Re: [Ltru] Test suite for language tags?

[Addison Phillips <addison at yahoo-inc.com>]

Doug Ewell wrote:
> Just a few minor nits here.  This is slightly more complex than meets 
> the eye, unfortunately.

Yep.
> > - singletons in the first position (except for 'x' and the 
> > grandfathered list)
>
> Sadly, the existence of the grandfathered list means that all 
> well-formed processors must also do a limited amount of validity 
> checking.  I don't question the importance of maintaining support for 
> the grandfathered tags, but this is a side effect.

Agreed: all RFC 3066bis processors have to have the list of 
grandfathered tags baked in.
> > - missing subtag ("--")
> > - a dangling hyphen ("foo-bar-baz-") or initial hyphen ("-foo-bar-baz")
>
> The second is really just a special case of the first: a missing subtag 
> at the end or beginning, respectively.  One thing I found useful, when 
> building my validator, was to parse out the subtags first and check them 
> for validity afterward, so the hyphens never become part of the validity 
> checking per se.

I did the same thing. However, one must check the hyphens. Tokenizers 
sometimes do not return "empty" tokens and can miss these cases.
> > "ab-x-abc-x-abc" // anything goes after x
>
> Not quite anything, of course:  1*("-" (1*8alphanum))

Yes. At least one alphanumeric subtag must follow 'x' and it cannot 
exceed eight characters in length.

Addison

--
Addison Phillips
Globalization Architect -- Yahoo! Inc.

Internationalization is an architecture.
It is not a feature.

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