jackrabbit-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alexandre Maurel <alexandre.mau...@gmail.com>
Subject tika xml mime type vs jackrabbit xml mime type
Date Fri, 03 Sep 2010 11:20:10 GMT

Importing a xml document into jackrabbit throught DAV result in creating 
a node with mime type "application/xml" (see below comment of 
XML_MIMETYPE_ALT : the alternative xml mimetype. tika detects xml as 
this ("application/xml")).

As a consequence, the method "canExport(ExportContext context, boolean 
isCollection)" will return false (see below) , and any http request that 
retreive the xml file will get an empty file.

Is this the behavior that jackrabbit is supposed to have ?

The method canExport could return true in both case , when mimetype is 
"text/xml" or "application/xml".

public class XmlHandler extends DefaultHandler {

     private static Logger log = LoggerFactory.getLogger(XmlHandler.class);

      * the xml mimetype
     public static final String XML_MIMETYPE = "text/xml";

      * the alternative xml mimetype. tika detects xml as this.
     public static final String XML_MIMETYPE_ALT = "application/xml";

     private static final Set<String> supportedTypes;
     static {
         supportedTypes = new HashSet<String>();


  public boolean canExport(ExportContext context, boolean isCollection) {
         if (super.canExport(context, isCollection)) {
             String mimeType = null;
             try {
                 Node contentNode = getContentNode(context, isCollection);
                 if (contentNode.hasProperty(JcrConstants.JCR_MIMETYPE)) {
                     mimeType = 
                 } else {
                     mimeType = detect(context.getExportRoot().getName());
             } catch (RepositoryException e) {
                 // ignore and return false
             return XML_MIMETYPE.equals(mimeType);
         return false;

View raw message