RE: Raw HTTP Headers

OK, I think the confusion is that there _isn't_ a verbatim copy. Even if
you switch logging of the RAW stream on, the headers that are logged are
actually reconstructed from HttpClient serialising the ones it has
constructed.

If we want to capture the raw headers we will need to do a little tweak
to the code to make it squirrel the incoming stream away somewhere.

Jo

> -----Original Message-----
> From: Sean Owen [mailto:srowen@google.com]
> Sent: 25 May 2007 16:24
> To: Jo Rabin
> Cc: public-mobileok-checker@w3.org
> Subject: Re: Raw HTTP Headers
> 
> On 5/25/07, Jo Rabin <jrabin@mtld.mobi> wrote:
> >
> >
> >
> > >
> > > How does it modify the headers -- just trimming whitespace?
> > >
> > > My view on this is that we should record the entire response,
> > > verbatim, in a CDATA section, Base64 encoded per EARL (or not),
and
> > > then feel free to embrace all kinds of normalization and parsing
in
> > > the moki document.
> > >
> > > You've got the original stashed away for the record, and then the
> > > useful, parsed version in the moki document.
> > >
> > > So: no I'd rather not be bothered by this.
> >
> > I'm sorry, I don't understand, you seem to be at odds with yourself
on
> > this. You both don't want to do it and want to have the verbatim
record?
> 
> There are two copies, no?
> 
> We have the verbatim response recorded somewhere, then, we also have
> this processed version with <header> and <element> tags. I'm not
> bothered if the latter is normalized in some ways, since we're already
> obviously processing it quite a bit. Any question of audit trail is
> answered by the verbatim copy. That seems to satisfy all requirements.

Received on Friday, 25 May 2007 15:30:44 UTC