If you’ve had an email telling you that some of your document files couldn’t be indexed, keep reading to find out what that means and what you can do about it.
What indexing does
Your library searches the text inside your files, not just the titles and descriptions you’ve typed. When you upload a file, we read the text out of it and add it to your library’s search. For example, a visitor searching for a phrase that only appears on page 12 of a long PDF will still find the right document.
Some files can’t be read this way. When that happens, we email you the list so you can decide whether to do anything about it.
Your documents still work normally
A file that couldn’t be indexed is still in your library, and visitors can still find it and download it as usual. Its title, description, categories, tags and author are all searchable. If a word appears only inside the file and nowhere else, a search for that word won’t bring the document up.
If people mostly find your documents by their titles, you can ignore the email.
Why a file can’t be indexed
The PDF is password-protected or encrypted
A PDF can carry two kinds of password: One that you need in order to open the file at all, and one that blocks printing, copying or editing. Either kind encrypts the file, so the text can’t be read out of it.
The second kind of password is often added by whoever created the file, so you may not know the PDF has one. To fix it, open the file in whatever created it and export a fresh copy with no security settings, then upload that copy in place of the old one.
The file is a scan, or a PDF made of images
A scanned page is a picture of text rather than text, so there’s nothing in the file to read. The same is true of a PDF exported from a design app as flattened images.
To fix it, run the file through optical character recognition (OCR) before you upload it. Adobe Acrobat does this, and so do several free online tools. Your scanner may also have a “searchable PDF” setting, which does the same thing as you scan.
The file type isn’t supported
We can read the text inside PDFs, Word documents (docx, docm, dotx, dotm), Excel workbooks (xlsx, xlsm, xltx, xltm, xlam), PowerPoint presentations (pptx, pptm, ppsx, ppsm, potx, potm, ppam, sldx, sldm) and OpenDocument files (odt, ods, odp, odg, odc, odb, odf).
The old binary Office formats from before 2007, which are the ones ending in doc, xls and ppt, can’t be read. If you have those, open each one in Word, Excel or PowerPoint and save it again as the modern version of the same format.
The file is damaged
Sometimes a file is incomplete or corrupted, usually because something went wrong when it was copied or downloaded. Try opening it on your own computer. If it won’t open there either, upload a fresh copy.
The document links to a file somewhere else
If a document points at a URL rather than a file you’ve uploaded, there’s nothing in your library to read. The document still works, and the link still takes visitors to the file. To make its contents searchable, upload the file to the document instead of linking to it.
Seeing which files are affected
Go to Documents > Failed indexing in your library. The page lists every file we couldn’t read, along with the reason for each one. The email we send you links straight to it.
Once you’ve replaced a file, we read the new one automatically. Nothing on that page needs clearing by hand.
Still stuck?
If a file opens perfectly on your computer and still won’t index, or you’re not sure which of the reasons above applies, contact support and we’ll take a look at the file for you.