I am trying to create a multi-line text column in a document library, to hold raw textual content of OCR'd fax documents so that it can be crawled, indexed, and searched.

I have been unable to determine how to configure the text column to be indexed and the stored text data to be included in search results.

The source documents are faxes which may be TIF's, JPG's, or PDF's, if that is helpful.

I have been able to configure other types of columns (non-text) that do get included in search results.


what you need to do is the following:

  1. make sure that your content is indexed
  2. lookup the crawled properties, and find your custom column in it. It starts with ows_ and it ends with the internal name of your custom column.
  3. create a managed property that is mapped to your crawled property
  4. re-index
  5. go to your search center, add your managed property to the retrieved columns of the core results webpart
  6. alter your xslt to display the results.
  • All columns from SharePoint should automatically be searchable after indexing as the SharePoint category for crawled props will map the data to the index by default. No need to map them (unless you want to set importance level or use it in a property query or refiner). Feb 5 '12 at 19:58
  • How do I determine what the internal name of the custom column is? Feb 6 '12 at 14:50
  • Answered my own follow up question: allaboutmoss.com/2010/05/11/… Feb 6 '12 at 15:08
  • Step 3 from above is what I was missing. My custom column was being crawled, but was not being included in the index. Thanks! Feb 6 '12 at 15:25

You should look into IFilter, which seems to be the proper interface to do this. There are several PDF IFilters available for text (non-OCR) PDF, and some for OCR PDFs and TIFFs, that I can see on Google. Developing a custom IFilter involves several pitfalls.

But it's probably more effort and cost than necessary if a multiple-line-of-text column does the trick for you :).

  • Yes, I have tried a couple different IFilters for PDF's, the downside is that these PDF's are just scanned fascimile images that have been converted to PDF. The IFilters work great for regular PDF's, though. Feb 6 '12 at 14:51

Some points to consider:

  1. Using iFilter solutions searches within the PDF documents, while using a unlimited multiline text field does not need iFilter installed.

  2. Also search syntax differs: when using iFilter, OOTB you can simply type in the words you are looking for in a search field present on many pages e.g. word1 OR word2 versus multilinefield:word1 or multilinefield:word2. The second needing some user training.

  3. Using iFilter you store the OCR text in a hidden field in the PDF document - this sometimes causes the document to grow MUCH larger than expected.

  4. storing the OCR text in the PDF document allows a regular text search when using adobe acrobat and you have the document open, using a multi line text field, you lose the metadata when you move the document out of SharePoint. You also lose acrobat text search.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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