Help with Text Analytics on a load of PDF docs!

Hi,

Am looking for some guidance w.r.t text analytics. We are trying to use HANA as a datamart.

Have tried out FULL-TEXT index on structured data and used the 'Contains' functions and fuzzy search - This works as described. However a questoin here is: is there a general rule of the thumb on what will be size of the index table when FULL-TEXT index is switched on? (If I have a 1 GB table where approx 100 MB is a text column, what will the index table size add up to?)

Also, Can you guide me with the below:

1. How to upload PDF files into HANA?

2. How to generate text indices on the PDF files

3. How to do select queries on the resultant data?

Thanks

Sudarshan