🚀 go-pugleaf

Partial answer to my question:

1. Pages that look like text may be graphical images of text, thus not searchable as text.

Trying to select as text within the PDF file fails completely.

So, if I select and copy as a graphic and paste into WordPad, saving as plain text, WordPad will show the textual graphic, but won't find anything if I search for a text string I can see on the page.

2. In PDF documents in which text can be selected, white space often turns up in odd places within words, even though you don't see it in the PDF. It might be an artifact of typesetting, kerning, tracking, or another typographic issue; I don't know. Seeing white space might require copying text and pasting it into a word processor, Notepad, or some such. The white space interferes with searching.

Hypothetical example: "treasure" in a PDF page might show up as "tre asu re" in Notepad. If that happens, a search in the PDF for "treasure" will get nothing.

Generally, very short strings, less than a word, might get you something, although along with a lot of irrelevant results. In this case, long search strings might backfire, because they'd miss what's in the PDF.

EBSCOhost and other database publishers might be listing PDFs in response to a search string when that same search string yields nothing when applied within the PDF because the database publishers might be searching HTML files that are equivalent to the PDFs, and then returning PDFs in the search results.

Also, if database publishers produce indexes that accompany PDFs and apply a user's search to the index, but don't supply the index with the PDF when we download the PDF, they might be able to search the PDF while preventing us users from searching within the downloaded PDF.

As yet, I'd still like a way to search for text in any PDF, but I suspect there isn't one.

(I hope this isn't stiff writing. Adobe says we can't start a line with "b", "i", or "c" and expect that letter to appear, so I edited the above for my pronoun, etc.)

-- Nick