+1 for option 3, but I'm not sure how much time it takes to implement this
option.

(If it's too much effort, option 2 looks okay to me)

Regards,
Jan-Kees


2011/7/4 Rudy De Busscher <rdebussc...@gmail.com>

> I can agree with jacob that "Suffix mapping is bad for resource-requests "
> but the choosen option shouldn't block developers from using suffix mapping
> for pages.
>
> As far as I can understand the discussion -> +1 for option 2 (option 3 if
> we want to have an advanced config version)
> Regards
> Rudy
> On 3 July 2011 02:33, Bruno Aranda <brunoara...@gmail.com> wrote:
>
>> +1 for 3
>>
>> Between 2 and 4, I still prefer a filter. For me an init param to deal
>> with such a specific case is more obscure than a filter, but it may be my
>> intuition,
>>
>> Cheers,
>>
>> Bruno
>>
>>
>> On 3 July 2011 00:20, Gerhard Petracek <gerhard.petra...@gmail.com>wrote:
>>
>>> i agree with martin and jakob.
>>>
>>> regards,
>>> gerhard
>>>
>>>
>>> http://www.irian.at
>>>
>>> Your JSF powerhouse -
>>> JSF Consulting, Development and
>>> Courses in English and German
>>>
>>> Professional Support for Apache MyFaces
>>>
>>>
>>>
>>> 2011/7/2 Jakob Korherr <jakob.korh...@gmail.com>
>>>
>>>> Hi,
>>>>
>>>> I totally agree with Martin on the preferred options and the filter
>>>> question.
>>>>
>>>> IMO there should not be any filter. Furthermore I really don't
>>>> understand why you want suffix mapping to work so badly, Leonardo.
>>>> Suffix mapping is bad for resource-requests (maybe even an epic fail),
>>>> because a css file is accessed via e.g. style.css.jsf. If the
>>>> mime-type is stripped or ignored or whatever, the browser (note there
>>>> are pretty bad browsers out there) might think this is a *.jsf file..
>>>> And there are some more reasons, that I can explain on request to
>>>> everyone interested.
>>>>
>>>> Regards,
>>>> Jakob
>>>>
>>>> 2011/7/1 Leonardo Uribe <lu4...@gmail.com>:
>>>>  > Hi Martin
>>>> >
>>>> > 2011/7/1 Martin Marinschek <mmarinsc...@apache.org>:
>>>> >> Hi Leo,
>>>> >>
>>>> >> how is 4 better than 2?
>>>> >>
>>>> >
>>>> > I was thinking on a scenario where some user want some other feature
>>>> > in myfaces-commons-resourcehandler like gzip compression, i18n locale
>>>> > appended to request path, library relocation of provide a custom
>>>> > request path pointing to a Content Delivery Network or just a
>>>> > directory inside the web application, and he/she is not interested in
>>>> > solution to the issue presented before.
>>>> >
>>>> > In such case, suffix mapping alone should work. Note 2 requires a
>>>> > prefixed mapping (note the assumption that /faces), but 4 does not
>>>> > enforce that, so it will work on both prefix and suffix mapping, but
>>>> > if you want a solution for the previous problem you just add the
>>>> > filter and problem solved. A filter is a simple solution to implement,
>>>> > and it will work without problem in any scenario. Note in this case
>>>> > the filter will be used only when suffix mapping is used.
>>>> >
>>>> > best regards,
>>>> >
>>>> > Leonardo Uribe
>>>> >
>>>> >> 2 is my preferred option, 3 if someone has the time to invest in
>>>> this.
>>>> >> I don't see the additional value of 4.
>>>> >>
>>>> >> best regards,
>>>> >>
>>>> >> Martin
>>>> >>
>>>> >> On 6/30/11, Leonardo Uribe <lu4...@gmail.com> wrote:
>>>> >>> +1 for 3.
>>>> >>>
>>>> >>> Option 4. doesn't cause any conflict, so we can just keep that code
>>>> as is.
>>>> >>>
>>>> >>> 2011/6/30 Leonardo Uribe <lu4...@gmail.com>:
>>>> >>>> Hi
>>>> >>>>
>>>> >>>> To reference images inside a css file in JSF 2.0, users have to
>>>> change
>>>> >>>> its code from this:
>>>> >>>>
>>>> >>>> .someclass
>>>> >>>> {
>>>> >>>> ...
>>>> >>>>    background-image:url('myimage.gif');
>>>> >>>> ...
>>>> >>>> }
>>>> >>>>
>>>> >>>> to this:
>>>> >>>>
>>>> >>>> .someclass
>>>> >>>> {
>>>> >>>> ...
>>>> >>>>    background-image:url(#{resource['mylib:myimage.gif']});
>>>> >>>> ...
>>>> >>>> }
>>>> >>>>
>>>> >>>> This means a lot of changes, including override css files and copy
>>>> >>>> images to other locations.
>>>> >>>>
>>>> >>>> Some months ago, a new module was added in MyFaces commons, with
>>>> the
>>>> >>>> objective of handle that problem in a gracefully way (just don't
>>>> >>>> change anything on the css file and make JSF load the images). But
>>>> >>>> there are different points of view about how to handle it on the
>>>> >>>> implementation of that module.
>>>> >>>>
>>>> >>>> Things works well when prefix mapping is used for FacesServlet. But
>>>> >>>> with suffix mapping, by default all resources have an additional
>>>> >>>> suffix added on its request path. So, the resource url looks
>>>> something
>>>> >>>> like this:
>>>> >>>>
>>>> >>>> http://
>>>> [server][port]/[webapp]/javax.faces.resource/mylib/image.gif.jsf
>>>> >>>>
>>>> >>>> breaking the css file.
>>>> >>>>
>>>> >>>> The intention is allow suffix mapping for jsf files, but prefix
>>>> >>>> mapping for resource links. There are the following alternatives:
>>>> >>>>
>>>> >>>> 1. Enforce prefix mapping for jsf applications using this module
>>>> and
>>>> >>>> do not support suffix mapping at all.
>>>> >>>>
>>>> >>>> 2. Add a prefix mapping entry for FacesServlet and create a web
>>>> config
>>>> >>>> init param to indicate that mapping will be used to handle
>>>> resources.
>>>> >>>> If such param is no present, assume "/faces" as prefix mapping
>>>> used.
>>>> >>>>
>>>> >>>> 3. Add a prefix mapping entry for FacesServlet and detect it
>>>> >>>> automatically, parsing web.xml file and in servlet 3.0 use
>>>> >>>> ServletRegistration. A web config init param anyway should be
>>>> provided
>>>> >>>> for handle portlets case.
>>>> >>>>
>>>> >>>> 4. Use a special filter and detect if was setup automatically,
>>>> looking
>>>> >>>> on application map if the filter was set or not and a web config
>>>> init
>>>> >>>> param to know the mapping used, without parse xml files or servlet
>>>> 3.0
>>>> >>>> specific code.
>>>> >>>>
>>>> >>>> Please vote about which one you think is the best alternative, and
>>>> >>>> should be done in that module.
>>>> >>>>
>>>> >>>> Please note: This vote is "majority approval" over the choice
>>>> selected
>>>> >>>> (see [1]).
>>>> >>>>
>>>> >>>> Thanks,
>>>> >>>> Leonardo Uribe
>>>> >>>>
>>>> >>>> [1] http://www.apache.org/foundation/voting.html#ReleaseVotes
>>>> >>>>
>>>> >>>
>>>> >>
>>>> >>
>>>> >> --
>>>> >>
>>>> >> http://www.irian.at
>>>> >>
>>>> >> Your JSF powerhouse -
>>>> >> JSF Consulting, Development and
>>>> >> Courses in English and German
>>>> >>
>>>> >> Professional Support for Apache MyFaces
>>>> >>
>>>> >
>>>>
>>>>
>>>>
>>>> --
>>>> Jakob Korherr
>>>>
>>>> blog: http://www.jakobk.com
>>>> twitter: http://twitter.com/jakobkorherr
>>>> work: http://www.irian.at
>>>>
>>>
>>>
>>
>
>
> --
> Rudy De Busscher
> http://www.c4j.be
>
>

Reply via email to