Merge branch 'issue-681' of https://github.com/UW-Madison-Library/jhove…

… into merge/madlib/issue-681

jhove-core/src/main/java/edu/harvard/hul/ois/jhove/JhoveBase.java

@@ -1115,6 +1115,17 @@ public void setCurrentThread(Thread t) {
         _currentThread = t;
     }
 
+    /**
+     * Sets the maximum number of bytes to check, for modules that look for
+     * an indefinitely positioned signature or check the first sigBytes bytes
+     * in lieu of a signature.
+     *
+     * @param sigBytes max number of bytes to check
+     */
+    public void setSigBytes(int sigBytes) {
+        _sigBytes = sigBytes;
+    }
+
     /**
      * Resets the abort flag. This must be called at the beginning of any
      * activity for which the abort flag may subsequently be set.

jhove-modules/xml-hul/src/main/java/edu/harvard/hul/ois/jhove/module/XmlModule.java

@@ -24,6 +24,7 @@
 import java.io.*;
 import java.net.URI;
 import java.net.URISyntaxException;
+import java.nio.charset.StandardCharsets;
 import java.text.MessageFormat;
 import java.util.*;
 import javax.xml.parsers.SAXParserFactory;
@@ -919,11 +920,11 @@ public void checkSignatures(File file, InputStream stream, RepInfo info)
 		int sigidx = 0;
 		JhoveBase jb = getBase();
 		int sigBytes = jb.getSigBytes();
-		DataInputStream dstream = new DataInputStream(stream);
+		Reader reader = new BufferedReader(new InputStreamReader(stream, StandardCharsets.UTF_8));
 		int charsRead = 0;
 		try {
 			while (charsRead < sigBytes) {
-				char ch = dstream.readChar();
+				char ch = (char) reader.read();
 				++charsRead;
 				// Skip over all whitespace till we reach "xml"
 				if (sigidx <= 2 && Character.isWhitespace(ch)) {

jhove-modules/xml-hul/src/test/java/edu/harvard/hul/ois/jhove/module/XmlModuleTest.java

@@ -0,0 +1,43 @@
+package edu.harvard.hul.ois.jhove.module;
+
+import edu.harvard.hul.ois.jhove.JhoveBase;
+import edu.harvard.hul.ois.jhove.JhoveException;
+import edu.harvard.hul.ois.jhove.RepInfo;
+import org.junit.Before;
+import org.junit.Test;
+
+import java.io.File;
+import java.io.FileInputStream;
+import java.io.IOException;
+
+import static org.junit.Assert.assertEquals;
+
+public class XmlModuleTest {
+
+    private static final String RESOURCE_DIR = "src/test/resources/edu/harvard/hul/ois/jhove/module/";
+    private static final String MODULE_NAME = "XML-hul";
+    private static final int SIG_BYTES = 1024;
+
+    private XmlModule module;
+
+    @Before
+    public void setup() throws JhoveException {
+        JhoveBase base = new JhoveBase();
+        base.setSigBytes(SIG_BYTES);
+        module = new XmlModule();
+        module.setBase(base);
+    }
+
+    @Test
+    public void shouldDetectXmlWhenHasDeclarationAndNotWellFormed() throws IOException {
+        File file = new File(RESOURCE_DIR + "not-well-formed.xml");
+        RepInfo info = new RepInfo(file.toURI().toString());
+
+        module.checkSignatures(file, new FileInputStream(file), info);
+
+        assertEquals(1, info.getSigMatch().size());
+        assertEquals(MODULE_NAME, info.getSigMatch().get(0));
+        assertEquals(RepInfo.TRUE, info.getWellFormed());
+    }
+
+}

jhove-modules/xml-hul/src/test/resources/edu/harvard/hul/ois/jhove/module/not-well-formed.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<outer>
+    <inner>
+        blah
+    </outer>
+</inner>