Simon,

On Wed, Aug 13, 2008 at 10:53 AM, Simon Kitching <[EMAIL PROTECTED]> wrote:
> Hi,
>
> Just a reminder of the email I posted a few weeks ago: I'm still very
> concerned about the recent ExtensionsFilter refactoring.
>
> The new code now parses the web.xml to extract some configuration
> information. This just feels wrong to me.
> The new code also is based around a custom FacesContext wrapper. I'm worried
> about possible interactions between this and other libraries that also wrap
> FacesContext. And I'm worried about the portability of this code to the
> upcoming JSF2.0.
>
> I know this stuff is needed to get tomahawk useable for portlets. But it's
> very important not to break existing users.
>
> What I would like to see is:
> * separate out the resource-serving stuff into a separate utilities module,
> that can be called from either a filter or a FacesContext.
>  And call the common code from both ExtensionsFilter and
> TomahawkFacesContextWrapper.
> * have the TomahawkFacesContextFactory create a FacesContext wrapper only in
> the case where the filter is not configured, ie
>  look in the request-scope for a flag placed there by the filter. That means
> there is no chance of breaking existing setups where
>  the filter is defined explicitly.
> * ideally, also have some way of turning off the faces-context wrapping even
> when the filter is not present.
> * don't parse the web.xml at all. I know you said earlier that it is needed,
> but I just don't see why. In any case, I think we *must*
>  find some other solution; the current approach is really hard to maintain.
>
> Until this is fixed, I would definitely vote -1 on any tomahawk release.
> It's a major issue.

was there a jira issue for it ?
or can you point me to the svn rev?

-M

>
> Regards, Simon
>
>



-- 
Matthias Wessendorf

Need JSF and Web 2.0?
http://code.google.com/p/facesgoodies

further stuff:
blog: http://matthiaswessendorf.wordpress.com/
sessions: http://www.slideshare.net/mwessendorf
mail: matzew-at-apache-dot-org

Reply via email to