NF: informative error when trying to slice images

From suggestion by Eric L.

Author: matthew-brett

nibabel/spatialimages.py

@@ -43,6 +43,9 @@
    * instance_to_filename(img, fname) - save ``img`` instance to
      filename ``fname``.
 
+You cannot slice an image, and trying to slice an image generates an
+informative TypeError.
+
 There are several ways of writing data.
 =======================================
 
@@ -734,3 +737,10 @@ def from_image(klass, img):
                      img.affine,
                      klass.header_class.from_header(img.header),
                      extra=img.extra.copy())
+
+    def __getitem__(self):
+        ''' No slicing or dictionary interface for images
+        '''
+        raise TypeError("Cannot slice image objects; consider slicing image "
+                        "array data with `img.dataobj[slice]` or "
+                        "`img.get_data()[slice]`")

nibabel/tests/test_image_api.py

@@ -16,6 +16,7 @@
 * ``img.uncache()`` (``img.get_data()`` is allowed to cache the result of the
   array creation.  If it does, this call empties that cache.  Implement this
   as a no-op if ``get_data()`` does not cache.
+* ``img[something]`` generates an informative TypeError
 """
 from __future__ import division, print_function, absolute_import
 
@@ -234,6 +235,11 @@ def validate_filenames(self, imaker, params):
             assert_almost_equal(img.get_data(), rt_img.get_data())
             del rt_img # to allow windows to delete the directory
 
+    def validate_no_slicing(self, imaker, params):
+        img = imaker()
+        assert_raises(TypeError, img.__getitem__, 'string')
+        assert_raises(TypeError, img.__getitem__, slice(None))
+
 
 class LoadImageAPI(GenericImageAPI):
     # Callable returning an image from a filename