I hope that leo can help here. Am Donnerstag, 28. Dezember 2017 schrieb Paul Nicolucci :
> Any updates here? We should work to make a decision on this soon so we can > get 2.3.0 completed. > > Any thoughts on what changes are required? > > Thanks, > > Paul Nicolucci > > > [image: Inactive hide details for Thomas Andraschko ---12/09/2017 03:08:38 > PM---I asked why this feature is actually needed and become]Thomas > Andraschko ---12/09/2017 03:08:38 PM---I asked why this feature is actually > needed and become further feedback: The idea is that OmniFaces > > From: Thomas Andraschko <[email protected]> > To: MyFaces Development <[email protected]> > Date: 12/09/2017 03:08 PM > Subject: Re: Dev Discussion - JSF 2.3 ResourceVisitOption.TOP_LEVEL_VIEWS_ONLY > different between MyFaces and Mojarra > ------------------------------ > > > > I asked why this feature is actually needed and become further feedback: > > The idea is that OmniFaces will eventually throw out it's now proprietary > FacesViews feature in favour of the standardised APIs, and perhaps will > only add some extra things. > Perhaps it's possible to add a clarification to the spec for JSF 2.4 about > the intention of the various view hints, with some examples etc. > > 2017-12-08 20:28 GMT+01:00 Thomas Andraschko < > *[email protected]* <[email protected]>>: > > I talked with Arjan about that topic. > Here is the statement: > > I: > How was it desinged? I actually agree with Leo and the MyFaces impl > that views are actually only files that can be served by the browser and > they must be placed inside the webapp directory. Templates, includes and > else can be placed in jars but is not a "view" actually. > WDYT? > > Arjan: > The above is not entirely correct. The design (contract) is that it > returns whatever the installed VDL(s) recognise as a view. The feature > should delegate to the VDL and negotiate with that. For instance, the > default Facelet VDL is able to load views from a jar since JSF 2.2. > > Another custom VDL might load views from a database, from an external > folder, or what have you. > > So the idea is not to make any assumptions about where views van > reside or not, but ask the VDL, and return whatever the VDL supports. > > > WDYT Leo? > > 2017-11-24 2:00 GMT+01:00 Leonardo Uribe <*[email protected]* > <[email protected]>>: > Hi > > I think MyFaces behavior is correct here. The reason is you will never > add views inside META-INF or WEB-INF folders, but you could add templates > there. But a template is not a view. That is what I understand with "top > level views". > > regards, > > Leonardo Uribe > > 2017-11-23 19:41 GMT-05:00 Thomas Andraschko < > *[email protected]* <[email protected]>>: > I think we should align myfaces here. A issue + patch would be > great. > > > Am Samstag, 18. November 2017 schrieb Paul Nicolucci : > The javadoc for ResourceVisitOption.html says the following: > > *https://javaee.github.io/javaee-spec/javadocs/javax/faces/application/ResourceVisitOption.html* > > <https://urldefense.proofpoint.com/v2/url?u=https-3A__javaee.github.io_javaee-2Dspec_javadocs_javax_faces_application_ResourceVisitOption.html&d=DwMFaQ&c=jf_iaSHvJObTbx-siA1ZOg&r=n-2RUhyQkncKQNTGKy9UmSKIHKSZzEVYEqiy1H7hEwA&m=dXsCs4V2Evbu59Npakfxsz9tC6S4wyYSQBgwBj0dmTw&s=HIb7TRcn2ip64lyIIpODzOYa1lrZKkFjl9IwRzRHMUk&e=> > > public static final *ResourceVisitOption* > > <https://urldefense.proofpoint.com/v2/url?u=https-3A__javaee.github.io_javaee-2Dspec_javadocs_javax_faces_application_ResourceVisitOption.html&d=DwMFaQ&c=jf_iaSHvJObTbx-siA1ZOg&r=n-2RUhyQkncKQNTGKy9UmSKIHKSZzEVYEqiy1H7hEwA&m=dXsCs4V2Evbu59Npakfxsz9tC6S4wyYSQBgwBj0dmTw&s=HIb7TRcn2ip64lyIIpODzOYa1lrZKkFjl9IwRzRHMUk&e=> > TOP_LEVEL_VIEWS_ONLY > Only visit resources that are top level views, i.e. views that can > be used to serve a request as opposed to those that can only be used for > includes. > > Thanks, > > Paul Nicolucci > > > [image: Inactive hide details for Thomas Andraschko ---11/18/2017 > 07:22:32 AM---Did you checked the spec texts? 2017-11-17 19:56 > GMT+01]Thomas > Andraschko ---11/18/2017 07:22:32 AM---Did you checked the spec texts? > 2017-11-17 19:56 GMT+01:00 Paul Nicolucci <[email protected]>: > > From: Thomas Andraschko <[email protected]> > To: MyFaces Development <[email protected]> > Date: 11/18/2017 07:22 AM > Subject: Re: Dev Discussion - JSF 2.3 > ResourceVisitOption.TOP_LEVEL_VIEWS_ONLY > different between MyFaces and Mojarra > ------------------------------ > > > > Did you checked the spec texts? > > 2017-11-17 19:56 GMT+01:00 Paul Nicolucci <*[email protected] > <[email protected]>*>: > Hello, > > I was testing out the ResourceHandler.getViewResources() > today and I noticed that we have quite a behavior different > between the two > implementations. > > Take the following application for example: > > testApplication > - /depth2/index.xhtml > -META-INF/index.xhtml > -WEB-INF/index.xhtml > - index.xhtml > - test > > Mojarra getViewResources( call with ResourceVisitOptions ) > /index.xhtml /depth2/index.xhtml > > Mojarra getViewResources ( call without ResourceVisitOptions ) > /index.xhtml /depth2/index.xhtml META-INF/index.xhtml > WEB-INF/index.xhtml > > MyFaces getViewResources( call with ResourceVisitOptions ) > /index.xhtml /depth2/index.xhtml > > MyFaces getViewResources( call without ResourceVisitOptions ) > /index.xhtml /test /depth2/index.xhtml > > In MyFaces if we use the ResourceVisitOptions then we filter > out any views that don't contain a valid suffix ( in the above > case /test > ). In addition MyFaces never returns any views in WEB-INF and > META-INF > > In Mojarra if we use the ResourceVisitOptions then anything > in WEB-INF and META-INF is not included. In addition Mojarra > never returns > any views without a valid suffix. > > I think we need a dev discussion to determine if we want to > stick with our current behavior or change it. > > Thanks, > > Paul Nicolucci > > > > > > > >
