From 8cec977d3323f7fdb2f7c84992a34e8328e42be0 Mon Sep 17 00:00:00 2001
From: Robert Applin <40830825+robertapplin@users.noreply.github.com>
Date: Tue, 9 Oct 2018 14:44:44 +0100
Subject: [PATCH] Refs #23463. More unit tests
---
.../Indirect/test/IndirectFittingModelTest.h | 225 ++++++++++++++----
1 file changed, 181 insertions(+), 44 deletions(-)
diff --git a/qt/scientific_interfaces/Indirect/test/IndirectFittingModelTest.h b/qt/scientific_interfaces/Indirect/test/IndirectFittingModelTest.h
index b36a44b8d4d..0d512d6482e 100644
--- a/qt/scientific_interfaces/Indirect/test/IndirectFittingModelTest.h
+++ b/qt/scientific_interfaces/Indirect/test/IndirectFittingModelTest.h
@@ -59,6 +59,20 @@ struct SetUpADSWithWorkspace {
AnalysisDataService::Instance().addOrReplace(inputWSName, workspace);
};
+ template <typename T>
+ void addOrReplace(std::string const &workspaceName, T const &workspace) {
+ AnalysisDataService::Instance().addOrReplace(workspaceName, workspace);
+ }
+
+ bool doesExist(std::string const &workspaceName) {
+ return AnalysisDataService::Instance().doesExist(workspaceName);
+ }
+
+ MatrixWorkspace_sptr retrieveWorkspace(std::string const &workspaceName) {
+ return boost::dynamic_pointer_cast<MatrixWorkspace>(
+ AnalysisDataService::Instance().retrieve(workspaceName));
+ }
+
~SetUpADSWithWorkspace() { AnalysisDataService::Instance().clear(); };
};
@@ -92,15 +106,6 @@ std::unique_ptr<DummyModel> getEmptyModel() {
return std::make_unique<DummyModel>();
}
-std::unique_ptr<DummyModel> createModelWithSingleInstrumentWorkspace(
- std::string const &workspaceName, int const &xLength, int const &yLength) {
- auto model = getEmptyModel();
- SetUpADSWithWorkspace ads(workspaceName,
- createWorkspaceWithInstrument(xLength, yLength));
- model->addWorkspace(workspaceName);
- return model;
-}
-
std::unique_ptr<DummyModel>
createModelWithSingleWorkspace(std::string const &workspaceName,
int const &numberOfSpectra) {
@@ -132,11 +137,20 @@ createModelWithMultipleWorkspaces(int const &numberOfSpectra,
return model;
}
+std::unique_ptr<DummyModel> createModelWithSingleInstrumentWorkspace(
+ std::string const &workspaceName, int const &xLength, int const &yLength) {
+ auto model = getEmptyModel();
+ SetUpADSWithWorkspace ads(workspaceName,
+ createWorkspaceWithInstrument(xLength, yLength));
+ model->addWorkspace(workspaceName);
+ return model;
+}
+
} // namespace
class IndirectFittingModelTest : public CxxTest::TestSuite {
public:
- // WorkflowAlgorithms do not appear in the FrameworkManager without this line
+ /// WorkflowAlgorithms do not appear in the FrameworkManager without this line
IndirectFittingModelTest() { FrameworkManager::Instance(); }
static IndirectFittingModelTest *createSuite() {
@@ -156,10 +170,8 @@ public:
void test_that_a_workspace_is_stored_correctly_in_the_ADS() {
SetUpADSWithWorkspace ads("WorkspaceName", createWorkspace(3));
- TS_ASSERT(AnalysisDataService::Instance().doesExist("WorkspaceName"));
- MatrixWorkspace_sptr storedWorkspace =
- boost::dynamic_pointer_cast<MatrixWorkspace>(
- AnalysisDataService::Instance().retrieve("WorkspaceName"));
+ TS_ASSERT(ads.doesExist("WorkspaceName"));
+ auto const storedWorkspace = ads.retrieveWorkspace("WorkspaceName");
TS_ASSERT_EQUALS(storedWorkspace->getNumberHistograms(), 3);
}
@@ -197,8 +209,8 @@ public:
auto model = getEmptyModel();
auto const workspace1 = createWorkspace(3);
auto const workspace2 = createWorkspace(3);
- SetUpADSWithWorkspace ads1("WorkspaceName1", workspace1);
- SetUpADSWithWorkspace ads2("WorkspaceName2", workspace2);
+ SetUpADSWithWorkspace ads("WorkspaceName1", workspace1);
+ ads.addOrReplace("WorkspaceName2", workspace2);
model->addWorkspace("WorkspaceName1");
model->addWorkspace("WorkspaceName2");
@@ -389,8 +401,7 @@ public:
auto const modelWorkspace = model->getWorkspace(0);
SetUpADSWithWorkspace ads("Name", modelWorkspace);
- auto alg =
- setupConvolutionSequentialFitAlgorithm(model, modelWorkspace, "Name");
+ auto const alg = getSetupFitAlgorithm(model, modelWorkspace, "Name");
TS_ASSERT(alg->isInitialized());
}
@@ -400,23 +411,28 @@ public:
auto const modelWorkspace = model->getWorkspace(0);
SetUpADSWithWorkspace ads("Name", modelWorkspace);
- auto alg =
- setupConvolutionSequentialFitAlgorithm(model, modelWorkspace, "Name");
+ auto const alg = getSetupFitAlgorithm(model, modelWorkspace, "Name");
TS_ASSERT_THROWS_NOTHING(alg->execute());
TS_ASSERT(alg->isExecuted());
}
- void
- test_that_isPreviouslyFit_returns_true_if_the_spectrum_has_been_fitted_previously() {
+ void test_that_addOutput_adds_the_output_of_a_fit_into_the_model() {
auto model = createModelWithSingleInstrumentWorkspace("__ConvFit", 6, 5);
auto const modelWorkspace = model->getWorkspace(0);
SetUpADSWithWorkspace ads("__ConvFit", modelWorkspace);
- auto const alg = executeConvolutionSequentialFitAlgorithm(
- model, modelWorkspace, "__ConvFit");
+ auto const alg =
+ getExecutedFitAlgorithm(model, modelWorkspace, "__ConvFit");
model->addOutput(alg);
+ TS_ASSERT(model->getResultWorkspace());
+ TS_ASSERT(model->getResultGroup());
+ }
+
+ void
+ test_that_isPreviouslyFit_returns_true_if_the_spectrum_has_been_fitted_previously() {
+ auto const model = getModelWithOutputFitData();
TS_ASSERT(model->isPreviouslyFit(0, 0));
}
@@ -464,8 +480,7 @@ public:
auto const modelWorkspace = model->getWorkspace(0);
SetUpADSWithWorkspace ads("Name", modelWorkspace);
- auto alg =
- setupConvolutionSequentialFitAlgorithm(model, modelWorkspace, "Name");
+ (void)getSetupFitAlgorithm(model, modelWorkspace, "Name");
TS_ASSERT(!model->isInvalidFunction());
}
@@ -501,8 +516,8 @@ public:
auto const modelWorkspace = model->getWorkspace(0);
SetUpADSWithWorkspace ads("__ConvFit", modelWorkspace);
- auto const alg = executeConvolutionSequentialFitAlgorithm(
- model, modelWorkspace, "__ConvFit");
+ auto const alg =
+ getExecutedFitAlgorithm(model, modelWorkspace, "__ConvFit");
model->addOutput(alg);
TS_ASSERT(!model->getFitParameterNames().empty());
@@ -607,26 +622,139 @@ public:
auto const modelWorkspace = model->getWorkspace(0);
SetUpADSWithWorkspace ads("Name", modelWorkspace);
- auto const alg =
- setupConvolutionSequentialFitAlgorithm(model, modelWorkspace, "Name");
+ (void)getSetupFitAlgorithm(model, modelWorkspace, "Name");
model->setDefaultParameterValue("Amplitude", 1.5, 0);
auto const parameters = model->getDefaultParameters(0);
TS_ASSERT_EQUALS(parameters.at("f1.f1.f0.Amplitude").value, 1.5);
}
+ void
+ test_that_getParameterValues_returns_an_empty_map_if_the_dataIndex_is_out_of_range() {
+ auto const model = getModelWithOutputFitData();
+ TS_ASSERT(model->getParameterValues(1, 0).empty());
+ }
+
+ void
+ test_that_getParameterValues_returns_the_default_parameters_if_there_are_no_fit_parameters() {
+ auto model = createModelWithSingleInstrumentWorkspace("__ConvFit", 6, 5);
+ auto const modelWorkspace = model->getWorkspace(0);
+ SetUpADSWithWorkspace ads("__ConvFit", modelWorkspace);
+
+ (void)getSetupFitAlgorithm(model, modelWorkspace, "__ConvFit");
+ model->setDefaultParameterValue("Amplitude", 1.5, 0);
+
+ auto const parameters = model->getParameterValues(0, 0);
+ TS_ASSERT_EQUALS(parameters.at("f1.f1.f0.Amplitude").value, 1.5);
+ }
+
+ void
+ test_that_getParameterValues_returns_the_fit_parameters_after_a_fit_has_been_executed() {
+ auto const model = getModelWithOutputFitData();
+
+ auto const parameters = model->getParameterValues(0, 0);
+ TS_ASSERT_EQUALS(parameters.at("f1.f1.f0.Amplitude").value, 1.0, 0.0001);
+ TS_ASSERT_EQUALS(parameters.at("f1.f1.f0.FWHM").value, 0.0175, 0.0001);
+ TS_ASSERT(!parameters.empty());
+ }
+
+ void test_getFitParameters_returns_an_empty_map_when_there_is_no_fitOutput() {
+ auto model = createModelWithSingleInstrumentWorkspace("__ConvFit", 6, 5);
+ auto const modelWorkspace = model->getWorkspace(0);
+ SetUpADSWithWorkspace ads("__ConvFit", modelWorkspace);
+
+ (void)getSetupFitAlgorithm(model, modelWorkspace, "__ConvFit");
+
+ TS_ASSERT(model->getFitParameters(0, 0).empty());
+ }
+
+ void test_getFitParameters_returns_the_fitParameters_after_a_fit() {
+ auto const model = getModelWithOutputFitData();
+
+ auto const parameters = model->getFitParameters(0, 0);
+ TS_ASSERT_EQUALS(parameters.at("f1.f1.f0.Amplitude").value, 1.0, 0.0001);
+ TS_ASSERT_EQUALS(parameters.at("f1.f1.f0.FWHM").value, 0.0175, 0.0001);
+ TS_ASSERT(!parameters.empty());
+ }
+
+ void
+ test_getDefaultParameters_returns_an_empty_map_when_the_dataIndex_is_out_of_range() {
+ auto const model = getModelWithOutputFitData();
+ TS_ASSERT(model->getDefaultParameters(1).empty());
+ }
+
+ void
+ test_getDefaultParameters_returns_the_default_parameters_which_have_been_set() {
+ auto const model = getModelWithOutputFitData();
+
+ model->setDefaultParameterValue("Amplitude", 1.5, 0);
+
+ auto const parameters = model->getDefaultParameters(0);
+ TS_ASSERT(!parameters.empty());
+ TS_ASSERT_EQUALS(parameters.at("f1.f1.f0.Amplitude").value, 1.5, 0.0001);
+ }
+
+ void test_that_getResultLocation_returns_a_location_for_the_output_data() {
+ auto const model = getModelWithOutputFitData();
+ TS_ASSERT(model->getResultLocation(0, 0));
+ }
+
+ void test_that_saveResult_does_not_throw_when_saving_data_from_a_fit() {
+ auto const model = getModelWithOutputFitData();
+ TS_ASSERT_THROWS_NOTHING(model->saveResult());
+ }
+
+ void
+ test_that_cleanFailedRun_removes_the_temporary_workspace_from_the_ADS_when_a_fit_fails() {
+ /// Fails the fit algorithm on purpose by providing an invalid function
+ auto model =
+ createModelWithSingleInstrumentWorkspace("__ConvFitResolution", 6, 5);
+ auto const modelWorkspace = model->getWorkspace(0);
+ SetUpADSWithWorkspace ads("__ConvFitResolution", modelWorkspace);
+
+ std::string const functionString =
+ "name=Convolution;name=Resolution,Workspace=__ConvFit"
+ "Resolution,WorkspaceIndex=0;";
+ auto alg = setupConvolutionSequentialFitAlgorithm(
+ modelWorkspace, "__ConvFit", functionString);
+ alg->execute();
+
+ TS_ASSERT(ads.doesExist("__ConvolutionFitSequential_ws1"));
+ model->cleanFailedRun(alg);
+ TS_ASSERT(!ads.doesExist("__ConvolutionFitSequential_ws1"));
+ }
+
+ void
+ test_that_cleanFailedSingleRun_removes_the_temporary_workspace_from_the_ADS_when_a_fit_fails_for_a_specific_workspaceIndex() {
+ /// Fails the fit algorithm on purpose by providing an invalid function
+ auto model =
+ createModelWithSingleInstrumentWorkspace("__ConvFitResolution", 6, 5);
+ auto const modelWorkspace = model->getWorkspace(0);
+ SetUpADSWithWorkspace ads("__ConvFitResolution", modelWorkspace);
+
+ std::string const functionString =
+ "name=Convolution;name=Resolution,Workspace=__ConvFit"
+ "Resolution,WorkspaceIndex=0;";
+ auto alg = setupConvolutionSequentialFitAlgorithm(
+ modelWorkspace, "__ConvFit", functionString);
+ alg->execute();
+
+ TS_ASSERT(ads.doesExist("__ConvolutionFitSequential_ws1"));
+ model->cleanFailedSingleRun(alg, 0);
+ TS_ASSERT(!ads.doesExist("__ConvolutionFitSequential_ws1"));
+ }
+
private:
void setFittingFunction(std::unique_ptr<DummyModel> &model,
- std::string const &functionString) {
+ std::string const &functionString) const {
auto const function =
FunctionFactory::Instance().createInitialized(functionString);
model->setFitFunction(function);
}
- IAlgorithm_sptr
- setupConvolutionSequentialFitAlgorithm(MatrixWorkspace_sptr workspace,
- std::string const &workspaceName,
- std::string const &functionString) {
+ IAlgorithm_sptr setupConvolutionSequentialFitAlgorithm(
+ MatrixWorkspace_sptr workspace, std::string const &workspaceName,
+ std::string const &functionString) const {
auto alg = boost::make_shared<ConvolutionFitSequential>();
TS_ASSERT_THROWS_NOTHING(alg->initialize());
alg->setProperty("InputWorkspace", workspace);
@@ -644,10 +772,9 @@ private:
return alg;
}
- IAlgorithm_sptr
- setupConvolutionSequentialFitAlgorithm(std::unique_ptr<DummyModel> &model,
- MatrixWorkspace_sptr workspace,
- std::string const &workspaceName) {
+ IAlgorithm_sptr getSetupFitAlgorithm(std::unique_ptr<DummyModel> &model,
+ MatrixWorkspace_sptr workspace,
+ std::string const &workspaceName) const {
std::string const function =
"name=LinearBackground,A0=0,A1=0,ties=(A0=0.000000,A1=0.0);"
"(composite=Convolution,FixResolution=true,NumDeriv=true;"
@@ -663,14 +790,24 @@ private:
}
IAlgorithm_sptr
- executeConvolutionSequentialFitAlgorithm(std::unique_ptr<DummyModel> &model,
- MatrixWorkspace_sptr workspace,
- std::string const &workspaceName) {
- auto const alg =
- setupConvolutionSequentialFitAlgorithm(model, workspace, workspaceName);
+ getExecutedFitAlgorithm(std::unique_ptr<DummyModel> &model,
+ MatrixWorkspace_sptr workspace,
+ std::string const &workspaceName) const {
+ auto const alg = getSetupFitAlgorithm(model, workspace, workspaceName);
alg->execute();
return alg;
}
+
+ std::unique_ptr<DummyModel> getModelWithOutputFitData() {
+ auto model = createModelWithSingleInstrumentWorkspace("__ConvFit", 6, 5);
+ auto const modelWorkspace = model->getWorkspace(0);
+ SetUpADSWithWorkspace ads("__ConvFit", modelWorkspace);
+
+ auto const alg =
+ getExecutedFitAlgorithm(model, modelWorkspace, "__ConvFit");
+ model->addOutput(alg);
+ return model;
+ }
};
#endif
--
GitLab