(I know that I should add more conditions to prevent unnecessary conversions)
I am copying a PDF into Inbox from the command line. The file is recognized and the automatic rule is being executed. After execution both the original and the created file both have the suffix “PDF document”, i.e. not the original “PDF document” and a target “PDF+Text”
Disable Make text in PDF documents searchable. It’s vision-processing the document on import and you don’t need that option enabled if you’re doing OCR.
Also, there’s no bug in having two files at the end of your first rule. OCRing to a specific format generates a new file, and always has. OCR > Apply does not generate extra files.
I forgot to answer that. I will put the criteria in there, just not whilst I was developing the rule. Once I’m convinced I have the actions figured out, my next step will be to add the criteria.
I would strongly recommend you start with criteria as overmatching items can be a big problem. I would even say you should stack the deck by matching items, then adding e.g., a tag or color label, to further restrict the matches. Then you have a correct set of documents to work with , but a smaller (and safer) subset of them.
