Hi, Thomas,

I learned a lot from your latest answer. Thanks!

caniuse.com's approach is exactly what I want. Can you give more details on
how the redirecting is done for the comment below?
"So redirecting from /foo to #foo might be better (caniuse.com does just
that: redirect /history to #feat=history) "?

-maq

On Mon, Sep 5, 2011 at 9:15 AM, Thomas Broyer <[email protected]> wrote:

>
>
> On Monday, September 5, 2011 1:52:37 PM UTC+2, maq wrote:
>>
>> Thanks Y2i and Thomas.
>>
>> One of the reason is I wonder if I  can redirect a full domain name to a
>> internal state of my application. For example , I go to a domain hosting to
>> redirect
>>          http://someotherdomain.com
>> to
>>          
>> http://mything.com/#**someotherdomain<http://mything.com/#someotherdomain>
>> But the domain hosting service complains about the URL format, it has to
>> be without the hash sign.
>>          
>> http://mything.com/**someotherdomain<http://mything.com/someotherdomain>
>>
>>
>> Thomas, is your proposal only applicable to new browsers?
>
>
> See http://caniuse.com/history
>
>
>> I wish a solution applies to old browser as well.
>>
>
> In your entry-point, you could look at the current history token, and if
> empty, then look at the path (treating /foo the same as #foo, i.e. a history
> token of "foo"). I.e. do it manually, and only at "initialization" time. If
> you then navigate into your app, then you'll have
> http://example.com/foo#bar URLs, which can be confusing.
> So redirecting from /foo to #foo might be better (caniuse.com does just
> that: redirect /history to #feat=history). Then, if you want, you could use
> replaceState in browsers that support it to redirect once more to /foo,
> though on the client-side this time, without request to the server for this
> history change.
>
> As for your redirect issue, request.getrequestDisplatcher() won't help
> redirecting the client-side, it will "rewrite" the URL on the server-side,
> but the client won't be aware of that: it requests a URL and is sent a
> response back. With sendRedirect, a "meta refresh" (as you did), or a JS
> location.replace() or location assignment, it requests a URL, and is told
> (at different levels: HTTP, HTML, JS) to rather request another one.
> You might want to try several alternatives in many browsers though, as
> redirecting to a URL containing a "hash part" might not always work as
> expected (search the GWT issue tracker, I seem to remember an issue with IE)
>
> --
> You received this message because you are subscribed to the Google Groups
> "Google Web Toolkit" group.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/google-web-toolkit/-/clDEoUhFlMcJ.
>
> To post to this group, send email to [email protected].
> To unsubscribe from this group, send email to
> [email protected].
> For more options, visit this group at
> http://groups.google.com/group/google-web-toolkit?hl=en.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/google-web-toolkit?hl=en.

Reply via email to