I'm using for my projects a rather complete UI auto-generation tool from database entities for windows forms and asp.net I wrote. Now I've built a working prototype UI auto-generation tool from domain objects. Right now it is in early stage of development and by reflection it generates user interface for creating and updating domain objects.
I searched a bit and I didn't find other opensource projects that have the same goal. Why? This type of tool is not useful? Is this idea fundamentally flawed?
Thanks.
Edit:
In my actual implementation you can drive the ui generation by putting predetermined attributes in the fields, properties and classes. You can decide the position, if a field is visible, if it is the primary key of that object, the default method to get a collection of that object type, the max size, and so on...
There is no reason at all not to re-invent the wheel.
Seriously.
This is one of the best ways to learn.
Take an honest run at it, take surface looks at other implementations.
Once you arrive at either a) a beta that does what you want or b) a deep dark corner where there be dragons - pull and read the source for an existing implementation.
Not only will you have a better perspective for appreciating the code, you will understand better why some things are done certain ways.
There is no reason at all not to re-invent the wheel.