ASSERT_TRUE(SparsifyModel(input_model, &output_builder).ok());
  auto output_fbm = tflite::FlatBufferModel::BuildFromBuffer(
      reinterpret_cast<const char*>(output_builder.GetCurrentBufferPointer()),
      output_builder.GetSize());
  tflite::ModelT output_model;
  output_fbm->GetModel()->UnPackTo(&output_model);

  // Extract output metadata
  std::map<std::string, std::string> output_metadata;

  tflite::FinishModelBuffer(input_builder, input_model_location);

  std::string serialized_model(
      reinterpret_cast<const char*>(input_builder.GetBufferPointer()),
      input_builder.GetSize());

  OwningOpRef<mlir::ModuleOp> module = tflite::FlatBufferToMlir(
      serialized_model, &context, UnknownLoc::get(&context));
  if (!module) {
    LOG(ERROR) << "Couldn't import flatbuffer to MLIR.";

	defer closeDisk()

	s := folderScanner{
		root:                  basePath,
		getSize:               getSize,
		oldCache:              cache,
		newCache:              dataUsageCache{Info: cache.Info},
		updateCache:           dataUsageCache{Info: cache.Info},
		dataUsageScannerDebug: false,
		healObjectSelect:      0,
		scanMode:              scanMode,

      fb_builder.CreateVector(subgraphs.begin(), subgraphs.size()));
  fb_builder.Finish(metadata);

  return std::string(
      reinterpret_cast<const char*>(fb_builder.GetBufferPointer()),
      fb_builder.GetSize());
}

void Verify(const RuntimeMetadata* result, const RuntimeMetadata* expected) {
  EXPECT_EQ(result->subgraph_metadata()->size(),
            expected->subgraph_metadata()->size());

        }
        return c;
    }

    private Conflict registerConflict(K target, K replacedBy) {
        return registerConflict(singletonList(target), replacedBy);
    }

    public int getSize() {
        return conflicts.size();
    }

    public Conflict popConflict() {
        assert !conflicts.isEmpty();
        Conflict conflict = conflicts.pop();

  tflite::FinishModelBuffer(input_builder, input_model_location);

  std::string serialized_model(
      reinterpret_cast<const char*>(input_builder.GetBufferPointer()),
      input_builder.GetSize());

  OwningOpRef<mlir::ModuleOp> module = tflite::FlatBufferToMlir(
      serialized_model, &context, UnknownLoc::get(&context));

  // Apply quantization passes.

                            fb_builder.CreateVector(subgraphs_metadata));
  fb_builder.Finish(runtime_metadata);
  return std::string(
      reinterpret_cast<const char*>(fb_builder.GetBufferPointer()),
      fb_builder.GetSize());
}

  if (!status.ok()) {
    error_reporter->exception();
    return nullptr;
  }
  return tflite::python_utils::ConvertToPyString(
      reinterpret_cast<const char*>(builder.GetCurrentBufferPointer()),
      builder.GetSize());
}

PyObject* RegisterCustomOpdefs(PyObject* list) {
  if (!PyList_Check(list)) {
    PyErr_SetString(PyExc_TypeError, "Expected list in argument");
    return nullptr;
  }

  // check if Flatbuffer builder can no longer hold the given amount of the data
  inline bool IsModelBiggerThan2GB(const uint64_t data_size) {
    return data_size > flatbuffer_size_max - builder_.GetSize();
  }

  // helper function for build stablehlo operators
  std::optional<BufferOffset<tflite::Operator>>
  BuildStablehloOperatorwithoutOptions(Operation* inst,

                            tflite::Model::Pack(input_builder, model));

  const std::string input_buffer(
      reinterpret_cast<const char*>(input_builder.GetBufferPointer()),
      input_builder.GetSize());
  auto status = mlir::lite::QuantizeModel(
      input_buffer, input_type, output_type, inference_tensor_type,
      /*operator_names=*/{}, disable_per_channel, fully_quantize, output_buffer,