Introduction

Perl's predecessors in the scripting world were primarily shell scripts. These programs provided some programming constructs, but had very little functionality built in to them; they were mostly just a way to run other commands, see what they did, and then make a decision about what to do next. While the ability to write scripts in this way is definitely very useful, the idea of firing off many programs from a script to achieve a sometimes simple goal is very inefficient. One of the things that perl tries to do is to have some of the functionality of many other programs, so that it can get the job done without relying on lots of outside programs. Still, there are times when running an operating system command is either necessary, or at least the easiest solution to a problem at hand.

We've already seen one way to interact with the operating system in the discussion of filehandles (Section ). By preceding a filename with the ``pipe'' symbol (|), we can send input to a program running independently of perl, and by following a filename with the symbol, we can read input from a program. Perl also offers a number of other ways of interacting with the operating system, as the following sections will outline.