example code for parsing p21 dataset (incomplete)

package de.sekmi.histream.etl.config;

import java.io.IOException;

import javax.xml.stream.XMLStreamException;

import de.sekmi.histream.etl.ParseException;

import de.sekmi.histream.etl.XMLExport;

public class TestP21 {

	public static void main(String[] args) throws IOException, ParseException, XMLStreamException {

		XMLExport.descriptorToXML(Class.class.getResource("/data/test-p21-datasource.xml"), System.out);

	}

}

<?xml version="1.0" encoding="UTF-8"?>

<datasource version="1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" >

	<meta>

		<id>test-1</id>

		<etl-strategy>replace-source</etl-strategy>

	</meta>

	<patient-table>

		<source xsi:type="csv-file">

			<url>p21khg/_PAT.csv</url>

			<separator>;</separator>

		</source>

		<idat>

			<patient-id column="Patientennummer"/>

			<!-- TODO allow concatenation/splitting of columns to join separate columns for birth year and month -->

			<birthdate format="u" na="" column="Geburtsjahr"/>

			<gender column="Geschlecht" na="">

				<map> <!-- maps a column -->

					<case value="w" set-value="female"/>

					<case value="m" set-value="male"/>

					<otherwise set-value="" log-warning="Unexpected gender value"/>

				</map>

			</gender>

		</idat>

		<ignore xsi:type="string" column="*"/>

	</patient-table>

	<!-- optional -->

	<visit-table>

		<source xsi:type="csv-file">

			<url>p21khg/FALL.csv</url>

			<separator>;</separator>

		</source>

		<idat>

			<patient-id column="Patientennummer"/>

			<visit-id column="KH-internes-Kennzeichen"/>

			<start column="Aufnahmedatum" format="uuuuMMddHHmm" na=""/>

			<end column="Entlassungsdatum" format="uuuuMMddHHmm" zone="Europe/Berlin" na=""/>

		</idat>

		<mdat>

			<!-- in/out code -->

			<concept id="AG">

				<value column="Aufnahmegrund" xsi:type="string"/>

			<start column="Aufnahmedatum" format="uuuuMMddHHmm" na=""/>

			</concept>

		</mdat>

		<ignore xsi:type="string" column="*"/>

	</visit-table>

	<wide-table>

		<source xsi:type="csv-file">

			<url>p21khg/FAB.csv</url>

			<separator>;</separator>

		</source>

		<idat>

			<visit-id column="KH-internes-Kennzeichen"/>

		</idat>

		<mdat>

			<!-- wechsel zwischen fachabteilung  -->

			<concept id="FAB">

				<value column="FAB" xsi:type="string" />

				<start column="FAB-Aufnahmedatum" format="uuuuMMddHHmm" na="" />

				<end column="FAB-Aufnahmedatum" format="uuuuMMddHHmm" na="" />

				<modifier id="ENTG">

					<value column="Entgeltbereich" xsi:type="string" na-action="drop-fact" />

				</modifier>

			</concept>

		</mdat>

	</wide-table>

	<wide-table>

		<source xsi:type="csv-file">

			<url>p21khg/ICD.csv</url>

			<separator>;</separator>

		</source>

		<idat>

			<visit-id column="KH-internes-Kennzeichen"/>

		</idat>

		<mdat>

			<!-- coded diagnoses -->

			<concept id="ICD">

				<!-- TODO code should be part of the concept -->

				<value column="ICD-Kode" xsi:type="string" />

				<modifier id="art">

					<value column="Diagnoseart" xsi:type="string">

						<map>

							<case value="HD" set-value="" set-concept="DIAG:H"/>

							<case value="ND" set-value="" set-concept="DIAG:N"/>

							<otherwise action="drop-fact" />

						</map>

					</value>

				</modifier>

				<modifier id="ICDVer">

					<value column="ICD-Version" xsi:type="string" />

				</modifier>

				<modifier id="lokality">

					<value column="Lokalisation" xsi:type="string">

						<map>

							<case value="L" set-value="" set-concept="DIAG:LO:L"/>

							<case value="R" set-value="" set-concept="DIAG:LO:R"/>

							<otherwise action="drop-fact" /><!-- removes the modifier if none of the listed values is specified -->

						</map>

					</value>

				</modifier>

			</concept>

		</mdat>

	</wide-table>

</datasource>