W3C home > Mailing lists > Public > www-talk@w3.org > January to February 1996

Re: bug, or "feature"?

From: Daniel W. Connolly <connolly@beach.w3.org>
Date: Mon, 01 Jan 1996 14:17:19 -0500
Message-Id: <m0tWpjU-0002UgC@beach.w3.org>
To: Larry Masinter <masinter@parc.xerox.com>
Cc: sjk@amazon.com, www-talk@w3.org
In message <95Dec29.204102pst.2733@golden.parc.xerox.com>, Larry Masinter write
>> I've just discovered that with some popular browsers, including the
>> latest from Netscape, if you do a GET on a URL with a fragment (an
>> anchor, a "#something" on the end), and the server issues a redirect
>> (302) so that a different document is fetched, then the browser will
>> NOT seek to the anchor to the ultimately viewed document -- you end up
>> staring at the beginning of the page.
>I think it's a bug. I suppose this should be clarified in the next
>revision of the URL documents.

I agree that it's not the way I'd like it to work. However, according
RFC1808, it's quite clearly a feature:


5.  Examples and Recommended Practice

   Within an object with a well-defined base URL of

      Base: <URL:http://a/b/c/d;p?q#f>

   the relative URLs would be resolved as follows:

5.1.  Normal Examples

      g:h        = <URL:g:h>
      g          = <URL:http://a/b/c/g>
      ./g        = <URL:http://a/b/c/g>
      g/         = <URL:http://a/b/c/g/>
      /g         = <URL:http://a/g>
      //g        = <URL:http://g>
      ?y         = <URL:http://a/b/c/d;p?y>
      g?y        = <URL:http://a/b/c/g?y>
      g?y/./x    = <URL:http://a/b/c/g?y/./x>
      #s         = <URL:http://a/b/c/d;p?q#s>
      g#s        = <URL:http://a/b/c/g#s>
      g#s/./x    = <URL:http://a/b/c/g#s/./x>
      g?y#s      = <URL:http://a/b/c/g?y#s>
      ;x         = <URL:http://a/b/c/d;x>
      g;x        = <URL:http://a/b/c/g;x>
      g;x?y#s    = <URL:http://a/b/c/g;x?y#s>
      .          = <URL:http://a/b/c/>
      ./         = <URL:http://a/b/c/>
      ..         = <URL:http://a/b/>
      ../        = <URL:http://a/b/>
      ../g       = <URL:http://a/b/g>
      ../..      = <URL:http://a/>
      ../../     = <URL:http://a/>
      ../../g    = <URL:http://a/g>

I think there's a whole bunch of edge cases that haven't been
fully explored in the URI specs. I started creating a test suite
a long time ago:


I hope to get resources to complete that test suite some time

Received on Monday, 1 January 1996 14:17:39 UTC

This archive was generated by hypermail 2.4.0 : Monday, 20 January 2020 16:08:19 UTC