NIFI-1617 Add source filename metadata to IdentifyMimeType

Signed-off-by: Matt Burgess <mattyb149@apache.org>

nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/pom.xml

@@ -244,6 +244,7 @@ language governing permissions and limitations under the License. -->
                         <exclude>src/test/resources/TestEncryptContent/text.txt</exclude>
                         <exclude>src/test/resources/TestEncryptContent/text.txt.asc</exclude>
                         <exclude>src/test/resources/TestIdentifyMimeType/1.txt</exclude>
+                        <exclude>src/test/resources/TestIdentifyMimeType/1.csv</exclude>
                         <exclude>src/test/resources/TestJson/json-sample.json</exclude>
                         <exclude>src/test/resources/TestJson/control-characters.json</exclude>
                         <exclude>src/test/resources/TestMergeContent/demarcate</exclude>

nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/src/main/java/org/apache/nifi/processors/standard/IdentifyMimeType.java

@@ -45,6 +45,7 @@ import org.apache.tika.config.TikaConfig;
 import org.apache.tika.detect.Detector;
 import org.apache.tika.io.TikaInputStream;
 import org.apache.tika.metadata.Metadata;
+import org.apache.tika.metadata.TikaMetadataKeys;
 import org.apache.tika.mime.MediaType;
 import org.apache.tika.mime.MimeType;
 import org.apache.tika.mime.MimeTypeException;
@@ -117,6 +118,7 @@ public class IdentifyMimeType extends AbstractProcessor {
 
         final ProcessorLog logger = getLogger();
         final ObjectHolder<String> mimeTypeRef = new ObjectHolder<>(null);
+        final String filename = flowFile.getAttribute(CoreAttributes.FILENAME.key());
 
         session.read(flowFile, new InputStreamCallback() {
             @Override
@@ -124,6 +126,10 @@ public class IdentifyMimeType extends AbstractProcessor {
                 try (final InputStream in = new BufferedInputStream(stream)) {
                     TikaInputStream tikaStream = TikaInputStream.get(in);
                     Metadata metadata = new Metadata();
+                    // Add filename if it exists
+                    if (filename != null) {
+                        metadata.add(TikaMetadataKeys.RESOURCE_NAME_KEY, filename);
+                    }
                     // Get mime type
                     MediaType mediatype = detector.detect(tikaStream, metadata);
                     mimeTypeRef.set(mediatype.toString());

nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/src/test/java/org/apache/nifi/processors/standard/TestIdentifyMimeType.java

@@ -58,6 +58,7 @@ public class TestIdentifyMimeType {
         expectedMimeTypes.put("1.7z", "application/x-7z-compressed");
         expectedMimeTypes.put("1.mdb", "application/x-msaccess");
         expectedMimeTypes.put("1.txt", "text/plain");
+        expectedMimeTypes.put("1.csv", "text/csv");
         expectedMimeTypes.put("1.txt.bz2", "application/x-bzip2");
         expectedMimeTypes.put("1.txt.gz", "application/gzip");
         expectedMimeTypes.put("1.zip", "application/zip");
@@ -76,6 +77,7 @@ public class TestIdentifyMimeType {
         expectedExtensions.put("1.7z", ".7z");
         expectedExtensions.put("1.mdb", ".mdb");
         expectedExtensions.put("1.txt", ".txt");
+        expectedExtensions.put("1.csv", ".csv");
         expectedExtensions.put("1.txt.bz2", ".bz2");
         expectedExtensions.put("1.txt.gz", ".gz");
         expectedExtensions.put("1.zip", ".zip");

nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/src/test/resources/TestIdentifyMimeType/1.csv

@@ -0,0 +1,2 @@
+id,name
+1,"Jane Smith"