switch in.Stack.Text() {
		case "else", "endif", "ifdef", "ifndef", "line":
			// Press on.
		default:
			return false
		}
	}
	switch in.Stack.Text() {
	case "define":
		in.define()
	case "else":
		in.else_()
	case "endif":
		in.endif()
	case "ifdef":
		in.ifdef(true)
	case "ifndef":
		in.ifdef(false)
	case "include":
		in.include()
	case "line":
		in.line()

// TFE_ContextEndStep while ops that rely on the step container may be running.
TF_CAPI_EXPORT extern void TFE_ContextEndStep(TFE_Context* ctx);

#ifdef __cplusplus
} /* end extern "C" */
#endif

#ifdef __cplusplus
// A workaround to ease conversion to and from numpy objects and
// TFE_TensorHandle's.
//
// TODO(ashankar): Figure out an alternative scheme that precludes the need for

// Experimental C API for TensorFlow.
//
// The API here is subject to changes in the future.
// --------------------------------------------------------------------------

#ifdef __cplusplus
extern "C" {
#endif

// When `enable` is true, set
// tensorflow.ConfigProto.OptimizerOptions.global_jit_level to ON_1, and also
// set XLA flag values to prepare for XLA compilation. Otherwise set

limitations under the License.
==============================================================================*/
#ifndef TENSORFLOW_C_EAGER_C_API_EXPERIMENTAL_H_
#define TENSORFLOW_C_EAGER_C_API_EXPERIMENTAL_H_

#include "tensorflow/c/c_api.h"
#include "tensorflow/c/c_api_macros.h"
#include "tensorflow/c/eager/c_api.h"

#ifdef __cplusplus
extern "C" {
#endif

// Resets `op_to_reset` with `op_or_function_name` and `raw_device_name`. This

See the License for the specific language governing permissions and
limitations under the License.
==============================================================================*/
#ifndef TENSORFLOW_C_EAGER_IMMEDIATE_EXECUTION_CONTEXT_H_
#define TENSORFLOW_C_EAGER_IMMEDIATE_EXECUTION_CONTEXT_H_

#include <functional>
#include <map>
#include <memory>
#include <vector>

                              TF_Status* status) {
  status->status = MessageToBuffer(func->record->fdef(), output_func_def);
}

TF_Function* TF_FunctionImportFunctionDef(const void* proto, size_t proto_len,
                                          TF_Status* status) {
  tensorflow::FunctionDef fdef;
  bool success = fdef.ParseFromArray(proto, proto_len);
  if (!success) {
    status->status = InvalidArgument(

        text_proto);
    return {};
  }
  std::vector<UniqueFuncPtr> ret;
  for (const FunctionDef& fdef : fdef_lib.function()) {
    // Make a copy so that we can mutate it.
    FunctionDef fdef_to_load = fdef;
    if (mutate_proto_func) {
      (*mutate_proto_func)(&fdef_to_load);
    }
    VLOG(1) << "Adding func to graph: " << fdef_to_load.DebugString();

                                   graph_outputs.size(), graph_outputs.data(),
                                   nullptr, nullptr, name_.data(), s);
    *f = new GraphFunction(std::move(func->record->fdef()));
    TF_DeleteFunction(func);
    TF_RETURN_IF_ERROR(StatusFromTF_Status(s));
    TF_DeleteStatus(s);
    return absl::OkStatus();
  }

  absl::Status RegisterFunction(AbstractFunction* func) override {

		for delKey := range strings.FieldsSeq(inputs[1]) {
			_, ok := currKVS.Lookup(delKey)
			if !ok {
				return Error[ErrConfigNotFound]("key %s doesn't exist", delKey)
			}
			defVal, isDef := defKVS.Lookup(delKey)
			if isDef {
				currKVS.Set(delKey, defVal)
			} else {
				currKVS.Delete(delKey)
			}
		}
		c[subSys][tgt] = currKVS
	} else {
		delete(c[subSys], tgt)
	}
	return nil
}