Here, we present detailed experimental results for our evaluation of pretraining data detection methods on code. AUC measures overall discrimination performance, while TPR@5%FPR measures detection sensitivity when the false-positive rate is restricted to 5%. We additionally report confidence intervals for focal AUC comparisons and expanded analysis experiments on CodeLlama-13B.
Embedded Files
- Main AUC Results
The above shows the AUC scores of 8 backbone models in Setting 1(Member-only Mutation.).
The above shows the AUC scores of 8 backbone models in Setting 2(Clone-Type-Aware Mutations).
The above shows the AUC scores of 4 biggest models in each series in Setting 3(Isolated Mutation Impact).
2. Statistical and Low-FPR Results
2. Statistical and Low-FPR Results
The following supplementary tables provide uncertainty estimates for focal AUC differences and report TPR at a fixed 5% false-positive rate. For all TPR tables, higher values indicate that more member samples are correctly detected while limiting false positives to 5%.
Supplementary Table S1. AUC differences and two-sided 95% confidence intervals for the focal RCM, RQ1, and RQ2 comparisons. The order shown in each comparison defines the sign of ΔAUC. Positive ReCaLL − PPL values favor ReCaLL, while negative Hybrid − Original values indicate lower performance under Hybrid mutation. A confidence interval that excludes zero supports the reported direction of the difference. The results support ReCaLL’s advantage over PPL in RQ1, the recovery provided by RCM under Hybrid mutation, and the performance degradation caused by Hybrid mutation in RQ2.
Supplementary Table S2. TPR@5%FPR for RQ1 on original Python functions across eight CodeLLMs and the evaluated detection methods. Rows correspond to target models and columns correspond to detection methods. The final row reports the aggregated values presented in the revised manuscript. A dash for Ref indicates that a suitable target-reference model pair is unavailable. The results show that ReCaLL provides the strongest low-FPR detection signal overall.
Supplementary Table S3. TPR@5%FPR for RQ2 Setting 1 on four representative CodeLLMs. Each panel corresponds to one target model; rows represent detection methods, and columns progress from the original data through individual mutations to the Hybrid mutation. Under Setting 1, only member samples are mutated while non-member samples remain unchanged. The table therefore shows how mutations of member code affect detection sensitivity at a fixed false-positive rate. ReCaLL remains the strongest method overall, although its TPR generally decreases under stronger mutations.
Supplementary Table S4. TPR@5%FPR for RQ2 Setting 2 on four representative CodeLLMs. Each panel corresponds to one target model; rows represent detection methods, and columns represent the evaluated clone definitions and Type-3 similarity thresholds. Membership labels and evaluation sets are constructed according to the corresponding clone definition. ReCaLL continues to provide the strongest low-FPR signal overall, while performance varies across clone types and mutation levels.
3. Additional Evaluation on CodeLlama-13B
3. Additional Evaluation on CodeLlama-13B
To examine whether the RQ3 and RQ4 findings extend beyond the model used in the main analysis, we additionally evaluate CodeLlama-13B using the same code-length ranges, programming languages, and mutation settings.
Supplementary Table S5. AUC results of CodeLlama-13B on original Python functions across five code-length ranges. Rows represent detection methods and columns represent code-length ranges. Reading each row from left to right shows how the method changes as code length increases. Performance generally improves from the shortest range to the next range, but further increases in code length do not consistently improve detection.
Supplementary Table S6. AUC results of CodeLlama-13B on C++ and Java under Settings 1 and 2. The four panels correspond to the two programming languages and two evaluation settings. Rows represent detection methods, columns represent original or mutated conditions, and bold values identify the best result in each column. ReCaLL achieves the highest AUC across all evaluated conditions, while its performance still decreases under member-only mutation. These results are consistent with the cross-language findings reported for StarCoder2-7B.
4. ReCaLL Prefix Examples
4. ReCaLL Prefix Examples
def convert_rate_to_percent(rate: float) -> str: if rate == 1.0: return '+0%' percent = round((rate - 1.0) * 100) if percent > 0: return f'+{percent}%' else: return f'{percent}%'async def _install(self, package: str, *, upgrade: bool) -> bool: if upgrade: return self.run(['rye', 'sync', '--update', *split_packages(package)]) return self.run(['rye', 'add', *split_packages(package)])def flow_warp(feature, flow, mask=False, padding_mode='zeros'): (b, c, h, w) = feature.size() assert flow.size(1) == 2 grid = coords_grid(b, h, w).to(flow.device) + flow return bilinear_sample(feature, grid, padding_mode=padding_mode, return_mask=mask)def test_resolve_array_noop(executing_kernel: Kernel) -> None: del executing_kernel array = ui.array([ui.text(), ui.slider(1, 10)]) registry = get_context().ui_element_registry assert registry.resolve_lens(array._id, {'0': 'hello world'}) == (array._id, {'0': 'hello world'})def _set_initial_state(self, state: State | None, max_iterations: int=MAX_ITERATIONS): if state is None: self.state = State(inputs={}, max_iterations=max_iterations) else: self.state = statedef draw_text_det_res(dt_boxes, img_path): src_im = cv2.imread(img_path) for box in dt_boxes: box = np.array(box).astype(np.int32).reshape(-1, 2) cv2.polylines(src_im, [box], True, color=(255, 255, 0), thickness=2) return src_imdef __init__(self, connection: Catalog, engine_name: Optional[VariableName]=None) -> None: super().__init__(connection, engine_name) self.default_database = self.get_default_database() self.default_schema = self.get_default_schema()def _check_edges(error: Error, expected_edges: Sequence[EdgeWithVar]) -> None: assert isinstance(error, CycleError) assert len(error.edges_with_vars) == len(expected_edges) for edge in expected_edges: assert edge in error.edges_with_varsdef easy_data(csv): from opensora.registry import DATASETS, build_module dataset = build_module({'type': 'VariableVideoTextDataset', 'transform_name': 'resize_crop', 'data_path': csv}, DATASETS) return dataset['0-113-360-640']async def monitor_idle_memory(duration_seconds: int=60, interval_seconds: int=1): end_time = time.time() + duration_seconds while time.time() < end_time: log_current_memory() await asyncio.sleep(interval_seconds)def test_extract_model_and_provider_dot_format(self): model = 'anthropic.claude-3-7' result = extract_model_and_provider(model) assert result['provider'] == 'anthropic' assert result['model'] == 'claude-3-7' assert result['separator'] == '.'def test_read_session_view_no_path(self): view = SessionView() manager = SessionCacheManager(view, None, 0.1) assert manager.read_session_view(SessionCacheKey(codes=tuple(), marimo_version='-1')) == view
The above is the ID prefix of the Python language that we use in the ReCaLL method.
def requires_grad(model, flag=True): """ Set requires_grad flag for all parameters in a model. """ for p in model.parameters(): p.requires_grad = flagdef test_stream_text(client): chat = client.chats.create(model='gemini-1.5-flash') chunks = 0 for chunk in chat.send_message_stream( 'tell me a story in 100 words', ): chunks += 1
assert chunks > 2def fetch_readme() -> str: """ This function reads the README.md file in the current directory.
Returns: The lines in the README file. """ with open("README.md", encoding="utf-8") as f: return f.read()def test_code_execution_result_empty_parts(): response = types.GenerateContentResponse( candidates=[types.Candidate(content=types.Content())] )
assert response.code_execution_result is Nonedef test_t_bytes(client): assert t.t_bytes(client._api_client, _RAW_BYTES) == base64.b64encode( _RAW_BYTES ).decode('ascii') assert t.t_bytes(client._api_client, 'string') == 'string'def fuse_qkv_projections(self) -> None: r"""Enables fused QKV projections.""" self.fusing_transformer = True self.transformer.fuse_qkv_projections()def get_current_weather(location: str) -> str: """Returns the current weather.
Args: location: The location of a city and state, e.g. "San Francisco, CA". """ return 'windy'def timeit(self, name): if name in self.timers: timer = self.timers[name] else: timer = self.dummy_timer
with timer: yielddef get(self) -> T: self.wait() assert self.result is not None if self.result.exception is not None: raise self.result.exception return self.result.valuedef _UrlContext_to_vertex( api_client: BaseApiClient, from_object: Union[dict[str, Any], object], parent_object: Optional[dict[str, Any]] = None,) -> dict[str, Any]: to_object: dict[str, Any] = {}
return to_objectdef get_reflection_pattern_score(res): reflection_pattern_dict = check_reflection_pattern(res) reflection_pattern_num = sum(reflection_pattern_dict.values()) return reflection_pattern_numdef test_list_of_parts(): assert t.t_contents(None, [ types.Part(text='hello'), types.Part(text='world'), ]) == [ types.UserContent(parts=[ types.Part(text='hello'), types.Part(text='world'), ]), ]
assert chunks > 2def fetch_readme() -> str: """ This function reads the README.md file in the current directory.
Returns: The lines in the README file. """ with open("README.md", encoding="utf-8") as f: return f.read()def test_code_execution_result_empty_parts(): response = types.GenerateContentResponse( candidates=[types.Candidate(content=types.Content())] )
assert response.code_execution_result is Nonedef test_t_bytes(client): assert t.t_bytes(client._api_client, _RAW_BYTES) == base64.b64encode( _RAW_BYTES ).decode('ascii') assert t.t_bytes(client._api_client, 'string') == 'string'def fuse_qkv_projections(self) -> None: r"""Enables fused QKV projections.""" self.fusing_transformer = True self.transformer.fuse_qkv_projections()def get_current_weather(location: str) -> str: """Returns the current weather.
Args: location: The location of a city and state, e.g. "San Francisco, CA". """ return 'windy'def timeit(self, name): if name in self.timers: timer = self.timers[name] else: timer = self.dummy_timer
with timer: yielddef get(self) -> T: self.wait() assert self.result is not None if self.result.exception is not None: raise self.result.exception return self.result.valuedef _UrlContext_to_vertex( api_client: BaseApiClient, from_object: Union[dict[str, Any], object], parent_object: Optional[dict[str, Any]] = None,) -> dict[str, Any]: to_object: dict[str, Any] = {}
return to_objectdef get_reflection_pattern_score(res): reflection_pattern_dict = check_reflection_pattern(res) reflection_pattern_num = sum(reflection_pattern_dict.values()) return reflection_pattern_numdef test_list_of_parts(): assert t.t_contents(None, [ types.Part(text='hello'), types.Part(text='world'), ]) == [ types.UserContent(parts=[ types.Part(text='hello'), types.Part(text='world'), ]), ]
The above is the OOD prefix of the Python language that we use in the ReCaLL method.
inline Quat<T> Quat<T>::inv(QuatAssumeType assumeUnit) const{ if (assumeUnit) { return conjugate(); } T norm2 = dot(*this); if (norm2 < CV_QUAT_EPS) { CV_Error(Error::StsBadArg, "This quaternion do not have inverse quaternion"); } return conjugate() / norm2;}
static GMetaArgs getOutMeta(const GMetaArgs &in_meta, const GArgs &in_args) { return getOutMeta_impl(in_meta, in_args, typename detail::MkSeq<sizeof...(Ins)>::type()); }
EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE const TensorCwiseBinaryOp<internal::scalar_difference_op<Scalar>, const Derived, const OtherDerived> operator-(const OtherDerived& other) const { return binaryExpr(other.derived(), internal::scalar_difference_op<Scalar>()); }
inlineT poly_eval( const Polynomials& poly, const T& x ){ typedef typename NumTraits<T>::Real Real;
if( numext::abs2( x ) <= Real(1) ){ return poly_eval_horner( poly, x ); } else { T val=poly[0]; T inv_x = T(1)/x; for( DenseIndex i=1; i<poly.size(); ++i ){ val = val*inv_x + poly[i]; }
return numext::pow(x,(T)(poly.size()-1)) * val; }}
void setMinute(int minute) { if (minute < 0) { throw std::invalid_argument("input minute must be greater than zero"); } if (minute > MAX_MINUTE) { throw std::invalid_argument("input minute must be less than DateTime::MAX_MINUTE"); } minute_ = minute; }
explicit NAPIHandleScope(LEPUSContext *ctx) : env_(nullptr), ctx_(ctx), handle_tail_(nullptr), reset_napi_env(nullptr) { is_gc = ctx_ == nullptr ? false : LEPUS_IsGCMode(ctx_); if (is_gc) { prev_ = reinterpret_cast<NAPIHandleScope *>(GetNapiScope(ctx_)); SetNapiScope(ctx_, this); } }
void UITextArea::FrameDidChanged() { UIBaseInput::FrameDidChanged(); NodeManager::Instance().SetAttributeWithNumberValue(input_node_, NODE_POSITION, 0, 0); NodeManager::Instance().SetAttributeWithNumberValue(input_node_, NODE_WIDTH, width_);}
EIGEN_DEVICE_FUNC inline void compute_inverse_size2_helper( const MatrixType& matrix, const typename ResultType::Scalar& invdet, ResultType& result){ result.coeffRef(0,0) = matrix.coeff(1,1) * invdet; result.coeffRef(1,0) = -matrix.coeff(1,0) * invdet; result.coeffRef(0,1) = -matrix.coeff(0,1) * invdet; result.coeffRef(1,1) = matrix.coeff(0,0) * invdet;}
static void test_basic_eventcount(){ MaxSizeVector<EventCount::Waiter> waiters(1); waiters.resize(1); EventCount ec(waiters); EventCount::Waiter& w = waiters[0]; ec.Notify(false); ec.Prewait(&w); ec.Notify(true); ec.CommitWait(&w); ec.Prewait(&w); ec.CancelWait(&w);}
T get_param(const IndexParams& params, cv::String name){ IndexParams::const_iterator it = params.find(name); if (it != params.end()) { return it->second.cast<T>(); } else { FLANN_THROW(cv::Error::StsBadArg, cv::String("Missing parameter '")+name+cv::String("' in the parameters given")); }}
int GDigit::drawpic(const char* picfn){ if(!m_status)return -1000; std::string picfile(picfn); JMat* mat_pic = NULL; frameSource->popVidRecyle(&mat_pic); if(!mat_pic)mat_pic = new JMat(); int rst = mat_pic->load(picfn); if(rst){ delete mat_pic; return rst*10000; }
MediaData md(mat_pic); frameSource->pushVidFrame(&md);
return 0;}
auto det(const NdArray<dtype>& inArray) { STATIC_ASSERT_ARITHMETIC(dtype);
const Shape inShape = inArray.shape(); if (!inShape.issquare()) { THROW_INVALID_ARGUMENT_ERROR("input array must be square."); }
return detail::det(inArray, inShape.rows); }
static GMetaArgs getOutMeta(const GMetaArgs &in_meta, const GArgs &in_args) { return getOutMeta_impl(in_meta, in_args, typename detail::MkSeq<sizeof...(Ins)>::type()); }
EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE const TensorCwiseBinaryOp<internal::scalar_difference_op<Scalar>, const Derived, const OtherDerived> operator-(const OtherDerived& other) const { return binaryExpr(other.derived(), internal::scalar_difference_op<Scalar>()); }
inlineT poly_eval( const Polynomials& poly, const T& x ){ typedef typename NumTraits<T>::Real Real;
if( numext::abs2( x ) <= Real(1) ){ return poly_eval_horner( poly, x ); } else { T val=poly[0]; T inv_x = T(1)/x; for( DenseIndex i=1; i<poly.size(); ++i ){ val = val*inv_x + poly[i]; }
return numext::pow(x,(T)(poly.size()-1)) * val; }}
void setMinute(int minute) { if (minute < 0) { throw std::invalid_argument("input minute must be greater than zero"); } if (minute > MAX_MINUTE) { throw std::invalid_argument("input minute must be less than DateTime::MAX_MINUTE"); } minute_ = minute; }
explicit NAPIHandleScope(LEPUSContext *ctx) : env_(nullptr), ctx_(ctx), handle_tail_(nullptr), reset_napi_env(nullptr) { is_gc = ctx_ == nullptr ? false : LEPUS_IsGCMode(ctx_); if (is_gc) { prev_ = reinterpret_cast<NAPIHandleScope *>(GetNapiScope(ctx_)); SetNapiScope(ctx_, this); } }
void UITextArea::FrameDidChanged() { UIBaseInput::FrameDidChanged(); NodeManager::Instance().SetAttributeWithNumberValue(input_node_, NODE_POSITION, 0, 0); NodeManager::Instance().SetAttributeWithNumberValue(input_node_, NODE_WIDTH, width_);}
EIGEN_DEVICE_FUNC inline void compute_inverse_size2_helper( const MatrixType& matrix, const typename ResultType::Scalar& invdet, ResultType& result){ result.coeffRef(0,0) = matrix.coeff(1,1) * invdet; result.coeffRef(1,0) = -matrix.coeff(1,0) * invdet; result.coeffRef(0,1) = -matrix.coeff(0,1) * invdet; result.coeffRef(1,1) = matrix.coeff(0,0) * invdet;}
static void test_basic_eventcount(){ MaxSizeVector<EventCount::Waiter> waiters(1); waiters.resize(1); EventCount ec(waiters); EventCount::Waiter& w = waiters[0]; ec.Notify(false); ec.Prewait(&w); ec.Notify(true); ec.CommitWait(&w); ec.Prewait(&w); ec.CancelWait(&w);}
T get_param(const IndexParams& params, cv::String name){ IndexParams::const_iterator it = params.find(name); if (it != params.end()) { return it->second.cast<T>(); } else { FLANN_THROW(cv::Error::StsBadArg, cv::String("Missing parameter '")+name+cv::String("' in the parameters given")); }}
int GDigit::drawpic(const char* picfn){ if(!m_status)return -1000; std::string picfile(picfn); JMat* mat_pic = NULL; frameSource->popVidRecyle(&mat_pic); if(!mat_pic)mat_pic = new JMat(); int rst = mat_pic->load(picfn); if(rst){ delete mat_pic; return rst*10000; }
MediaData md(mat_pic); frameSource->pushVidFrame(&md);
return 0;}
auto det(const NdArray<dtype>& inArray) { STATIC_ASSERT_ARITHMETIC(dtype);
const Shape inShape = inArray.shape(); if (!inShape.issquare()) { THROW_INVALID_ARGUMENT_ERROR("input array must be square."); }
return detail::det(inArray, inShape.rows); }
The above is the ID prefix of the C++ language that we use in the ReCaLL method.
@Override protected Object callWithOneArg(String methodName, String fieldName) { if ("getFromInstance".equals(methodName)) { return "foo"; } throw new UnsupportedOperationException("don't support method: " + methodName); }
private void readAndWrite07(File file) { EasyExcel.write(file, TemplateData.class) .withTemplate(TestFileUtil.readFile("template" + File.separator + "template07.xlsx")).sheet() .doWrite(data()); EasyExcel.read(file, TemplateData.class, new TemplateDataListener()).headRowNumber(3).sheet().doRead(); }
@Test public void hh1() throws IOException { ExcelReader reader = EasyExcel.read(new FileInputStream("D:\\test\\hg2.xls"), LockData.class, new RepeatListener()).headRowNumber(0).build(); ReadSheet r2 = EasyExcel.readSheet(0).build(); reader.read(r2); reader.finish(); }
@Bean @DependsOn({"TaskMappers"}) public MaestroGateTask maestroGateTask( MaestroStepInstanceDao stepInstanceDao, @Qualifier(MAESTRO_QUALIFIER) ObjectMapper objectMapper) { LOG.info("Creating Maestro gateTask within Spring boot..."); return new MaestroGateTask(stepInstanceDao, objectMapper); }
@Test public void testScheduleWithoutDelay() { var future = Mockito.mock(ScheduledFuture.class); when(context.schedule(any(), anyLong())).thenReturn(future);
groupActor.schedule(Action.GROUP_START, 0); verify(context, times(0)).schedule(any(), anyLong()); verifyActions(groupActor, Action.GROUP_START); }
@JsonSetter("actions") public void deserializeActions(final Set<String> actionsStr) { if (actionsStr != null && !actionsStr.isEmpty()) { actions = new HashSet<>(); actionsStr.forEach(s -> actions.add(Action.valueOf(s.toUpperCase(Locale.US)))); } }
@Override protected @Nullable Object executeImpl(VirtualFrame frame) { var repetitions = (long) frame.getArguments()[2]; for (long i = 0; i < repetitions; i++) { blackholeNode.executeGeneric(frame); } LoopNode.reportLoopCount( this, repetitions > Integer.MAX_VALUE ? Integer.MAX_VALUE : (int) repetitions); return null; }
@Test public void isEmpty() { Set<ConstraintViolation<TestId>> violations = validator.validate(new TestId("")); assertEquals(1, violations.size()); ConstraintViolation<TestId> violation = violations.iterator().next(); assertEquals("", violation.getInvalidValue()); assertEquals("[maestro id or name reference] cannot be null or empty", violation.getMessage()); }
public ViewHolder(@NonNull View itemView) { super(itemView); showSystemApps = itemView.findViewById(R.id.showSystemApps); disabled = itemView.findViewById(R.id.disabled); blacklist = itemView.findViewById(R.id.blacklist); }
@Test public void testGetFlowsFromWithError() { Mockito.doThrow(new MaestroInternalError("test")).when(flowDao).getFlows(group, 10, ""); assertNull(context.getFlowsFrom(group, 10, "")); verify(flowDao, times(1)).getFlows(group, 10, ""); }
private StepRuntimeSummary cloneSummary(StepRuntimeSummary summary) { StepAction latestAction = summary.getPendingAction(); StepRuntimeSummary cloned = objectMapper.convertValue(summary, StepRuntimeSummary.class); cloned.setPendingAction(latestAction); return cloned; }
@Test public void testRunForActionStopForActiveRunningTask() { taskActor.runForAction(Action.TASK_STOP); verify(context, times(1)).cancel(any(), any()); assertEquals(Set.of(Action.TASK_STOP), taskActor.getScheduledActions().keySet()); verifyEmptyAction(taskActor); assertTrue(taskActor.isRunning()); }
private void readAndWrite07(File file) { EasyExcel.write(file, TemplateData.class) .withTemplate(TestFileUtil.readFile("template" + File.separator + "template07.xlsx")).sheet() .doWrite(data()); EasyExcel.read(file, TemplateData.class, new TemplateDataListener()).headRowNumber(3).sheet().doRead(); }
@Test public void hh1() throws IOException { ExcelReader reader = EasyExcel.read(new FileInputStream("D:\\test\\hg2.xls"), LockData.class, new RepeatListener()).headRowNumber(0).build(); ReadSheet r2 = EasyExcel.readSheet(0).build(); reader.read(r2); reader.finish(); }
@Bean @DependsOn({"TaskMappers"}) public MaestroGateTask maestroGateTask( MaestroStepInstanceDao stepInstanceDao, @Qualifier(MAESTRO_QUALIFIER) ObjectMapper objectMapper) { LOG.info("Creating Maestro gateTask within Spring boot..."); return new MaestroGateTask(stepInstanceDao, objectMapper); }
@Test public void testScheduleWithoutDelay() { var future = Mockito.mock(ScheduledFuture.class); when(context.schedule(any(), anyLong())).thenReturn(future);
groupActor.schedule(Action.GROUP_START, 0); verify(context, times(0)).schedule(any(), anyLong()); verifyActions(groupActor, Action.GROUP_START); }
@JsonSetter("actions") public void deserializeActions(final Set<String> actionsStr) { if (actionsStr != null && !actionsStr.isEmpty()) { actions = new HashSet<>(); actionsStr.forEach(s -> actions.add(Action.valueOf(s.toUpperCase(Locale.US)))); } }
@Override protected @Nullable Object executeImpl(VirtualFrame frame) { var repetitions = (long) frame.getArguments()[2]; for (long i = 0; i < repetitions; i++) { blackholeNode.executeGeneric(frame); } LoopNode.reportLoopCount( this, repetitions > Integer.MAX_VALUE ? Integer.MAX_VALUE : (int) repetitions); return null; }
@Test public void isEmpty() { Set<ConstraintViolation<TestId>> violations = validator.validate(new TestId("")); assertEquals(1, violations.size()); ConstraintViolation<TestId> violation = violations.iterator().next(); assertEquals("", violation.getInvalidValue()); assertEquals("[maestro id or name reference] cannot be null or empty", violation.getMessage()); }
public ViewHolder(@NonNull View itemView) { super(itemView); showSystemApps = itemView.findViewById(R.id.showSystemApps); disabled = itemView.findViewById(R.id.disabled); blacklist = itemView.findViewById(R.id.blacklist); }
@Test public void testGetFlowsFromWithError() { Mockito.doThrow(new MaestroInternalError("test")).when(flowDao).getFlows(group, 10, ""); assertNull(context.getFlowsFrom(group, 10, "")); verify(flowDao, times(1)).getFlows(group, 10, ""); }
private StepRuntimeSummary cloneSummary(StepRuntimeSummary summary) { StepAction latestAction = summary.getPendingAction(); StepRuntimeSummary cloned = objectMapper.convertValue(summary, StepRuntimeSummary.class); cloned.setPendingAction(latestAction); return cloned; }
@Test public void testRunForActionStopForActiveRunningTask() { taskActor.runForAction(Action.TASK_STOP); verify(context, times(1)).cancel(any(), any()); assertEquals(Set.of(Action.TASK_STOP), taskActor.getScheduledActions().keySet()); verifyEmptyAction(taskActor); assertTrue(taskActor.isRunning()); }
The above is the ID prefix of the Java language that we use in the ReCaLL method.
Page updated
Google Sites
Report abuse