Merge branch '279-parseContent' into dev

Author: Scott Stafford

file/build.gradle

@@ -2,7 +2,9 @@ apply plugin: "com.jfrog.bintray"
 
 dependencies {
     compile "org.springframework.batch:spring-batch-core:$springBatchVersion"
-    testCompile "junit:junit:4.12"
+    compile "com.marklogic:marklogic-client-api:4.0.3"
+    compile "org.apache.tika:tika-parsers:1.17"
+    compile project(":infrastructure")
 }
 
 ext {

file/src/main/java/com/marklogic/spring/batch/item/file/TikaParserItemProcessor.java

@@ -0,0 +1,32 @@
+package com.marklogic.spring.batch.item.file;
+
+import com.marklogic.client.io.InputSourceHandle;
+import com.marklogic.client.io.marker.AbstractWriteHandle;
+import com.marklogic.spring.batch.item.file.support.TikaParser;
+import com.marklogic.spring.batch.item.processor.AbstractMarkLogicItemProcessor;
+import com.marklogic.spring.batch.item.processor.support.UriGenerator;
+import org.springframework.core.io.Resource;
+import org.xml.sax.InputSource;
+
+import java.io.StringReader;
+
+public class TikaParserItemProcessor extends AbstractMarkLogicItemProcessor<Resource> {
+
+    public TikaParserItemProcessor() {
+        super();
+    }
+
+    public TikaParserItemProcessor(UriGenerator uriGenerator) {
+        super(uriGenerator);
+    }
+
+    @Override
+    public AbstractWriteHandle getContentHandle(Resource item) throws Exception {
+        String parsedContent = TikaParser.parseToXML(item.getInputStream());
+        InputSource inputSource = new InputSource(new StringReader(parsedContent.toString()));
+        InputSourceHandle handle = new InputSourceHandle(inputSource);
+        return handle;
+    }
+
+
+}

file/src/main/java/com/marklogic/spring/batch/item/file/support/TikaParser.java

@@ -0,0 +1,22 @@
+package com.marklogic.spring.batch.item.file.support;
+
+import org.apache.tika.exception.TikaException;
+import org.apache.tika.metadata.Metadata;
+import org.apache.tika.parser.AutoDetectParser;
+import org.apache.tika.sax.ToXMLContentHandler;
+import org.xml.sax.ContentHandler;
+import org.xml.sax.SAXException;
+
+import java.io.IOException;
+import java.io.InputStream;
+
+public class TikaParser {
+
+    public static String parseToXML(InputStream inputStream) throws IOException, SAXException, TikaException {
+        ContentHandler handler = new ToXMLContentHandler();
+        AutoDetectParser parser = new AutoDetectParser();
+        Metadata metadata = new Metadata();
+        parser.parse(inputStream, handler, metadata);
+        return handler.toString();
+    }
+}

file/src/test/java/com/marklogic/spring/batch/item/file/support/TikaParserTest.java

@@ -0,0 +1,34 @@
+package com.marklogic.spring.batch.item.file.support;
+
+import org.junit.Before;
+import org.junit.Test;
+import org.springframework.core.io.ClassPathResource;
+import org.springframework.core.io.Resource;
+
+import static org.hamcrest.MatcherAssert.assertThat;
+import static org.hamcrest.CoreMatchers.*;
+
+public class TikaParserTest {
+
+    private Resource wordDocument;
+    private String parsedXml;
+
+    @Test
+    public void parseWordDocumentTest() throws Exception {
+        givenWordDocument("word/test-1.docx");
+        whenDocumentIsParsed();
+        thenContainsText();
+    }
+
+    public void givenWordDocument(String path) {
+        wordDocument = new ClassPathResource(path);
+    }
+
+    public void whenDocumentIsParsed() throws Exception {
+        parsedXml = TikaParser.parseToXML(wordDocument.getInputStream());
+    }
+
+    public void thenContainsText() {
+        assertThat(parsedXml, containsString("The quick brown fox"));
+    }
+}