How to Index Anything

You might want to build custom indices of documents for many reasons. A widely cited one is to supply search functionality to a web site, but you also may want to index your e-mail or technical documents. Anyone who has looked into implementing such a functionality has probably found it's not as easy as it might seem. Various factors conspire to make searching difficult.

The venerable and indispensable grep and its ilk are effective for scanning through lines of text. But grep, egrep and their relations won't do everything for you. They won't search across lines, they won't show search results in a ranked order and their linear search algorithms don't lend themselves to searching larger volumes of data.

HTML doesn't help the situation either. Its display-oriented features, idiosyncratic grammar and bevy of formatting and entity tags make it fairly difficult to parse correctly.

At the other end of the data storage spectrum is data slotted into a database. The ubiquitous example is that of the SQL database, which allows somewhat sophisticated search facilities but usually is not particularly fast for searching. Some database engines, notably MySQL 4, address this issue by allowing fast and ranked searches, but they may not be as customizable as desired.

In this article, we explore ways to create custom indices using SWISH-E, Perl and XML on Linux. Through examples, we show how SWISH-E can be used to build indices of HTML files, PDF files and man pages.

SWISH-E (simple web indexing system for humans—enhanced) is a descendant of SWISH, which was created in 1994 by Kevin Hughes. SWISH was transferred in 1996 to the UC Berkeley Library to fix bugs and add features, and the result was licensed under the GPL and renamed SWISH-E. Development continues, spearheaded by current project maintainer Bill Moseley and assisted by a team of developers.

Here at SkateboardDirectory.com, we happened upon SWISH-E when researching indexing toolkits. We found that it offers a unique combination of features that make it attractive for our purposes. Not only does SWISH-E offer a fast and robust toolkit with which to build and query indices, but it is also well documented, undergoes active development and bug fixes and includes a Perl interface. We also liked that maintainer Moseley and other experienced SWISH-E users and developers are usually prompt when addressing questions and bugs brought up on the SWISH-E mailing list.

For our examples, we started with a stock Red Hat 7.3 workstation with the Software Development bundle of packages installed. We also tested the examples on a Red Hat 6.2 workstation and a Debian Woody.

Currently, installing SWISH-E on Red Hat means installing from source, and the zlib and libxml2 libraries are required to build SWISH-E fully. If you find you need to install either, you probably can find packages provided with your distribution. We also use the xpdf package in our examples, so you may want to install that now if it isn't already. Our reference Red Hat 7.3 workstation setup had all of SWISH-E's prerequisites installed.

Here, we describe the use of SWISH-E 2.4, which according to the development team, should be released by the time you read this article. You can fetch and set up SWISH-E with the following sequence of commands, substituting the current version for (x.x):

% wget \
  http://swish-e.org/Download/swish-e-x.x.tar.gz
% tar zxf swish-e-x.x.tar.gz
% cd swish-e-x.x
% ./configure
% make
% make test

To install the SWISH-E binary, C libraries and man pages into their default locations in /usr/local, type make install as root. This installs the SWISH-E executable into /usr/local/bin. If this directory isn't in your PATH, either change your appropriate dot file to include /usr/local/bin in your PATH, or always call the swish-e executable by full pathname, like /usr/local/bin/swish-e.

Now, let's build and install the SWISH::API Perl module from the Perl directory in the source. We'll need it later when we build a Perl client for our index of man pages. SWISH::API is set up by the normal Perl module install process:

% cd perl
% perl Makefile.PL
% make
% make test

Then, install the SWISH-E Perl module by typing make install as root.

Now that SWISH-E and the SWISH::API Perl module are installed fully, let's build a simple index of HTML files to test SWISH-E. For this example, we index the HTML, one-page-per-section versions of the Linux Documentation Project (LDP) HOWTOs, which we've unpacked into ~/HOWTO-htmls/. The tarballs of LDP documents used in this article come from www.tldp.org/docs.html.