If you have a lot of text files, slide shows, and spreadsheets on your computer, you will need, sooner or later, to know quickly which files contain certain words or sentences. You might also want to use that information to perform some other actions automatically, like sending email notifications or adding new records to a database. Sometimes, you can do this with the Recoll desktop search engine described in the previous issue of Linux Pro Magazine [1]. Should you, however, want something lighter or more flexible than Recoll, try odfgrep: It not only might work better, but also teach you other, very efficient ways to manage all your office documents.

What and Why

A really basic knowledge of the command line and Bash syntax is helpful, but not mandatory: The code is short and explained as accurately as possible, to help you learn some basics of shell programming, if needed.

In fact, the hardest part of this whole tutorial may not be the code itself, but figuring out why you might want to learn and use it. In a nutshell, learning how to search or otherwise process ODF files from the command line, with odfgrep or similar tools, can help you to become a much more productive desktop user, able to delegate to your computer many more otherwise very time-consuming tasks. That's it, really.

What Is grep?

The Unix world, to which Linux belongs, has been using and improving tools for automatic processing of plain text files for decades. The grep command-line program is one of those tools and is one of the reasons why Linux is so great at text processing. By default the grep utility searches for lines that match a given pattern in all the files passed to it and then prints the lines or counts the occurrences. The grep options you are most likely to use are:

• -c (count): Print the number of lines matching the pattern.
• -l (list): Print only the name of each input file that contains the pattern.
• -v (invert match): Print only the lines that do not match the pattern.

All Hail ODF!

ODF is more than just a really open standard, which of course is an extremely important thing in and of itself. Compared with Microsoft Office file formats, or to almost any other format with comparable features, ODF is also very, very simple to analyze or generate automatically. In fact, as you can see in Figure 1, any ODF text, presentation, or spreadsheet is nothing but a ZIP archive of eXtensible Markup Language (XML) files, each with a predefined name and purpose, and pictures. XML is very verbose, but it is plain text, with tons of Free Software libraries, programs, and documentation to easily process it. At the end of this tutorial, for example, I include a link that contains my own little scripts for automatically generating ODF invoices or slide shows.

Figure 1: An ODF file is just an ordinary ZIP archive of images, folders, and XML files, each with its own, fully documented purpose.