Re: [Ltru] matching: Language Range Lists

as a chair: Please use 'matching:' in the subject of all mails
concerning the matching draft. Thanks!
[I have done this in this subject, and also made the subject
to be closer to the actual topic.]

At 04:31 05/09/29, Mark Davis wrote:
>The main issue I have with the document as it stands is that it neglects a 
>very important case. The general case of language matching involves having 
>a *list* of language ranges. You take a list like "en-US en fr it" and use 
>it for either matching or lookup. So I would suggest adding the following 
>text. (It probably needs fleshing out a bit, but this is a start.).

without my chair hat:

This is already defined, in quite a bit more detail and
with some slight differences (i.e. allowing a q value) in
the HTTP 1.1 spec. See e.g.
http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.4

Do we need anything more?

Please also note that according to the HTTP rules,
zh-Hant-CN

in the 'language range list' does NOT match zh or zh-Hant.
(whereas the other way round, a document of zh-Hant-CN IS
matched by a request for zh or zh-Hant).
This was done to avoid false positives. In the example belows,
zh is in third place, but if the user explicitly specifies
zh-Hant-CN, this probably means (or should be reserved to
mean) that she doesn't read simplified,... In that case,
it seems that en-US or fr should have higher priority,
because the user actually understands that.

Regards,    Martin.
>2.3 Language Range Lists
>
>A language range list is a space-delimited list of zero or more language 
>ranges. An extended language range list contains extended language ranges; 
>a basic language range list is limited to containing basic language 
>ranges. In matching, anything that matches any element of the list 
>according to the above options is included. (With distance metrics, the 
>shortest distance from any element in the list is the one used for the 
>overall distance).
>
>In lookup, the list is examined from first to last for a match, according 
>to the above options. For example, starting with the list range 
>"zh-Hant-CN en-US fr", and using Lookup with truncation as in 2.1.1, the 
>lookup would progressively search for content as shown below:
>
>
>List Range to match: zh-Hant-CN en-US fr
>1. zh-Hant-CN
>2. zh-Hant
>3. zh
>4. en-US
>5. en
>6. fr
>7. (default content or the empty tag)
>
>
>
>2.3ツ Meaning of Language Tags and Ranges [RENUMBER to 2.4...]
>
>_______________________________________________
>Ltru mailing list
>Ltru@ietf.org
>https://www1.ietf.org/mailman/listinfo/ltru

_______________________________________________
Ltru mailing list
Ltru@ietf.org
https://www1.ietf.org/mailman/listinfo/ltru