Skip to main content

[jsr339-experts] Re: [jax-rs-spec users] Re: Re: Default Servlet Mapping?

  • From: Marek Potociar <marek.potociar@...>
  • To: jsr339-experts@...
  • Subject: [jsr339-experts] Re: [jax-rs-spec users] Re: Re: Default Servlet Mapping?
  • Date: Wed, 21 Mar 2012 20:37:30 +0100


On Mar 21, 2012, at 8:08 PM, Bill Burke wrote:

> 
> 
> On 3/21/12 3:01 PM, Marek Potociar wrote:
>
>> On Mar 19, 2012, at 3:01 PM, Bill Burke wrote:
>
>>> 
>>> 
>>> On 3/15/12 4:15 PM, Santiago Pericas-Geertsen wrote:
>>>> 
>>>> On Mar 15, 2012, at 4:00 PM, Bill Burke wrote:
>>>> 
>>>>> 
>>>>> 
>>>>> On 3/15/12 3:54 PM, Santiago Pericas-Geertsen wrote:
>>>>>> 
>>>>>> On Mar 15, 2012, at 3:31 PM, Bill Burke wrote:
>>>>>> 
>>>>>>> I would say do nothing because it would break existing 1.1 
>>>>>>> applications when deployed.
>>>>>> 
>>>>>>  How so? Before you had to always specify the mapping, and that will 
>>>>>> continue to take precedence. Can you elaborate?
>>>>>> 
>>>>> We assumed everything could be scanned, you could have an empty web.xml 
>>>>> file, and no Application class.  This was an incorrect assumption?  I 
>>>>> know a lot of people use us in this manner.  Maybe I just interpreted 
>>>>> the 1.1 spec wrong?
>>>> 
>>>>  Class scanning and servlet mapping are orthogonal. The only way you can 
>>>> get the servlet mapping in 1.1 is from (i) a web.xml or (ii) the 
>>>> @ApplicationPath annotation on an Application subclass AFAICT, with (i) 
>>>> overriding (ii) if both are present. For the other cases, 1.1 states 
>>>> that "the application MUST be packaged with a web.xml that specifies a 
>>>> servlet mapping for the added servlet".
>>>> 
>>>>  Perhaps you're already using a default in Resteasy?
>>>> 
>>> 
>>> Well, then the spec should allow the case for *no* Application class and 
>>> no web.xml listing.  The default mapping should be "/*" and should work 
>>> with static content too.  This way its very simple for users and no 
>>> thought has to be put into anything.
>>> 
>
>> I think so far we can all agree, that a default mapping is good. I want to 
>> however point out that suggested "/*" as a default mapping is too 
>> aggressive in connection with any other technology that uses some default 
>> mapping as well (e.g. JSF). For that reason, I suggest to choose a named 
>> default mapping directly under the root path e.g. the "/webapi/*" as 
>> proposed earlier.
>
> 
> Please tell me why "/*" is too aggressive?  its easily handled if your 
> implementation is Filter based.

Didn't I just tell you? Other JavaEE specs (afaik JSF uses "/faces") (and 
possibly other EE technologies) define their default mappings. Users using 
combination of these technologies in a single application should not be 
required to resolve any conflicts caused by the default mappings. Reserving 
conservatively a single named subtree of the application URI space is a 
better option ("live and let live") with a positive impact on the end-user 
convenience.

> But, this is orthogonal to the issue of requiring a web.xml or Application 
> class.  I do not think either of these artifacts should be required to 
> deploy a JAX-RS service.

I agree.

Marek

> 
> Bill
> 
> -- 
> Bill Burke
> JBoss, a division of Red Hat
> http://bill.burkecentral.com



[jsr339-experts] Re: Default Servlet Mapping?

(continued)

[jsr339-experts] Re: Default Servlet Mapping?

Sergey Beryozkin 03/14/2012

[jsr339-experts] Re: Default Servlet Mapping?

Marek Potociar 03/14/2012

[jsr339-experts] Re: Default Servlet Mapping?

Sergey Beryozkin 03/14/2012

[jsr339-experts] Re: Default Servlet Mapping?

Bill Burke 03/15/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Default Servlet Mapping?

Santiago Pericas-Geertsen 03/15/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Default Servlet Mapping?

Bill Burke 03/15/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Default Servlet Mapping?

Santiago Pericas-Geertsen 03/15/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Default Servlet Mapping?

Bill Burke 03/19/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Default Servlet Mapping?

Marek Potociar 03/21/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Default Servlet Mapping?

Bill Burke 03/21/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Re: Default Servlet Mapping?

Marek Potociar 03/21/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Re: Default Servlet Mapping?

Bill Burke 03/21/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Re: Re: Default Servlet Mapping?

Santiago Pericas-Geertsen 03/22/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Default Servlet Mapping?

Sergey Beryozkin 03/21/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Default Servlet Mapping?

Bill Burke 03/22/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Default Servlet Mapping?

Sergey Beryozkin 03/22/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Re: Default Servlet Mapping?

Santiago Pericas-Geertsen 03/23/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Re: Default Servlet Mapping?

Sergey Beryozkin 03/23/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Default Servlet Mapping?

Marek Potociar 03/15/2012

[jsr339-experts] Re: [jax-rs-spec users] Re: Default Servlet Mapping?

Bill Burke 03/19/2012
 
 
Close
loading
Please Confirm
Close