I'm trying to create a simple Automator workflow to find all non-searchable PDFs then OCR them with PDFpen. I can't make much sense of the one or two other posts here (like this one), nor do I have Hazel.

Outside of Automator, I can use Finder to show non-searchable PDFs by searching "kind:pdf NOT intext:"a"" but I don't know how to translate that to an Automator workflow, pass those files on, scripting, etc.

Thanks for any help!

  • I don't have PDFpen, so I can't provide a solution nor do I know what PDFpen expects as input, the file as an alias or a POSIX path. However as far as finding the image PDF files, if in Finder searching for kind:pdf NOT intext:"a" does the trick, you could first do that and the select all that it finds. Then in the Workflow you can get the selected files in more then one way. At which point you can continue to process. If you want to automate it totally then you might have to convert kind:pdf NOT intext:"a" to the form the mdfind command can use and then process its output accordingly. – user3439894 Jan 16 '18 at 16:58
  • What version of OS X/macOS and PDFpen are you using? – user3439894 Jan 16 '18 at 23:02
  • PDFpen 9.2.3 and 10.13.2 macOS. I've just now stumbled upon a semi-solution here which creates an item in the services menu to OCR PDFs that I select in Finder. It seems a little buggy depending on the PDF, but it seems to get the job done. Thanks for your feedback on this! – william Jan 17 '18 at 19:25

You must log in to answer this question.

Browse other questions tagged .