst@…
Message:

implemented Addison's proposal from http://lists.w3.org/Archives/Public/public-iri/2010Oct/0039.html, with a slight tweak to align with previous paragraph

File:
1 edited

Legend:

Unmodified
Added
Removed
  • draft-ietf-iri-3987bis/draft-ietf-iri-3987bis.xml

    r21 r22  
    589589 encoding such as UTF-8 or UTF-16) should be left as is and not
    590590 normalized (see (see <xref target="normalization"/>).</t>
    591  
    592 <t>If the IRI or IRI reference is an octet stream in some known
    593  non-Unicode character encoding, convert the IRI to a sequence of
    594  characters from the UCS; this sequence SHOULD also be normalized
    595  according to Unicode Normalization Form C (NFC, <xref
    596  target="UTR15"/>). In this case, retain the original character
    597  encoding as the "document character encoding". (DESIGN QUESTION:
    598  NOT WHAT MOST IMPLEMENTATIONS DO, CHANGE? ) </t>
     591
     592  <t>An IRI or IRI reference is a sequence of characters from the UCS.
     593    For IRIs that are not already in a Unicode form
     594    (as when written on paper, read aloud, or represented in a text stream
     595    using a legacy character encoding), convert the IRI to Unicode.
     596    Note that some character encodings or transcriptions can be converted
     597    to or represented by more than one sequence of Unicode characters.
     598    Ideally the resulting IRI would use a normalized form,
     599    such as Unicode Normalization Form C <xref target="UTR15"/>
     600    (see <xref target='#ladder'/> Normalization and Comparison),
     601    since that ensures a stable, consistent representation
     602    that is most likely to produce the intended results.
     603    Implementers and users are cautioned that, while denormalized character sequences are valid,
     604    they might be difficult for other users or processes to reproduce
     605    and might lead to unexpected results.
     606  </t>
    599607
    600608<t> In other cases (written on paper, read aloud, or otherwise
Note: See TracChangeset for help on using the changeset viewer.