UNIX for Beginners Questions & Answers

Guys: I have a customer using the 'pdftotext' utility under Linux. PDFs are received via email, converted to text, etc. and it has worked nicely for years. They received a PDF from a customer and the utility will not read it. The text file is created but it's either empty or has 1-2 bytes of garbage in it. Acrobat renders the document correctly. I did a FILE -> SAVE AS TEXT inside of Acrobat and the same thing happened... an empty file. I tried it on another PDF and it worked. So why would pdftotext have an issue with a certain PDF? Could it be encrypted? Most of these files are 17 to 20kb in size. The one that will not process is a whopping 426kb. Is there a size limitation? I have used pdftotext but I have only 'used it'... I am not an expert on all its abilities. Thanks for reading and thanks for the help.

EDIT: I can upload the PDF here if anyone is interested. Thx.

No. The file itself has problems. If you can read it with Acrobat, try to save the file as a different filename. If Acrobat cannot save it there not much you can do.

Help:

This is probably a stupid question, but is your PDF rendering text or an image, such a a scanned-in page, or text rendered to an image before being imported into the PDF? I'm not quite sure how you can check that. If your copy of Acrobat has the capability, can you perform an OCR on the document?

Andrew

The file is more-than-likely corrupted. but we need more info.

Have you ran that PDF file though a PDF checker, like this one:

Code:

https://www.datalogics.com/products/pdftools/pdf-checker/

That site will provide you a zip file with the text results.

Thank you all for the reply. I did wonder if the entire document is an image but it's very hard to tell. It does not look like a scan. There are some barcode-like images in it as well as some text that is overlapping which is a bit unusual. I will try to save it to another filename. I will also attach it here in case any of you would like to see it.

Here is the PDF in question. Thanks again. I will report back after some further testing.

Mmm, apparently my version of Acrobat will not let me check the accessibility of the document. I feel like the barcodes are NOT the issue because we have PDFs come in with logos, etc. and those items are just ignored by pdftotext and the actual readable text is what comes out in the text file. Just the sheer size of the file suggests that it could be an image/scan which I could see causing problems... no actual text exists in the document. Also, I saved the document in Acrobat as TTS_NEW2.pdf and it saved without issue.