Occasionally DT identify some PDF vector documents, as PDFs (not searchable).
If you try OCR such PDF vector document into searchable, the result is rough. any idea?
If you are doing OCR on a document with text, it will generate an image of the text, hence the ragged appearance. However, you shouldn’t run OCR on a document if it doesn’t need it.
Try opening the PDF in Preview (via control-click > Open With), then holding the Option key and choosing File > Save As. Set Quartz Filter to Create Generic PDFX-3 Document and overwrite the file. After it overwrites, check it in DEVONthink again.
1 Like
Thanks for the suggestion.
Try two PDFs with the steps - replace the current ones with generic PDFX-3 via Preview, and DT still thinks the same, DT logs enclosed.
Hold the Option key and choose Help > Report bug to start a support ticket and attach the PDF in question. Thanks.
Sorry for this late reply - but the issue is too obvious.
Almost 10% of the previous indexed PDF document (vector/true) under DT 3.0.1, when reindexing under DT 3.0.3, will become non-searchable, such as this one
Did you start a support ticket, per my instructions?
Did you modify the files in the meantime or did you upgrade macOS?
The files were not modified I think, but macOS could have been upgraded during now and then - my macOS now 10.15.2 which was released early December.
Could you send such a document to cgrunenberg - at - devon-technologies.com? Then we could check this on both Catalina and Mojave. Thank you.