new project for export functionality

histream-export/.gitignore

+.settings/
+.classpath
+.project
+target/

histream-export/README.txt

+Algorithm
+---------
+
+1. Read all concept references. This can be actual concept codes 
+   or wildcard concept codes.
+
+2. The facts will be processed grouped by encounter. E.g. as 
+   a list of all concepts for the given encounter.
+   
+   If the patient or visit is different from the previous one
+   create new row in visit or encounter. Process all columns
+   which are NOT bound to any fact (not for fact tables which
+   are always bound to a fact).
+
+3. For each fact:
+3.1	Find all matching columns.
+3.2	
+
+or
+
+3. For each column:
+3.1	Find all matching facts
\ No newline at end of file

histream-export/pom.xml

+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
+	<modelVersion>4.0.0</modelVersion>
+	<name>HIStream : Export</name>
+	<description>
+	Export streams or collections of facts to table files
+	like CSV.
+	</description>
+
+	<groupId>de.sekmi.histream</groupId>
+	<artifactId>histream-export</artifactId>
+	<version>0.8-SNAPSHOT</version>
+
+	<parent>
+		<groupId>de.sekmi.histream</groupId>
+		<artifactId>histream</artifactId>
+		<version>0.8-SNAPSHOT</version>
+	</parent>
+
+	<build>
+		<plugins>
+		</plugins>
+	</build>
+	<dependencies>
+		<dependency>
+			<groupId>de.sekmi.histream</groupId>
+			<artifactId>histream-core</artifactId>
+			<version>0.8-SNAPSHOT</version>
+		</dependency>
+		<dependency>
+			<groupId>junit</groupId>
+			<artifactId>junit</artifactId>
+		</dependency>
+	</dependencies>
+</project>
\ No newline at end of file

histream-export/src/main/java/de/sekmi/histream/export/TableExportFactory.java

+package de.sekmi.histream.export;
+
+import de.sekmi.histream.ObservationSupplier;
+
+public class TableExportFactory {
+
+	public void export(ObservationSupplier supplier){
+		
+	}
+}

histream-export/src/main/java/de/sekmi/histream/export/config/AbstractColumn.java

+package de.sekmi.histream.export.config;
+
+import javax.xml.bind.annotation.XmlAccessType;
+import javax.xml.bind.annotation.XmlAccessorType;
+import javax.xml.bind.annotation.XmlAttribute;
+import javax.xml.bind.annotation.XmlSeeAlso;
+import javax.xml.bind.annotation.XmlTransient;
+
+@XmlTransient
+@XmlAccessorType(XmlAccessType.NONE)
+@XmlSeeAlso({SequenceColumn.class, IdColumn.class})
+public abstract class AbstractColumn {
+
+	/**
+	 * Column header name
+	 */
+	@XmlAttribute(required=true)
+	String header;
+
+
+	/**
+	 * Column value to use to indicate NA.
+	 * This will be used if null values are
+	 * encountered by the column. If not specified,
+	 * the empty string {@code ''} will be used.
+	 */
+	@XmlAttribute
+	String na;
+	
+	/**
+	 * Prepare the column for the next row. This
+	 * method is called once for every row. 
+	 * <p> The default implementation does nothing.</p>
+	 */
+	protected void prepareRow(){
+		
+	}
+	
+	/**
+	 * Initializes the column. This method will be called
+	 * only once and before any other method.
+	 */
+	protected void initialize(){
+		
+	}
+	
+	/**
+	 * Get the column value for the current row.
+	 * @return String value or {@code null} for NA.
+	 */
+	public abstract Object getValueString();
+}

histream-export/src/main/java/de/sekmi/histream/export/config/AbstractTable.java

+package de.sekmi.histream.export.config;
+
+import javax.xml.bind.annotation.XmlAccessType;
+import javax.xml.bind.annotation.XmlAccessorType;
+import javax.xml.bind.annotation.XmlTransient;
+
+@XmlTransient
+@XmlAccessorType(XmlAccessType.NONE)
+public abstract class AbstractTable {
+
+	public abstract String[] getHeaders();
+	public abstract AbstractColumn getColumn(int index);
+	public abstract AbstractColumn getColumnByHeader(String header);
+}

histream-export/src/main/java/de/sekmi/histream/export/config/ExportDescriptor.java

+package de.sekmi.histream.export.config;
+
+import javax.xml.bind.annotation.XmlElement;
+import javax.xml.bind.annotation.XmlRootElement;
+
+@XmlRootElement(name="export")
+public class ExportDescriptor {
+
+	@XmlElement
+	PatientTable patient;
+	
+	@XmlElement
+	VisitTable visit;
+	
+	@XmlElement(name="table")
+	FactTable[] tables;
+}

histream-export/src/main/java/de/sekmi/histream/export/config/FactTable.java

+package de.sekmi.histream.export.config;
+
+public class FactTable {
+
+}

histream-export/src/main/java/de/sekmi/histream/export/config/IdColumn.java

+package de.sekmi.histream.export.config;
+
+/**
+ * Column which outputs intrinsic ID values. This
+ * column can only be used in {@link PatientTable} and
+ * {@link VisitTable}.
+ * 
+ * @author R.W.Majeed
+ *
+ */
+public class IdColumn extends AbstractColumn{
+
+	@Override
+	public Object getValueString() {
+		// TODO Auto-generated method stub
+		return null;
+	}
+
+}

histream-export/src/main/java/de/sekmi/histream/export/config/PatientTable.java

+package de.sekmi.histream.export.config;
+
+import javax.xml.bind.annotation.XmlElement;
+
+public class PatientTable {
+
+	@XmlElement(name="column")
+	AbstractColumn[] columns;
+}

histream-export/src/main/java/de/sekmi/histream/export/config/SequenceColumn.java

+package de.sekmi.histream.export.config;
+
+import javax.xml.bind.annotation.XmlElement;
+import javax.xml.bind.annotation.XmlRootElement;
+
+@XmlRootElement(name="sequence")
+public class SequenceColumn extends AbstractColumn{
+	@XmlElement(required=true)
+	String id;
+	
+	@XmlElement
+	Long start;
+	
+	long value;
+	
+	protected SequenceColumn(){
+	}
+	public SequenceColumn(String id){
+		this.id = id;
+	}
+	
+	@Override
+	protected void initialize(){
+		if( start != null ){
+			value = start;
+		}else{
+			value = 1;
+		}
+	}
+	@Override
+	public Object getValueString() {
+		return value;
+		// TODO increment value somewhere else
+	}
+}

histream-export/src/main/java/de/sekmi/histream/export/config/VisitTable.java

+package de.sekmi.histream.export.config;
+
+public class VisitTable {
+
+}

histream-export/src/main/java/de/sekmi/histream/export/config/package-info.java

+/**
+ * Export configuration with JAXB annotations
+ * 
+ */
+@javax.xml.bind.annotation.XmlSchema(
+		namespace="http://sekmi.de/ns/histream/export"
+)
+package de.sekmi.histream.export.config;
\ No newline at end of file

histream-export/src/test/java/de/sekmi/histream/export/config/TestMarshal.java

+package de.sekmi.histream.export.config;
+
+import javax.xml.bind.JAXBContext;
+import javax.xml.bind.JAXBException;
+import javax.xml.bind.Marshaller;
+
+import org.junit.Test;
+
+public class TestMarshal {
+
+	@Test
+	public void verifyMarshal() throws JAXBException{
+		ExportDescriptor e = new ExportDescriptor();
+		e.patient = new PatientTable();
+		e.patient.columns = new AbstractColumn[2];
+		e.patient.columns[0] = new IdColumn();
+		e.patient.columns[0].header = "pid";
+		e.patient.columns[1] = new SequenceColumn("seq1");
+		e.patient.columns[1].header = "seq";
+		
+		JAXBContext j = JAXBContext.newInstance(ExportDescriptor.class);
+		Marshaller m = j.createMarshaller();
+		m.setProperty(Marshaller.JAXB_FORMATTED_OUTPUT, true);
+		m.marshal(e, System.out);
+		
+	}
+}

histream-export/src/test/resources/export1.xml

+<?xml version="1.0" encoding="UTF-8"?>
+<export xmlns="http://sekmi.de/ns/histream/export">
+	<patient>
+		<column header="pid" type="id"/>
+		<column header="seq" type="sequence">
+			<id>seq1</id>
+		</column>
+	</patient>
+	<visit>
+		<column header="pid" type="patient-ref"/>
+		<column header="visit" type="id"/>
+		<!-- concepts for the visit table must occur only once
+		per visit and may not repeat -->
+		<column header="start" type="attribute">
+			<property>start</property>
+		</column>
+		<!-- What is better? -->
+		<!-- (a) First column, then concepts with value -->
+		<column header="diagnostik_labort_ts" na="NULL" type="fact">
+			<concept code="LOINC:26436-6"/>
+			<concept code="LOINC:26436-6:NEG"/>
+			<value type="script">
+					<eval>fact.conceptId.substring(6)</eval>
+			</value>
+		</column>
+		<!-- (b) first concepts then columns with value -->
+		<group>
+			<concept code="LOINC:26436-6"/>
+			<concept code="LOINC:26436-6:NEG"/>
+			<column header="lab_ts" type="attribute">
+				<attribute>start</attribute>
+			</column>
+			<!-- We want this: 
+					6:NEG -> not tested
+					6/mod=OPB -> tested non pathological
+					6/mod=PB -> tested pathological 
+			-->
+			<column header="lab_ergebnis" type="attribute">
+				<modifier>PB</modifier>
+				<attribute>start</attribute>
+			</column>
+		</group>
+	</visit>
+	<table id="diagnosen">
+		<!-- makes more sense for fact tables to specify
+			the list of concepts first. -->
+		<concept wildcard="ICD10:*"/>
+		<concept code="ICD9:123"/>
+		<column header="pid" type="patient-ref"/>
+		<column header="visit" type="visit-ref"/>
+		<column header="start" type="attribute">
+			<attribute>start</attribute>
+		</column>
+		<column header="primary" type="attribute" na="nein">
+			<modifier>fuehrend</modifier>
+			<!-- if a modifier element is provided, the property
+			 will use the context of that modifier -->
+			<constant-value>ja</constant-value>
+		</column>
+		<column header="text" type="attribute">
+			<modifier>originalText</modifier>
+			<property>value</property>
+		</column>
+		<!-- reference sequence from patient -->
+		<column header="seq1" type="sequence">
+			<sequence ref="seq1"/>
+		</column>
+	</table>
+</export>
\ No newline at end of file

pom.xml

@@ -57,6 +57,8 @@
 		<!--<module>histream-hl7</module>-->
 		<module>histream-skos</module>
 		<module>histream-import</module>
+		<module>histream-export</module>
+		<module>histream-js</module>
 		<module>i2b2-integration-tests</module>
 		<module>distribution</module>
 	</modules>