I discussed the whole SCSF recipe issue with Ward. His summary of the result:
Some SCSF recipes will work against Cabana; some won't.
I think adding a business module should work. I suspect that this recipe is looking for something that isn't there - that the recipe is broken. For example, I have no idea what "CommonProject" is; it is NOT among the projects generated by the 2006 SCSF (I just looked). Maybe he neglected a value in the wizard? In any case, he may want to test the recipe against a clean SCSF generated application before turning to Cabana.
Looks like you've successfully accomplished this --Bill J.
But read on ... because the real problem here is that recipes should not be used at all.
[BTW, this output is just ONE of the reasons to avoid recipes - you have NO idea what's going wrong. At least with templates the developer stands a chance]
There are no SCSF recipes worth running. That sounds strong but I can't think of one that I would want to run that produces anything I should want.
Consider the recipe in question. It adds a business module in a manner that would not be helpful.
If I remember correctly, it creates a new project, puts classes in it called Module and ModuleController, and registers the module in the ProfileCatalog. That is barely a start. Such modules don't look like Cabana modules and don't know how to hook into our navigation (or any navigation for that matter).
To take another example, the view recipes produce files that derive from UserControl and the scsf Presenter<T> class ... bad ideas both.
We haven't tackled the intricacies of GAT recipe development because ... they're ridiculously complex and hard to maintain. The recipes as shipped are largely useless (or counterproductive) so I haven't bothered trying to make Cabana work with them.
[I'm not trying to be dismissive of the SCSF recipes. They were offered as a demo. No one should build a real application that way. It was always intended that the application architect would develop recipes appropriate to the application. Unfortunately, that has turned out to be prohibitively difficult to do. The guys at Cambar have made some progress there but I came away more convinced then ever that one should avoid GAT recipe building at this stage of their evolution.]
We're hoping the forthcoming Visual Studio templates answer the need.
So there!
BTW, I suggest you take a look at our new templates/wizards (available real soon now) before deciding if they're useful.
Thanks,
Bill J.