JDBC Connector in combination with FAST Search Server RRS feed

  • Question

  • Hello,

    I had a problem with SharePoint 2010, Fast Search Server and JDBC connector.

    I'm inserting Database URLs from SharePoint Task Items into a MS SQL DB and I want to index the whole item. In the pipelineconfig.xml I found a process Document Retriever and I think this process connect to SharePoint and reads out the Item information, but this doesn’t work. The JDBC connector retrieves the data from the database. The Fast Server creates the crawled property “getpath(Text)and I map this crawled property to a managed property with the name getpath, after this I got only badhits from the Fast Search Server. I changed the managed property to "url" but the Fast indexed only the URLs to the fields like title..

    Crawling is no alternative. How can I solve this problem? Is it possible to use the Document Retriever process to get the SharePoint Item?

    Best regards

    Wednesday, May 9, 2012 11:25 AM

All replies

  • Hello,

    I am not sure if this will work, but does the only thing problem you may run into is the document retriever cannot utilize a proxy.  Does the access to the URL links require a proxy?

    Best Wishes,

    Michael Puangco | Senior Support Escalation Engineer |US Customer Service & Support

    Customer Service & Support                      Microsoft| Services

    Thursday, May 17, 2012 3:23 AM
  • Hello,

    Unfortunately, there is no native mapping for the getpath as there is with the url(sounds like a bug to me).

    The problem is that the getpath is a managed property and the crawled property getpath that you are trying to map is actually happening way too late in the pipeline. The alternative would be to create another instance of the documentretriever and have it map to the JDBC getpath crawled property(you need to input the GUID and field type as a prefix in addition to the name of the property) in the input, but I'm not sure this scenario is supported.

    Best regards,


    • Edited by PVenInt Wednesday, May 23, 2012 8:15 AM
    Wednesday, May 23, 2012 8:12 AM
  • I hope that this answers doesn't came too late, but you can achieve this using BCS ( Using BCS, you can join file content and metadata from database in one Fast document.

    Best Regards,

    Thursday, October 18, 2012 6:41 PM