I'm sorry I have not been able to reply before now, we have been working on the RichFaces 4.0.0.CR1 code freeze.
The research I did was based on looking at the special tomcat 7 connector that MyFaces currently uses, and the one that the Weld team did to work around this problem ( links in jira ). I also did a lot of searching to see if there was some setting or configuration I was missing tomcat 7 to turn it back on, but it did not look like it. I think the real answer is going to be MyFaces creating their own instance manager that handles the injections - much like what Weld did. -Jay On Fri, Feb 18, 2011 at 11:39 AM, Scott O'Bryan <[email protected]> wrote: > Yeah, I agree Jay. That looks like a pretty big issue. You said they > changed how annotations are processed in Tomcat 7. I'm not on that > list but can you tell me how? Also is there an equivalent ticket over > in the Tomcat Queue? > > On Feb 18, 2011, at 6:59 AM, "[email protected]" <[email protected]> wrote: > > > Hi Guys, > > > > We are working through some last minute issues before we release > RichFaces > > 4. On tomcat 7 with MyFaces 2.0.4 we have been running into some issues > > with annotation processing. > > > > Please see https://issues.apache.org/jira/browse/MYFACES-3035 for some > > details. It basically seems that @PostConstruct is being called before > > managed properties are processed. This is obviously causing big issues. > > > > It is only with tomcat 7 since they changed the way they handle > annotations. > > Unless I'm missing a work around, or something else this seems like a > show > > stopper for MyFaces on Tomcat 7. > > > > Has anyone else had this issue, or know of a way around it? I'd also > like > > to discuss any plans for reviewing this issue. > > > > Thanks, > > Jay Balunas > > RichFaces project lead > > > > -- > > blog: http://in.relation.to/Bloggers/Jay > -- blog: http://in.relation.to/Bloggers/Jay
