Here is my latest patch which add :

gzip reply compression support
gzip request compression support (tested with Apache 2.x / mod_deflate)
basic user auth




On Wed, 9 Mar 2005 12:16:52 +0100, Henri Gomez <[EMAIL PROTECTED]> wrote:
> As a tomcat commiter, I take a look at what Remy does in coyote HTTP
> 1.1 gzip support and also at what is allready done in Apache 2.x
> mod_deflate :
> 
> BTW : in Coyote HTTP11 the code is looking for gzip in Accept-Encoding :
> 
>         // Check if browser support gzip encoding
>         MessageBytes acceptEncodingMB =
>             request.getMimeHeaders().getValue("accept-encoding");
> 
>         if ((acceptEncodingMB == null)
>             || (acceptEncodingMB.indexOf("gzip") == -1))
>             return false;
> 
> indexOf gzip :-))
> 
> On Tue, 08 Mar 2005 21:25:07 +0100, Jochen Wiedmann
> <[EMAIL PROTECTED]> wrote:
> > Henri Gomez wrote:
> >
> > >Nobody to revue my GZIP patch and commit to HEAD ?
> >
> > Henri, some comments:
> >
> > 1.) The "Accept-Encoding" header transmitted by the client seems to me
> > to be RFC compliant, so that is fine.
> >
> > However, if the server is simply looking for the word "gzip" in the
> > header, that seems to me to be insufficient. For example, the following
> > might be handled wrongly:
> >
> >      Accept-Encoding: gzip;q=0.0, identity; q=0.5, *;q=0
> >
> > (I admit, that the example might be academic.) However, I'd recommend to
> > change the patch allong the lines of
> >
> >      public static boolean isUsingGzipEncoding(String pHeaderValue) {
> >                 if (pHeaderValue == null) {
> >                         return false;
> >          }
> >          for (StringTokenizer st = new StringTokenizer(pHeaderValue,
> >                                                        ",");
> >               st.hasMoreTokens();  ) {
> >              String encoding = st.nextToken();
> >              int offset = encoding.indexOf(';');
> >              if (offset >= 0) {
> >                  encoding = encoding.substring(0, offset);
> >              }
> >              if ("gzip".equals(encoding)) {
> >                 return true;
> >              }
> >          }
> >          return false;
> >      }
> >
> > 2.) Any change, that you handle compression of the requests as well?
> 
> I could do that if you want, I allready added the basic authentification ...
> 
> > 3.) Isn't your change worth a patch for the docs?
> 
> I'm not a commiter of xml-rpc project :)
>

Reply via email to