Save point: [skip ci]

* Moved iptc reference info to model rather than having a separate models directory
* Use InputStream and OutputStream in metadata extractors rather than Files.

engines/base/src/main/java/org/alfresco/transform/base/TransformManager.java

@@ -38,7 +38,8 @@ public interface TransformManager
 {
     /**
      * Allows a {@link CustomTransformer} to use a local source {@code File} rather than the supplied {@code InputStream}.
-     * The file will be deleted once the request is completed.
+     * The file will be deleted once the request is completed. To avoid creating extra files, if a File has already
+     * been created by the base t-engine, it is returned.
      * If possible this method should be avoided as it is better not to leave content on disk.
      * @throws IllegalStateException if this method has already been called.
      */
@@ -46,14 +47,14 @@ public interface TransformManager
 
     /**
      * Allows a {@link CustomTransformer} to use a local target {@code File} rather than the supplied {@code OutputStream}.
-     * The file will be deleted once the request is completed.
+     * The file will be deleted once the request is completed. To avoid creating extra files, if a File has already
+     * been created by the base t-engine, it is returned.
      * If possible this method should be avoided as it is better not to leave content on disk.
      * @throws IllegalStateException if this method has already been called. A call to {@link #respondWithFragment(Integer)}
      *         allows the method to be called again.
      */
     File createTargetFile();
 
-    // TODO: Do we want to support the following?
     /**
      * Allows a single transform request to have multiple transform responses. For example images from a video at
      * different time offsets or different pages of a document. Following a call to this method a transform response is

engines/base/src/main/java/org/alfresco/transform/base/metadataExtractors/AbstractMetadataExtractor.java

@@ -33,7 +33,6 @@ import org.alfresco.transform.base.CustomTransformer;
 import org.alfresco.transform.base.TransformManager;
 import org.slf4j.Logger;
 
-import java.io.File;
 import java.io.IOException;
 import java.io.InputStream;
 import java.io.OutputStream;
@@ -64,19 +63,17 @@ import static org.alfresco.transform.base.metadataExtractors.AbstractMetadataExt
  *
  * The transform results in a Map of extracted properties encoded as json being returned to the content repository.
  * <ul>
- *   <li>The content repository will use a transform in preference to any metadata extractors it might have defined
- *   locally for the same MIMETYPE.</li>
- *   <li>The T-Engine's Controller class will call a method in a class that extends {@link AbstractMetadataExtractor}
- *   based on the source and target mediatypes in the normal way.</li>
- *   <li>The method extracts ALL available metadata is extracted from the document and then calls
- *   {@link #mapMetadataAndWrite(File, Map, Map)}.</li>
+ *   <li>The method extracts ALL available metadata from the document with
+ *   {@link #extractMetadata(String, InputStream, String, OutputStream, Map, TransformManager)} and then calls
+ *   {@link #mapMetadataAndWrite(OutputStream, Map, Map)}.</li>
  *   <li>Selected values from the available metadata are mapped into content repository property names and values,
  *   depending on what is defined in a {@code "<classname>_metadata_extract.properties"} file.</li>
  *   <li>The selected values are set back to the content repository as a JSON representation of a Map, where the values
  *   are applied to the source node.</li>
  * </ul>
  * To support the same functionality as metadata extractors configured inside the content repository,
- * extra key value pairs may be returned from {@link #extractMetadata}. These are:
+ * extra key value pairs may be returned from {@link #extractMetadata(String, InputStream, String, OutputStream, Map, TransformManager)}.
+ * These are:
  * <ul>
  *     <li>{@code "sys:overwritePolicy"} which can specify the
  *     {@code org.alfresco.repo.content.metadata.MetadataExtracter.OverwritePolicy} name. Defaults to "PRAGMATIC".</li>
@@ -89,7 +86,8 @@ import static org.alfresco.transform.base.metadataExtractors.AbstractMetadataExt
  * If a transform specifies that it can convert from {@code "<MIMETYPE>"} to {@code "alfresco-metadata-embed"}, it is
  * indicating that it can embed metadata in {@code <MIMETYPE>}.
  *
- * The transform results in a new version of supplied source file that contains the metadata supplied in the transform
+ * The transform calls {@link #embedMetadata(String, InputStream, String, OutputStream, Map, TransformManager)}
+ * which should results in a new version of supplied source file that contains the metadata supplied in the transform
  * options.
  * 
  * @author Jesper Steen Møller
@@ -162,24 +160,13 @@ public abstract class AbstractMetadataExtractor implements CustomTransformer
         }
         else
         {
-            extractMetadata(sourceMimetype, inputStream, targetMimetype, outputStream, transformOptions, transformManager);
+            extractMapAndWriteMetadata(sourceMimetype, inputStream, targetMimetype, outputStream, transformOptions, transformManager);
         }
     }
 
-    public void embedMetadata(String sourceMimetype, InputStream inputStream,
-            String targetMimetype, OutputStream outputStream,
-            Map<String, String> transformOptions, TransformManager transformManager) throws Exception
-    {
-        File sourceFile = transformManager.createSourceFile();
-        File targetFile = transformManager.createTargetFile();
-        embedMetadata(sourceMimetype, targetMimetype, transformOptions, sourceFile, targetFile);
-    }
-
-    public void embedMetadata(String sourceMimetype, String targetMimetype, Map<String, String> transformOptions,
-                              File sourceFile, File targetFile) throws Exception
-    {
-        // Default nothing, as embedding is not supported in most cases
-    }
+    public abstract void embedMetadata(String sourceMimetype, InputStream inputStream, String targetMimetype,
+            OutputStream outputStream, Map<String, String> transformOptions, TransformManager transformManager)
+            throws Exception;
 
     protected Map<String, Serializable> getMetadata(Map<String, String> transformOptions)
     {
@@ -507,31 +494,18 @@ public abstract class AbstractMetadataExtractor implements CustomTransformer
         return true;
     }
 
-    public void extractMetadata(String sourceMimetype, InputStream inputStream,
-            String targetMimetype, OutputStream outputStream,
-            Map<String, String> transformOptions, TransformManager transformManager) throws Exception
+    private void extractMapAndWriteMetadata(String sourceMimetype, InputStream inputStream, String targetMimetype,
+            OutputStream outputStream, Map<String, String> transformOptions, TransformManager transformManager)
+            throws Exception
     {
-        File sourceFile = transformManager.createSourceFile();
-        File targetFile = transformManager.createTargetFile();
-        extractMetadata(sourceMimetype, transformOptions, sourceFile, targetFile);
-    }
-
-    /**
-     * The {@code transformOptions} may contain a replacement set of mappings. These will be used in place of the
-     * default mappings from read from file if supplied.
-     */
-    public void extractMetadata(String sourceMimetype, Map<String, String> transformOptions, File sourceFile,
-                                File targetFile) throws Exception
-    {
-        Map<String, Set<String>> mapping = getExtractMappingFromOptions(transformOptions, defaultExtractMapping);
-
         // Use a ThreadLocal to avoid changing method signatures of methods that currently call getExtractMapping.
+        Map<String, Set<String>> mapping = getExtractMappingFromOptions(transformOptions, defaultExtractMapping);
         try
         {
             extractMapping.set(mapping);
-            Map<String, Serializable> metadata = extractMetadata(sourceMimetype, transformOptions, sourceFile);
-            mapMetadataAndWrite(targetFile, metadata, mapping);
-
+            Map<String, Serializable> metadata = extractMetadata(sourceMimetype, inputStream, targetMimetype,
+                    outputStream, transformOptions, transformManager);
+            mapMetadataAndWrite(outputStream, metadata, mapping);
         }
         finally
         {
@@ -539,8 +513,9 @@ public abstract class AbstractMetadataExtractor implements CustomTransformer
         }
     }
 
-    public abstract Map<String, Serializable> extractMetadata(String sourceMimetype, Map<String, String> transformOptions,
-            File sourceFile) throws Exception;
+    public abstract Map<String, Serializable> extractMetadata(String sourceMimetype, InputStream inputStream,
+            String targetMimetype, OutputStream outputStream, Map<String, String> transformOptions,
+            TransformManager transformManager) throws Exception;
 
     private Map<String, Set<String>> getExtractMappingFromOptions(Map<String, String> transformOptions, Map<String,
             Set<String>> defaultExtractMapping)
@@ -561,17 +536,7 @@ public abstract class AbstractMetadataExtractor implements CustomTransformer
         return defaultExtractMapping;
     }
 
-    /**
-     * @deprecated use {@link #extractMetadata(String, Map, File, File)} rather than calling this method.
-     * By default call the overloaded method with the default {@code extractMapping}.
-     */
-    @Deprecated
-    public void mapMetadataAndWrite(File targetFile, Map<String, Serializable> metadata) throws IOException
-    {
-        mapMetadataAndWrite(targetFile, metadata, defaultExtractMapping);
-    }
-
-    public void mapMetadataAndWrite(File targetFile, Map<String, Serializable> metadata,
+    public void mapMetadataAndWrite(OutputStream outputStream, Map<String, Serializable> metadata,
                                     Map<String, Set<String>> extractMapping) throws IOException
     {
         if (logger.isDebugEnabled())
@@ -581,7 +546,7 @@ public abstract class AbstractMetadataExtractor implements CustomTransformer
         }
 
         metadata = mapRawToSystem(metadata, extractMapping);
-        writeMetadata(targetFile, metadata);
+        writeMetadata(outputStream, metadata);
     }
 
     /**
@@ -633,9 +598,9 @@ public abstract class AbstractMetadataExtractor implements CustomTransformer
         return new TreeMap<String, Serializable>(systemProperties);
     }
 
-    private void writeMetadata(File targetFile, Map<String, Serializable> results)
+    private void writeMetadata(OutputStream outputStream, Map<String, Serializable> results)
             throws IOException
     {
-        jsonObjectMapper.writeValue(targetFile, results);
+        jsonObjectMapper.writeValue(outputStream, results);
     }
 }

engines/misc/src/main/java/org/alfresco/transform/misc/metadataExtractors/HtmlMetadataExtractor.java

@@ -29,7 +29,6 @@ package org.alfresco.transform.misc.metadataExtractors;
 import org.alfresco.transform.base.CustomTransformer;
 import org.alfresco.transform.base.TransformManager;
 import org.alfresco.transform.base.metadataExtractors.AbstractMetadataExtractor;
-import org.alfresco.transform.common.TransformException;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
 
@@ -87,8 +86,17 @@ public class HtmlMetadataExtractor extends AbstractMetadataExtractor implements
     }
 
     @Override
-    public Map<String, Serializable> extractMetadata(String sourceMimetype, Map<String, String> transformOptions,
-                                                     File sourceFile) throws Exception
+    public void embedMetadata(String sourceMimetype, InputStream inputStream, String targetMimetype,
+            OutputStream outputStream, Map<String, String> transformOptions, TransformManager transformManager)
+            throws Exception
+    {
+        // Only used for extract, so may be empty.
+    }
+
+    @Override
+    public Map<String, Serializable> extractMetadata(String sourceMimetype, InputStream inputStream,
+            String targetMimetype, OutputStream outputStream, Map<String, String> transformOptions,
+            TransformManager transformManager) throws Exception
     {
         final Map<String, Serializable> rawProperties = new HashMap<>();
 
@@ -175,10 +183,10 @@ public class HtmlMetadataExtractor extends AbstractMetadataExtractor implements
             rawProperties.clear();
             Reader r = null;
 
-            try (InputStream cis = new FileInputStream(sourceFile))
+            try
             {
                 // TODO: for now, use default charset; we should attempt to map from html meta-data
-                r = new InputStreamReader(cis, charsetGuess);
+                r = new InputStreamReader(inputStream, charsetGuess);
                 HTMLEditorKit.Parser parser = new ParserDelegator();
                 parser.parse(r, callback, tries > 0);
                 break;

engines/misc/src/main/java/org/alfresco/transform/misc/metadataExtractors/RFC822MetadataExtractor.java

@@ -29,7 +29,6 @@ package org.alfresco.transform.misc.metadataExtractors;
 import org.alfresco.transform.base.CustomTransformer;
 import org.alfresco.transform.base.TransformManager;
 import org.alfresco.transform.base.metadataExtractors.AbstractMetadataExtractor;
-import org.alfresco.transform.common.TransformException;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
 
@@ -89,102 +88,108 @@ public class RFC822MetadataExtractor extends AbstractMetadataExtractor implement
     }
 
     @Override
-    public Map<String, Serializable> extractMetadata(String sourceMimetype, Map<String, String> transformOptions,
-                                                     File sourceFile) throws Exception
+    public void embedMetadata(String sourceMimetype, InputStream inputStream, String targetMimetype,
+            OutputStream outputStream, Map<String, String> transformOptions, TransformManager transformManager)
+            throws Exception
+    {
+        // Only used for extract, so may be empty.
+    }
+
+    @Override
+    public Map<String, Serializable> extractMetadata(String sourceMimetype, InputStream inputStream,
+            String targetMimetype, OutputStream outputStream, Map<String, String> transformOptions,
+            TransformManager transformManager) throws Exception
     {
         final Map<String, Serializable> rawProperties = new HashMap<>();
 
-        try (InputStream is = new FileInputStream(sourceFile))
+        MimeMessage mimeMessage = new MimeMessage(null, inputStream);
+
+        if (mimeMessage != null)
         {
-            MimeMessage mimeMessage = new MimeMessage(null, is);
+            /**
+             * Extract RFC822 values that doesn't match to headers and need to be encoded.
+             * Or those special fields that require some code to extract data
+             */
+            String tmp = InternetAddress.toString(mimeMessage.getFrom());
+            tmp = tmp != null ? MimeUtility.decodeText(tmp) : null;
+            putRawValue(KEY_MESSAGE_FROM, tmp, rawProperties);
 
-            if (mimeMessage != null)
+            tmp = InternetAddress.toString(mimeMessage.getRecipients(RecipientType.TO));
+            tmp = tmp != null ? MimeUtility.decodeText(tmp) : null;
+            putRawValue(KEY_MESSAGE_TO, tmp, rawProperties);
+
+            tmp = InternetAddress.toString(mimeMessage.getRecipients(RecipientType.CC));
+            tmp = tmp != null ? MimeUtility.decodeText(tmp) : null;
+            putRawValue(KEY_MESSAGE_CC, tmp, rawProperties);
+
+            putRawValue(KEY_MESSAGE_SENT, mimeMessage.getSentDate(), rawProperties);
+
+            /**
+             * Received field from RFC 822
+             *
+             * "Received"    ":"        ; one per relay
+             *   ["from" domain]        ; sending host
+             *   ["by"   domain]        ; receiving host
+             *   ["via"  atom]          ; physical path
+             *  ("with" atom)           ; link/mail protocol
+             *   ["id"   msg-id]        ; receiver msg id
+             *   ["for"  addr-spec]     ; initial form
+             * ";"    date-time         ; time received
+             */
+            Date rxDate = mimeMessage.getReceivedDate();
+
+            if(rxDate != null)
             {
-                /**
-                 * Extract RFC822 values that doesn't match to headers and need to be encoded.
-                 * Or those special fields that require some code to extract data
-                 */
-                String tmp = InternetAddress.toString(mimeMessage.getFrom());
-                tmp = tmp != null ? MimeUtility.decodeText(tmp) : null;
-                putRawValue(KEY_MESSAGE_FROM, tmp, rawProperties);
-
-                tmp = InternetAddress.toString(mimeMessage.getRecipients(RecipientType.TO));
-                tmp = tmp != null ? MimeUtility.decodeText(tmp) : null;
-                putRawValue(KEY_MESSAGE_TO, tmp, rawProperties);
-
-                tmp = InternetAddress.toString(mimeMessage.getRecipients(RecipientType.CC));
-                tmp = tmp != null ? MimeUtility.decodeText(tmp) : null;
-                putRawValue(KEY_MESSAGE_CC, tmp, rawProperties);
-
-                putRawValue(KEY_MESSAGE_SENT, mimeMessage.getSentDate(), rawProperties);
-
-                /**
-                 * Received field from RFC 822
-                 *
-                 * "Received"    ":"        ; one per relay
-                 *   ["from" domain]        ; sending host
-                 *   ["by"   domain]        ; receiving host
-                 *   ["via"  atom]          ; physical path
-                 *  ("with" atom)           ; link/mail protocol
-                 *   ["id"   msg-id]        ; receiver msg id
-                 *   ["for"  addr-spec]     ; initial form
-                 * ";"    date-time         ; time received
-                 */
-                Date rxDate = mimeMessage.getReceivedDate();
-
-                if(rxDate != null)
+                // The email implementation extracted the received date for us.
+                putRawValue(KEY_MESSAGE_RECEIVED, rxDate, rawProperties);
+            }
+            else
+            {
+                // the email implementation did not parse the received date for us.
+                String[] rx = mimeMessage.getHeader("received");
+                if(rx != null && rx.length > 0)
                 {
-                    // The email implementation extracted the received date for us.
-                    putRawValue(KEY_MESSAGE_RECEIVED, rxDate, rawProperties);
-                }
-                else
-                {
-                    // the email implementation did not parse the received date for us.
-                    String[] rx = mimeMessage.getHeader("received");
-                    if(rx != null && rx.length > 0)
+                    String lastReceived = rx[0];
+                    lastReceived = MimeUtility.unfold(lastReceived);
+                    int x = lastReceived.lastIndexOf(';');
+                    if(x > 0)
                     {
-                        String lastReceived = rx[0];
-                        lastReceived = MimeUtility.unfold(lastReceived);
-                        int x = lastReceived.lastIndexOf(';');
-                        if(x > 0)
-                        {
-                            String dateStr = lastReceived.substring(x + 1).trim();
-                            putRawValue(KEY_MESSAGE_RECEIVED, dateStr, rawProperties);
-                        }
+                        String dateStr = lastReceived.substring(x + 1).trim();
+                        putRawValue(KEY_MESSAGE_RECEIVED, dateStr, rawProperties);
                     }
                 }
+            }
 
-                String[] subj = mimeMessage.getHeader("Subject");
-                if (subj != null && subj.length > 0)
+            String[] subj = mimeMessage.getHeader("Subject");
+            if (subj != null && subj.length > 0)
+            {
+                String decodedSubject = subj[0];
+                try
                 {
-                    String decodedSubject = subj[0];
-                    try
-                    {
-                        decodedSubject = MimeUtility.decodeText(decodedSubject);
-                    }
-                    catch (UnsupportedEncodingException e)
-                    {
-                        logger.warn(e.toString());
-                    }
-                    putRawValue(KEY_MESSAGE_SUBJECT, decodedSubject, rawProperties);
+                    decodedSubject = MimeUtility.decodeText(decodedSubject);
                 }
-
-                /*
-                 * Extract values from all header fields, including extension fields "X-"
-                 */
-                Set<String> keys = getExtractMapping().keySet();
-                @SuppressWarnings("unchecked")
-                Enumeration<Header> headers = mimeMessage.getAllHeaders();
-                while (headers.hasMoreElements())
+                catch (UnsupportedEncodingException e)
                 {
-                    Header header = (Header) headers.nextElement();
-                    if (keys.contains(header.getName()))
-                    {
-                        tmp = header.getValue();
-                        tmp = tmp != null ? MimeUtility.decodeText(tmp) : null;
+                    logger.warn(e.toString());
+                }
+                putRawValue(KEY_MESSAGE_SUBJECT, decodedSubject, rawProperties);
+            }
 
-                        putRawValue(header.getName(), tmp, rawProperties);
-                    }
+            /*
+             * Extract values from all header fields, including extension fields "X-"
+             */
+            Set<String> keys = getExtractMapping().keySet();
+            @SuppressWarnings("unchecked")
+            Enumeration<Header> headers = mimeMessage.getAllHeaders();
+            while (headers.hasMoreElements())
+            {
+                Header header = (Header) headers.nextElement();
+                if (keys.contains(header.getName()))
+                {
+                    tmp = header.getValue();
+                    tmp = tmp != null ? MimeUtility.decodeText(tmp) : null;
+
+                    putRawValue(header.getName(), tmp, rawProperties);
                 }
             }
         }

engines/tika/src/main/java/org/alfresco/transform/tika/metadataExtractors/AbstractTikaMetadataExtractor.java

@@ -51,7 +51,6 @@ import org.xml.sax.Attributes;
 import org.xml.sax.ContentHandler;
 import org.xml.sax.Locator;
 
-import java.io.File;
 import java.io.FileInputStream;
 import java.io.InputStream;
 import java.io.OutputStream;
@@ -215,95 +214,93 @@ public abstract class AbstractTikaMetadataExtractor extends AbstractMetadataExtr
     }
 
     @Override
-    public Map<String, Serializable> extractMetadata(String sourceMimetype, Map<String, String> transformOptions,
-                                                     File sourceFile) throws Exception
+    public Map<String, Serializable> extractMetadata(String sourceMimetype, InputStream inputStream,
+            String targetMimetype, OutputStream outputStream, Map<String, String> transformOptions,
+            TransformManager transformManager) throws Exception
     {
         Map<String, Serializable> rawProperties = new HashMap<>();
 
-        try (InputStream is = new FileInputStream(sourceFile))
+        Parser parser = getParser();
+
+        Metadata metadata = new Metadata();
+        metadata.add(Metadata.CONTENT_TYPE, sourceMimetype);
+
+        ParseContext context = buildParseContext(metadata, sourceMimetype);
+
+        ContentHandler handler;
+        Map<String,String> headers = null;
+        if (needHeaderContents())
         {
-            Parser parser = getParser();
-
-            Metadata metadata = new Metadata();
-            metadata.add(Metadata.CONTENT_TYPE, sourceMimetype);
-
-            ParseContext context = buildParseContext(metadata, sourceMimetype);
-
-            ContentHandler handler;
-            Map<String,String> headers = null;
-            if (needHeaderContents())
-            {
-                MapCaptureContentHandler headerCapture =
-                        new MapCaptureContentHandler();
-                headers = headerCapture.tags;
-                handler = new HeadContentHandler(headerCapture);
-            }
-            else
-            {
-                handler = new NullContentHandler();
-            }
-
-            parser.parse(is, handler, metadata, context);
-
-            // First up, copy all the Tika metadata over
-            // This allows people to map any of the Tika
-            //  keys onto their own content model
-            for (String tikaKey : metadata.names())
-            {
-                // TODO review this change (part of MNT-15267) - should we really force string concatenation here !?
-                putRawValue(tikaKey, getMetadataValue(metadata, Property.internalText(tikaKey)), rawProperties);
-            }
-
-            // Now, map the common Tika metadata keys onto
-            //  the common Alfresco metadata keys. This allows
-            //  existing mapping properties files to continue
-            //  to work without needing any changes
-
-            // The simple ones
-            putRawValue(KEY_AUTHOR, getMetadataValue(metadata, TikaCoreProperties.CREATOR), rawProperties);
-            putRawValue(KEY_TITLE, getMetadataValue(metadata, TikaCoreProperties.TITLE), rawProperties);
-            putRawValue(KEY_COMMENTS, getMetadataValue(metadata, TikaCoreProperties.COMMENTS), rawProperties);
-
-            // Tags
-            putRawValue(KEY_TAGS, getMetadataValues(metadata, KEY_TAGS), rawProperties);
-
-            // Get the subject and description, despite things not
-            //  being nearly as consistent as one might hope
-            String subject = getMetadataValue(metadata, TikaCoreProperties.SUBJECT);
-            String description = getMetadataValue(metadata, TikaCoreProperties.DESCRIPTION);
-            if (subject != null && description != null)
-            {
-                putRawValue(KEY_DESCRIPTION, description, rawProperties);
-                putRawValue(KEY_SUBJECT, subject, rawProperties);
-            }
-            else if (subject != null)
-            {
-                putRawValue(KEY_DESCRIPTION, subject, rawProperties);
-                putRawValue(KEY_SUBJECT, subject, rawProperties);
-            }
-            else if (description != null)
-            {
-                putRawValue(KEY_DESCRIPTION, description, rawProperties);
-                putRawValue(KEY_SUBJECT, description, rawProperties);
-            }
-
-            // Try for the dates two different ways too
-            if (metadata.get(TikaCoreProperties.CREATED) != null)
-            {
-                putRawValue(KEY_CREATED, metadata.get(TikaCoreProperties.CREATED), rawProperties);
-            }
-            else if (metadata.get(TikaCoreProperties.MODIFIED) != null)
-            {
-                putRawValue(KEY_CREATED, metadata.get(TikaCoreProperties.MODIFIED), rawProperties);
-            }
-
-            // If people created a specific instance
-            //  (eg OfficeMetadataExtractor), then allow that
-            //  instance to map the Tika keys onto its
-            //  existing namespace so that older properties
-            //  files continue to map correctly
-            rawProperties = extractSpecific(metadata, rawProperties, headers);
+            MapCaptureContentHandler headerCapture =
+                    new MapCaptureContentHandler();
+            headers = headerCapture.tags;
+            handler = new HeadContentHandler(headerCapture);
         }
+        else
+        {
+            handler = new NullContentHandler();
+        }
+
+        parser.parse(inputStream, handler, metadata, context);
+
+        // First up, copy all the Tika metadata over
+        // This allows people to map any of the Tika
+        //  keys onto their own content model
+        for (String tikaKey : metadata.names())
+        {
+            // TODO review this change (part of MNT-15267) - should we really force string concatenation here !?
+            putRawValue(tikaKey, getMetadataValue(metadata, Property.internalText(tikaKey)), rawProperties);
+        }
+
+        // Now, map the common Tika metadata keys onto
+        //  the common Alfresco metadata keys. This allows
+        //  existing mapping properties files to continue
+        //  to work without needing any changes
+
+        // The simple ones
+        putRawValue(KEY_AUTHOR, getMetadataValue(metadata, TikaCoreProperties.CREATOR), rawProperties);
+        putRawValue(KEY_TITLE, getMetadataValue(metadata, TikaCoreProperties.TITLE), rawProperties);
+        putRawValue(KEY_COMMENTS, getMetadataValue(metadata, TikaCoreProperties.COMMENTS), rawProperties);
+
+        // Tags
+        putRawValue(KEY_TAGS, getMetadataValues(metadata, KEY_TAGS), rawProperties);
+
+        // Get the subject and description, despite things not
+        //  being nearly as consistent as one might hope
+        String subject = getMetadataValue(metadata, TikaCoreProperties.SUBJECT);
+        String description = getMetadataValue(metadata, TikaCoreProperties.DESCRIPTION);
+        if (subject != null && description != null)
+        {
+            putRawValue(KEY_DESCRIPTION, description, rawProperties);
+            putRawValue(KEY_SUBJECT, subject, rawProperties);
+        }
+        else if (subject != null)
+        {
+            putRawValue(KEY_DESCRIPTION, subject, rawProperties);
+            putRawValue(KEY_SUBJECT, subject, rawProperties);
+        }
+        else if (description != null)
+        {
+            putRawValue(KEY_DESCRIPTION, description, rawProperties);
+            putRawValue(KEY_SUBJECT, description, rawProperties);
+        }
+
+        // Try for the dates two different ways too
+        if (metadata.get(TikaCoreProperties.CREATED) != null)
+        {
+            putRawValue(KEY_CREATED, metadata.get(TikaCoreProperties.CREATED), rawProperties);
+        }
+        else if (metadata.get(TikaCoreProperties.MODIFIED) != null)
+        {
+            putRawValue(KEY_CREATED, metadata.get(TikaCoreProperties.MODIFIED), rawProperties);
+        }
+
+        // If people created a specific instance
+        //  (eg OfficeMetadataExtractor), then allow that
+        //  instance to map the Tika keys onto its
+        //  existing namespace so that older properties
+        //  files continue to map correctly
+        rawProperties = extractSpecific(metadata, rawProperties, headers);
 
         return rawProperties;
     }