Ctrl F does not find words, but OCR works and can select text

[thefluffybunny]

Odd thing just happened.  I got a PDF from someone, and the CTRL+F search function does not work. The document is not scanned (or an image) and does not need OCR as I can insert/replace/change text.  But when I try and find a word it says "No matches were found" even though I can plainly see the word in front of me. Then, as a test, I tried to find merely the letter "a" and it would select multiple sentences at a time, jumping whole paragraphs. Even more puzzling is that I can convert the document to Microsoft word and the search feature works fine and there are no missing clauses/words.  I can also select the text, but when I try and copy-paste it into a word doc it comes up in some weird format.  For example, the original text that says "This Agreement covers the employment of the above" it comes out as "_____2_                                                ___,        ____                       _5_         _____    __,          "

There is no password protection in the document, and I can't see any restrictions.   I am running Acrobat X Pro.  I looked at the PDF's inner workings and it appears to have been created by "PScript5.dll Version 5.2.2."  It says the PDF producer is Acrobat Distiller 8.2.0 (Windows).  The fonts are Arial, Free3of9 and Times, and the encoding is "Custom."

Any ideas as to what could be causing this weird behavior? And how can I fix it? Converting it to word just to perform a basic search function is annoying. 

[thefluffybunny]

Oh, also.  I tried flattening the document but there don't appear to be any layers. I also converted it to several other PDF formats to the same results.

[Knottsav]

I have encountered the same thing....hoping for an answer to your post.

Were you using ClearScan or Searchable Image for your OCR?

[AbhigyanModi]

It's most likely a problem with the document. It doesn't contain necessary font information which would allow correct copy/paste or Find.

[browniebodrum]

I don't know if this will work, but if I were in your position and the document were not huge, I would experiment by printing out one page from the pdf (choose a page which you know to be dodgy) and then re-scanning the paper copy.  I would enlarge it at the same time, if possible, either by printing it out at a larger size or enlarging the hard copy using the copy machine, if you have that facility.  I would also try changing the orientation of the page when you scan it again e.g. if you've doubled its size, cut it in half and then re-scan the half page in landscape format rather than portrait. When you scan and then run OCR on the new pdf, it might overcome Adobe 'thinking' it has already captured and recognised all the data correctly.

[thefluffybunny]

This seems like an awful lot of work considering I can already convert it to word, which is far less work.

[apangasa]

Can you try rasterizing (converting the document to tiff/bmp) and then importing those image files (tiff/bmp) back in Acrobat.

This would allow document to be re-OCRed and you will be able to search text in it. You can use lossless export/import settings to preserve the quality of ducment.

For exporting the file:

1: Go to File > Save As Other > Image > Tiff.

2: Click on Settings button in the Save As Dialog and set the file settings as ZIP for Monochrome, Grayscale and Color.

Now for importing the files back:

1: Press Ctrl + K

2: In the Preferences dialog, select Convert To PDF option in Categories.

3: Choose tiff or bmp (as per the format used while exporting) and un-check Scan Optimization and OCR.

4: Set Color and Grayscale to Zip and Monochrome to CCITT G4.

5: Save the settings.

6: Now import all the pages in Acrobat.

7: You can use Combine Files feature to merge all the image files.

[Shayan5CB0]

hi

I use acrobat pro DC

how should i do step six and seven?

thank you

[Shayan5CB0]

i also have to say the document language is persian (farsi).

thanks for your support.

[Gabriela24366130yhlw]

THIS WAS BEYOND AND ABOVE, SOOOO HELPFUL, TY!!

[CtDave]

For find / search to function the fonts in a PDF must map to Unicode.

"Free3of9"  -- odds are this font 's characters don't map to Unicode.

Be well...

[BigRyg]

This is the problem I'm having with an inDesign project.  Ctrl F was working and it would find words.  But then, on one find, I accidentaly started typing directly the desired replacement word which started with the letter N. That seemed to change everything.  Now my Finds tell me no matches even though I can clearly see the word I'm searching.

[TSP25645612tzg2]

Going to Printing and selecting Microsoft Print to PDF helped me