Hi Matthias, Thanks for taking a look at the patch. When I apply it to my local copy of current12/core I am able to use JSF in a JSP without breaking any JSTL apps. Without the patch JSPs that use JSTL without also using JSF produce the NullPointerException mentioned earlier. In that case org.apache.myfaces.application.ApplicationImpl has created several resolvers that participate in JSTL's variable resolution while assuming that the MyFaces context factory has already put a FacesContext into the ELContext.
Does the MyFaces project have a private list where details about TCK are discussed that I could subscribe to? Alternatively I can provide you with the address for Geronimo's TCK list that you can subscribe to if you have signed the NDA. Best wishes, Paul On 2/17/07, Matthias Wessendorf <[EMAIL PROTECTED]> wrote:
Paul, looks ok for me. What are the TCK tests saying ? Better now ? Also are the full JSF TCK tests part of the Java EE TCK ? -M On 2/16/07, Paul McMahan <[EMAIL PROTECTED]> wrote: > On 2/16/07, Paul McMahan <[EMAIL PROTECTED]> wrote: > > Question 2: > > After integrating MyFaces into Geronimo's Tomcat container (v6.0.8) I > > found that I could deploy JSF applications and pass many tests. > > However, I found that several of the JSTL tests no longer worked. It > > seems that when MyFaces has been enabled it assumes that all the JSPs > > in a web app will produce a FacesContext. But that's not the case for > > JSPs like the following: > [...] > > I created MYFACES-1536 for this and attached a patch to it that checks > for null after calling elContext.getContext(FacesContext.class). > Please commit if it looks ok or advise if it needs more work. > > Best wishes, > Paul > -- Matthias Wessendorf http://tinyurl.com/fmywh further stuff: blog: http://jroller.com/page/mwessendorf mail: mwessendorf-at-gmail-dot-com
