Implementing Boolean Operators in Searches.

Hi,

I have a large archive of scanned documents in PDF format and I'm testing the (new to me) OCR search functionality. The recognition quality seems good and I can find items using specific unique terms easily. However, I can't figure out how to do anything more than enter a basic search string - and even that isn't always interpreted the way I want it.

Here's a simple example. I see a charge on a credit card statement of GBP 31.34. I'd like to find the receipt document that includes that value. So I search for 31.34. The result set is large, and it includes documents named "2017-08-03-15-31-34.pdf" or similar. When I access documents from the result set, it is very hard to see if "31.34" even appears in them anywhere - there's no highlighting of the search term, and I'm pretty sure some of the result documents don't even contain that term. Btw, the actual document I'm looking for is part of the list - but it takes 30 minutes of manual browsing to identify it.

Now, this is a simple example. It seems impossible right now to specify an exact search string - when I say "31.34", I don't mean any document that contains either 31 or 34 (which is what Dropbox seems to be doing at this time). In reality I would really like to be able to combine terms:

"Customer Ltd" AND (invoice OR quote)

"Customer Ltd" AND (NOT "Project Beta")

This kind of stuff is rather standard and I don't think further examples are needed. I'm impressed by the document recognition and indexing capabilities, but without a search engine that supports more than the most basic auto-interpreted search terms, the value of the entire feature shrinks drastically.

Maybe I'm missing something? 

Thanks

Oli