----- Original Message -----
From: "André Warnier" <a...@ice-sa.com>
To: "Tomcat Users List" <users@tomcat.apache.org>
Sent: Sunday, September 05, 2010 11:38 AM
Subject: Re: URL Rewrite
michel wrote:
...
André, I am not sure that I understand but I think that I do. In this
case, I believe that HTML does interpret links in a natural way.
Normally, the HTML picks up the base href from the toolbar. In the case
of a forward with a clean URL in the toolbar, we already have an
unnatural HTML process going on. But I am only using the base href on
that single page anyway.
I will take your last phrase above first.
If that is true, and if it remains that way, then it is not a big problem.
But if you find yourself in a situation where you start having to use the
"base href" mechanism in more pages, then stop and rethink the whole thing
before you do that.
Because it is the kind of thing where you start by putting your little
finger, and before you know it, your whole arm will be sucked into it
also.
Now about the rest.
The browser does not really look at the toolbar or the URL input box.
The browser gets a page from a server, and it interprets that page.
When it does that, the browser knows one thing : from which absolute URL
it received the current page. That is its starting point to interpret the
links that are contained in the page.
In other words, *the URL from which it received the current page* is its
"natural base href".
But when inside that page, you use the <base href=...> directive, then you
force the browser to ignore where-from it has received the current page,
and to use this new "base href" instead.
That is a very blunt instrument, because it influences the way in which
the browser is now going to interpret all the relative links inside the
page.
So it is like a big hammer to hit a small nail, and you should be very
careful with it.
For example, if you give a copy of all your website pages to a website
graphic designer, to improve the look, then it means that this graphic
designer will not be able to easily "simulate" your website on his own
computer, just because all the links will be interpreted wrongly (they
will keep linking to your original website, instead of his copy).
For example also, it means that if one day you decide that your clients
have to go through a proxy server to access some of your pages, you may
have problems because browsers will try, for some links, to keep accessing
the original server directly.
You get the idea.
What I mean in general, is that fixing your problem using a "base href"
may be a quick solution in this particular case, but it is probably not
the "right" solution in general, and may cause you more problems later on.
The right solution would be to make sure that all the relative links in
your pages, when they are interpreted by the browser and requested from
the server, are also being caught by the rewriting mechanism on the
server, and properly redirected to where they should.
And preferably, that this would happen internally on the server, without
sending a redirect response to the browser and forcing the browser to make
a second request for everything.
André, I think that you are correct, but I have limited skills in URL
rewrite, and I found that getting this part to work was a major pain. I
should, in time, return to this. But there is also the case (in this case at
least) where the base url is in a JSP and is dynamically generated by a
bean, so I am not so sure that generating it from a URL rewrite is any
better.
Michel
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org
