Fix error messages in variables.py

I've focused on the messages that are relevant to resource variables or migration rather than the 1.x-only symbols / reference variables.

Removes an error message for tf.Variable.__iter__. It now just reads from the variable and iterates over the result, which is what you'd expect it to do. I think that's better than trying to fix an error message that shouldn't exist in the first place.

Fixes tensorflow#23185.

PiperOrigin-RevId: 395965288
Change-Id: Id5ff6ed2545c4ef5e67acccf3fcbc4094d289e4f

tensorflow/python/distribute/values_v2_test.py

@@ -243,8 +243,7 @@ def testEqualityGraph(self):
 
   def testIteration(self):
     v = self.create_variable([1.])
-    with self.assertRaises(TypeError):
-      v.__iter__()
+    self.assertEqual([1.], list(iter(v)))
 
   def testProperties(self):
     v = self.create_variable()

tensorflow/python/kernel_tests/resource_variable_ops_test.py

@@ -1600,5 +1600,11 @@ def _create_and_delete_variable():
     checker.report()
     checker.assert_no_leak_if_all_possibly_except_one()
 
+  @test_util.run_v2_only
+  def testIterateVariable(self):
+    v = variables.Variable([1., 2.])
+    self.assertAllClose([1., 2.], list(iter(v)))
+
+
 if __name__ == "__main__":
   test.main()

tensorflow/python/kernel_tests/variables_test.py

@@ -121,14 +121,25 @@ def testCyclicInitializer(self):
           "test", cyclic)
       self.assertIs(initial_value, cyclic)
 
-  def testIterable(self):
-    with self.assertRaisesRegex(TypeError, "not iterable"):
+  @test_util.run_deprecated_v1
+  def testIterableV1(self):
+    with self.assertRaisesRegex(TypeError, "not allowed in Graph"):
       for _ in variables.Variable(0.0):
         pass
-    with self.assertRaisesRegex(TypeError, "not iterable"):
+    with self.assertRaisesRegex(TypeError, "not allowed in Graph"):
       for _ in variables.Variable([0.0, 1.0]):
         pass
 
+  @test_util.run_v2_only
+  def testIterableV2(self):
+    with self.assertRaisesRegex(TypeError, "scalar tensor"):
+      for _ in variables.Variable(0.0):
+        pass
+    values = []
+    for v in variables.Variable([0.0, 1.0]):
+      values.append(v)
+    self.assertAllClose([0., 1.], values)
+
   @test_util.run_deprecated_v1
   def testAssignments(self):
     with self.cached_session():

tensorflow/python/ops/variables.py

@@ -1038,8 +1038,8 @@ def _TensorConversionFunction(v, dtype=None, name=None, as_ref=False):  # pylint
     _ = name
     if dtype and not dtype.is_compatible_with(v.dtype):
       raise ValueError(
-          "Incompatible type conversion requested to type '%s' for variable "
-          "of type '%s'" % (dtype.name, v.dtype.name))
+          f"Incompatible type conversion requested to type '{dtype.name}' for "
+          f"variable of type '{v.dtype.name}' (Variable: {v}).")
     if as_ref:
       return v._ref()  # pylint: disable=protected-access
     else:
@@ -1082,8 +1082,9 @@ def _run_op(a, *args, **kwargs):
 
   def __hash__(self):
     if ops.Tensor._USE_EQUALITY and ops.executing_eagerly_outside_functions():  # pylint: disable=protected-access
-      raise TypeError("Variable is unhashable. "
-                      "Instead, use tensor.ref() as the key.")
+      raise TypeError(
+          "Variable is unhashable. "
+          f"Instead, use variable.ref() as the key. (Variable: {self})")
     else:
       return id(self)
 
@@ -1106,18 +1107,8 @@ def __ne__(self, other):
       return self is not other
 
   def __iter__(self):
-    """Dummy method to prevent iteration.
-
-    Do not call.
-
-    NOTE(mrry): If we register __getitem__ as an overloaded operator,
-    Python will valiantly attempt to iterate over the variable's Tensor from 0
-    to infinity.  Declaring this method prevents this unintended behavior.
-
-    Raises:
-      TypeError: when invoked.
-    """
-    raise TypeError("'Variable' object is not iterable.")
+    """When executing eagerly, iterates over the value of the variable."""
+    return iter(self.read_value())
 
   # NOTE(mrry): This enables the Variable's overloaded "right" binary
   # operators to run when the left operand is an ndarray, because it
@@ -1783,7 +1774,9 @@ def _init_from_args(self,
       # Ensure that we weren't lifted into the eager context.
       if context.executing_eagerly():
         raise RuntimeError(
-            "RefVariable not supported when eager execution is enabled. ")
+            "Reference variables are not supported when eager execution is "
+            "enabled. Please run `tf.compat.v1.enable_resource_variables()` to "
+            "switch to resource variables.")
       with ops.name_scope(name, "Variable",
                           [] if init_from_fn else [initial_value]) as name: