Hi Claus,
Really, we have missed PDF Mime type in the EmptyParser, just thought
that switch off this line is enough:
<parser class="org.apache.jackrabbit.core.query.pdf.PDFParser">
<mime>application/pdf</mime> </parser>
But still, after adding PDF Mime type in the EmptyParser, indexing still
exists.
Talking about tikaConfigPath, it seems there is no such parameter at
all. When we use it we have the following error:
Caused by: com.sun.appserv.connectors.internal.api.PoolingException:
Failed to create session: Invalid repository configuration:
{org.apache.jackrabbit.repository.home=f:/jackrabbit_fms,
org.apache.jackrabbit.repository.conf=f:/jackrabbit_fms/repository.xml,
rep.home=f:/jackrabbit_fms}
We use Jackrabbit 2.2.11, maybe this parameter is valid for earlier
releases only?
Regards,
Anton
Hi Anton,
First the name of the attribute to point to the tika-config.xml is
<param name="tikaConfigPath" value="${wsp.home}/tika-config.xml"/>
and not
<param name="indexingConfiguration" value="${wsp.home}/tika-config.xml"/>
And second in your tika-config.xml there is the PDF Mime Type not in the
EmptyParser
<parser class="org.apache.tika.parser.EmptyParser">
....
<mime>application/pdf</mime>
....
</parser>
greets
claus
