Re: [xhr] statusText is underdefined

On 2012-03-28 10:57, Boris Zbarsky wrote:
> On 3/28/12 2:48 AM, Anne van Kesteren wrote:
>> Wow, "inflating" is what the specification currently requires for header
>> fields, for what it's worth.
>
> Yes, Gecko has different behavior here for the status text and for
> header fields. I have no idea whether that's on purpose; I suspect it's
> mostly historical accident.
>
> -Boris

FWIW, when we discussed this years ago in HTTPbis, we found that certain 
servers indeed send localized reason phrases, and I believe we saw at 
least one example of ISO-8859-1.

Anyway. This is an edge case of an edge case. "Inflating" IMHO is good 
enough.

It would be much more interesting to fix other problems, like resetting 
a header field value, or the topic of field value encoding.

Best regards, Julian

Received on Thursday, 29 March 2012 07:01:34 UTC