Search in sources :

Example 1 with ClassInfo

use of org.hl7.elm_modelinfo.r1.ClassInfo in project quality-measure-and-cohort-service by Alvearie.

the class SparkCqlEvaluator method createDataTypeAliases.

private Map<String, String> createDataTypeAliases(List<ContextDefinition> filteredContexts, CqlToElmTranslator translator) {
    Set<String> dataTypes = filteredContexts.stream().map(ContextDefinition::getPrimaryDataType).collect(Collectors.toSet());
    Collection<ModelInfo> modelInfos = translator.getRegisteredModelInfos().values();
    Map<String, String> dataTypeAliases = new HashMap<>();
    for (ModelInfo modelInfo : modelInfos) {
        modelInfo.getTypeInfo().stream().filter(ClassInfo.class::isInstance).map(ClassInfo.class::cast).filter(classInfo -> dataTypes.contains(classInfo.getName())).forEach(info -> {
            String dataType = info.getName();
            QName baseType = ModelUtils.getBaseTypeName(modelInfo, info);
            if (baseType != null) {
                // for inheritance types
                dataTypeAliases.put(dataType, baseType.getLocalPart());
            }
            Collection<String> choiceTypes = ModelUtils.getChoiceTypeNames(info);
            // for choice types
            choiceTypes.forEach(choiceType -> dataTypeAliases.put(dataType, choiceType));
        });
    }
    return dataTypeAliases;
}
Also used : ModelInfo(org.hl7.elm_modelinfo.r1.ModelInfo) Arrays(java.util.Arrays) CqlDataProvider(com.ibm.cohort.cql.data.CqlDataProvider) CqlToElmTranslator(com.ibm.cohort.cql.translation.CqlToElmTranslator) FileSystem(org.apache.hadoop.fs.FileSystem) ZonedDateTime(java.time.ZonedDateTime) LoggerFactory(org.slf4j.LoggerFactory) EncodedParametersCache(com.ibm.cohort.cql.spark.util.EncodedParametersCache) DataRowDataProvider(com.ibm.cohort.datarow.engine.DataRowDataProvider) Format(com.ibm.cohort.cql.library.Format) ClassInfo(org.hl7.elm_modelinfo.r1.ClassInfo) ColumnRuleCreator(com.ibm.cohort.cql.spark.aggregation.ColumnRuleCreator) ConfigurableOutputColumnNameEncoder(com.ibm.cohort.cql.spark.data.ConfigurableOutputColumnNameEncoder) DefaultDatasetRetriever(com.ibm.cohort.cql.spark.data.DefaultDatasetRetriever) SparkSchemaCreator(com.ibm.cohort.cql.spark.data.SparkSchemaCreator) ContextDefinition(com.ibm.cohort.cql.spark.aggregation.ContextDefinition) Map(java.util.Map) Path(org.apache.hadoop.fs.Path) CqlEvaluationRequest(com.ibm.cohort.cql.evaluation.CqlEvaluationRequest) ModelUtils(com.ibm.cohort.cql.spark.optimizer.ModelUtils) ConstraintViolation(javax.validation.ConstraintViolation) StructType(org.apache.spark.sql.types.StructType) HadoopPathOutputMetadataWriter(com.ibm.cohort.cql.spark.metadata.HadoopPathOutputMetadataWriter) Collection(java.util.Collection) Set(java.util.Set) Validator(javax.validation.Validator) CqlLibraryProvider(com.ibm.cohort.cql.library.CqlLibraryProvider) Reader(java.io.Reader) Tuple2(scala.Tuple2) Collectors(java.util.stream.Collectors) FileNotFoundException(java.io.FileNotFoundException) Serializable(java.io.Serializable) ContextRetriever(com.ibm.cohort.cql.spark.aggregation.ContextRetriever) List(java.util.List) CqlEvaluationRequests(com.ibm.cohort.cql.evaluation.CqlEvaluationRequests) CqlTerminologyProvider(com.ibm.cohort.cql.terminology.CqlTerminologyProvider) QName(javax.xml.namespace.QName) SparkOutputColumnEncoder(com.ibm.cohort.cql.spark.data.SparkOutputColumnEncoder) Parameter(com.ibm.cohort.cql.evaluation.parameters.Parameter) MapUtils(com.ibm.cohort.cql.util.MapUtils) ExternalFunctionProvider(org.opencds.cqf.cql.engine.data.ExternalFunctionProvider) ByteArrayOutputStream(java.io.ByteArrayOutputStream) Dataset(org.apache.spark.sql.Dataset) SerializableConfiguration(org.apache.spark.util.SerializableConfiguration) CqlEvaluationResult(com.ibm.cohort.cql.evaluation.CqlEvaluationResult) SparkDataRow(com.ibm.cohort.cql.spark.data.SparkDataRow) PriorityCqlLibraryProvider(com.ibm.cohort.cql.library.PriorityCqlLibraryProvider) R4FileSystemFhirTerminologyProvider(com.ibm.cohort.cql.terminology.R4FileSystemFhirTerminologyProvider) HashMap(java.util.HashMap) ValidatorFactory(javax.validation.ValidatorFactory) HadoopBasedCqlLibraryProvider(com.ibm.cohort.cql.library.HadoopBasedCqlLibraryProvider) SparkTypeConverter(com.ibm.cohort.cql.spark.data.SparkTypeConverter) Function(java.util.function.Function) ArrayList(java.util.ArrayList) CustomMetricSparkPlugin(com.ibm.cohort.cql.spark.metrics.CustomMetricSparkPlugin) CollectionUtils(org.apache.commons.collections.CollectionUtils) EvaluationError(com.ibm.cohort.cql.spark.errors.EvaluationError) DataRowRetrieveProvider(com.ibm.cohort.datarow.engine.DataRowRetrieveProvider) Validation(javax.validation.Validation) TranslatingCqlLibraryProvider(com.ibm.cohort.cql.translation.TranslatingCqlLibraryProvider) NoSuchElementException(java.util.NoSuchElementException) EvaluationSummary(com.ibm.cohort.cql.spark.metadata.EvaluationSummary) DataRow(com.ibm.cohort.datarow.model.DataRow) SparkSession(org.apache.spark.sql.SparkSession) PrintStream(java.io.PrintStream) CqlEvaluator(com.ibm.cohort.cql.evaluation.CqlEvaluator) ClasspathCqlLibraryProvider(com.ibm.cohort.cql.library.ClasspathCqlLibraryProvider) SaveMode(org.apache.spark.sql.SaveMode) Logger(org.slf4j.Logger) CqlDebug(com.ibm.cohort.cql.evaluation.CqlDebug) Iterator(java.util.Iterator) RowFactory(org.apache.spark.sql.RowFactory) JCommander(com.beust.jcommander.JCommander) ObjectMapper(com.fasterxml.jackson.databind.ObjectMapper) IOException(java.io.IOException) Row(org.apache.spark.sql.Row) CollectionAccumulator(org.apache.spark.util.CollectionAccumulator) InputStreamReader(java.io.InputStreamReader) JavaPairRDD(org.apache.spark.api.java.JavaPairRDD) LongAccumulator(org.apache.spark.util.LongAccumulator) CqlExpressionConfiguration(com.ibm.cohort.cql.evaluation.CqlExpressionConfiguration) AnyColumnFunctions(com.ibm.cohort.cql.functions.AnyColumnFunctions) MDC(org.slf4j.MDC) UnsupportedTerminologyProvider(com.ibm.cohort.cql.terminology.UnsupportedTerminologyProvider) CohortExternalFunctionProvider(com.ibm.cohort.cql.functions.CohortExternalFunctionProvider) ContextDefinitions(com.ibm.cohort.cql.spark.aggregation.ContextDefinitions) Collections(java.util.Collections) OutputMetadataWriter(com.ibm.cohort.cql.spark.metadata.OutputMetadataWriter) ModelInfo(org.hl7.elm_modelinfo.r1.ModelInfo) HashMap(java.util.HashMap) QName(javax.xml.namespace.QName) ClassInfo(org.hl7.elm_modelinfo.r1.ClassInfo)

Example 2 with ClassInfo

use of org.hl7.elm_modelinfo.r1.ClassInfo in project quality-measure-and-cohort-service by Alvearie.

the class ModelUtilsTest method testGetBaseTypeNameIncludedNamespacePrefixMissing.

@Test
public void testGetBaseTypeNameIncludedNamespacePrefixMissing() {
    ModelSpecifier otherModel = new ModelSpecifier().withName("Other").withVersion("1.2.3").withUrl("urn:oid:Other");
    ModelInfo modelInfo = new ModelInfo();
    modelInfo.setName("Dummy");
    modelInfo.setVersion("5.4.3");
    modelInfo.setUrl("urn:oid:Dummy");
    modelInfo.getRequiredModelInfo().add(otherModel);
    ClassInfo typeInfo = new ClassInfo();
    typeInfo.setName("MyType");
    typeInfo.setBaseType("Missing.BaseType");
    IllegalArgumentException iex = assertThrows(IllegalArgumentException.class, () -> ModelUtils.getBaseTypeName(modelInfo, typeInfo));
    assertTrue(iex.getMessage(), iex.getMessage().contains("Missing"));
    assertTrue(iex.getMessage(), iex.getMessage().contains(modelInfo.getName()));
    assertTrue(iex.getMessage(), iex.getMessage().contains(modelInfo.getVersion()));
}
Also used : ModelInfo(org.hl7.elm_modelinfo.r1.ModelInfo) ModelSpecifier(org.hl7.elm_modelinfo.r1.ModelSpecifier) ClassInfo(org.hl7.elm_modelinfo.r1.ClassInfo) Test(org.junit.Test)

Example 3 with ClassInfo

use of org.hl7.elm_modelinfo.r1.ClassInfo in project quality-measure-and-cohort-service by Alvearie.

the class ModelUtilsTest method testGetBaseTypeNameNoNamespacePrefix.

@Test
public void testGetBaseTypeNameNoNamespacePrefix() {
    ModelInfo modelInfo = new ModelInfo();
    modelInfo.setName("Dummy");
    modelInfo.setUrl("urn:oid:Dummy");
    ClassInfo typeInfo = new ClassInfo();
    typeInfo.setName("MyType");
    typeInfo.setBaseType("BaseType");
    QName qname = ModelUtils.getBaseTypeName(modelInfo, typeInfo);
    assertEquals(new QName(modelInfo.getUrl(), "BaseType"), qname);
}
Also used : ModelInfo(org.hl7.elm_modelinfo.r1.ModelInfo) QName(javax.xml.namespace.QName) ClassInfo(org.hl7.elm_modelinfo.r1.ClassInfo) Test(org.junit.Test)

Example 4 with ClassInfo

use of org.hl7.elm_modelinfo.r1.ClassInfo in project quality-measure-and-cohort-service by Alvearie.

the class ModelUtilsTest method testGetBaseTypeNameSystemNamespacePrefix.

@Test
public void testGetBaseTypeNameSystemNamespacePrefix() {
    ModelInfo modelInfo = new ModelInfo();
    modelInfo.setName("Dummy");
    modelInfo.setUrl("urn:oid:Dummy");
    ClassInfo typeInfo = new ClassInfo();
    typeInfo.setName("MyType");
    typeInfo.setBaseType("System.BaseType");
    QName qname = ModelUtils.getBaseTypeName(modelInfo, typeInfo);
    assertEquals(new QName(CqlConstants.SYSTEM_MODEL_URI, "BaseType"), qname);
}
Also used : ModelInfo(org.hl7.elm_modelinfo.r1.ModelInfo) QName(javax.xml.namespace.QName) ClassInfo(org.hl7.elm_modelinfo.r1.ClassInfo) Test(org.junit.Test)

Example 5 with ClassInfo

use of org.hl7.elm_modelinfo.r1.ClassInfo in project quality-measure-and-cohort-service by Alvearie.

the class SparkSchemaCreator method getDataTypeForContextKey.

private Tuple2<String, DataType> getDataTypeForContextKey(String contextName, Set<Tuple2<String, String>> usingInfos) {
    ContextDefinition contextDefinition = contextDefinitions.getContextDefinitionByName(contextName);
    String primaryDataType = contextDefinition.getPrimaryDataType();
    String primaryKeyColumn = contextDefinition.getPrimaryKeyColumn();
    DataType keyType = null;
    ModelManager modelManager = translator.newModelManager();
    // Try to find the key column's type information from a single model info.
    for (Tuple2<String, String> usingInfo : usingInfos) {
        VersionedIdentifier modelInfoIdentifier = new VersionedIdentifier().withId(usingInfo._1()).withVersion(usingInfo._2());
        ModelInfo modelInfo = modelManager.getModelInfoLoader().getModelInfo(modelInfoIdentifier);
        // Look for a ClassInfo element matching primaryDataType for the context
        List<ClassInfo> classInfos = getClassInfos(primaryDataType, modelInfo);
        if (!classInfos.isEmpty()) {
            if (classInfos.size() == 1) {
                ClassInfo classInfo = classInfos.get(0);
                List<ClassInfoElement> elements = classInfo.getElement().stream().filter(x -> x.getName().equals(primaryKeyColumn)).collect(Collectors.toList());
                // check base type
                String baseType = classInfo.getBaseType();
                if (classInfo.getBaseType() != null) {
                    List<ClassInfo> baseClassInfos = getClassInfos(baseType, modelInfo);
                    baseClassInfos.stream().map(ClassInfo::getElement).flatMap(List::stream).filter(element -> element.getName().equals(primaryKeyColumn)).forEach(elements::add);
                }
                // check choice types
                Collection<String> choiceTypes = ModelUtils.getChoiceTypeNames(classInfo);
                choiceTypes.stream().map(type -> getClassInfos(type, modelInfo)).flatMap(List::stream).map(ClassInfo::getElement).flatMap(List::stream).filter(element -> element.getName().equals(primaryKeyColumn)).findFirst().ifPresent(elements::add);
                // A future ModelInfo file may contain the information
                if (elements.isEmpty()) {
                    continue;
                } else if (elements.size() == 1) {
                    String elementType = elements.get(0).getElementType();
                    // store it
                    if (keyType == null) {
                        keyType = getSparkTypeForSystemValue(elementType);
                    } else {
                        throw new IllegalArgumentException("Multiple definitions found for " + primaryDataType + "." + primaryKeyColumn + " in the provided ModelInfo files. Cannot infer key type for context: " + contextName);
                    }
                } else if (elements.size() > 1) {
                    throw new IllegalArgumentException("ModelInfo " + modelInfoIdentifier + " contains multiple element definitions for " + primaryKeyColumn + " for type " + primaryDataType);
                }
            } else {
                throw new IllegalArgumentException("ModelInfo " + modelInfoIdentifier + " contains multiple definitions for type " + primaryDataType);
            }
        }
    }
    if (keyType == null) {
        throw new IllegalArgumentException("Could not locate type information for " + primaryDataType + "." + primaryKeyColumn + " in the provided ModelInfo files. Cannot infer key type for context: " + contextName);
    }
    return new Tuple2<>(contextDefinition.getPrimaryKeyColumn(), keyType);
}
Also used : DataType(org.apache.spark.sql.types.DataType) ModelInfo(org.hl7.elm_modelinfo.r1.ModelInfo) CqlToElmTranslator(com.ibm.cohort.cql.translation.CqlToElmTranslator) HashMap(java.util.HashMap) Format(com.ibm.cohort.cql.library.Format) ExpressionDef(org.cqframework.cql.elm.execution.ExpressionDef) ClassInfo(org.hl7.elm_modelinfo.r1.ClassInfo) HashSet(java.util.HashSet) ContextDefinition(com.ibm.cohort.cql.spark.aggregation.ContextDefinition) Map(java.util.Map) CqlLibraryReader(org.opencds.cqf.cql.engine.execution.CqlLibraryReader) CqlEvaluationRequest(com.ibm.cohort.cql.evaluation.CqlEvaluationRequest) ModelUtils(com.ibm.cohort.cql.spark.optimizer.ModelUtils) DataTypes(org.apache.spark.sql.types.DataTypes) StructField(org.apache.spark.sql.types.StructField) StructType(org.apache.spark.sql.types.StructType) ModelManager(org.cqframework.cql.cql2elm.ModelManager) Collection(java.util.Collection) VersionedIdentifier(org.hl7.elm.r1.VersionedIdentifier) Set(java.util.Set) CqlLibrary(com.ibm.cohort.cql.library.CqlLibrary) CqlLibraryProvider(com.ibm.cohort.cql.library.CqlLibraryProvider) Tuple2(scala.Tuple2) Collectors(java.util.stream.Collectors) CqlLibraryDescriptor(com.ibm.cohort.cql.library.CqlLibraryDescriptor) List(java.util.List) CqlEvaluationRequests(com.ibm.cohort.cql.evaluation.CqlEvaluationRequests) Library(org.cqframework.cql.elm.execution.Library) ClassInfoElement(org.hl7.elm_modelinfo.r1.ClassInfoElement) QName(javax.xml.namespace.QName) ContextDefinitions(com.ibm.cohort.cql.spark.aggregation.ContextDefinitions) ModelInfo(org.hl7.elm_modelinfo.r1.ModelInfo) ContextDefinition(com.ibm.cohort.cql.spark.aggregation.ContextDefinition) ModelManager(org.cqframework.cql.cql2elm.ModelManager) ClassInfoElement(org.hl7.elm_modelinfo.r1.ClassInfoElement) VersionedIdentifier(org.hl7.elm.r1.VersionedIdentifier) Tuple2(scala.Tuple2) DataType(org.apache.spark.sql.types.DataType) List(java.util.List) ClassInfo(org.hl7.elm_modelinfo.r1.ClassInfo)

Aggregations

ModelInfo (org.hl7.elm_modelinfo.r1.ModelInfo)11 ClassInfo (org.hl7.elm_modelinfo.r1.ClassInfo)10 Test (org.junit.Test)9 QName (javax.xml.namespace.QName)6 TypeInfo (org.hl7.elm_modelinfo.r1.TypeInfo)3 CqlEvaluationRequest (com.ibm.cohort.cql.evaluation.CqlEvaluationRequest)2 CqlEvaluationRequests (com.ibm.cohort.cql.evaluation.CqlEvaluationRequests)2 CqlLibraryProvider (com.ibm.cohort.cql.library.CqlLibraryProvider)2 Format (com.ibm.cohort.cql.library.Format)2 ContextDefinition (com.ibm.cohort.cql.spark.aggregation.ContextDefinition)2 ContextDefinitions (com.ibm.cohort.cql.spark.aggregation.ContextDefinitions)2 ModelUtils (com.ibm.cohort.cql.spark.optimizer.ModelUtils)2 File (java.io.File)2 HashMap (java.util.HashMap)2 JCommander (com.beust.jcommander.JCommander)1 ObjectMapper (com.fasterxml.jackson.databind.ObjectMapper)1 CqlDataProvider (com.ibm.cohort.cql.data.CqlDataProvider)1 CqlDebug (com.ibm.cohort.cql.evaluation.CqlDebug)1 CqlEvaluationResult (com.ibm.cohort.cql.evaluation.CqlEvaluationResult)1 CqlEvaluator (com.ibm.cohort.cql.evaluation.CqlEvaluator)1