lynx-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: LYNX-DEV more BASEness


From: Foteos Macrides
Subject: Re: LYNX-DEV more BASEness
Date: Mon, 19 Jan 1998 16:02:15 -0500 (EST)

Al Gilman <address@hidden> wrote:
>Using Lynx 2.7 -- OK, so it's antique.
>
>G)o to www.irs.gov -- it's that time of year.
>
>Netscape server is browser-sniffing, I guess, because I get to
>start on the text-only page.  This page contains relative URLs
>starting with "../" which would be illegal if the page knew
>it were served as <http://www.irs.gov/>.
>
>On the other hand, this page thinks it is
><http://www.irs.gov/plain/cover.html> and it can be retrieved by
>this URL and if you had so accessed it, all the relative URLs
>would work.
>
>I perused a trace, looking for a redirect transaction, and I
>could find none.
>
>The Netscape software would seem to be assuming that ../ in
>excess of enough to back up to the root of the hierarchical
>domain in URL space will be ignored.
>[...]

        This was vigoroursly debated and the current URL/URI syntax
draft has a compromise position.  The parsing rules SHOULD yield the
lead relative elements in the resolved absolute URL.  The browser MAY
apply "error recovery" if the result is invalid.  In Lynx v2.7.2, any
lead relative elements in the resolved absolute URL are stripped for
http(s) URLs (because no http(s) server would accept them), but not
for other URLs (e.g., ftp, for which they could be intended).  Roy
Fielding communicated to me privately that he agreed that's the best
way to handle the problem, but doubts, as do I, that the Big Two
browsers will stopped stripping them for all URLs.

                                Fote

=========================================================================
 Foteos Macrides            Worcester Foundation for Biomedical Research
 address@hidden         222 Maple Avenue, Shrewsbury, MA 01545
=========================================================================
 

reply via email to

[Prev in Thread] Current Thread [Next in Thread]