Has anyone experienced a phenomenon wherein Preview's "AI" for PDF text extraction sometimes strips spaces out of a line or two of text?

I'm working with some college documents and need to extract snippets of texts for various purposes. In some (thankfully rare but still hugely annoying) instances, Preview will copy a section of text and it will come out something like

"ThistextispartofaPDFdocumentthathasbeenaccessedusingPreviewonMacOSX.Ifyou try to copy the text sometimes things don't work right."

This effect seems to happen most often on the first line of a given paragraph, section, etc. and particularly on lines that are part of bulleted lists or the first line of an entire chapter/section where some text styling may be in use. Example:

"INTHISCHAPTERWEwillbetalkingaboutthePreviewapplicationforMacOS X. This application lets you..."

When I try to manually select the text character by character I can see the effect at work. The selection indicator will cover both the final letter in a word and the space following it at the same time.

This has mostly occurred on university-provided academic documents, so unfortunately I can't share an example (copyright) and I haven't found a document online that exhibits the problem yet. However, I do know for sure that this is Preview-specific, because the same document in Adobe Acrobat and SumatraPDF on Windows works just fine, i.e. the spaces are not stripped out of any lines. (However, Acrobat on Windows adds hard line-breaks at the end of each physical line of text, so then the arduous task becomes stripping out the excess line breaks - pretty much as bad as adding the spaces back in.)

I'm sure this is somehow related to the "AI" that Apple added to Preview to help it deal with columns and other strange PDF anomalies. However, no solution I have found is sufficiently trouble-free so as not to be just as annoying.

Anyone else seen this behavior?


You must log in to answer this question.

Browse other questions tagged .