Generate also CSV from CultureGraph as hbz-RVK concordance (#1085)

pom.xml

@@ -67,6 +67,11 @@
             <artifactId>metafacture-elasticsearch</artifactId>
             <version>6.0.0</version>
         </dependency>
+        <dependency>
+            <groupId>org.metafacture</groupId>
+            <artifactId>metafacture-csv</artifactId>
+            <version>6.0.0</version>
+        </dependency>
         <dependency>
             <groupId>org.metafacture</groupId>
             <artifactId>metafacture-flowcontrol</artifactId>

src/main/java/org/lobid/resources/run/CulturegraphXmlFilterHbzRvkToCsv.java

@@ -0,0 +1,47 @@
+/* Copyright 2020 hbz, Pascal Christoph. Licensed under the EPL 2.0*/
+
+package org.lobid.resources.run;
+
+import java.io.File;
+import java.io.IOException;
+
+import org.metafacture.biblio.marc21.MarcXmlHandler;
+import org.metafacture.csv.CsvEncoder;
+import org.metafacture.io.FileOpener;
+import org.metafacture.io.ObjectWriter;
+import org.metafacture.xml.XmlDecoder;
+import org.metafacture.metafix.Metafix;
+
+/**
+ * Filter resources with hbz holdings from culturegraph's MARCXML while tranform it with reject()
+ * into a CSV file.
+ *
+ * @author Pascal Christoph (dr0i)
+ * @author Tobias Bülte (TobiasNx)
+ **/
+public final class CulturegraphXmlFilterHbzRvkToCsv {
+	private static String OUTPUT_FILE="cg-concordance.csv";
+
+	public static void main(String... args) {
+		String XML_INPUT_FILE = new File(args[0]).getAbsolutePath();
+
+		if (args.length > 1) OUTPUT_FILE = args[1];
+
+		final FileOpener opener = new FileOpener();
+		try {
+			opener.setReceiver(new XmlDecoder()).setReceiver(new MarcXmlHandler())
+					.setReceiver(new Metafix("src/main/resources/fix-cg-to-es.fix"))
+					.setReceiver(new CsvEncoder())
+					.setReceiver(new ObjectWriter<>(OUTPUT_FILE));
+		} catch (IOException e) {
+			e.printStackTrace();
+		}
+		opener.process(
+				new File(XML_INPUT_FILE).getAbsolutePath());
+		try {
+			opener.closeStream();
+		} catch (final NullPointerException e) {
+			// ignore, see https://github.com/hbz/lobid-resources/issues/1030
+		}
+	}
+}

src/test/java/org/lobid/resources/CulturegraphXmlFilterHbzRvkToCsvTest.java

@@ -0,0 +1,45 @@
+/* Copyright 2020 hbz, Pascal Christoph. Licensed under the EPL 2.0*/
+
+package org.lobid.resources;
+
+import static org.junit.Assert.assertEquals;
+
+import java.io.File;
+import java.io.IOException;
+
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import org.junit.Test;
+import org.lobid.resources.run.CulturegraphXmlFilterHbzRvkToCsv;
+
+/**
+ * Test of filtering resources with hbz holdings from culturegraph MARCXML,
+ * tranforming into a CSV file.
+ *
+ * @author Pascal Christoph(dr0i)
+ **/
+public final class CulturegraphXmlFilterHbzRvkToCsvTest {
+
+	private static final Logger LOG =
+			LoggerFactory.getLogger(CulturegraphXmlFilterHbzRvkToCsvTest.class);
+
+	private static final String PATH_TO_TEST = "src/test/resources/";
+	public static final String OUTPUT_FILE =
+			PATH_TO_TEST + "cg/output.csv";
+
+	private static final String XML_INPUT_FILE = "cg/aggregate_20240507_example.marcxml";
+
+	@SuppressWarnings("static-method")
+	@Test
+	public void testExtractLookupTableFromCgAsHbzRvk() {
+		CulturegraphXmlFilterHbzRvkToCsv.main(PATH_TO_TEST + XML_INPUT_FILE,
+				OUTPUT_FILE);
+	}
+
+	/**private static void ingest() throws IOException {
+		File jsonFile = new File(OUTPUT_FILE);
+	}*/
+
+
+}

src/test/java/org/lobid/resources/CulturegraphXmlFilterHbzToJsonTest.java

@@ -49,9 +49,9 @@ public final class CulturegraphXmlFilterHbzToJsonTest {
 
 	private static final String PATH_TO_TEST = "src/test/resources/";
 	public static final String JSON_OUTPUT_FILE =
-			PATH_TO_TEST + "jsonld-cg/bulk.ndjson";
+			PATH_TO_TEST + "cg/output-es-bulk.ndjson";
 
-	private static final String XML_INPUT_FILE = "aggregate_20240507_example.marcxml";
+	private static final String XML_INPUT_FILE = "cg/aggregate_20240507_example.marcxml";
 	private static PluginConfigurableNode node;
 	private static Client client;
 	private static final int ELASTICSEARCH_HTTP_PORT = 19200;