`<FieldC>A.C.C</FieldC>` +
			`<EmbedB>` +
			`<FieldB>A.B.B</FieldB>` +
			`<FieldA>` +
			`<A1>A.B.C.A1</A1>` +
			`<A2>A.B.C.A2</A2>` +
			`</FieldA>` +
			`<FieldC>A.B.C.C</FieldC>` +
			`</EmbedB>` +
			`<FieldA>A.A</FieldA>` +
			`<FieldE>A.D.E</FieldE>` +
			`</EmbedA>`,
	},

	// Anonymous struct pointer field which is nil
	{
		Value:     &EmbedB{},

			},
		},
		{name: "preserve unknown fields",
			obj: map[string]interface{}{
				"withUnknown": map[string]interface{}{
					"field1": "a",
					"field2": "b",
				},
				"withUnknownList": []interface{}{
					map[string]interface{}{
						"field1": "a",
						"field2": "b",
					},
					map[string]interface{}{
						"field1": "x",
						"field2": "y",
					},
					map[string]interface{}{

        "tf.Yield"() : () -> ()
        }, {
        "tf.Yield"() : () -> ()
       }) {is_stateless = true} : (tensor<i1>) -> ()
       "tf.Yield"(%1) : (tensor<i1>) -> ()
    }, {
    ^bb0(%arg1: tensor<i32>, %arg2: tensor<?xf32>):
      "tf.IfRegion"(%0) ({
        %2 = "tf.B"() : () -> (tensor<f32>)
        "tf.Yield"() : () -> ()
        }, {
        "tf.Yield"() : () -> ()

       // CHECK: "tf.Yield"(%[[CONST]])
       %constant = "tf.Const"() {value = dense<0.0> : tensor<4xf32>} : () -> tensor<4xf32>
       "tf.AssignVariableOp"(%0, %constant) : (tensor<*x!tf_type.resource<tensor<4xf32>>>, tensor<4xf32>) -> ()
       "tf.Yield"() : () -> ()
      }, {
       // CHECK: "tf.Yield"(%[[READ]])
       "tf.Yield"() : () -> ()

        rng.uniform(low=0.0, high=1.0, size=static_input_shape).astype(
            np.float32
        )
    )

    def data_gen() -> repr_dataset.RepresentativeDataset:
      for _ in range(100):
        yield {
            'input_tensor': rng.uniform(
                low=0.0, high=1.0, size=static_input_shape
            ).astype(np.float32)
        }

    dataset_path = self.create_tempfile('tfrecord').full_path

		}
		_ = stmt.Schema.ParseIndexes()
		field := stmt.Schema.LookUpField(fieldName)
		if field == nil {
			t.Fatalf("%v: failed to find column %q", utils.FileWithLineNum(), fieldName)
		}
		if field.Unique != unique {
			t.Fatalf("%v: %q column %q unique should be %v but got %v", utils.FileWithLineNum(), stmt.Schema.Table, fieldName, unique, field.Unique)
		}
		if field.UniqueIndex != uniqueIndex {

  %graph = tf_executor.graph {
    %island, %ctrl = tf_executor.island {
      %pred = "tf.SomeOp"(%arg2) : (tensor<f32>) -> tensor<i32>
      tf_executor.yield %pred : tensor<i32>
    }
    tf_executor.fetch %island : tensor<i32>
  }
  func.return %graph : tensor<i32>
}

// CHECK-LABEL:   func @simple_independent_chains

  ^bb0(%barg0: tensor<f32>, %barg1: tensor<i32>):
    // CHECK: %[[PRED:.*]] = "tf._SomeOp"(%[[BARG1]])
    // CHECK: "tf.Yield"(%[[PRED]])
    %pred = "tf._SomeOp"(%barg1) : (tensor<i32>) -> tensor<i1>
    "tf.Yield" (%pred) : (tensor<i1>) -> ()
  }, {
  // CHECK: ^bb0(%[[BARG0:.*]]: tensor<f32>, %[[BARG1:.*]]: tensor<i32>):
  ^bb0(%barg0: tensor<f32>, %barg1: tensor<i32>):

module {
func.func @simpleWhile(%arg0: tensor<i32>) -> tensor<i32> {
  %0 = "tfl.while"(%arg0) ({
  ^bb0(%block: tensor<i32>):
    "tfl.yield"(%block) : (tensor<i32>) -> ()
  },{
  ^bb0(%block: tensor<i32>):
    "tfl.yield"(%block) : (tensor<i32>) -> ()
  }) {tac.device = "CPU", fused_activation_function = "RELU6", tac.inference_type = "FLOAT"} : (tensor<i32>) -> tensor<i32>
  func.return %0 : tensor<i32>
}

  regions accept the current value of the iteration variables as inputs.

  The condition region yields a tensor<i1> which, if false, will exit the loop.
  It can also, optionally and additionally, yield the iteration variables, which
  must be unchanged.

  The body region always has to yield the (possibly updated) iteration variables.

  The iteration variables are initialized to the Op input, and the results of the