(personal response) > > The WebApps WG believes that removing the redundant repetitions in > a > standardized way may avoid interop issues. Having said that, in the > latest editors' draft, the rightmost occurrences are removed (as > suggested above).
Actually, I don't believe that there are interoperability issues to avoid. If two identical ranges appear in a language priority list, the second occurrence will produce the same results as the first one. If you are processing the second one, presumably the result will be no match. Note that, with the lookup algorithm, this also applies to ranges that are prefixes of other ranges. For example, consider the list "en-us-boont,en". The second item (en) mirrors the last match attempted with "en-us-boont". It is an optimization to omit these ranges, but won't change the results. > > Please see for the updated algorithm: > http://dev.w3.org/2006/waf/widgets/#step-5--derive-the-user-agents- > locale > Thanks. We'll review it. Addison
