3

I have a fileserver, containing 500'000 documents which should be crawled. Upon search, the result page displays the title, part of document content and the fileserver path. So far so god.

But I want to use refiners on the search result start page based on metadata following the path on the fileserver. As an example, the fileserver path structure looks like this:

C:\region\customer\project\

This structure is followed without deviation. I can extraxt the folder structure and import it to term store, but there would be no use, since the documents aren't tagged with metadata.

So is there a way to simply add metadata to the index on each document based on the location on the fileserver?

Improve this question

2 Answers 2

Reset to default
4

What you're looking for is Content Enrichment.

SharePoint 2013 enables you to add your own modules to the content processing pipeline and thus enrich your content before it gets stored in the index.

You can add new refiners based on existing managed properties (the file path).

Hope this helps.

Luis

Improve this answer
2
  • Looks like just what I need! I'll try this in a few hours. Thanks!
    – Benny Skogberg ♦
    Commented Oct 4, 2013 at 3:42
  • 1
    The thing is I would like to have a no-code, not even WCF Web Service, solution - which seems impossible.
    – Benny Skogberg ♦
    Commented Oct 4, 2013 at 14:02
2

If you just need a refiner and don't plan to use the metadata in any other way, you could also just write a better refiner that solves the refinement issue. With a few lines of javascript you can change the presentation of the refiner rather than writing a Content Enrichment Service.

Improve this answer
2
  • Thanks for your tip! The only reason for content enrichment is to implement a better search experience.
    – Benny Skogberg ♦
    Commented Oct 4, 2013 at 14:04
  • 1
    I guess you have to define "better". As I said, in the file path you have all the info you need to create the refiner you defined. I would only consider Content Enrichment in the event I could not find another solution. It adds overhead to the crawl time. Even 500ms per document can really add up on large repositories.
    – Matthew McDermott
    Commented Oct 5, 2013 at 14:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.