[Ltru] status of RFC 3066 or RFC 3066bis in relation to HTTP Accept-Language

[Peter Constable]

> From: mrc at pangtzu.panda.com [mailto:mrc at pangtzu.panda.com] On Behalf
Of Mark
> Crispin


> > Peter - I think you're on your own - and note that RFC 1766
> > doesn't exactly gracefully prepare programmers for 'script'
> > subtags in the second position followed by 'region' subtags
> > in the third position
> 
> Sounds like a good argument to abolish script subtags, and to tell the
> people who want script identity that they are screwed, now and
forever,
> and will just have to live with it because someone may be excessively
> pedantic about RFC 1766.

One more time... If anyone is to be excessively pedantic about RFC 1766,
then *far, far* more is allowed in a 1766 tag than is allowed under
3066bis -- under 1766, there is *very little* that can be assumed about
the internal structure of tags. That doesn't in any way suggest
abolishing subtags; it just means that if all you know is RFC 1766 then
you don't know very much.

RFC 1766 allowed tags like az-Cyrl-AZ; it just didn't call that out as a
specific scenario.


I guess I've figured out the answer to my own question: if you only care
about 1766, you can test for a few possible cases and must fail
gracefully for a very large set of other cases. But if it's of any
interest to you, then you can anticipate that an Accept-Language header
can contain any of the tags permitted by 3066bis and you can parse them
assuming that 3066bis is applying rather than the
wild-west-anything-goes rules of 1766.



Peter Constable