Please report the bug here: Feature Request/Bug Report Form
The Acrobat Forums are mainly used by users of Adobe's applications, so it's not the best way to make Adobe aware of bugs. I would upload the test file to the Document Cloud, and then share the link to the file in the bug report. Here are some instructions about sharing a file via DC: Share Documents via Adobe's Document Cloud - KHKonsulting LLC
Thanks for your helpful suggestions. I submitted a forum post because the Adobe web site said that this was my only support option. Also, I expect that other users might benefit from my little workaround.
The file link is: Shared Files - Acrobat.com
this file contains "renderable text". You can see that when you try to start OCR via the Tools>Enhance Scans>Recognize Text function. You will end up with this error message:
"Renderable text" means normal text, text that is not part of an image. Acrobat for whatever reason does not want to OCR a page that already contains "normal" text. You can see this "text" when you bring up the "Contents" pane on the left side:
In your case, when you expand this text element, you will see that it's actually empty. This means that you can easily remove this text element without changing the document. You do this by highlighting the "Text" element, and then using the delete key, or by right-clicking and selecting to delete from the menu. Once you do that, the OCR function should be executed without a problem when you try to edit the document (at least it was here, several times in a row).
Thanks for your detailed and insightful analysis. First, it's curious (or, rather, problematic) that entering Edit PDF mode does not engender the "This page contains renderable text" error, as one would expect; instead OCR just stalls out without any message. (Yes, we shouldn't enter into the absurdity that an OCR program would choke when a page contains - horrors! - actual text!) Second, one wonders why Acrobat has created this useless and debilitating empty text element while scanning. Third, my silly (but effective) workaround apparently deletes this ghost text element when flipping the selected image (you can watch it disappear from the Content panel in real time). Fourth, doing so does NOT trigger another OCR attempt as it does when actually deleting the text element (following your advice); I'm forced to click Add Text and then Edit.
In short, Acrobat is clearly buggy in a number of respects. I submitted a bug report.
Thanks again for all your help!
Happy New Year,
Tom, only Adobe can explain we you cannot OCR a page when it already contains renderable text.
Here is how I see Acrobat's OCR functionality: It's a great way to OCR simple documents, the results are pretty good (and has been improved over the years), and it's built right into an application that does a ton of other stuff. However, when it comes to more challenging OCR jobs, I keep a copy of Abbyy's FineReader around (e.g. more than one language, languages not supported by Acrobat and of course pages that already contain renderable text )