Hi Chiara,
Jeff is right, this usually happens when the PDF is either made of bitmap images (view text is empty) or in an encrypted format (view text is full of crazy symbols). I'll add some more details for everyone's benefit.
PDFs allow for mixed content as well, so sometime protions of the text will be extracted and some other will not (either because they are images or because they are encrypted). There is no real solution for such situations:
If the PDF is made out of scanned images rather than text, to extract the text we would need to automatically extract the images and feed them into a text recognition algorithm (OCR), this is possible in theory, but it is certain to be error-prone and is very complex to put toghether. We don't have plans to write this feature for the time being.
If the PDF is encrypted, then there is no solution at all: encryption is used precisely to stop people from working with the textual content. People may read it and sometimes print it, but all other features are effectively blocked.
You can verify if a troublesome PDF falls into one of the two categories in this way:
1) open up the pdf outside of EPPI-Reviewer in adobe reader (or equivalent),
2) select some text, copy it and try to paste it into a program like Word.
3a) If you couldn't select, copy and and paste anything at all, then it is likely that the PDF is made out of images.
3b) If the pasted text is made out of symbols, then the file is encrypted.
3c) if the text can be transferred into Word, but EPPI-Reviewer does not import it, then you hit one of the rare cases were some small error in the PDF itself prevents our system to correctly extract the text. In this case, you can usually "Save as Text" the PDF from inside Acrobat Reader. You can then upload the symple text file into EPPI-Reviewer as you would upload ordinary PDFs.
I hope this helps,
Sergio