Thanks again for the followup.
I can _force_ the problem to occur on the apache site, but a quick
scan of the site doesn't find any actual errors. For example, if I
manually type in:
http://www.apache.org/index.html/
I witness the behavior. Of course, your HTML generation avoids the
construct, so knowbody should be encountering this.
As background, we discovered this because of log file entries,
apparently generated by robots. It's unclear to me why the robots
would be trying to add a trailing slash, but they are in some cases.
- Marc
---- In reference to:
Marc Evans wrote:
>
> Thank you for your followup. I am curious, do you consider the
> browsers as having a bug in that the images referenced by the URL
> that are referenced as reletive links can not be found, apprently
> due to how the browser interprets the URL? In other words, the
> browsers present different results depending upon if the trailing
> slash is included or not.
An interesting point. But no, I don't think that's a bug -- the
browsers are behaving correctly. They have no way of knowing where
the path-info begins.
If a resource is capable of accepting path-info, it really should
have the responsibility of correctly coding internal relative
references so they appear as absolute to the client, and avoid
this ambiguity.
Do you have specific cases on the Apache site where this is
done? Because they should be fixed if so..
--
#ken P-)}
Ken Coar <http://Web.Golux.Com/coar/>
Apache Software Foundation <http://www.apache.org/>
"Apache Server for Dummies" <http://ASFD.MeepZor.Com/>
