On 1/18/12 1:41 PM, Mark Thomas wrote:
> On 18/01/2012 21:25, Shing Wai Chan wrote:
>> On 1/14/12 2:33 AM, Mark Thomas wrote:
>>> On 13/01/2012 23:02, Shing Wai Chan wrote:
>>>> Hi,
>>>>
>>>> I am looking at
>>>> http://java.net/jira/browse/SERVLET_SPEC-27
>>>> ("sendRedirect Javadoc prevents use of protocol relative URLs")
>>>>
>>>> According to
>>>> http://blog.httpwatch.com/2010/02/10/using-protocol-relative-urls-to-switch-between-http-and-https/
>>>>
>>>>
>>>> ,
>>>> it seems that it tries to workaround an issue in IE.
>>>>
>>>> I am not sure if it is worth to make it as a standard.
>>>>
>>>> Any comments?
>>> Protocol relative redirects are standard. [1] is just one example of
>> Can you clarify what "standard" is for protocol relative URLs?
> RFC 3986. Section 4.2. Strictly, it should be called a "network-path
> reference".
In this case, we should use the term "network-path reference" in javadoc
(with reference to the rfc).
Shing Wai Chan
>
> Mark
>
>> Shing Wai Chan
>>> where they might be used. Currently, the wording of the sendRedirect()
>>> Javadoc prevents the use of protocol relative redirects. They may not be
>>> as widely used but they are standard and the Servlet spec should not
>>> prevent their use. The fix is a simple change to the Javadoc.
>>>
>>> Mark
>>>
>>> [1]
>>> http://blog.httpwatch.com/2010/02/10/using-protocol-relative-urls-to-switch-between-http-and-https/
>>>