Windows Search is a a built-in feature of the operating system which, if I recall correctly, has been around since Windows XP. It can find which files (“documents” in Windows-speak) contain certain words or phrases. It is very handy if, for example, you know you have the shopping list somewhere on your computer but can’t remember exactly where it is or what the file is called.

The trouble with Windows Search is that it can only search certain kinds of documents. Not too long ago I spent more money than I care to admit on a bunch of PDFs for Ars Magica; I already owned most of those books in hard copy but the corpus of material for that game has grown so large, and is scattered across so many books, that I don’t have the time or the patience to flip through all the possible books that might contain an obscure rule. (This is compounded by the publisher’s choice not to include an index in most of the recent books, but that’s another rant.) I thought that with the electronic rule books, that problem would be solved. Imagine my disappointment when I typed “magus” (the word for wizards in Ars Magica) and got no results!

It turns out, Windows Search doesn’t work on PDF documents, only MS Office documents and plain-text files. At least, not out of the box. But it is easy to add that capability.

The Technical FAQ for Windows Search contains the answer, and it is easy enough to find on Google — once the possibility comes to mind.

Windows Search can handle the file formats Microsoft controls, plus ASCII text. PDF is a proprietary file format, although there are free programs that can read and produce it. Ergo, there are licensing restrictions on who can include what PDF-related functionality in their software. (As an aside, this is why proprietary standards are annoying and harmful; they stop people from doing things that would obviously be useful.) If I had to guess, it may have been licensing reasons that stopped Microsoft from including PDF functionality in Windows’ search feature.

Windows Search is designed to be extensible, though. Software vendors can write a drop-in component called an “IFilter” that users can install, and then Windows Search will be able to process a new file format. This is pretty similar to the idea of a browser plug-in like we use to play Flash video or view PDFs in a Web browser, though instead of plugging into the browser it plugs into the Windows Search software. (Why Microsoft didn’t just call it a Search plug-in is beyond me.)

An IFilter for PDF documents is available from either Adobe or Foxit. You only need one of the two, so take your pick. I downloaded the Adobe version and voilà! I can find the word “magus” in that big folder full of e-books I just bought.

Search results in a folder full of PDF e-books.

Windows Search can find text inside PDF documents like these e-books, if you download and install the right IFilter.