On Tue, May 07, 2002 at 10:55:14AM +0100, Ian Collier wrote: [...] > > Curiously, the server doesn't return a Content-Location at all if you ask > for /index.html instead of / . > > imc
On Tue, May 07, 2002 at 08:50:27AM -0400, Thomas Dickey wrote: > On Tue, May 07, 2002 at 02:37:29PM +0200, Serge Munhoven wrote: [...] > > > > Maybe this can be overridden by an explicit > > <base href="http://www.srsb.org.uk/";> in the <head> of the document. > > maybe - but there is none. I know. I just wanted to suggest a possible workaround to what IMHO is a server bug or misconfiguration. Since I don't have/use IIS, I could not test it. On Wed, May 08, 2002 at 07:05:06PM +0100, Peter Norrington wrote: > > Dear Serge, > > Thanks for the help. (RFC's are a bit over my head at the moment. However...) I > tried the suggestion on one test page only. I could now access all the other > unaltered pages. Is this unexpected? (I'm not sure I'll understand the answer, but > it may help lynx-dev to know this.) > > The difficulty with this is that I would have to change ALL pages (in case the > site is not entered by the front page) with an explicit <base href=...> , making > the site hard to test off-line. If it's a choice between this and nothing, no > problem, but any other ideas would be welcome. > As Ian wrote, the problem only occurs when accessing http://www.srsb.org.uk or http://www.srsb.org.uk/ and not when accessing http://www.srsb.org.uk/index.html. It seems not to occur with any page containing an explicit filename (i.e. not pointing to a directory, in which case the server has to decide according to its configuration which file to return, if any). In short, in this case, "fixing" only the file index.html corresponding to http://www.srsb.org.uk/ should be enough. If there were any subdirectories with an index.html file the same would apply to those too. As you remarked, off-line testing needs the <base href= ...> to be removed. Best would of course be to have your ISP fix its server ... Hope this helps, - Serge PS: For those interested, the following message about a related issue discusses a bit how "de facto" (or rather not) this "standard" feature is: http://lists.zope.org/pipermail/zope-coders/2001-November/000460.html ; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to [EMAIL PROTECTED]
