converting to searchable PDF and trouble with help

DEVONthink Tips
1

First, the usual tip of the hat – DT is amazing. I’m running DTPO 2.0.1.

I’m having trouble converting an RTF I imported into a searchable PDF, and am not sure why it’s not working. The Convert… pop-up menu doesn’t show “…to searchable PDF” when I control-click on the item, and the convert to PDF option is grayed out when I try to accomplish this action from from Action pull-down button or the Data menu.

A connected problem: I tried to search the Help menu for info on this, but when I click on a help menu item, a blank Help window pops up. Did I miss an import of the help files when I installed the new 2.0.1 version?

2

The ‘Convert>to Searchable PDF’ command in DTPO is to convert an image-only PDF to a PDF + text document using the OCR capabilities. There is no option to convert from other file formats to searchable PDF. You could always print the RTF document to a PDF file and import that to DTPO.

The help files should come along with the program. What happens if you select ‘DEVONthink Pro Office Help’ from the Help menu without entering any text to search on?

3

Actually it also works for PDF+Text documents, not just for image-only PDFs. That is, you can select a PDF+Text document and “recognize to OCR”, which throws away the text, rasterizes each page and performs OCR on the result.

This can be very useful, especially since Acrobat Pro, for instance, refuses to perform OCR on PDFs containing text unless you first convert them to images, while simultaneously providing no way to automate this conversion.

4

This all makes sense now as far as PDF conversion goes — thanks for weighing in!

5

rez–this is a late response, but might help. I’m not sure why one would want to convert an rtf document to “searchable PDF.” A searchable PDF, if I understand things, is a pdf with a bit-image for display and text “underneath” it which can be selected. If you want a pdf that also includes selectable text, why not open the document in, say, OpenOffice and then export as pdf. Then you will get a document with actual font glyphs (which display faster and take less storage), as well as having selectable text. And, it won’t have any OCR errors. (Print to PDF also solves this problem, but more information is retained if you export as pdf in OpenOffice).

I can imagine some occasions where outputting as an image and then OCR’ing might be be desirable (e.g., depending on the source text, “é” might come out as “´e”), but in general I think you’ll like this solution better.