# itself is verified in the common graphdef converter, so here just assert
# it is being invoked.
# CHECK: module
# CHECK-SAME: tf.versions
# CHECK-SAME: bad_consumers
# CHECK-SAME: min_consumer
# CHECK-SAME: producer

# CHECK:      func {{@[a-zA-Z_0-9]+}}(
# CHECK-SAME:   [[ARG:%.*]]: tensor<*xf32> {tf_saved_model.index_path = ["x"]}

# CHECK: [[shape:%.*]] = "tf.Shape"([[ARG]])

// CKECK-SAME: <{calibration_method = 5 : i32, id = "skipping_id", num_bins = 32 : i32, max_percentile = 0.000000e+00 : f32, min_percentile = 0.000000e+00 : f32}> : (tensor<1x3x4x3xf32>) -> (tensor<1x3x4x3xf32>, tensor<f32>, tensor<f32>, tensor<512xi64>)
// CHECK: %[[CUSTOM_AGGREGATOR_1:.*]], %[[MIN_1:.*]], %[[MAX_1:.*]], %[[HISTOGRAM_1:.*]] = "tf.CustomAggregator"

class ChannelIdentifierTest extends Specification {
    def "equals and hash code"() {
        def id = new ChannelIdentifier("channel")
        def same = new ChannelIdentifier("channel")
        def different = new ChannelIdentifier("other")

        expect:
        id == same
        id.hashCode() == same.hashCode()
        id != different
        id != null
        id != "channel"
    }

# CHECK: tf_executor._SwitchN
# CHECK-SAME: of 3 : tensor<*xi32>
# CHECK-SAME: T = i32
# CHECK-SAME: loc(fused["_SwitchN:", "Case/branch_index/_3"])
# CHECK: tf_executor._SwitchN
# CHECK-SAME: of 2 : tensor<*xf32>
# CHECK-SAME: T = f32
# CHECK-SAME: loc(fused["_SwitchN:", "Case/Case/input_0/_7"])

node {
  name: "Case/branch_index"
  op: "Const"
  attr {

# CHECK:      func {{@[a-zA-Z_0-9]+}}(
# CHECK-SAME:   [[ARG0:%.*]]: tensor<!tf_type.resource<tensor<5x3xf32>>> {tf_saved_model.bound_input = @[[VAR0]]},
# CHECK-SAME:   [[ARG1:%.*]]: tensor<!tf_type.resource<tensor<3x5xf32>>> {tf_saved_model.bound_input = @[[VAR1]]})
# CHECK-SAME:             -> (tensor<5x5xf32> {{{.*}}})
# CHECK-SAME: attributes {{.*}} tf_saved_model.exported_names = ["key"]

namespace tensorflow {
namespace ops {

// Return a tensor with the same shape and contents as the input tensor or
// value.
Status Identity(AbstractContext* ctx, AbstractTensorHandle* const input,
                AbstractTensorHandle** output, const char* name = nullptr,
                const char* raw_device_name = nullptr);

// Returns a list of tensors with the same shapes and contents as the input
Status IdentityN(AbstractContext* ctx,

  // Check that Conv2D computed in NCHW format, and all redundant transpose
  // operations removed from the function.

  // CHECK: %[[CONV:[0-9]*]] = "tf.Conv2D"(%arg0, %arg1)
  // CHECK-SAME: data_format = "NCHW"
  // CHECK-SAME: -> tensor<1x8x32x32xf32>

  // CHECK: return %[[CONV]]

  func.return %4 : tensor<1x8x32x32xf32>

}

// -----

// CHECK-LABEL: func @no_args
// CHECK-SAME: (%arg0: tensor<!tf_type.resource<tensor<f32>>> {tf.resource_name = "x"})
// CHECK-NOT: "tf.VarHandleOp"
func.func @no_args() {
  %0 = "tf.VarHandleOp"() {container = "", shape = "tfshape$", shared_name = "x"} : () -> tensor<!tf_type.resource<tensor<f32>>>
  func.return
}

// CHECK-LABEL: func @some_args

    value {
      type: DT_INT32
    }
  }
}
versions {
  producer: 27
}

# CHECK-LABEL: func @main
# CHECK-SAME:  (%[[ARG_0:[a-z0-9]+]]: tensor<10xi32>, %[[ARG_1:[a-z0-9]+]]: tensor<10xi32>) -> tensor<*xi32>
# CHECK-SAME:  control_outputs = ""
# CHECK-SAME:  inputs = "input0,input1"
# CHECK-SAME:  outputs = "Sub"

    func.return
  }

  // CHECK-LABEL: func @main
  // CHECK-SAME:    %[[ARG0:.*]]: tensor<10xi32>, %[[ARG1:.*]]: tensor<10xi32>
  func.func @main(%arg0: tensor<10xi32>, %arg1: tensor<10xi32>) -> tensor<10xi32> {
    // CHECK:      %[[RESULT:.*]] = "tf.XlaCallModule"(%[[ARG0]], %[[ARG1]])
    // CHECK-NOT:    function_list
    // CHECK-SAME:   module = ""
    // CHECK-SAME:   _entry_function = @main_0,

    // `module` is stablehlo bytecode for: