If so, the film should also be textured. In this case you can still do that, even by using regular 2theta-omega scan. However, that will only give you the information of planes that are parallel to the substrate plane, which means information of some peaks will disappear (not grains of all orientations are present since it is textured). In this case you can do Chi scan (which you tilt a substrate before do the scan) or grazing angle Incidence XRD since it will minimize the information you get from the substrate.

it is not impossible. Check Sections 5.7 and 5.8 in the attached review "X-ray analysis of thin films and multilayers" by Paul F. Fewster

What do you mean with XRD analysis? Phase analysis? Structure refinement?

The major problem are the additional uncertainties which are already huge for non-textured materials. But this does not mean that it does not work at all. It is simply limited for specific conditions, i.e. for phases where the overdetermined system of equations is that big that some more parameter are not that dramatic. 

If so, the film should also be textured. In this case you can still do that, even by using regular 2theta-omega scan. However, that will only give you the information of planes that are parallel to the substrate plane, which means information of some peaks will disappear (not grains of all orientations are present since it is textured). In this case you can do Chi scan (which you tilt a substrate before do the scan) or grazing angle Incidence XRD since it will minimize the information you get from the substrate.