Re: Problems with RAW GET

Hi,

First, I did try this.  However, I would also like to download the source
for XML-pages like http://slashdot.org/slashdot.xml.  How can I download
these if I'm not using WWW_RAW?  Or is there a way to get rid of the block
counts or force HTTP/1.0?

Kind regards,

                Mark Wormgoor

----- Original Message -----
From: Roger McCalman <r.mccalman@elsevier.co.uk>
To: <www-lib@w3.org>
Sent: Thursday, November 25, 1999 10:14 AM
Subject: Re: Problems with RAW GET


> I would say that the results you are seeing are due to the HTTP 1.1
chunked
> encoding. Each block is preceeded by a count. Try using WWW_SOURCE rather
> than WWW_RAW.
>
> Cheers, Roger
>
> On Thu, Nov 25, 1999 at 10:04:51AM +0100, Mark Wormgoor wrote:
> > Hi,
> >
> > Using libwww I am trying to write a small application that will fetch
> > newsheaders from sites like slashdot and such.  For this reason I'm
using a
> > raw get of the page (slashdot and freshmeat use xml).  The platform is
> > Redhat 6.0 with libwww 5.2.8.  I'v attached test-source to the program.
> >
> > The problem is this.  When I try to fetch the URL in the sourcecode (a
> > Dutch newssite), it contains strange characters in the middle of the
raw
> > output, for example:
> > <img src=
> > 19c
> > '../grafx/nw_letter_nieuws.gif'
> > When I download the same page in Netscape, it prints:
> > <img src='../grafx/nw_letter_nieuws.gif'
> > which is the correct code.  Every time I download the page, these
things
> > appear at the same place.  When the page changes, I get different
> > characters at different locations.
> >
> > If somebody knows what's causing this, I would really like to know.
> > BTW, I compile this using:
> > gcc -O6 `libwww-config --cflags` -Wall `libwww-config --libs` -o test
> > test.c
> >
> > Kind regards,
> >
> >                 Mark Wormgoor
> >
>
>

Received on Thursday, 25 November 1999 04:25:38 UTC